28
Quantitative seismic geomorphology of the middle Frio uvial systems, south Texas, United States Hamed Z. El-Mowafy and Kurt J. Marfurt ABSTRACT In this paper we use three-dimensional seismic attribute imaging and well data to reveal the previously unknown quantitative measures, directionality, and spatial locations of the Oligo- cene middle Frio uvial channel systems within an area of 254 km 2 (98 mi 2 ) that covers two oil and gas elds in the Texas Gulf Coast Basin of the United States. The objective of this study is to apply quantitative seismic geomorphology techniques to quantify the morphometric parameters im- portant to building predictive geologic models for uvial reservoirs. Three categories of channel systems are differ- entiated based on their geomorphology, seismic signature, and the mode of transport. The rst, category 1, includes channel systems of high-amplitude, moderate- to high- sinuosity, mixed load channels. Category 2 channel systems are high-amplitude, straight to low-sinuosity, bed load channels with both category 1 and 2 channels lled with coarse-grained sandstone deposits. Category 3 crevasse channel systems are low-amplitude, highly sinuous, suspended load channels lled with ne-grained deposits. These uvial system categories were found to show unique morphometric charac- teristics such as channel width, meander belt width, and me- ander length. Analysis of the middle Frio channel systems imaged in the south Texas study area revealed a signicant downstream decrease of channel belt width along the length of the channel belts. The creation of a quantitative morpho- metric database for the middle Frio uvial reservoirs in the basin would be very useful for exploration and development purposes. The results of this study may have general appli- cability to the Texas Gulf Coast Basin and to similar uvial reservoirs worldwide. AUTHORS Hamed Z. El-Mowafy ~ Geoscientist Consultant, 6327 Borg Breakpoint Dr., Spring, Texas 77379; [email protected] Hamed Z. El-Mowafy is a petroleum geosci- entist with extensive domestic (United States) and international oil industry experience. His professional interests include seismic and sequence stratigraphy, seismic geomorphology, sedimentology, interpretation of seismic data, and reservoir characterization. He is also an associate professor of petroleum geosciences at Al-Azhar University, Egypt. Kurt J. Marfurt ~ ConocoPhillips School of Geology and Geophysics, University of Oklahoma, 100 E Boyd, Norman, Oklahoma 73071; [email protected] Kurt J. Marfurt is a professor of geophysics at the University of Oklahoma leading a research effort in the development, calibration, and application of seismic attributes to aid in seismic geomorphology, structural analysis, and quantitative seismic interpretation, with a current emphasis on resource plays. ACKNOWLEDGMENTS The authors would like to thank Anadarko Petroleum Corporation for providing the Union Pacic Resources three-dimensional seismic survey that was very important to the completion of this work. The rst author is grateful to the Geosciences Department of The University of Tulsa, especially Dennis Kerr and Christopher Liner (now at the University of Arkansas) for their help and advice. We would like to thank William Galloway, Dallas Dunlap, and Jay Busch for reviewing this manuscript. The authors are grateful to reviewers Andrew Miall and Tobi Payenberg, AAPG Editor Michael Sweet, AAPG Senior Associate Editor Colin North, and AAPG consulting geologist Frances Whitehurst, whose comments greatly improved the original manuscript. EDITOR S NOTE Color versions of Figures 3, 4, 611, and 14 can be seen in the online version of this paper. Copyright ©2016. The American Association of Petroleum Geologists. All rights reserved. Manuscript received July 1, 2015; provisional acceptance September 10, 2015; revised manuscript received December 7, 2015; nal acceptance February 1, 2016. DOI:10.1306/02011615136 AAPG Bulletin, v. 100, no. 4 (April 2016), pp. 537564 537

Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Embed Size (px)

Citation preview

Page 1: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Quantitative seismicgeomorphology of the middle Friofluvial systems south TexasUnited StatesHamed Z El-Mowafy and Kurt J Marfurt

ABSTRACT

In this paper we use three-dimensional seismic attribute imagingand well data to reveal the previously unknown quantitativemeasures directionality and spatial locations of the Oligo-cene middle Frio fluvial channel systems within an area of254 km2 (98 mi2) that covers two oil and gas fields in theTexas Gulf Coast Basin of the United States The objective ofthis study is to apply quantitative seismic geomorphologytechniques to quantify the morphometric parameters im-portant to building predictive geologic models for fluvialreservoirs Three categories of channel systems are differ-entiated based on their geomorphology seismic signatureand the mode of transport The first category 1 includeschannel systems of high-amplitude moderate- to high-sinuosity mixed load channels Category 2 channel systemsare high-amplitude straight to low-sinuosity bed loadchannels with both category 1 and 2 channels filled withcoarse-grained sandstone deposits Category 3 crevasse channelsystems are low-amplitude highly sinuous suspended loadchannels filled with fine-grained deposits These fluvial systemcategories were found to show unique morphometric charac-teristics such as channel width meander belt width and me-ander length Analysis of the middle Frio channel systemsimaged in the south Texas study area revealed a significantdownstream decrease of channel belt width along the lengthof the channel belts The creation of a quantitative morpho-metric database for the middle Frio fluvial reservoirs in thebasin would be very useful for exploration and developmentpurposes The results of this study may have general appli-cability to the Texas Gulf Coast Basin and to similar fluvialreservoirs worldwide

AUTHORS

Hamed Z El-Mowafy ~ GeoscientistConsultant 6327 Borg Breakpoint Dr SpringTexas 77379 helmowafyyahoocom

Hamed Z El-Mowafy is a petroleum geosci-entist with extensive domestic (United States)and international oil industry experienceHis professional interests include seismicand sequence stratigraphy seismicgeomorphology sedimentology interpretationof seismic data and reservoir characterizationHe is also an associate professor of petroleumgeosciences at Al-Azhar University Egypt

Kurt J Marfurt ~ ConocoPhillips Schoolof Geology and Geophysics University ofOklahoma 100 E Boyd Norman Oklahoma73071 kmarfurtouedu

Kurt J Marfurt is a professor of geophysicsat the University of Oklahoma leading aresearch effort in the development calibrationand application of seismic attributes to aid inseismic geomorphology structural analysisand quantitative seismic interpretation with acurrent emphasis on resource plays

ACKNOWLEDGMENTS

The authors would like to thank AnadarkoPetroleum Corporation for providing theUnion Pacific Resources three-dimensionalseismic survey that was very important to thecompletion of this work The first author isgrateful to the Geosciences Department ofThe University of Tulsa especially Dennis Kerrand Christopher Liner (now at the Universityof Arkansas) for their help and advice Wewould like to thank William Galloway DallasDunlap and Jay Busch for reviewing thismanuscript The authors are grateful toreviewers Andrew Miall and Tobi PayenbergAAPG Editor Michael Sweet AAPG SeniorAssociate Editor Colin North and AAPGconsulting geologist Frances Whitehurstwhose comments greatly improved theoriginal manuscript

EDITOR rsquoS NOTE

Color versions of Figures 3 4 6ndash11 and 14can be seen in the online version of thispaper

Copyright copy2016 The American Association of Petroleum Geologists All rights reserved

Manuscript received July 1 2015 provisional acceptance September 10 2015 revised manuscript receivedDecember 7 2015 final acceptance February 1 2016DOI10130602011615136

AAPG Bulletin v 100 no 4 (April 2016) pp 537ndash564 537

INTRODUCTION

Recently deep water channel systems and turbiditesplays have been the primary focus of hydrocarbonexploration with emphasis on seismic geomor-phology sedimentary processes and their tectoniccontrols Although fluvialndashdeltaic reservoirs havelong produced oil and gas in the United States GulfCoast few published studies show their expressionusing modern three-dimensional (3-D) seismic inter-pretation workflows and tools

Quantitative seismic geomorphology is a tech-nique in seismic interpretation that can be used toimprove our understanding of reservoir geometry anddistribution in a complex depositional system Theessence of this technique is to collect quantitativedimensional data of fluvial and deltaic systems from3-D seismic data maximizing time-equivalent lateralvariability (Posamentier 2002 Carter 2003 Davieset al 2007 Posamentier et al 2007 Wood 2007Wood and Mize-Spansky 2009) Because of theirrelevance for hydrocarbon exploration and produc-tion quantitative data on the dimensions of fluvialdeposits provide statistical insight into the morpho-metric characteristics of these clastic systems In thisstudy the term morphometric refers to the measuredquantitative parameters of the shape and dimen-sionality of the channel belts and associated sandbodies that make up the Oligocene middle Frio fluvialdepositional sequences

Although the geometry of channel belts has re-ceived much attention (Gibling 2006) data on the3-D variations of channel belt sand body geometry isrelatively scarce Some field studies (eg Shanley andMcCabe 1993 Ryseth et al 1998 Tye et al 1999Holbrook et al 2006) have focused onunderstandingthe architecture (geometry and spatial distribution) ofchannel belts and overbank deposits in fluvial suc-cessions Miall (2002) conducted a study on a non-marine Pleistocene section in theMalay BasinGulf ofThailand using 3-D seismic surveys He analyzedseismic time slice images that revealed five types offluvial systems ofwidely varying style and dimensionsThese fluvial systems range from braided systemswith channel belt widths of more than 4 km (25 mi)to small-scale meandering systems with meander beltwidths (MBWs) of a few hundred meters Gaininginsight into the variation of channel belt dimensionscould improve alluvial architecturemodels and lead to

better predictions of hydrocarbon reserves There-fore this study focuses on analyzing the middle Friofluvial channel systems imaged in the 3-D seismicdata to (1) quantify the geometric variability ofthe middle Frio channel belts and (2) test if along-stream geometric variations of channel belts is com-mon in the middle Frio sequences

Study Area and Geologic Setting

The Frio Formation represents a sediment supplyndashdriven progradational pulse into the northwesternpassivemargin of theGulf ofMexico basin (Gallowayet al 1982a) that lasted approximately 8 my duringtheOligocene (ca 33ndash25Ma)Galloway et al (1982b)subdivide the Frio into three operational unitsthe progradational lower aggradational middle andretrogradational uppermembers This study focuseson the basal part of the middle Frio aggradationalunit The Stratton and Agua Dulce fields are locatedin south Texas (Figure 1) Production from thetwo fields comes mainly from the Frio FormationThe thickness of the Frio Formation penetrated bythe Union Production Company 7A Driscoll wellin the study area is ~1005 m (~3300 ft) The middleFrio Formation consists of vertically stacked reservoirsequences (Figure 2) This study focuses on the ag-gradational fluvial sandstone reservoirs of the deeperF-series referred to as basal middle Frio which maybe unconformably resting above the progradationalupward coarsening deltaic reservoir sequences of thelower Frio Formation Galloway (1989) indicated thepresence of inner coastal plain unconformity betweenthe middle and the lower Frio Formation Kerr (1990)noticed that the lenticular fluvial sandstone depositsof the middle Frio Formation is in marked contrast tothe underlying laterally extensive sandstones of thelower Frio Formation deposited in a lower coastal plainto inner shelf setting (Galloway et al 1982b) Thieset al (1993) interpreted the middle Frio to uncon-formably overlie the lower Frio in Orange CountyTexas The interval of detailed analysis of this studycovers the interval from the top of the G2 reservoirto above the top of the F11 reservoir which has anaverage total thickness of 152 m (500 ft) (Figure 2)

The gas reservoirs in the Vicksburg and lowerndashmiddle Frio Formations are affected by southerndipping growth faults The two contiguous fields are

538 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

included in a large rollover anticline truncated bynormal faults which are synthetic and antithetic tothe Vicksburg and Agua Dulce major growth faultsOver most of the study area the deepest part ofthe middle Frio section is affected by these structures(El-Mowafy and Marfurt 2008)

The Frio Formation in south Texas comprises theGueydan fluvial system which drained a semiarid

source areas in the desert southwest (Galloway et al1982a) and entered the Gulf of Mexico through theRio Grande embayment TheGueydan fluvial systemis interpreted as having been deposited by mixedload to bed load slightly sinuous streams with broadwell-developed natural levees (Galloway 1977)Single-story (storey of Friend et al 1979) channelsandstone bodies are 3ndash10 m (10ndash33 ft) thick but

Figure 1 Index map showingthe location of the contiguousStratton and Agua Dulce fieldsand locations of the two three-dimensional (3-D) seismic sur-veys used in this study The twofields are represented schemati-cally by one polygon (not toscale) which approximates thestructure closure at the middleFrio stratigraphic level The twofields occur next to each otherand they are defined by parallel-to-the fault northeastndashsouthwest-trending large rollover anticlinethat is related to the Vicksburgand Agua Dulce faultsrsquo hangingwall deformation Stratton fieldextends from northern KlebergCounty into southern NuecesCounty and Agua Dulce fieldextends farther to the northeastand is mostly covered by theUnion Pacific Resources (UPR)3-D seismic survey Polygon Adelineates the interpreted areaof the UPR 3-D seismic surveyPolygon B delineates area withwell control inside the UPR 3-Dseismic survey Polygon C definesarea covered by the smallerUniversity of Texas Bureau ofEconomic Geology (UT-BEG) 3-Dseismic survey The FR-4 gas playis the prolific Frio gas trend insouth Texas with production fromFrio and Vicksburg fluvial anddeltaic sandstone reservoirs re-spectively along the Vicksburgfault zone Stratton and AguaDulce fields are located in this

play (modified from Kosters et al 1989) Location of Union Pacific Resources Company (UPRC) 175 Wardner well with vertical seismicprofile (VSP) is shown on themap Line of the stratigraphic section in Figure 4 connecting the UPRC 182 and 185Wardner GP wells is shownCorpus Christi bay is also marked on this map Reprinted by permission of UT-BEG whose permission is required for further use

EL-MOWAFY aND MARFURT 539

they commonly amalgamate into units as thick as30 m (100 ft) These amalgamated multistory sand-stone bodies develop into multilateral belts as muchas several miles wide Crevasse splay deposits arewidespread and can extend as far as several thousandfeet from the main channel (Galloway 1981)

Data Description

Two stacked slightly overlapping 3-D seismic sur-veys have been used in this study for the inter-pretation of lateral and vertical variability of themiddle Frio fluvial deposits The signal-to-noise ratio

is high and nomultiples or coherent noise is apparentin the data The first survey (area A in Figure 1) wasacquired in 1993 by Union Pacific Resources (UPR)and processed byWestern Geophysical resulting in afrequency bandwidth of 10ndash90 Hz with a dominantfrequency of 50Hz The size of this survey is 233 km2

(90 mi2) The second survey (area C in Figure 1)was acquired in 1992 by the University of Texas atAustinndashBureau of Economic Geology (UT-BEG) andwas reprocessed by Mercury International Technol-ogy Company in Tulsa Oklahoma resulting in afrequency bandwidth of 10ndash60 Hz with a dominantfrequency of 35 Hz The size of this survey is197 km2 (76 mi2) Using the average frequencies ofthe two 3-D seismic data sets and the average intervalvelocity of 10500 ftsec (3200 msec) of the middleFrio interval obtained from sonic logs the averagevertical resolution of the UT-BEG survey is 23 m(75 ft) and of the UPR survey is 158 m (52 ft) Wegenerated several wedge models to estimate andconfirm the minimum resolved thickness of theseismic data Ricker wavelet with 35-Hz dominantfrequencywas used in the generation of thesemodelsThemodels indicate that theminimum thickness thatcan be resolved seismically in the UPR survey is152 m (50 ft) The detectability limits are estimatedto be 26 and 17m (85 and 56 ft) for the UT-BEG andthe UPR 3-D surveys respectively We also useddigital and hard copies of well logs for 171 wells aswell as one core described by Kerr and Jirik (1990)

Well-to-Seismic Tie

The tools used in associating seismic data with geo-logic horizons are the vertical seismic profile (VSP) ofthe Union Pacific Resource Company (UPRC) 175Wardner well (Figure 3) and synthetic seismogramsThese tools are used to define the predicted two-waytravel time for each of the depositional surfaces ofinterest in the two 3-D seismic data sets The twopoststack 3-D seismic volumes (UT-BEG and UPR)partially overlap in the northern Stratton field area(Figure 1) The UPRC 175 Wardner well with theVSP lies in the heart of the smaller UT-BEG 3-Dseismic survey but on the southern border of the UPRsurvey The eastndashwest crossline 204 was extractedfrom the UT-BEG survey that passes through theUPRC 175Wardner well the VSPwasmatched with

Figure 2 Type log from the Union Producing Company 7ADriscoll well showing the middle Frio reservoir groups and no-menclature at Stratton (left) and AguaDulce (right) fields (modifiedfrom Kerr 1990) Reprinted by permission of the AAPG Gulf CoastAssociation of Geological Societies whose permission is requiredfor further use BMF = basal middle Frio ResD = deep resistivitySP = spontaneous potential U Vicksburg = Upper Vicksburg

540 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the seismic data and events of interest were markedon this crossline The E41 F11 F39 and G2 eventswere translated from crossline 204 into inlines 190 and210 from the UT-BEG survey as shown in Figure 3The stratigraphic positions of the basal middle Frioreflectors are in good agreementwith respect to traveltime To tie the geologic horizons of interest (E41F11 F39 and G2 Figures 2 3 4) with the UPR 3-Dseismic volume the northndashsouth inline 1238 (theclosest seismic inline to the location of UPRC 175Wardner well) was extracted from the UPR 3-Dseismic volume and events of interest from the VSPwere tied to that seismic inline

Seismic Attributes

Amplitude horizon slices extracted every sample(4 ms) away from the interval of interest show in-dications for channelized features and incised valleysbut were not sharply resolved For this reason wegenerated multiple attribute volumes in an attemptto clearly delineate and resolve fluvial channel fea-tures to enable greater analysis Seismic attributescombine amplitude values at adjacent time samples

and traces to quantify amplitude phase or frequencyof the seismic data Single-trace attributes use ad-jacent samples in a given trace Multitrace or geo-metric attributesmeasure lateral changes in waveformamplitude and phase We find simple root-mean-square (RMS) amplitude (the RMS of the seismicamplitudes) within a 10-ms window to be particularlysensitive to thin bed tuning effects associated with ourfluvial systems The 3-D seismic attributes that bestrevealed the middle Frio fluvial architectural elementsare the RMS amplitude extracted within a 10-mswindow coherent amplitude gradients energy andvolumetric curvature (Chopra and Marfurt 2007)These seismically enhanced delineations of the archi-tectural elements of the middle Frio fluvial systems aidin themeasure of theirmorphologic characteristics andreservoir parameters

Methods

Commercial interpretation packages are used for 3-Dseismic interpretation and attribute extraction TheE41 F11 F39 andG2 seismic horizons (Figure 3) arepicked in the two 3-D seismic data setsmade available

Figure 3 Vertical seismic profile (VSP) display of Union Pacific Resources Company 175 Wardner well (modified from Hardage et al1994) The horizons of interest E41 F11 and F39 are located at 145 158 and 165 sec respectively The peaks (black-filled to the right)are top of reservoir units Crossline 204 from the University of Texas Bureau of Economic Geology three-dimensional (3-D) seismic volumewas tied with inlines (ILs) 190 and 210 and then tied with IL 1238 from the Union Pacific Resources (UPR) 3-D seismic volume Reprinted bypermission of the Society of Exploration Geophysicists whose permission is required for further use

EL-MOWAFY aND MARFURT 541

to this study and a series of stratigraphic horizonslices are produced in one-sample (4-ms) incrementsabove and below the reference surfaces an intervalcorresponding in these data to approximately 6 m(20 ft) of strata Some of the channel systems shownin this article are imaged on window attribute mapseg Figures 5 and 6 and other channel systemsare imaged on horizon slices eg Figures 7 and 8 Itshould be noted that the morphometric measure-ments should be considered minimum because theyare restricted to the resolution of the seismic images

MIDDLE FRIO FRAMEWORK ANDGEOMORPHOLOGY

Sequence Stratigraphy

The sequence models for fluvial systems is docu-mented in detail by Miall (2014 chapter 6) Se-quence stratigraphic reconstruction of the middleFrio in the study area is awork in progress A sequencestratigraphic workflow was applied to reconstructthe architecture and framework of the middle Friosequences In this workflow four architectural ele-ment levels are evaluated facies channel beltssystems tracts and the middle Frio depositional se-quences In this paper we focus on the channel beltelement Sequence stratigraphic subdivisions of themiddle Frio are based to a greater extent onwell logsin addition to one core description and the two 3-Dseismic data volumes used in this study We estab-lished the log response for different types of themiddle Frio deposits by comparing the core char-acteristics and the corresponding well log shapesderived from spontaneous potential (SP) gamma ray(GR) and resistivity logs In the study area the welllog profile of the channel bodies is characterized byboth blocky and bell-shaped or upward-fining pat-terns Crevasse splays are recognized by a funnel-shaped or upward-coarsening pattern Levee bodiesare represented by a spiky pattern and floodplainmudstones are dominated by a baseline pattern Basedon the well log signatures and stacking patterns theinterval between the two sand-dominated channelbelt complexes F11 and F39 is interpreted as a high-stand systems tract (HST) (Figure 4) The deposits ofthe lowstand systems tracts (LST) of the basal middleFrio depositional sequences I and II in the study area

Figure 4 Stratigraphic well log cross section flattened on thetop of F11 showing the subdivisions of the basal middle Frio (BMF)sequences into system tracts at the Union Pacific ResourcesCompany (UPRC) 182 and 185 Wardner well locations Lowstandsystems tracts (LSTs) are characterized by amalgamated multistorysandstone bodies Highstand systems tracts (HSTs) are charac-terized by single-story and multistory channel bodies and single-story crevasse splay and levee bodies encased within floodplainmudstones and siltstones The F11 channel incision at the UPRC 182Wardner well marks the upper boundary of the BMF depositionalsequence I Connected sand bodies are interpreted based on welllog sequence analysis No pressure data were made available to thisstudy Location of the cross section XX9 is shown in Figure 1 deposeq = depositional sequence GR = gamma ray ILD = induction logdeep SP = spontaneous potential TVD = true vertical depth

542 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

consist mostly of single-story and multistory mul-tilateral channel bodies deposited above sequenceboundaries whereas those of the HST are made upof single-story and multistory channel bodies cre-vasse splays and levee bodies isolated within flood-plain mudstones (Figure 4)

The criteria used for the recognition of sequenceboundaries that bound the basal middle Frio se-quences in the study area (Figure 4) are as followsFirst the coarse-grained deposits that accumulateon the channel floor form a channel lag This lag liesabove the basal erosion surface and consists of lo-cally derived material such as mud clasts and blockseroded from the channel banks and bottom plantdebris and coarse-grained gravel and sand Thischannel lag may represent a sequence boundarysurface in fluvial strata The sequence boundary canthen be traced at the top of paleosol horizons thatare correlative to the unconformities generated inthe channel subenvironment (Wright and Marriott

1993 Galloway and Hobday 1996) Themudstoneintraclasts and the paleosol layermdashmarked by car-bonate nodules and root molds from a core cut inthe UPRC 184 Wardner wellmdashrepresent a channellag (refer to Kerr and Jirik 1990 for detailed coredescription) This lag may indicate the upperbounding surface of the basal middle Frio deposi-tional sequence I at the base of the F11 interval(Figure 4) Second abrupt deflections to the left ofthe GR and SP log curves indicate erosional basesof the F11 and F39 channel belt bodies additionallythe local incisions (eg base F11 in the UPRC 182Wardner well Figure 4) could be interpreted torepresent sequence boundaries Third the presenceof the low- and high-sinuosity channel belts imagedon the 3-D seismic attribute maps at the F11 andF39 stratigraphic intervals might also indicate se-quenceboundaries Fourth an incised valley imagedona most-negative curvature attribute map (not shownongoing sequence stratigraphic work) just below the

Figure 5 Root-mean-square(RMS) seismic attribute mapgenerated within a 10-ms windowaround the F11 horizon from theUnion Pacific Resources three-dimensional seismic survey(area A in Figure 1) Several fluvialarchitectural elements are shownon this map See close-up viewsin Figures 6 9 and 10 Dottedcircles indicate locations of pos-sible crevassing andor nodalavulsion points controlled bygrowth faults The coordinatereference system shown in thismap and in Figures 6 9 and10 is the Universal TransverseMercator grid (X = eastingY = northing) zone 14 north

EL-MOWAFY aND MARFURT 543

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 2: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

INTRODUCTION

Recently deep water channel systems and turbiditesplays have been the primary focus of hydrocarbonexploration with emphasis on seismic geomor-phology sedimentary processes and their tectoniccontrols Although fluvialndashdeltaic reservoirs havelong produced oil and gas in the United States GulfCoast few published studies show their expressionusing modern three-dimensional (3-D) seismic inter-pretation workflows and tools

Quantitative seismic geomorphology is a tech-nique in seismic interpretation that can be used toimprove our understanding of reservoir geometry anddistribution in a complex depositional system Theessence of this technique is to collect quantitativedimensional data of fluvial and deltaic systems from3-D seismic data maximizing time-equivalent lateralvariability (Posamentier 2002 Carter 2003 Davieset al 2007 Posamentier et al 2007 Wood 2007Wood and Mize-Spansky 2009) Because of theirrelevance for hydrocarbon exploration and produc-tion quantitative data on the dimensions of fluvialdeposits provide statistical insight into the morpho-metric characteristics of these clastic systems In thisstudy the term morphometric refers to the measuredquantitative parameters of the shape and dimen-sionality of the channel belts and associated sandbodies that make up the Oligocene middle Frio fluvialdepositional sequences

Although the geometry of channel belts has re-ceived much attention (Gibling 2006) data on the3-D variations of channel belt sand body geometry isrelatively scarce Some field studies (eg Shanley andMcCabe 1993 Ryseth et al 1998 Tye et al 1999Holbrook et al 2006) have focused onunderstandingthe architecture (geometry and spatial distribution) ofchannel belts and overbank deposits in fluvial suc-cessions Miall (2002) conducted a study on a non-marine Pleistocene section in theMalay BasinGulf ofThailand using 3-D seismic surveys He analyzedseismic time slice images that revealed five types offluvial systems ofwidely varying style and dimensionsThese fluvial systems range from braided systemswith channel belt widths of more than 4 km (25 mi)to small-scale meandering systems with meander beltwidths (MBWs) of a few hundred meters Gaininginsight into the variation of channel belt dimensionscould improve alluvial architecturemodels and lead to

better predictions of hydrocarbon reserves There-fore this study focuses on analyzing the middle Friofluvial channel systems imaged in the 3-D seismicdata to (1) quantify the geometric variability ofthe middle Frio channel belts and (2) test if along-stream geometric variations of channel belts is com-mon in the middle Frio sequences

Study Area and Geologic Setting

The Frio Formation represents a sediment supplyndashdriven progradational pulse into the northwesternpassivemargin of theGulf ofMexico basin (Gallowayet al 1982a) that lasted approximately 8 my duringtheOligocene (ca 33ndash25Ma)Galloway et al (1982b)subdivide the Frio into three operational unitsthe progradational lower aggradational middle andretrogradational uppermembers This study focuseson the basal part of the middle Frio aggradationalunit The Stratton and Agua Dulce fields are locatedin south Texas (Figure 1) Production from thetwo fields comes mainly from the Frio FormationThe thickness of the Frio Formation penetrated bythe Union Production Company 7A Driscoll wellin the study area is ~1005 m (~3300 ft) The middleFrio Formation consists of vertically stacked reservoirsequences (Figure 2) This study focuses on the ag-gradational fluvial sandstone reservoirs of the deeperF-series referred to as basal middle Frio which maybe unconformably resting above the progradationalupward coarsening deltaic reservoir sequences of thelower Frio Formation Galloway (1989) indicated thepresence of inner coastal plain unconformity betweenthe middle and the lower Frio Formation Kerr (1990)noticed that the lenticular fluvial sandstone depositsof the middle Frio Formation is in marked contrast tothe underlying laterally extensive sandstones of thelower Frio Formation deposited in a lower coastal plainto inner shelf setting (Galloway et al 1982b) Thieset al (1993) interpreted the middle Frio to uncon-formably overlie the lower Frio in Orange CountyTexas The interval of detailed analysis of this studycovers the interval from the top of the G2 reservoirto above the top of the F11 reservoir which has anaverage total thickness of 152 m (500 ft) (Figure 2)

The gas reservoirs in the Vicksburg and lowerndashmiddle Frio Formations are affected by southerndipping growth faults The two contiguous fields are

538 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

included in a large rollover anticline truncated bynormal faults which are synthetic and antithetic tothe Vicksburg and Agua Dulce major growth faultsOver most of the study area the deepest part ofthe middle Frio section is affected by these structures(El-Mowafy and Marfurt 2008)

The Frio Formation in south Texas comprises theGueydan fluvial system which drained a semiarid

source areas in the desert southwest (Galloway et al1982a) and entered the Gulf of Mexico through theRio Grande embayment TheGueydan fluvial systemis interpreted as having been deposited by mixedload to bed load slightly sinuous streams with broadwell-developed natural levees (Galloway 1977)Single-story (storey of Friend et al 1979) channelsandstone bodies are 3ndash10 m (10ndash33 ft) thick but

Figure 1 Index map showingthe location of the contiguousStratton and Agua Dulce fieldsand locations of the two three-dimensional (3-D) seismic sur-veys used in this study The twofields are represented schemati-cally by one polygon (not toscale) which approximates thestructure closure at the middleFrio stratigraphic level The twofields occur next to each otherand they are defined by parallel-to-the fault northeastndashsouthwest-trending large rollover anticlinethat is related to the Vicksburgand Agua Dulce faultsrsquo hangingwall deformation Stratton fieldextends from northern KlebergCounty into southern NuecesCounty and Agua Dulce fieldextends farther to the northeastand is mostly covered by theUnion Pacific Resources (UPR)3-D seismic survey Polygon Adelineates the interpreted areaof the UPR 3-D seismic surveyPolygon B delineates area withwell control inside the UPR 3-Dseismic survey Polygon C definesarea covered by the smallerUniversity of Texas Bureau ofEconomic Geology (UT-BEG) 3-Dseismic survey The FR-4 gas playis the prolific Frio gas trend insouth Texas with production fromFrio and Vicksburg fluvial anddeltaic sandstone reservoirs re-spectively along the Vicksburgfault zone Stratton and AguaDulce fields are located in this

play (modified from Kosters et al 1989) Location of Union Pacific Resources Company (UPRC) 175 Wardner well with vertical seismicprofile (VSP) is shown on themap Line of the stratigraphic section in Figure 4 connecting the UPRC 182 and 185Wardner GP wells is shownCorpus Christi bay is also marked on this map Reprinted by permission of UT-BEG whose permission is required for further use

EL-MOWAFY aND MARFURT 539

they commonly amalgamate into units as thick as30 m (100 ft) These amalgamated multistory sand-stone bodies develop into multilateral belts as muchas several miles wide Crevasse splay deposits arewidespread and can extend as far as several thousandfeet from the main channel (Galloway 1981)

Data Description

Two stacked slightly overlapping 3-D seismic sur-veys have been used in this study for the inter-pretation of lateral and vertical variability of themiddle Frio fluvial deposits The signal-to-noise ratio

is high and nomultiples or coherent noise is apparentin the data The first survey (area A in Figure 1) wasacquired in 1993 by Union Pacific Resources (UPR)and processed byWestern Geophysical resulting in afrequency bandwidth of 10ndash90 Hz with a dominantfrequency of 50Hz The size of this survey is 233 km2

(90 mi2) The second survey (area C in Figure 1)was acquired in 1992 by the University of Texas atAustinndashBureau of Economic Geology (UT-BEG) andwas reprocessed by Mercury International Technol-ogy Company in Tulsa Oklahoma resulting in afrequency bandwidth of 10ndash60 Hz with a dominantfrequency of 35 Hz The size of this survey is197 km2 (76 mi2) Using the average frequencies ofthe two 3-D seismic data sets and the average intervalvelocity of 10500 ftsec (3200 msec) of the middleFrio interval obtained from sonic logs the averagevertical resolution of the UT-BEG survey is 23 m(75 ft) and of the UPR survey is 158 m (52 ft) Wegenerated several wedge models to estimate andconfirm the minimum resolved thickness of theseismic data Ricker wavelet with 35-Hz dominantfrequencywas used in the generation of thesemodelsThemodels indicate that theminimum thickness thatcan be resolved seismically in the UPR survey is152 m (50 ft) The detectability limits are estimatedto be 26 and 17m (85 and 56 ft) for the UT-BEG andthe UPR 3-D surveys respectively We also useddigital and hard copies of well logs for 171 wells aswell as one core described by Kerr and Jirik (1990)

Well-to-Seismic Tie

The tools used in associating seismic data with geo-logic horizons are the vertical seismic profile (VSP) ofthe Union Pacific Resource Company (UPRC) 175Wardner well (Figure 3) and synthetic seismogramsThese tools are used to define the predicted two-waytravel time for each of the depositional surfaces ofinterest in the two 3-D seismic data sets The twopoststack 3-D seismic volumes (UT-BEG and UPR)partially overlap in the northern Stratton field area(Figure 1) The UPRC 175 Wardner well with theVSP lies in the heart of the smaller UT-BEG 3-Dseismic survey but on the southern border of the UPRsurvey The eastndashwest crossline 204 was extractedfrom the UT-BEG survey that passes through theUPRC 175Wardner well the VSPwasmatched with

Figure 2 Type log from the Union Producing Company 7ADriscoll well showing the middle Frio reservoir groups and no-menclature at Stratton (left) and AguaDulce (right) fields (modifiedfrom Kerr 1990) Reprinted by permission of the AAPG Gulf CoastAssociation of Geological Societies whose permission is requiredfor further use BMF = basal middle Frio ResD = deep resistivitySP = spontaneous potential U Vicksburg = Upper Vicksburg

540 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the seismic data and events of interest were markedon this crossline The E41 F11 F39 and G2 eventswere translated from crossline 204 into inlines 190 and210 from the UT-BEG survey as shown in Figure 3The stratigraphic positions of the basal middle Frioreflectors are in good agreementwith respect to traveltime To tie the geologic horizons of interest (E41F11 F39 and G2 Figures 2 3 4) with the UPR 3-Dseismic volume the northndashsouth inline 1238 (theclosest seismic inline to the location of UPRC 175Wardner well) was extracted from the UPR 3-Dseismic volume and events of interest from the VSPwere tied to that seismic inline

Seismic Attributes

Amplitude horizon slices extracted every sample(4 ms) away from the interval of interest show in-dications for channelized features and incised valleysbut were not sharply resolved For this reason wegenerated multiple attribute volumes in an attemptto clearly delineate and resolve fluvial channel fea-tures to enable greater analysis Seismic attributescombine amplitude values at adjacent time samples

and traces to quantify amplitude phase or frequencyof the seismic data Single-trace attributes use ad-jacent samples in a given trace Multitrace or geo-metric attributesmeasure lateral changes in waveformamplitude and phase We find simple root-mean-square (RMS) amplitude (the RMS of the seismicamplitudes) within a 10-ms window to be particularlysensitive to thin bed tuning effects associated with ourfluvial systems The 3-D seismic attributes that bestrevealed the middle Frio fluvial architectural elementsare the RMS amplitude extracted within a 10-mswindow coherent amplitude gradients energy andvolumetric curvature (Chopra and Marfurt 2007)These seismically enhanced delineations of the archi-tectural elements of the middle Frio fluvial systems aidin themeasure of theirmorphologic characteristics andreservoir parameters

Methods

Commercial interpretation packages are used for 3-Dseismic interpretation and attribute extraction TheE41 F11 F39 andG2 seismic horizons (Figure 3) arepicked in the two 3-D seismic data setsmade available

Figure 3 Vertical seismic profile (VSP) display of Union Pacific Resources Company 175 Wardner well (modified from Hardage et al1994) The horizons of interest E41 F11 and F39 are located at 145 158 and 165 sec respectively The peaks (black-filled to the right)are top of reservoir units Crossline 204 from the University of Texas Bureau of Economic Geology three-dimensional (3-D) seismic volumewas tied with inlines (ILs) 190 and 210 and then tied with IL 1238 from the Union Pacific Resources (UPR) 3-D seismic volume Reprinted bypermission of the Society of Exploration Geophysicists whose permission is required for further use

EL-MOWAFY aND MARFURT 541

to this study and a series of stratigraphic horizonslices are produced in one-sample (4-ms) incrementsabove and below the reference surfaces an intervalcorresponding in these data to approximately 6 m(20 ft) of strata Some of the channel systems shownin this article are imaged on window attribute mapseg Figures 5 and 6 and other channel systemsare imaged on horizon slices eg Figures 7 and 8 Itshould be noted that the morphometric measure-ments should be considered minimum because theyare restricted to the resolution of the seismic images

MIDDLE FRIO FRAMEWORK ANDGEOMORPHOLOGY

Sequence Stratigraphy

The sequence models for fluvial systems is docu-mented in detail by Miall (2014 chapter 6) Se-quence stratigraphic reconstruction of the middleFrio in the study area is awork in progress A sequencestratigraphic workflow was applied to reconstructthe architecture and framework of the middle Friosequences In this workflow four architectural ele-ment levels are evaluated facies channel beltssystems tracts and the middle Frio depositional se-quences In this paper we focus on the channel beltelement Sequence stratigraphic subdivisions of themiddle Frio are based to a greater extent onwell logsin addition to one core description and the two 3-Dseismic data volumes used in this study We estab-lished the log response for different types of themiddle Frio deposits by comparing the core char-acteristics and the corresponding well log shapesderived from spontaneous potential (SP) gamma ray(GR) and resistivity logs In the study area the welllog profile of the channel bodies is characterized byboth blocky and bell-shaped or upward-fining pat-terns Crevasse splays are recognized by a funnel-shaped or upward-coarsening pattern Levee bodiesare represented by a spiky pattern and floodplainmudstones are dominated by a baseline pattern Basedon the well log signatures and stacking patterns theinterval between the two sand-dominated channelbelt complexes F11 and F39 is interpreted as a high-stand systems tract (HST) (Figure 4) The deposits ofthe lowstand systems tracts (LST) of the basal middleFrio depositional sequences I and II in the study area

Figure 4 Stratigraphic well log cross section flattened on thetop of F11 showing the subdivisions of the basal middle Frio (BMF)sequences into system tracts at the Union Pacific ResourcesCompany (UPRC) 182 and 185 Wardner well locations Lowstandsystems tracts (LSTs) are characterized by amalgamated multistorysandstone bodies Highstand systems tracts (HSTs) are charac-terized by single-story and multistory channel bodies and single-story crevasse splay and levee bodies encased within floodplainmudstones and siltstones The F11 channel incision at the UPRC 182Wardner well marks the upper boundary of the BMF depositionalsequence I Connected sand bodies are interpreted based on welllog sequence analysis No pressure data were made available to thisstudy Location of the cross section XX9 is shown in Figure 1 deposeq = depositional sequence GR = gamma ray ILD = induction logdeep SP = spontaneous potential TVD = true vertical depth

542 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

consist mostly of single-story and multistory mul-tilateral channel bodies deposited above sequenceboundaries whereas those of the HST are made upof single-story and multistory channel bodies cre-vasse splays and levee bodies isolated within flood-plain mudstones (Figure 4)

The criteria used for the recognition of sequenceboundaries that bound the basal middle Frio se-quences in the study area (Figure 4) are as followsFirst the coarse-grained deposits that accumulateon the channel floor form a channel lag This lag liesabove the basal erosion surface and consists of lo-cally derived material such as mud clasts and blockseroded from the channel banks and bottom plantdebris and coarse-grained gravel and sand Thischannel lag may represent a sequence boundarysurface in fluvial strata The sequence boundary canthen be traced at the top of paleosol horizons thatare correlative to the unconformities generated inthe channel subenvironment (Wright and Marriott

1993 Galloway and Hobday 1996) Themudstoneintraclasts and the paleosol layermdashmarked by car-bonate nodules and root molds from a core cut inthe UPRC 184 Wardner wellmdashrepresent a channellag (refer to Kerr and Jirik 1990 for detailed coredescription) This lag may indicate the upperbounding surface of the basal middle Frio deposi-tional sequence I at the base of the F11 interval(Figure 4) Second abrupt deflections to the left ofthe GR and SP log curves indicate erosional basesof the F11 and F39 channel belt bodies additionallythe local incisions (eg base F11 in the UPRC 182Wardner well Figure 4) could be interpreted torepresent sequence boundaries Third the presenceof the low- and high-sinuosity channel belts imagedon the 3-D seismic attribute maps at the F11 andF39 stratigraphic intervals might also indicate se-quenceboundaries Fourth an incised valley imagedona most-negative curvature attribute map (not shownongoing sequence stratigraphic work) just below the

Figure 5 Root-mean-square(RMS) seismic attribute mapgenerated within a 10-ms windowaround the F11 horizon from theUnion Pacific Resources three-dimensional seismic survey(area A in Figure 1) Several fluvialarchitectural elements are shownon this map See close-up viewsin Figures 6 9 and 10 Dottedcircles indicate locations of pos-sible crevassing andor nodalavulsion points controlled bygrowth faults The coordinatereference system shown in thismap and in Figures 6 9 and10 is the Universal TransverseMercator grid (X = eastingY = northing) zone 14 north

EL-MOWAFY aND MARFURT 543

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 3: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

included in a large rollover anticline truncated bynormal faults which are synthetic and antithetic tothe Vicksburg and Agua Dulce major growth faultsOver most of the study area the deepest part ofthe middle Frio section is affected by these structures(El-Mowafy and Marfurt 2008)

The Frio Formation in south Texas comprises theGueydan fluvial system which drained a semiarid

source areas in the desert southwest (Galloway et al1982a) and entered the Gulf of Mexico through theRio Grande embayment TheGueydan fluvial systemis interpreted as having been deposited by mixedload to bed load slightly sinuous streams with broadwell-developed natural levees (Galloway 1977)Single-story (storey of Friend et al 1979) channelsandstone bodies are 3ndash10 m (10ndash33 ft) thick but

Figure 1 Index map showingthe location of the contiguousStratton and Agua Dulce fieldsand locations of the two three-dimensional (3-D) seismic sur-veys used in this study The twofields are represented schemati-cally by one polygon (not toscale) which approximates thestructure closure at the middleFrio stratigraphic level The twofields occur next to each otherand they are defined by parallel-to-the fault northeastndashsouthwest-trending large rollover anticlinethat is related to the Vicksburgand Agua Dulce faultsrsquo hangingwall deformation Stratton fieldextends from northern KlebergCounty into southern NuecesCounty and Agua Dulce fieldextends farther to the northeastand is mostly covered by theUnion Pacific Resources (UPR)3-D seismic survey Polygon Adelineates the interpreted areaof the UPR 3-D seismic surveyPolygon B delineates area withwell control inside the UPR 3-Dseismic survey Polygon C definesarea covered by the smallerUniversity of Texas Bureau ofEconomic Geology (UT-BEG) 3-Dseismic survey The FR-4 gas playis the prolific Frio gas trend insouth Texas with production fromFrio and Vicksburg fluvial anddeltaic sandstone reservoirs re-spectively along the Vicksburgfault zone Stratton and AguaDulce fields are located in this

play (modified from Kosters et al 1989) Location of Union Pacific Resources Company (UPRC) 175 Wardner well with vertical seismicprofile (VSP) is shown on themap Line of the stratigraphic section in Figure 4 connecting the UPRC 182 and 185Wardner GP wells is shownCorpus Christi bay is also marked on this map Reprinted by permission of UT-BEG whose permission is required for further use

EL-MOWAFY aND MARFURT 539

they commonly amalgamate into units as thick as30 m (100 ft) These amalgamated multistory sand-stone bodies develop into multilateral belts as muchas several miles wide Crevasse splay deposits arewidespread and can extend as far as several thousandfeet from the main channel (Galloway 1981)

Data Description

Two stacked slightly overlapping 3-D seismic sur-veys have been used in this study for the inter-pretation of lateral and vertical variability of themiddle Frio fluvial deposits The signal-to-noise ratio

is high and nomultiples or coherent noise is apparentin the data The first survey (area A in Figure 1) wasacquired in 1993 by Union Pacific Resources (UPR)and processed byWestern Geophysical resulting in afrequency bandwidth of 10ndash90 Hz with a dominantfrequency of 50Hz The size of this survey is 233 km2

(90 mi2) The second survey (area C in Figure 1)was acquired in 1992 by the University of Texas atAustinndashBureau of Economic Geology (UT-BEG) andwas reprocessed by Mercury International Technol-ogy Company in Tulsa Oklahoma resulting in afrequency bandwidth of 10ndash60 Hz with a dominantfrequency of 35 Hz The size of this survey is197 km2 (76 mi2) Using the average frequencies ofthe two 3-D seismic data sets and the average intervalvelocity of 10500 ftsec (3200 msec) of the middleFrio interval obtained from sonic logs the averagevertical resolution of the UT-BEG survey is 23 m(75 ft) and of the UPR survey is 158 m (52 ft) Wegenerated several wedge models to estimate andconfirm the minimum resolved thickness of theseismic data Ricker wavelet with 35-Hz dominantfrequencywas used in the generation of thesemodelsThemodels indicate that theminimum thickness thatcan be resolved seismically in the UPR survey is152 m (50 ft) The detectability limits are estimatedto be 26 and 17m (85 and 56 ft) for the UT-BEG andthe UPR 3-D surveys respectively We also useddigital and hard copies of well logs for 171 wells aswell as one core described by Kerr and Jirik (1990)

Well-to-Seismic Tie

The tools used in associating seismic data with geo-logic horizons are the vertical seismic profile (VSP) ofthe Union Pacific Resource Company (UPRC) 175Wardner well (Figure 3) and synthetic seismogramsThese tools are used to define the predicted two-waytravel time for each of the depositional surfaces ofinterest in the two 3-D seismic data sets The twopoststack 3-D seismic volumes (UT-BEG and UPR)partially overlap in the northern Stratton field area(Figure 1) The UPRC 175 Wardner well with theVSP lies in the heart of the smaller UT-BEG 3-Dseismic survey but on the southern border of the UPRsurvey The eastndashwest crossline 204 was extractedfrom the UT-BEG survey that passes through theUPRC 175Wardner well the VSPwasmatched with

Figure 2 Type log from the Union Producing Company 7ADriscoll well showing the middle Frio reservoir groups and no-menclature at Stratton (left) and AguaDulce (right) fields (modifiedfrom Kerr 1990) Reprinted by permission of the AAPG Gulf CoastAssociation of Geological Societies whose permission is requiredfor further use BMF = basal middle Frio ResD = deep resistivitySP = spontaneous potential U Vicksburg = Upper Vicksburg

540 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the seismic data and events of interest were markedon this crossline The E41 F11 F39 and G2 eventswere translated from crossline 204 into inlines 190 and210 from the UT-BEG survey as shown in Figure 3The stratigraphic positions of the basal middle Frioreflectors are in good agreementwith respect to traveltime To tie the geologic horizons of interest (E41F11 F39 and G2 Figures 2 3 4) with the UPR 3-Dseismic volume the northndashsouth inline 1238 (theclosest seismic inline to the location of UPRC 175Wardner well) was extracted from the UPR 3-Dseismic volume and events of interest from the VSPwere tied to that seismic inline

Seismic Attributes

Amplitude horizon slices extracted every sample(4 ms) away from the interval of interest show in-dications for channelized features and incised valleysbut were not sharply resolved For this reason wegenerated multiple attribute volumes in an attemptto clearly delineate and resolve fluvial channel fea-tures to enable greater analysis Seismic attributescombine amplitude values at adjacent time samples

and traces to quantify amplitude phase or frequencyof the seismic data Single-trace attributes use ad-jacent samples in a given trace Multitrace or geo-metric attributesmeasure lateral changes in waveformamplitude and phase We find simple root-mean-square (RMS) amplitude (the RMS of the seismicamplitudes) within a 10-ms window to be particularlysensitive to thin bed tuning effects associated with ourfluvial systems The 3-D seismic attributes that bestrevealed the middle Frio fluvial architectural elementsare the RMS amplitude extracted within a 10-mswindow coherent amplitude gradients energy andvolumetric curvature (Chopra and Marfurt 2007)These seismically enhanced delineations of the archi-tectural elements of the middle Frio fluvial systems aidin themeasure of theirmorphologic characteristics andreservoir parameters

Methods

Commercial interpretation packages are used for 3-Dseismic interpretation and attribute extraction TheE41 F11 F39 andG2 seismic horizons (Figure 3) arepicked in the two 3-D seismic data setsmade available

Figure 3 Vertical seismic profile (VSP) display of Union Pacific Resources Company 175 Wardner well (modified from Hardage et al1994) The horizons of interest E41 F11 and F39 are located at 145 158 and 165 sec respectively The peaks (black-filled to the right)are top of reservoir units Crossline 204 from the University of Texas Bureau of Economic Geology three-dimensional (3-D) seismic volumewas tied with inlines (ILs) 190 and 210 and then tied with IL 1238 from the Union Pacific Resources (UPR) 3-D seismic volume Reprinted bypermission of the Society of Exploration Geophysicists whose permission is required for further use

EL-MOWAFY aND MARFURT 541

to this study and a series of stratigraphic horizonslices are produced in one-sample (4-ms) incrementsabove and below the reference surfaces an intervalcorresponding in these data to approximately 6 m(20 ft) of strata Some of the channel systems shownin this article are imaged on window attribute mapseg Figures 5 and 6 and other channel systemsare imaged on horizon slices eg Figures 7 and 8 Itshould be noted that the morphometric measure-ments should be considered minimum because theyare restricted to the resolution of the seismic images

MIDDLE FRIO FRAMEWORK ANDGEOMORPHOLOGY

Sequence Stratigraphy

The sequence models for fluvial systems is docu-mented in detail by Miall (2014 chapter 6) Se-quence stratigraphic reconstruction of the middleFrio in the study area is awork in progress A sequencestratigraphic workflow was applied to reconstructthe architecture and framework of the middle Friosequences In this workflow four architectural ele-ment levels are evaluated facies channel beltssystems tracts and the middle Frio depositional se-quences In this paper we focus on the channel beltelement Sequence stratigraphic subdivisions of themiddle Frio are based to a greater extent onwell logsin addition to one core description and the two 3-Dseismic data volumes used in this study We estab-lished the log response for different types of themiddle Frio deposits by comparing the core char-acteristics and the corresponding well log shapesderived from spontaneous potential (SP) gamma ray(GR) and resistivity logs In the study area the welllog profile of the channel bodies is characterized byboth blocky and bell-shaped or upward-fining pat-terns Crevasse splays are recognized by a funnel-shaped or upward-coarsening pattern Levee bodiesare represented by a spiky pattern and floodplainmudstones are dominated by a baseline pattern Basedon the well log signatures and stacking patterns theinterval between the two sand-dominated channelbelt complexes F11 and F39 is interpreted as a high-stand systems tract (HST) (Figure 4) The deposits ofthe lowstand systems tracts (LST) of the basal middleFrio depositional sequences I and II in the study area

Figure 4 Stratigraphic well log cross section flattened on thetop of F11 showing the subdivisions of the basal middle Frio (BMF)sequences into system tracts at the Union Pacific ResourcesCompany (UPRC) 182 and 185 Wardner well locations Lowstandsystems tracts (LSTs) are characterized by amalgamated multistorysandstone bodies Highstand systems tracts (HSTs) are charac-terized by single-story and multistory channel bodies and single-story crevasse splay and levee bodies encased within floodplainmudstones and siltstones The F11 channel incision at the UPRC 182Wardner well marks the upper boundary of the BMF depositionalsequence I Connected sand bodies are interpreted based on welllog sequence analysis No pressure data were made available to thisstudy Location of the cross section XX9 is shown in Figure 1 deposeq = depositional sequence GR = gamma ray ILD = induction logdeep SP = spontaneous potential TVD = true vertical depth

542 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

consist mostly of single-story and multistory mul-tilateral channel bodies deposited above sequenceboundaries whereas those of the HST are made upof single-story and multistory channel bodies cre-vasse splays and levee bodies isolated within flood-plain mudstones (Figure 4)

The criteria used for the recognition of sequenceboundaries that bound the basal middle Frio se-quences in the study area (Figure 4) are as followsFirst the coarse-grained deposits that accumulateon the channel floor form a channel lag This lag liesabove the basal erosion surface and consists of lo-cally derived material such as mud clasts and blockseroded from the channel banks and bottom plantdebris and coarse-grained gravel and sand Thischannel lag may represent a sequence boundarysurface in fluvial strata The sequence boundary canthen be traced at the top of paleosol horizons thatare correlative to the unconformities generated inthe channel subenvironment (Wright and Marriott

1993 Galloway and Hobday 1996) Themudstoneintraclasts and the paleosol layermdashmarked by car-bonate nodules and root molds from a core cut inthe UPRC 184 Wardner wellmdashrepresent a channellag (refer to Kerr and Jirik 1990 for detailed coredescription) This lag may indicate the upperbounding surface of the basal middle Frio deposi-tional sequence I at the base of the F11 interval(Figure 4) Second abrupt deflections to the left ofthe GR and SP log curves indicate erosional basesof the F11 and F39 channel belt bodies additionallythe local incisions (eg base F11 in the UPRC 182Wardner well Figure 4) could be interpreted torepresent sequence boundaries Third the presenceof the low- and high-sinuosity channel belts imagedon the 3-D seismic attribute maps at the F11 andF39 stratigraphic intervals might also indicate se-quenceboundaries Fourth an incised valley imagedona most-negative curvature attribute map (not shownongoing sequence stratigraphic work) just below the

Figure 5 Root-mean-square(RMS) seismic attribute mapgenerated within a 10-ms windowaround the F11 horizon from theUnion Pacific Resources three-dimensional seismic survey(area A in Figure 1) Several fluvialarchitectural elements are shownon this map See close-up viewsin Figures 6 9 and 10 Dottedcircles indicate locations of pos-sible crevassing andor nodalavulsion points controlled bygrowth faults The coordinatereference system shown in thismap and in Figures 6 9 and10 is the Universal TransverseMercator grid (X = eastingY = northing) zone 14 north

EL-MOWAFY aND MARFURT 543

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 4: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

they commonly amalgamate into units as thick as30 m (100 ft) These amalgamated multistory sand-stone bodies develop into multilateral belts as muchas several miles wide Crevasse splay deposits arewidespread and can extend as far as several thousandfeet from the main channel (Galloway 1981)

Data Description

Two stacked slightly overlapping 3-D seismic sur-veys have been used in this study for the inter-pretation of lateral and vertical variability of themiddle Frio fluvial deposits The signal-to-noise ratio

is high and nomultiples or coherent noise is apparentin the data The first survey (area A in Figure 1) wasacquired in 1993 by Union Pacific Resources (UPR)and processed byWestern Geophysical resulting in afrequency bandwidth of 10ndash90 Hz with a dominantfrequency of 50Hz The size of this survey is 233 km2

(90 mi2) The second survey (area C in Figure 1)was acquired in 1992 by the University of Texas atAustinndashBureau of Economic Geology (UT-BEG) andwas reprocessed by Mercury International Technol-ogy Company in Tulsa Oklahoma resulting in afrequency bandwidth of 10ndash60 Hz with a dominantfrequency of 35 Hz The size of this survey is197 km2 (76 mi2) Using the average frequencies ofthe two 3-D seismic data sets and the average intervalvelocity of 10500 ftsec (3200 msec) of the middleFrio interval obtained from sonic logs the averagevertical resolution of the UT-BEG survey is 23 m(75 ft) and of the UPR survey is 158 m (52 ft) Wegenerated several wedge models to estimate andconfirm the minimum resolved thickness of theseismic data Ricker wavelet with 35-Hz dominantfrequencywas used in the generation of thesemodelsThemodels indicate that theminimum thickness thatcan be resolved seismically in the UPR survey is152 m (50 ft) The detectability limits are estimatedto be 26 and 17m (85 and 56 ft) for the UT-BEG andthe UPR 3-D surveys respectively We also useddigital and hard copies of well logs for 171 wells aswell as one core described by Kerr and Jirik (1990)

Well-to-Seismic Tie

The tools used in associating seismic data with geo-logic horizons are the vertical seismic profile (VSP) ofthe Union Pacific Resource Company (UPRC) 175Wardner well (Figure 3) and synthetic seismogramsThese tools are used to define the predicted two-waytravel time for each of the depositional surfaces ofinterest in the two 3-D seismic data sets The twopoststack 3-D seismic volumes (UT-BEG and UPR)partially overlap in the northern Stratton field area(Figure 1) The UPRC 175 Wardner well with theVSP lies in the heart of the smaller UT-BEG 3-Dseismic survey but on the southern border of the UPRsurvey The eastndashwest crossline 204 was extractedfrom the UT-BEG survey that passes through theUPRC 175Wardner well the VSPwasmatched with

Figure 2 Type log from the Union Producing Company 7ADriscoll well showing the middle Frio reservoir groups and no-menclature at Stratton (left) and AguaDulce (right) fields (modifiedfrom Kerr 1990) Reprinted by permission of the AAPG Gulf CoastAssociation of Geological Societies whose permission is requiredfor further use BMF = basal middle Frio ResD = deep resistivitySP = spontaneous potential U Vicksburg = Upper Vicksburg

540 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the seismic data and events of interest were markedon this crossline The E41 F11 F39 and G2 eventswere translated from crossline 204 into inlines 190 and210 from the UT-BEG survey as shown in Figure 3The stratigraphic positions of the basal middle Frioreflectors are in good agreementwith respect to traveltime To tie the geologic horizons of interest (E41F11 F39 and G2 Figures 2 3 4) with the UPR 3-Dseismic volume the northndashsouth inline 1238 (theclosest seismic inline to the location of UPRC 175Wardner well) was extracted from the UPR 3-Dseismic volume and events of interest from the VSPwere tied to that seismic inline

Seismic Attributes

Amplitude horizon slices extracted every sample(4 ms) away from the interval of interest show in-dications for channelized features and incised valleysbut were not sharply resolved For this reason wegenerated multiple attribute volumes in an attemptto clearly delineate and resolve fluvial channel fea-tures to enable greater analysis Seismic attributescombine amplitude values at adjacent time samples

and traces to quantify amplitude phase or frequencyof the seismic data Single-trace attributes use ad-jacent samples in a given trace Multitrace or geo-metric attributesmeasure lateral changes in waveformamplitude and phase We find simple root-mean-square (RMS) amplitude (the RMS of the seismicamplitudes) within a 10-ms window to be particularlysensitive to thin bed tuning effects associated with ourfluvial systems The 3-D seismic attributes that bestrevealed the middle Frio fluvial architectural elementsare the RMS amplitude extracted within a 10-mswindow coherent amplitude gradients energy andvolumetric curvature (Chopra and Marfurt 2007)These seismically enhanced delineations of the archi-tectural elements of the middle Frio fluvial systems aidin themeasure of theirmorphologic characteristics andreservoir parameters

Methods

Commercial interpretation packages are used for 3-Dseismic interpretation and attribute extraction TheE41 F11 F39 andG2 seismic horizons (Figure 3) arepicked in the two 3-D seismic data setsmade available

Figure 3 Vertical seismic profile (VSP) display of Union Pacific Resources Company 175 Wardner well (modified from Hardage et al1994) The horizons of interest E41 F11 and F39 are located at 145 158 and 165 sec respectively The peaks (black-filled to the right)are top of reservoir units Crossline 204 from the University of Texas Bureau of Economic Geology three-dimensional (3-D) seismic volumewas tied with inlines (ILs) 190 and 210 and then tied with IL 1238 from the Union Pacific Resources (UPR) 3-D seismic volume Reprinted bypermission of the Society of Exploration Geophysicists whose permission is required for further use

EL-MOWAFY aND MARFURT 541

to this study and a series of stratigraphic horizonslices are produced in one-sample (4-ms) incrementsabove and below the reference surfaces an intervalcorresponding in these data to approximately 6 m(20 ft) of strata Some of the channel systems shownin this article are imaged on window attribute mapseg Figures 5 and 6 and other channel systemsare imaged on horizon slices eg Figures 7 and 8 Itshould be noted that the morphometric measure-ments should be considered minimum because theyare restricted to the resolution of the seismic images

MIDDLE FRIO FRAMEWORK ANDGEOMORPHOLOGY

Sequence Stratigraphy

The sequence models for fluvial systems is docu-mented in detail by Miall (2014 chapter 6) Se-quence stratigraphic reconstruction of the middleFrio in the study area is awork in progress A sequencestratigraphic workflow was applied to reconstructthe architecture and framework of the middle Friosequences In this workflow four architectural ele-ment levels are evaluated facies channel beltssystems tracts and the middle Frio depositional se-quences In this paper we focus on the channel beltelement Sequence stratigraphic subdivisions of themiddle Frio are based to a greater extent onwell logsin addition to one core description and the two 3-Dseismic data volumes used in this study We estab-lished the log response for different types of themiddle Frio deposits by comparing the core char-acteristics and the corresponding well log shapesderived from spontaneous potential (SP) gamma ray(GR) and resistivity logs In the study area the welllog profile of the channel bodies is characterized byboth blocky and bell-shaped or upward-fining pat-terns Crevasse splays are recognized by a funnel-shaped or upward-coarsening pattern Levee bodiesare represented by a spiky pattern and floodplainmudstones are dominated by a baseline pattern Basedon the well log signatures and stacking patterns theinterval between the two sand-dominated channelbelt complexes F11 and F39 is interpreted as a high-stand systems tract (HST) (Figure 4) The deposits ofthe lowstand systems tracts (LST) of the basal middleFrio depositional sequences I and II in the study area

Figure 4 Stratigraphic well log cross section flattened on thetop of F11 showing the subdivisions of the basal middle Frio (BMF)sequences into system tracts at the Union Pacific ResourcesCompany (UPRC) 182 and 185 Wardner well locations Lowstandsystems tracts (LSTs) are characterized by amalgamated multistorysandstone bodies Highstand systems tracts (HSTs) are charac-terized by single-story and multistory channel bodies and single-story crevasse splay and levee bodies encased within floodplainmudstones and siltstones The F11 channel incision at the UPRC 182Wardner well marks the upper boundary of the BMF depositionalsequence I Connected sand bodies are interpreted based on welllog sequence analysis No pressure data were made available to thisstudy Location of the cross section XX9 is shown in Figure 1 deposeq = depositional sequence GR = gamma ray ILD = induction logdeep SP = spontaneous potential TVD = true vertical depth

542 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

consist mostly of single-story and multistory mul-tilateral channel bodies deposited above sequenceboundaries whereas those of the HST are made upof single-story and multistory channel bodies cre-vasse splays and levee bodies isolated within flood-plain mudstones (Figure 4)

The criteria used for the recognition of sequenceboundaries that bound the basal middle Frio se-quences in the study area (Figure 4) are as followsFirst the coarse-grained deposits that accumulateon the channel floor form a channel lag This lag liesabove the basal erosion surface and consists of lo-cally derived material such as mud clasts and blockseroded from the channel banks and bottom plantdebris and coarse-grained gravel and sand Thischannel lag may represent a sequence boundarysurface in fluvial strata The sequence boundary canthen be traced at the top of paleosol horizons thatare correlative to the unconformities generated inthe channel subenvironment (Wright and Marriott

1993 Galloway and Hobday 1996) Themudstoneintraclasts and the paleosol layermdashmarked by car-bonate nodules and root molds from a core cut inthe UPRC 184 Wardner wellmdashrepresent a channellag (refer to Kerr and Jirik 1990 for detailed coredescription) This lag may indicate the upperbounding surface of the basal middle Frio deposi-tional sequence I at the base of the F11 interval(Figure 4) Second abrupt deflections to the left ofthe GR and SP log curves indicate erosional basesof the F11 and F39 channel belt bodies additionallythe local incisions (eg base F11 in the UPRC 182Wardner well Figure 4) could be interpreted torepresent sequence boundaries Third the presenceof the low- and high-sinuosity channel belts imagedon the 3-D seismic attribute maps at the F11 andF39 stratigraphic intervals might also indicate se-quenceboundaries Fourth an incised valley imagedona most-negative curvature attribute map (not shownongoing sequence stratigraphic work) just below the

Figure 5 Root-mean-square(RMS) seismic attribute mapgenerated within a 10-ms windowaround the F11 horizon from theUnion Pacific Resources three-dimensional seismic survey(area A in Figure 1) Several fluvialarchitectural elements are shownon this map See close-up viewsin Figures 6 9 and 10 Dottedcircles indicate locations of pos-sible crevassing andor nodalavulsion points controlled bygrowth faults The coordinatereference system shown in thismap and in Figures 6 9 and10 is the Universal TransverseMercator grid (X = eastingY = northing) zone 14 north

EL-MOWAFY aND MARFURT 543

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 5: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

the seismic data and events of interest were markedon this crossline The E41 F11 F39 and G2 eventswere translated from crossline 204 into inlines 190 and210 from the UT-BEG survey as shown in Figure 3The stratigraphic positions of the basal middle Frioreflectors are in good agreementwith respect to traveltime To tie the geologic horizons of interest (E41F11 F39 and G2 Figures 2 3 4) with the UPR 3-Dseismic volume the northndashsouth inline 1238 (theclosest seismic inline to the location of UPRC 175Wardner well) was extracted from the UPR 3-Dseismic volume and events of interest from the VSPwere tied to that seismic inline

Seismic Attributes

Amplitude horizon slices extracted every sample(4 ms) away from the interval of interest show in-dications for channelized features and incised valleysbut were not sharply resolved For this reason wegenerated multiple attribute volumes in an attemptto clearly delineate and resolve fluvial channel fea-tures to enable greater analysis Seismic attributescombine amplitude values at adjacent time samples

and traces to quantify amplitude phase or frequencyof the seismic data Single-trace attributes use ad-jacent samples in a given trace Multitrace or geo-metric attributesmeasure lateral changes in waveformamplitude and phase We find simple root-mean-square (RMS) amplitude (the RMS of the seismicamplitudes) within a 10-ms window to be particularlysensitive to thin bed tuning effects associated with ourfluvial systems The 3-D seismic attributes that bestrevealed the middle Frio fluvial architectural elementsare the RMS amplitude extracted within a 10-mswindow coherent amplitude gradients energy andvolumetric curvature (Chopra and Marfurt 2007)These seismically enhanced delineations of the archi-tectural elements of the middle Frio fluvial systems aidin themeasure of theirmorphologic characteristics andreservoir parameters

Methods

Commercial interpretation packages are used for 3-Dseismic interpretation and attribute extraction TheE41 F11 F39 andG2 seismic horizons (Figure 3) arepicked in the two 3-D seismic data setsmade available

Figure 3 Vertical seismic profile (VSP) display of Union Pacific Resources Company 175 Wardner well (modified from Hardage et al1994) The horizons of interest E41 F11 and F39 are located at 145 158 and 165 sec respectively The peaks (black-filled to the right)are top of reservoir units Crossline 204 from the University of Texas Bureau of Economic Geology three-dimensional (3-D) seismic volumewas tied with inlines (ILs) 190 and 210 and then tied with IL 1238 from the Union Pacific Resources (UPR) 3-D seismic volume Reprinted bypermission of the Society of Exploration Geophysicists whose permission is required for further use

EL-MOWAFY aND MARFURT 541

to this study and a series of stratigraphic horizonslices are produced in one-sample (4-ms) incrementsabove and below the reference surfaces an intervalcorresponding in these data to approximately 6 m(20 ft) of strata Some of the channel systems shownin this article are imaged on window attribute mapseg Figures 5 and 6 and other channel systemsare imaged on horizon slices eg Figures 7 and 8 Itshould be noted that the morphometric measure-ments should be considered minimum because theyare restricted to the resolution of the seismic images

MIDDLE FRIO FRAMEWORK ANDGEOMORPHOLOGY

Sequence Stratigraphy

The sequence models for fluvial systems is docu-mented in detail by Miall (2014 chapter 6) Se-quence stratigraphic reconstruction of the middleFrio in the study area is awork in progress A sequencestratigraphic workflow was applied to reconstructthe architecture and framework of the middle Friosequences In this workflow four architectural ele-ment levels are evaluated facies channel beltssystems tracts and the middle Frio depositional se-quences In this paper we focus on the channel beltelement Sequence stratigraphic subdivisions of themiddle Frio are based to a greater extent onwell logsin addition to one core description and the two 3-Dseismic data volumes used in this study We estab-lished the log response for different types of themiddle Frio deposits by comparing the core char-acteristics and the corresponding well log shapesderived from spontaneous potential (SP) gamma ray(GR) and resistivity logs In the study area the welllog profile of the channel bodies is characterized byboth blocky and bell-shaped or upward-fining pat-terns Crevasse splays are recognized by a funnel-shaped or upward-coarsening pattern Levee bodiesare represented by a spiky pattern and floodplainmudstones are dominated by a baseline pattern Basedon the well log signatures and stacking patterns theinterval between the two sand-dominated channelbelt complexes F11 and F39 is interpreted as a high-stand systems tract (HST) (Figure 4) The deposits ofthe lowstand systems tracts (LST) of the basal middleFrio depositional sequences I and II in the study area

Figure 4 Stratigraphic well log cross section flattened on thetop of F11 showing the subdivisions of the basal middle Frio (BMF)sequences into system tracts at the Union Pacific ResourcesCompany (UPRC) 182 and 185 Wardner well locations Lowstandsystems tracts (LSTs) are characterized by amalgamated multistorysandstone bodies Highstand systems tracts (HSTs) are charac-terized by single-story and multistory channel bodies and single-story crevasse splay and levee bodies encased within floodplainmudstones and siltstones The F11 channel incision at the UPRC 182Wardner well marks the upper boundary of the BMF depositionalsequence I Connected sand bodies are interpreted based on welllog sequence analysis No pressure data were made available to thisstudy Location of the cross section XX9 is shown in Figure 1 deposeq = depositional sequence GR = gamma ray ILD = induction logdeep SP = spontaneous potential TVD = true vertical depth

542 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

consist mostly of single-story and multistory mul-tilateral channel bodies deposited above sequenceboundaries whereas those of the HST are made upof single-story and multistory channel bodies cre-vasse splays and levee bodies isolated within flood-plain mudstones (Figure 4)

The criteria used for the recognition of sequenceboundaries that bound the basal middle Frio se-quences in the study area (Figure 4) are as followsFirst the coarse-grained deposits that accumulateon the channel floor form a channel lag This lag liesabove the basal erosion surface and consists of lo-cally derived material such as mud clasts and blockseroded from the channel banks and bottom plantdebris and coarse-grained gravel and sand Thischannel lag may represent a sequence boundarysurface in fluvial strata The sequence boundary canthen be traced at the top of paleosol horizons thatare correlative to the unconformities generated inthe channel subenvironment (Wright and Marriott

1993 Galloway and Hobday 1996) Themudstoneintraclasts and the paleosol layermdashmarked by car-bonate nodules and root molds from a core cut inthe UPRC 184 Wardner wellmdashrepresent a channellag (refer to Kerr and Jirik 1990 for detailed coredescription) This lag may indicate the upperbounding surface of the basal middle Frio deposi-tional sequence I at the base of the F11 interval(Figure 4) Second abrupt deflections to the left ofthe GR and SP log curves indicate erosional basesof the F11 and F39 channel belt bodies additionallythe local incisions (eg base F11 in the UPRC 182Wardner well Figure 4) could be interpreted torepresent sequence boundaries Third the presenceof the low- and high-sinuosity channel belts imagedon the 3-D seismic attribute maps at the F11 andF39 stratigraphic intervals might also indicate se-quenceboundaries Fourth an incised valley imagedona most-negative curvature attribute map (not shownongoing sequence stratigraphic work) just below the

Figure 5 Root-mean-square(RMS) seismic attribute mapgenerated within a 10-ms windowaround the F11 horizon from theUnion Pacific Resources three-dimensional seismic survey(area A in Figure 1) Several fluvialarchitectural elements are shownon this map See close-up viewsin Figures 6 9 and 10 Dottedcircles indicate locations of pos-sible crevassing andor nodalavulsion points controlled bygrowth faults The coordinatereference system shown in thismap and in Figures 6 9 and10 is the Universal TransverseMercator grid (X = eastingY = northing) zone 14 north

EL-MOWAFY aND MARFURT 543

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 6: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

to this study and a series of stratigraphic horizonslices are produced in one-sample (4-ms) incrementsabove and below the reference surfaces an intervalcorresponding in these data to approximately 6 m(20 ft) of strata Some of the channel systems shownin this article are imaged on window attribute mapseg Figures 5 and 6 and other channel systemsare imaged on horizon slices eg Figures 7 and 8 Itshould be noted that the morphometric measure-ments should be considered minimum because theyare restricted to the resolution of the seismic images

MIDDLE FRIO FRAMEWORK ANDGEOMORPHOLOGY

Sequence Stratigraphy

The sequence models for fluvial systems is docu-mented in detail by Miall (2014 chapter 6) Se-quence stratigraphic reconstruction of the middleFrio in the study area is awork in progress A sequencestratigraphic workflow was applied to reconstructthe architecture and framework of the middle Friosequences In this workflow four architectural ele-ment levels are evaluated facies channel beltssystems tracts and the middle Frio depositional se-quences In this paper we focus on the channel beltelement Sequence stratigraphic subdivisions of themiddle Frio are based to a greater extent onwell logsin addition to one core description and the two 3-Dseismic data volumes used in this study We estab-lished the log response for different types of themiddle Frio deposits by comparing the core char-acteristics and the corresponding well log shapesderived from spontaneous potential (SP) gamma ray(GR) and resistivity logs In the study area the welllog profile of the channel bodies is characterized byboth blocky and bell-shaped or upward-fining pat-terns Crevasse splays are recognized by a funnel-shaped or upward-coarsening pattern Levee bodiesare represented by a spiky pattern and floodplainmudstones are dominated by a baseline pattern Basedon the well log signatures and stacking patterns theinterval between the two sand-dominated channelbelt complexes F11 and F39 is interpreted as a high-stand systems tract (HST) (Figure 4) The deposits ofthe lowstand systems tracts (LST) of the basal middleFrio depositional sequences I and II in the study area

Figure 4 Stratigraphic well log cross section flattened on thetop of F11 showing the subdivisions of the basal middle Frio (BMF)sequences into system tracts at the Union Pacific ResourcesCompany (UPRC) 182 and 185 Wardner well locations Lowstandsystems tracts (LSTs) are characterized by amalgamated multistorysandstone bodies Highstand systems tracts (HSTs) are charac-terized by single-story and multistory channel bodies and single-story crevasse splay and levee bodies encased within floodplainmudstones and siltstones The F11 channel incision at the UPRC 182Wardner well marks the upper boundary of the BMF depositionalsequence I Connected sand bodies are interpreted based on welllog sequence analysis No pressure data were made available to thisstudy Location of the cross section XX9 is shown in Figure 1 deposeq = depositional sequence GR = gamma ray ILD = induction logdeep SP = spontaneous potential TVD = true vertical depth

542 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

consist mostly of single-story and multistory mul-tilateral channel bodies deposited above sequenceboundaries whereas those of the HST are made upof single-story and multistory channel bodies cre-vasse splays and levee bodies isolated within flood-plain mudstones (Figure 4)

The criteria used for the recognition of sequenceboundaries that bound the basal middle Frio se-quences in the study area (Figure 4) are as followsFirst the coarse-grained deposits that accumulateon the channel floor form a channel lag This lag liesabove the basal erosion surface and consists of lo-cally derived material such as mud clasts and blockseroded from the channel banks and bottom plantdebris and coarse-grained gravel and sand Thischannel lag may represent a sequence boundarysurface in fluvial strata The sequence boundary canthen be traced at the top of paleosol horizons thatare correlative to the unconformities generated inthe channel subenvironment (Wright and Marriott

1993 Galloway and Hobday 1996) Themudstoneintraclasts and the paleosol layermdashmarked by car-bonate nodules and root molds from a core cut inthe UPRC 184 Wardner wellmdashrepresent a channellag (refer to Kerr and Jirik 1990 for detailed coredescription) This lag may indicate the upperbounding surface of the basal middle Frio deposi-tional sequence I at the base of the F11 interval(Figure 4) Second abrupt deflections to the left ofthe GR and SP log curves indicate erosional basesof the F11 and F39 channel belt bodies additionallythe local incisions (eg base F11 in the UPRC 182Wardner well Figure 4) could be interpreted torepresent sequence boundaries Third the presenceof the low- and high-sinuosity channel belts imagedon the 3-D seismic attribute maps at the F11 andF39 stratigraphic intervals might also indicate se-quenceboundaries Fourth an incised valley imagedona most-negative curvature attribute map (not shownongoing sequence stratigraphic work) just below the

Figure 5 Root-mean-square(RMS) seismic attribute mapgenerated within a 10-ms windowaround the F11 horizon from theUnion Pacific Resources three-dimensional seismic survey(area A in Figure 1) Several fluvialarchitectural elements are shownon this map See close-up viewsin Figures 6 9 and 10 Dottedcircles indicate locations of pos-sible crevassing andor nodalavulsion points controlled bygrowth faults The coordinatereference system shown in thismap and in Figures 6 9 and10 is the Universal TransverseMercator grid (X = eastingY = northing) zone 14 north

EL-MOWAFY aND MARFURT 543

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 7: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

consist mostly of single-story and multistory mul-tilateral channel bodies deposited above sequenceboundaries whereas those of the HST are made upof single-story and multistory channel bodies cre-vasse splays and levee bodies isolated within flood-plain mudstones (Figure 4)

The criteria used for the recognition of sequenceboundaries that bound the basal middle Frio se-quences in the study area (Figure 4) are as followsFirst the coarse-grained deposits that accumulateon the channel floor form a channel lag This lag liesabove the basal erosion surface and consists of lo-cally derived material such as mud clasts and blockseroded from the channel banks and bottom plantdebris and coarse-grained gravel and sand Thischannel lag may represent a sequence boundarysurface in fluvial strata The sequence boundary canthen be traced at the top of paleosol horizons thatare correlative to the unconformities generated inthe channel subenvironment (Wright and Marriott

1993 Galloway and Hobday 1996) Themudstoneintraclasts and the paleosol layermdashmarked by car-bonate nodules and root molds from a core cut inthe UPRC 184 Wardner wellmdashrepresent a channellag (refer to Kerr and Jirik 1990 for detailed coredescription) This lag may indicate the upperbounding surface of the basal middle Frio deposi-tional sequence I at the base of the F11 interval(Figure 4) Second abrupt deflections to the left ofthe GR and SP log curves indicate erosional basesof the F11 and F39 channel belt bodies additionallythe local incisions (eg base F11 in the UPRC 182Wardner well Figure 4) could be interpreted torepresent sequence boundaries Third the presenceof the low- and high-sinuosity channel belts imagedon the 3-D seismic attribute maps at the F11 andF39 stratigraphic intervals might also indicate se-quenceboundaries Fourth an incised valley imagedona most-negative curvature attribute map (not shownongoing sequence stratigraphic work) just below the

Figure 5 Root-mean-square(RMS) seismic attribute mapgenerated within a 10-ms windowaround the F11 horizon from theUnion Pacific Resources three-dimensional seismic survey(area A in Figure 1) Several fluvialarchitectural elements are shownon this map See close-up viewsin Figures 6 9 and 10 Dottedcircles indicate locations of pos-sible crevassing andor nodalavulsion points controlled bygrowth faults The coordinatereference system shown in thismap and in Figures 6 9 and10 is the Universal TransverseMercator grid (X = eastingY = northing) zone 14 north

EL-MOWAFY aND MARFURT 543

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 8: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

F39 interval may also be an indication for a sequenceboundary at the base of the basal middle Frio depo-sitional sequence I (Figure 4)

The controls on the middle Frio fluvial archi-tecture may be a function of several factors Miall(2015 p 4) argued that ldquosystematic changes in al-luvial architecture are not the product of changingavulsion rates and changes in fluvial style under theinfluence of variable rates of accommodation butreflect regional shifts in facies belts that themselvesare a response to tectonism and to changes in ac-commodation and other variables (eg dischargesediment supply bank materials Gibling 2006)rdquo

Seismic modeling indicates that high seismicamplitudes are related to a high content of coarse-grained sandstone deposits and low amplitudes arerelated to fine-grainedndashdominated layers These re-sults suggest the ability to use seismic morphometricdata to identify fill type within fluvial incisions Themiddle Frio channel belts incorporate both mainchannels and crevasse channels Based on the seismic

signature the main channels are high-amplitude fea-tures andnamedcategories 1 and2 (category2 channelsare expected to be filled with the coarsest grain sizerelative to category 1) whereas crevasse channels arelow-amplitude features and named category 3

Channel Belts

A channel belt can be defined as an array of con-tiguous channel deposits formed by lateral migrationof a single channel (Friend 1983) Based on thisdefinition a channel belt can be composed of mul-tistory channel bodies (multiple depositional epi-sodes) such as the succession of channel bodiescomposing the F11 channel belt described from thecore cut from the UPRC 184 Wardner well (fordetailed core description refer to Kerr and Jirik1990) A channel belt can be identified from welllogs by analyzing channel fill sandstone bodiesbounded by log breaks Channel belt dimensions

Figure 6 Close-up view of the northwest part of the F11 root-mean-square (RMS) amplitude map in Figure 5 showing a high-amplitudemeandering channel belt depicted by the circled 1 in (B) imaged in the footwall block of the Agua Dulce growth fault This channel belttrends in a northeastndashsouthwest direction parallel to and confined by the major Agua Dulce growth fault Note the bright amplitudes insidethe meander loops interpreted to represent point bars (A) Uninterpreted and (B) interpreted

544 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 9: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

and directions can be best estimated from the 3-Dseismic amplitude extractions A channel belt appearson the study arearsquos 3-D seismic maps as a sinuouschannel and the associated point bars or lateral accre-tion deposits that are represented by bright amplitudesinside its meander loops (Figures 5 6) Based on theseresults single channel belts (Figures 6 7 [feature 1] 8)can be resolved using 3-D seismic attributes

Abandoned channels are a common architecturalelement in the meandering fluvial systems which resultfrom avulsion processes Abandoned channels are com-monly filled with fine-grained sediments but sometimesmay also be filled with deposits equal in grain size tothe deposits of the main channel (Figure 9 feature 7)

The architectural elements found in the middleFrio interval indicate a great variety in channel beltdirectionality and dimensionality (eg Figures 6ndash10)The seismic extractions further indicate variations invertical and lateral stacking of amalgamated channelbelt deposits

Feature 8 in Figure 9 is situated in the footwallside of the Agua Dulce fault and has a distinctiveseismic character and morphology It could be in-terpreted as a segment of an incised valley because it isverywide comparedwith other channels of the fluvialsystem and contains internal channel segments Itcould also be a segment of a wide and highly sinuouschannel belt However the area of this feature is toosmall to be sure of the identification This feature is upto 3600m (11811 ft) in width and 1600 m (5250 ft)in meander arc height (MAH)

Crevasse Channels

Miall (1996) defined crevasse channels as small delta-like distributary systems up to a few thousand feetin width that become shallower away from the mainchannel and consist mainly of fine- tomedium-grainedsandstones and siltstones In this study we identify

Figure 7 Horizon slice 24 msabove the F11 horizon throughthe northndashsouth inline coherentamplitude gradient attribute Thetwo channels depicted by circlednumbers in the southeast part ofthe survey appear clearly and runin a northeastndashsouthwestdirection

EL-MOWAFY aND MARFURT 545

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 10: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

the crevasse channel architectural element fromseismic attribute maps In the southern and northernparts of the hanging wall side of the AguaDulce fault(Figures 9 10) we recognize two wide category 3low-amplitude crevassendashchannel systems composedof narrow individual channels Unfortunately nowell

data aremade available in this part of the study area tocalibrate with the seismic The branching of the low-amplitude crevasse channelsmay indicate that severalchannels are being imaged on the same map eitherby geologically cutting down through earlier featuresor by seismically mixing vertically stacked features

Figure 8 (A) The image to theleft is a coherence slice extractedfrom the Union Pacific Resourcesthree-dimensional seismic surveyat approximately the F39 strati-graphic interval near the base ofthe middle Frio The image to theright is an eastndashwest componentof the coherent energyndashweightedamplitude gradient attribute ex-tracted at the same level Thisattribute is less sensitive to faultswhich are generally incoherentand more sensitive to amplitudechanges Note the classic ex-pression of a channel in the ver-tical seismic section (upper leftcorner) where it is both verticallyand laterally confined with dif-ferent reflection strength At thislocation it is not fault controlled(B) Same as in Figure 8A but theimage to the right is overlain bya multiattribute image of peakfrequency (modulating the hue)and amplitude at the peakfrequency (modulating the light-ness) Blue corresponds to 5 Hzand red corresponds to 70 Hz Thechannel indicated by the magentaarrow depicted by number 3shows up as bright green imply-ing that it is tuned at about 40 Hzand has strong amplitude Twochannels depicted by numbers 1and 2 appear in the section (or-ange arrows) These are deeperand have very low-amplitude fillsuch that they do not show up inthe coherent energyndashweightedeastndashwest amplitude gradient Thedeep blue color indicates thatthese gouges are quite thickNote A color version of thisfigure appears in the onlineversion of this paper

546 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 11: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

through the band-limited seismic wavelet We in-terpret the category 3 crevasse channel systems to befilled with overbank fine-grained deposits resultingin a low-amplitude anomaly corresponding to a lowacoustic impedance contrast between these overbankdeposits and the surrounding channel fill coarse-grained sandstone deposits

Quantitative Seismic Geomorphology of theMiddle Frio Fluvial Systems

Quantitative seismic geomorphology is the quanti-tative analysis of landforms imaged in 3-D verticaland horizontal seismic sections with the objective ofunderstanding thehistory processes andfill architecture

Figure 9 Close-up view ofthe southern part of the F11root-mean-square (RMS) ampli-tude map in Figure 5 Eight fluvialarchitectural elements (depictedby numbers 1ndash8) are detected onthis map crevasse channelsabandoned meander loops andpossible segment of incised valleyor highly sinuous channel beltThe crevasse channels andabandoned channels have low-amplitude and high-amplitude fillrespectively Feature 8 is domi-nated by low-amplitude fill withinternal high-amplitude channelsDotted circle indicates location oftrunk channel crevassing andorpossible upstream nodal avulsionpoint controlled by Agua Dulcegrowth fault activity In the anal-ysis window the dominantlow-amplitude nature of feature 8may be attributed to fine-grainedlithologies near the base ofthe channelized feature (A) Un-interpreted and (B) interpreted

EL-MOWAFY aND MARFURT 547

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 12: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Figure 10 Close-up view ofthe northern part of the root-mean-square (RMS) amplitudemap in Figure 5 showing crevassechannel systems depicted bycircled numbers 1ndash3 in (B) on thehanging wall of the Agua Dulcefault The low-amplitude crevassechannels are interpreted to befilled with fine-grained faciesDotted circle indicates location ofcrevassing andor possible up-stream nodal avulsion pointcontrolled by Agua Dulce growthfault activity The interpreted tiechannel feature 2 and floodplaindepression feature 4 are com-mon architectural elements of ameandering fluvial system thatmay also be related to avulsionand abandonment A tie channelis a channel that transfers waterand sediment to floodplain de-pressions from the main riverchannel during high-flow events(Coffman et al 2010) (A) Un-interpreted and (B) interpreted

548 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 13: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

of a basin (Wood 2007) The 3-D seismic data wereused for the collection of deterministic quantitativedata on the middle Frio channel system morphologythat can be used for field development planning andreservoir modeling

To the knowledge of the authors no publishedwork is available on the quantitative aspects of themiddle Frio fluvial systems in the Texas Gulf CoastBasin and to some extent the general lack of suchdata are generally lacking in the global fluvial data-base Hammes et al (2005) analyzed deep seismicdata from an interval equivalent to the lower FrioUnit in the Corpus Christi area and demonstratedthe evolution from basin floor fans at the base of thesection to slope fans in the middle of the section toprograding wedge systems at the top of the section

Table 1 summarizes examples of published quan-titative morphometric data of global fluvial systemscompared with those of the Frio Formation in southTexas

The aim of applying quantitative seismic geo-morphology techniques to the study of the middleFrio fluvial systems in the Texas Gulf Coast Basin isto (1) collect key morphometric data derived from3-D seismic attribute maps which include channelwidth (CW) MBW MAH meander wavelength(ML) channel thalweg length sinuosity and pointbar length and width and (2) examine the spatialand temporal morphometric trends in the middleFrio fluvial architecture

Morphometric parameters of fluvial architec-tural elements (channels abandoned meanderloops and point bars) imaged in the study area(Figures 5ndash10) were measured The seismic at-tribute maps with spatial fluvial morphologies ofinterest were selected Each feature of interest(eg channel or point bar) in each map was tracedcarefully with a smooth line(s) or polygon(s) iden-tical with the measured feature in commercialseismic interpretation packages Then the lengthsof each of the lines andor polygons of each ar-chitectural element were measured and automaticallytranslated into numbers Each morphometric pa-rameter was measured as graphically illustrated inFigure 11

Cross plots of the 10 best-imaged channel sys-tems (Figure 6 feature 1 Figure 7 features 1 and 2Figure 8 features 1 2 and 3 Figure 9 features 1 and2 and Figure 10 features 1 and 3) that have

measurable morphology are used to assess sim-ilarities that would enable them to be categorizedinto families on the basis of their morphology(Figure 11) and to examine the relationship betweenthe different morphometric parameters The chan-nel systems are differentiated based on their geo-morphology (straight versus sinuous) seismiccharacter (low amplitude versus high amplitude)depth in the middle Frio sequence (shallow F11versus deep F39) and spatial locations of each ar-chitectural element

Quantitative Morphometric Analysis

The structure attitude of the middle Frio strata in thestudy area in south Texas is highly variable where itranges frommajor syndepositional growth fault (gt91m[300 ft] of vertical throw) deformation and associatedsediment rotation to hanging wall rollover anticlinesin addition to the preexisting topography The mainarchitectural elements of the middle Frio fluvialsystems imaged in the study area include straight tolow-sinuosity channels moderate- to high-sinuositychannels and associated point bars and abandonedmeander loops Each channel system was dividedinto segments Segments are defined as the lengthsof channels that display similarity with respect tochannel morphology or planform Dividing eachchannel system into segments is practical for segment-level analysis and comparing characteristic changesbetween different segments (Wood andMize-Spansky2009) In this study changes in the planform alongthe path of the same channel system are consideredthe characteristic feature and used to calculate seg-ment sinuosity

Several keymorphometric variables were derivedfrom 3-D seismic data including CW MBW MAHML and sinuosity (Figure 11) Because of limitationsin seismic resolution all the measurements should beconsidered asminimumRefer toWood (2007) aboutthe issues that can affect the ability of seismic data toreflect accurate measurements of the extent and di-mensions of the depositional morphology and fluvialarchitectural elements The morphometric parame-ters of these channel systems could assist in buildingaccurate geologic models for hydrocarbon productionand in reducing exploration risk in the study area insouth Texas

EL-MOWAFY aND MARFURT 549

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 14: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Table1

Exam

ples

ofPublished

QuantitativeMorphom

etric

Dataof

GlobalFluvialSystemsComparedwith

Thoseof

theFrioForm

ationinSouthTexas

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Thisstu

dyFrioForm

ationsouthTexas

3-Dseism

ic80ndash570

(262ndash1870)

023ndash2375

(014ndash148)

042ndash293

(026ndash176)

70ndash625

(230ndash2051)

105ndash18

7300ndash650

(984ndash2133)

930ndash1800

(3051ndash5906)

Nuse

etal

(2015)

CedarMountainForm

ation

Utah

Outcrops

008

(005)

15355

(445)

12

Kukulskietal

(2013)

LateJurassicndash

Early

CretaceousM

onteith

Form

ationAlbertaCanada

Wirelinelogs

and

cores

126ndash320

(413ndash1050)

0827ndash2851

(051ndash177)

Labrecqueetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

icand

wirelinelogs

500ndash584

(1640ndash1916)

24

5900 (19357)

Hubbardetal

(2011)

LCretaceous

McM

urray

AlbertaC

anada

3-Dseism

ic390ndash640

(1280ndash2100)

Gouw

and

Berendsen

(2007)

MississippiRiverchannel

beltUn

itedStates

Geom

apsand

borings

035ndash125

(022ndash078)

Wood2007

LateMiocenendashPliocene

north

ernGu

lfof

Mexico

Un

itedStates

3-Dseism

ic200ndash1800

(656ndash5906)

30ndash160

(186ndash99)

50ndash180

(31ndash1118)

500ndash5400

(1640ndash1171

7)10ndash235

Gibling(2006)

Fluvialchannelbodies

and

valleyfillsgeological

record

Seism

icwireline

logs

coresand

outcrops

lt10(33)

togt10000

(32808)

Carter(2003)

WiduriFieldJavaSea

Indonesia

3-Dseism

ic50ndash150

(164ndash492)

06ndash25

(037ndash155)

50ndash180

(164ndash591)

Zaleha

etal

(2001)

LakotaandCloverly

Form

ationsW

yoming

Wirelinelogs

and

outcrops

48ndash180

(157ndash591)

11ndash14

Reynolds

(1999)

Ancient

record

Surface

57ndash1400

(187ndash4593)

Alexanderetal

(1994)

ModernMadiso

nChannel

southw

estM

ontana

Surfaceground-

penetrating

radarandcores

50ndash100

(164ndash328)

05ndash16

(031ndash10)

012ndash04

(007ndash025)

15ndash178

(continued)

550 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 15: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Middle Frio Fluvial System Categories

Channel morphology can be related to severalfactors some of which include discharge of sedi-ment and water (Schumm 1960) sediment cali-ber (Schumm 1968) climate (Stanistreet et al1993) and river grade Schumm (1968) classifiedthe channels or rivers into three types bed loadmixed load and suspended load systems Eachhas its own range of geomorphologic charac-teristics fill and fill architecture Variations inchannel dimensions among many rivers may becaused by differences in sediment caliber with lower-sinuosity channels transporting coarser-grainedbed load material and highly sinuous streamstransporting finer-grained suspended load mate-rial Moderately sinuous streams are shown to trans-port a mixture of bed load and suspended loadmaterial

Ancient fluvial deposits preserved in the rockrecord document events of channel development andabandonment Horizontal seismic slices preservesuccessive instances in time sufficient to define theaverage pattern of a fluvial system over time (Wood2007)

Three types of channelized systems are visiblein the 3-D seismic attribute images taken fromthe study area Category 1 systems are interpreted asmeandering fluvial systems with moderate to highsinuosity large MBWs and large MAHs (Figures6 7) These systems can form extensive flood-plains with abandoned meander loops and meandercutoffs (Figures 9 10) Category 2 channel systemsare straight channels that have significantly lowersinuosity and small MAHs (Figure 8) Milliken et al(2012) conducted a study to test the scaling rela-tionships in fluvial depositional systems as related tobackwater effects They found a good correlationbetween the scales of modern fluvial systems andchannel belt scales interpreted in the ancientrecord In the middle Frio study the differencesin the scales of the channel belts range fromwidemdashcategory 1mdash on the footwall side of the AguaDulce fault (upstream) to narrowmdashcategory 2mdash onthe downthrown side of the fault (downstream)These changes may be caused by scaling relationshipdifferences of two different fluvial systems mean-dering versus braided or low sinuosity Category 3channel systems are represented by highly sinuousTa

ble1

Continued

Reference

Form

ationand

GeographicArea

DataUsed

for

Measurement

Channel

Width

(m[ft])

Meander

Belt

Width(km

[mi])

Meander

Length

(km

[mi])

Meander

ArcHeight

(m[ft])

Sinuosity

PointB

arDimensio

ns

Width(m

[ft])

Length(m

[ft])

Tylerand

Ethridge

(1983)

MorrisonC

olorado

Outcrops

100+

(328+)

20ndash100

(124ndash62)

Tye(1991)

TravisPeakeastTexas

Wirelinelogs

and

cores

48ndash96

(30ndash60)

Elliott(1976)

Exposedfluvialchannel

north

England

Outcrops

120(394)

15(93)

546

(34)

166

Busch(1974)

LittleCreekFieldMississippi

Wirelinelogs

6000

(19685)

Abbreviations3-D

5three-dimensionalL5

Lower

EL-MOWAFY aND MARFURT 551

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 16: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

and nearly anastomosing crevasse channels (Figures 910) that form narrow meander belts

Sinuosity and System Categories

The sinuosity of a channel can be defined as the ratioof channel thalweg length divided by valley length(the length measured down the axis of the meanderbelt containing the channel Figure 11) Galloway(1981) described the channel belts of the CenozoicGueydan fluvial system including the Frio For-mation of the Texas Gulf Coast from a combinationof outcrop and subsurface well log data as low-sinuosity rivers In this study the sinuosity anddirection of lateral continuity of the middle Friochannel belts are estimated from 3-D seismic at-tribute maps Sinuosity has been noted (eg Rust1977 Schumm 1981Miall 1996) to be influencedby a variety of parameters Schumm (1981) noted astrong relationship between sinuosity and dominantgrain size transported by the flow in subaerial chan-nels The sinuosity of bed load transported channelsranged from 10 to 14 the sinuosity of mixed loadchannels ranged from 14 to 20 and the sinuosity ofsuspended load channels was 20 and higher Themeasured sinuosities of the middle Frio channel sys-tems exhibit lowmoderate andhigh sinuosity ranges

The sinuosity of each of the 10 selected channelssystems (labeled in Figures 6ndash10) was calculatedand is one of the variables used to place them intothree separate categories (Figure 12) Category 1 has

moderate to high sinuosity (Figures 6 7) and is easilydifferentiated from the relatively straight category 2systems (Figure 8) The sinuosity of category 3 sys-tems (Figures 9 10) is higher than the category 1sinuosity range Because the classification of thesechannelized systems is not based on sinuosity alonethe range of sinuosity in categories 1 and 3 is some-what overlapping (Figure 12)

The average sinuosity of each of the 10 chan-nels selected for quantitative analysis is shown inFigure 12 It shows that all the data points fall within

Figure 12 Graph showing three middle Frio channel systemscategories and their sinuosities According to Schumm (1968) theboundaries of the sinuosity of bed load channel systems rangefrom 10 to 11 mixed load channels range from 14 to 17 andsuspended load channels are greater than 17 Some overlap existsbetween category 1 and category 3 channel systems

Figure 11 Zoom of thenorthwest part of the map viewin Figure 6 showing variousquantitative geomorphologicmeasures channel width mean-der arc height meander wave-length meander belt widthchannel thalweg length andpoint bar length and widthSinuosity is calculated as afunction of channel thalweglength and meander lengthRMS = root-mean-square

552 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 17: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

the range defined by the fluvial geomorphologistsas bed load mixed load and suspended load incharacter (Schumm 1968) Category 2 system sinu-osity ranges from 1 to 115 (bed load) category 1system sinuosity ranges from14 to 178 (mixed load)and category 3 system sinuosity is 164 and higher(suspended load)

Based on sequence stratigraphic reconstructionscategories 1 and 2 systems are interpreted as lowstandsystems tracts and contain most of the good qualitysandstone reservoir and exhibit sharp-based blockyand upward fining log character (eg Figure 4) Thecore-measured porosity and permeability of category1 system sandstone reservoirs range from 54 to257 and from 003 to 135 mD respectively(Figure 13) Category 3 systems were not evaluatedfor sand quality

Morphometric Measurements

Channel WidthThe fluvial CW is defined as a measure of the bank-to-bankwidthof a channel feature as indicatedby changesin seismic amplitude measured at its maximum spatialextent For example in Figure 8 we note the classicexpression of a middle Frio fluvial channel in thevertical seismic section (upper left corner) which isboth vertically and laterally confined with differentreflection strength In general themiddle Frio channelsare not well defined in the vertical seismic sectionspartly because of their shallow nature and also becauseof the limited vertical resolution capabilities of theseismic data The CWmeasurements of all the channelsystems imaged in Figures 6ndash10 are performed on theplanform of each channel as depicted in Figure 11 andare considered to represent the minimum bankfullwidth Each channel was divided into segments andeach segment is the channel distance between theapexes of two sequential meanders Within each seg-ment two width measurements were made one up-slope and one downslope (locations of two inflectionpoints shown in Figure 11) The two measurementsare combined to provide an average width of thesegment The minimum and maximum widths ofcategory 1 channel systems (Figures 6 7) range from110 to 560 m (360 to 1837 ft) category 2 channelsystems (Figure 8) range from 175 to 570 m (574 to1870 ft) and category 3 channel systems (Figures 9 10)range from 70 to 270 m (230 to 886 ft)

Meander Belt WidthTheMBW ismeasured in seismic images as the widthbetween two lines that bound outermost visiblemeander loop sets (Figure 11) and defines the con-tainer within which individual channels migrateWhen measured from seismic data it is consideredthe minimum width that might characterize thatmeander belt The minimum and maximum MBWs(a measure of the width between the minimum andmaximum deflections of the meander loops re-spectively) were measured for each segment of themiddle Frio fluvial channel systems The MBWs ofcategory 1 systems (Figures 6 7) imaged in the studyarea range from 670 to 2375 m (2198 to 7792 ft)category 2 channel belt widths (Figure 8) range from560 to 1275 m (1837 to 4183 ft) and category 3crevasse channel belt widths (Figures 9 10) rangefrom 140 to 835 m (459 to 2740 ft)

Meander WavelengthWood (2007) defined the ML as a measure of astraight line between updip-most and downdip-mostinflection points defining a single complete meander(Figure 11) The ML is related to the planform prop-erties of CW and the radius of curvature (Leopoldand Wolman 1960) In modern fluvial channels theratio of ML to CW is approximately 10 (Brice 1984)In the study area in south Texas the average middleFrio ML to CW is 10 for category 1 and 2 channelsystems (Figures 6ndash8) and 8 for the category 3 crevasse

Figure 13 Cross plot of core porosity versus permeability for thecored interval (F11ndashF15) from theWardner 184 well The plot showsthe channel fill sandstone reservoir facies exhibits good reservoirquality Depth and location of the core are shown in Figure 2

EL-MOWAFY aND MARFURT 553

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 18: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

channels systems (Figures 9 10) Reaches lacking acomplete meander were not measured The mini-mum and maximum MLs measured for the middleFrio category 1 channel systems (Figures 6 7) imagedin the study area range from 1025 to 2930 m (3363to 9613 ft) category 2 channel system wavelengths(Figure 8) range from 2240 to 2455m (7349 to 8055ft) and the lengths of the category 3 crevasse channelsystems (Figures 9 10) range from 280 to 1670 m(919 to 5479 ft)

Meander Arc HeightTheMAH ismeasured as a distance along a line drawnperpendicular to a line that bisects two inflectionpoints bounding updip and downdip limbs of a me-ander (Figure 11) The MAH could be used as ameasure of bend symmetry (Brice 1984) and thefairway within which the channel is migrating similarto the MBW (Wood and Mize-Spansky 2009) TheMAH of the category 1 channel systems ranges from285 to 625m (935 to 2051 ft) and category 2 channelsystems range from 205 to 470 m (673 to 1542 ft)The MAHs of category 3 crevasse channel systemsrange from 45 to 585 m (148 to 1919 ft)

Dimensions of Other Fluvial ArchitecturalElements

Point BarsInformation on sand body thickness and internallithofacies composition commonly comes from out-crops and subsurface wells These two sources yieldno direct information regarding the lateral continuityof the fluvial sand bodies During exploration andearly development phases well spacing generally isinadequate for accurate sand bodydelineation exceptin cases of tight well spacing during enhanced oilrecovery projects (Miall 1996) Lorenz et al (1985)suggested that given average sedimentation rates thewidth of a typical point bar would be approximatelyequivalent to the amplitude of the meanders

In the south Texas study area point bars havebeen identified in 3-D seismic attribute extractions ashigh-amplitude anomalies caused by coarse-grainedsandstone facies deposited inside meander loops Thepoint bar width is considered as equivalent to me-ander amplitude whereas the length is equivalent tothe diameter of the meander loop (Figure 11) The

width and the length of the interpreted point bar inFigure 5 are 650 and 930 m (2132 and 3051 ft)respectively The average width and length of thepoint bars in Figure 6 are 470 and 1800 m (1542 and5905 ft) whereas the width and the length of thepoint bar identified in Figure 9 (feature 7) are 300 and1200 m (984 and 3937 ft) respectively Whenprospecting in fluvial sandstone reservoirs in southTexas quantitative information on reservoir dimen-sions such as the width and the length of point barsshould be very useful for reservoir modeling and indetermining the best locations of development or infilldrilling

Floodplain DepressionsFloodplain depressions are the lowest areas of thevalley floor where water and sediment are storedduring and after overbank flow events Water andsediment are sometimes transferred to floodplaindepressions from the main river channel duringrelatively high-flow events through tie channels(Coffman et al 2010) In addition floodplain de-pressions can be associated with abandoned channels(Wilcox 1993)

The middle Frio crevasse channels and associ-ated splays (Figures 9 10) are likely formed duringflooding of the trunk channel systems that occupythe accommodation space created by the majorAgua Dulce fault Features 3 and 4 in Figure 10 areinterpreted as possible tie channel and floodplaindepression respectively The width and lengthof the interpreted floodplain depression imaged atthe F11 stratigraphic level in the northern part ofthe study area are 1667 and 3083 m (5469 and10115 ft) respectively

Comparison of Middle Frio Morphometricswith Global Fluvial Database

The morphometric parameters measured for themiddle Frio fluvial channels in the study area in southTexas are compared with some published examplesfrom the global fluvial database (Table 1) Somemiddle Friomorphometric parameters (eg sinuosityand CW) are similar or fall in the range of someavailable global examples whereas others such asMBW ML and point bar dimensions are differentGiblingrsquos (2006) compilations of the dimensions of

554 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 19: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

fluvial channel bodies from the ancient record(his table 6 p 741 based on the work of Reynolds1999) indicated that the widths of the fluvial chan-nel bodies range from 57 to 1400 m (187 to 4593 ft)In comparison the width dimensions of the Oligo-cene middle Frio fluvial channels range from 140 to2375 m (459 to 7792 ft) He also presented a re-vised classification of the channel bodies accordingto their size and form and found that the channelbodies range from very narrow ribbons less than10 m (3281 ft) to very wide sheets greater than10000 m (32810 ft) The differences in themorphometrics may be related to (1) local geology(eg riverbank lithology floodplain vegetationsediment regime supply and load and valley orriver gradient) (2) the type of the data sets usedeg outcrop versus subsurface well logs andor3-D seismic and (3) the accuracy in measurementsPrimarily 3-D seismic data are used in the case of themiddle Frio versus variable outcrop well logs and

3-D seismic data sets used in the published globalexamples and documented in Table 1

Middle Frio River Gradients

Multiple types of middle Frio channel systems (egcategories 1 and 3 in Figures 5 6 8 10) occur acrossthe study area The possible contemporaneous nat-ure of the channel systems suggests that controlson channel morphology may be a function of localvariables such as valley gradient sediment supplyand caliber In addition some of the channel systemsare preferentially oriented parallel to the direction oftectonic extension such as themeandering channel inFigure 6 that is structurally controlled and confined inthe footwall side of the Agua Dulce fault Category 3crevasse channel systems imaged in Figures 9 and10 may also be controlled by syndepositional growthfault activity

Figure 14 Seismic section from the Union Pacific Resources three-dimensional seismic survey showing steeper gradients at the deeperF39 stratigraphic level relative to shallower gradients at the F11 stratigraphic level It also shows the increase in the number of the sandbodies (black peaks) and overall thickness in the area closer to the Agua Dulce fault The interval between F11 and G2 consists of four layers(peaks) at the crest of the rollover anticline and nine layers (peaks) closer to the growth fault Possible onlap on the lower Frio G2 layer isinterpreted as evidence for sequence boundary between the lower Frio progradational deltaic sequence and the overlying middle Frioaggradational sequence Line index YY9 is shown in Figure 5 (A) Uninterpreted and (B) interpreted

EL-MOWAFY aND MARFURT 555

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 20: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Tectonic tilting associated with basin extensioninfluences the graded profile of a river leading tostream deflection and avulsion in the direction ofmaximum subsidence (Emery and Myers 1996)Leeder (1993) found that individual river channelsare highly susceptible to gradient changes caused bytectonic tilting This causes channel belt migrationincision or avulsion depending to some extent onthe magnitude of the gradients involved Leeder andAlexander (1987) noted that the form of the aban-doned meander loops within the Madison and SouthFork meander belts southwest Montana indicatesthat they were produced by gradual migration of theactive channels This progressive migration can berelated to tectonic tilting produced by regional ex-tensional faulting In the study area tilting caused bygrowth fault deformation increased accommodationspace and focused the position of channels on thehanging wall side (eg Figures 9 14) leading togreater channel deposit density and sand body in-terconnectedness (El-Mowafy and Marfurt 2008)Rotation of the middle Frio strata into the majorVicksburg and Agua Dulce growth faults super-imposes more tilting and steeper gradients of theolder lower andmiddle Frio strata Consequently thedeeper middle Frio category 2 channels (Figure 8)imaged at the F39 stratigraphic level are straight to

very low sinuosity reflecting higher river gradientsrelative to the shallower category 1 F11 channels(Figures 6 7) that exhibit moderate to high sinuosityand lower river gradients

Meander Arc Height versus Meander BeltWidth

The MAH is an important measure of a channelrsquostortuosity that affects the ability of flow to stayconfined to the channel (Wood and Mize-Spansky2009) In the study area these morphometric pa-rameters are assumed to be influenced by down-stream changes associated with local structures suchas growth faults and associated rollover anticlinesThe MAH values of the middle Frio deeper F39category 2 systems range from 205 to 470m (673 and1542 ft) These values increase to range from 285 to625m (935 to 2051 ft) for the shallower F11 category1 channels The sinuosity values of the category 2channels increase from about 106 at the deeper F39stratigraphic level to 178 for the category 1 channelsat the shallower F11 stratigraphic level

Hudson and Kesel (2000) have shown thathigh ratios of MAHMBW are associated with largemigration rates The morphometrics of the middleFrio channels in the study area (Figure 15) indicateroughly 31 and 151MBWMAH ratios exist for theF11 category 1 and F39 category 2 channel systemsrespectively These ratios indicate smaller migrationrates for the deeper F39 category 2 channel beltsrelative to the shallower F11 category 1 channel beltsThe cross plot in Figure 15 also indicates that increasingMAH directly correlates with increasing MBW al-though category 3 crevasse channels are more pre-dictable compared with other categories

Meander Belt Width versus MeanderWavelength

TheMBWdefines the extent of the areawithinwhichthe fluvial reservoir units can be deposited and lat-erally accrete (Figure 11) In seismic amplitude mapsMBW could be recognized by the maximum de-flection on both sides of high-amplitude or low-amplitude individual channels Figure 16 is a crossplot of MBW versus ML of the three categories ofchannel systems identified in the study area It shows

Figure 15 Cross plot of meander arc height versus meander beltwidth of the segments of three channel system categories It showsthat as the meander belt widths increase meander arc heightsincrease The envelope around the data points indicates a higherchance of predicting the actual channel body sizes as systemsbecome larger Note the overlap of the category 2 low-amplitudesystems (features 1 and 2 in Figure 8) and category 3 systems

556 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 21: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

the three channel families are distinct in size Channelcategories 1 and 2 show no tendency for MBWs togrow with increasing ML but the relationship isrelatively true for category 3 system Category 3systems show smallMBWs of less than 140m (459 ft)as well as small MLs of less than 280 m (918 ft)Category 1 channel systems show tight MBWs ofanywhere from 670 to 1750 m (2198 to 5742 ft)but high MLs ranging from 920 to 2930 m (3019to 9613 ft) The lower-sinuosity category 2 systemsshow narrow MBWs of 560ndash1275 m (1837ndash4183 ft)for the scale of their MLs of 2240ndash2405 m(7349ndash7890 ft) Relationships derived from dataanalyses of the middle Frio fluvial systems indicatethat in contrast to categories 1 and 2 category 3crevasse channel systems (Figures 9 10) have lessvariability in MBWs and are more predictable

Meander Arc Height versus Channel Width

TheMAH for all categories ranges from 45m (148 ft)to a maximum of 625 m (2051 ft) Widths of allchannel systems range from 70 m (230 ft) to amaximum of 570 m (1870 ft) Cross plot of MAHversus CW of the three system categories (Figure 17)

is intended to examine the empirical relationshipbetween these two variables Category 1 systems arerepresented by wide ranges of MAHs with valuesfrom 285 to 625 m (935 to 2051 ft) The CWs ofcategory 1 system range from 105 to 560 m (345 to1837 ft) The values of the MAH of category 2 sys-tems are lower and range from 205 to 470 m (673 to1542 ft) and their CWs range from 190 to 570 m(623 to 1870 ft) Category 3 systems are smallermorphologies with MAH ranges from 45 to 580 m(148 to 1903 ft) and CWs range from 70 to 270 m(230 to 886 ft) Increasing CW clearly correlates withincreasing MAH for all system categories Althoughthe three category systems show significant overlap inthe size of the MAHs they are distinct in the size oftheir widths with category 2 system having largerincisions Category 3 systems show low variations inCWs relative to categories 1 and 2

Meander Belt Width versus Channel Width

TheMBW is an important parameter for defining theextent within which a reservoir can develop TheMBW shows a large variability and defines the areawithin which the channel may migrate laterally The

Figure 17 Cross plot of meander arc height versus channelwidth of the segments of the three system categories The wid-ening envelope as the channel widths and meander arc heightsincrease indicates increasing uncertainty of predicting the actualchannel body sizes as systems become larger Although overlapexists the three systems categories fall into distinct provinces onthe plot that allows their differentiation Note that the category 2low-amplitude channel systems (features 1 and 2 in Figure 8)overlap of the category 3 low-amplitude systems

Figure 16 Cross plot of meander belt width versus meanderlength for the segments of the channel systems imaged within thestudy area Category 1 channels show a high meander belt widthand a moderate meander length indicative of larger sinuoussystems Category 2 channels show a low meander belt width anda high meander length indicative of straight to low-sinuositylarger systems Category 3 channels show a low meander beltwidth and a low meander length indicative of narrower systemsNote the partial overlap in the meander lengths of category 1 andcategory 3 channel systems

EL-MOWAFY aND MARFURT 557

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 22: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

CW is an important parameter for defining the crosssectional size of channel elements and fills (Wood andMize-Spansky 2009)

A cross plot of MBW versus CW is shown inFigure 18 The plot shows a wide scatter in the widthsof the category 1 channel systems where the datapoints cluster into two groups The shallowestchannels imaged above F11 (Figure 7) have a narrowrange (875ndash1625 m [2871ndash5331 ft]) of MBWs(cluster to the right) versus wider range (642ndash2375m[2106ndash7792 ft]) of MBWs of the channel systemsimaged at the F11 (Figure 6) stratigraphic level(cluster to the left) Changes inMBWs of category 1channels may be related to changes in substratelithology Channel belts with constant width tendto form on sandy substrate whereas channel beltswith variable widths tend to form on floodplaindeposits (Gouw and Berendsen 2007) Category 2channel systems have a narrower range (575ndash1275 m[1886ndash4183 ft]) of channel belt widths relativeto categories 1 and 3 Category 3 crevasse channelsystems are also clearly separated into two clusters(Figure 18) representing two crevasse channelcomplexes The CWs in the left cluster representthe crevasse channel complex shown in Figure 9

(features 1 and 2) and range from 70 to 130m (230 to427 ft) and MBWs range from 190 to 650 m (623 to2133 ft) The channel widths in the right clusterrepresents the crevasse channel complex shown inFigure 10 (features 1 and 3) and range from 170 to230m (558 to 755 ft) andMBWs range from 335 to835 m (1099 to 2740 ft)

Category 2 channels (Figure 8)measure from twoto three times wider than category 1 channel systems(Figures 6 7) In contrast to category 1 and 2 chan-nels category 3 crevasse channels (Figure 9 10) havevariable widths and show relatively persistent pat-terns of change Category 3 channels are generallynarrow as they progress downslope on the hangingwall side of the major Agua Dulce growth fault al-though they may widen again as they continue ba-sinward or down slope

Channel Width versus Meander Wavelength

In seismic amplitude maps ML is measured as astraight line between updip-most and downdip-mostinflection points (Figure 11) Figure 19 is a cross plotof CW versus ML of the three categories of channelsystems identified in the study area The plot showsthat channel categories 1 and 3 have limited and

Figure 18 Cross plot of me-ander belt width versus channelwidth of the segments of thethree channel system categoriesWide scatter in the widths of thecategory 1 channel systems mayreflect changes in substratesThe shallowest channels imagedabove F11 (Figure 7) have anarrow range or more or lessconstant meander belt width(cluster to the right) versus widerange of meander belt widths ofthe channel systems imaged atthe F11 (Figure 6) stratigraphiclevel Category 2 high-amplitudechannel systems have the big-gest channel widths relativeto the other two categoriesCategory 3 crevasse channelsystems are isolated into twoclusters representing two cre-vasse channel complexesimaged in Figures 9 and 10

558 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 23: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

overlapping MLs compared with category 2 Thecategory 1 channel system imaged at the F11 strati-graphic level (Figure 6) shows an increase of MLwith increasing CW where the MLs range from 928to 2928m (3045 to 9606 ft) andCWs range from107to 250 m (351 to 820 ft) Compared with thosein Figure 6 the cluster of category 1 channel systemsimaged at a shallower stratigraphic level (Figure 7)exhibits wider CWs but shorter MLs that range from343 to 562 m (1125 to 1844 ft) and from 750 to1875 m (2461 to 6152 ft) respectively Category 2channel systems have the highest channel widths andlengths ranging from175 to566m (574 to1857 ft) inwidth and from 2240 to 2452 m (7349 to 8045 ft) inlength Compared with categories 1 and 2 category 3channel systems exhibit the narrowest widths andthe shortest lengths The dimensions of the southerncrevasse channel complex shown in Figure 9 rangefrom 69 to 128 m (226 to 420 ft) in width and from279 to 1488 m (915 to 4882 ft) in length Howeverthe dimensions of northern crevasse channel compleximaged in Figure 10 are higher than those in Figure 9which range from 166 to 270 m (545 to 886 ft) inwidth and from 418 to 1666 m (1371 to 5466 ft) inlength A partial overlap exists in the dimensions ofcategory 1 and category 3 channel systems

DISCUSSION AND INTERPRETATION

Different architectural elements were recognized inthe middle Frio fluvial system and analyzed throughquantitative seismic geomorphology techniquesThese elements have different morphometrics (sinu-osity CW channel belt width meander length andMAH) and different fill type (bed load mixed loadand suspended load systems) The Gueydan fluvialsystem of the Frio formation is interpreted as havingbeen deposited by mixed load to bed load slightlysinuous streams with broad well-developed naturallevees (Galloway 1977) Eighteen fluvial systemschannel features (labeled in Figures 6ndash10) are ob-servedwithin themiddle Frio stratigraphic sequencesAs the channel systems evolved through time thenumber of the middle Frio channels increases fromthree straight to low-sinuosity channels at the deeperF39 stratigraphic level (Figure 8) to more than eightchannels at the shallower F11 stratigraphic intervals(Figures 6 9 10) The number of channels decreases

back to two channels (Figure 7) imaged in the inter-val above the F11 stratigraphic level The younger(shallower) channels aremore sinuous than the older(deeper) channels These channels change lateralpositions and exhibit different directionality Changesin the directions and lateral continuity of the middleFrio channels (Figures 6ndash10) observed in the studyarea in south Texas may be attributed to channelnodal avulsions caused by growth fault activity(Figures 9 10) and possible lateral migration overtime The highest channel segment sinuosity14ndash237 and total channel lengths 94 km (58 mi)are seen at the F11 stratigraphic level (Figures 6ndash9)versus 105ndash115 and 23 km (14 mi) at the F39 level(Figure 8) Both the F11 and the F39 intervals areapparently periods of high channel density up to 11channel features at the F11 (Figures 6 9 10) and 3 atthe F39 (Figure 8) which are interpreted as LSTs inthe study area (Figure 4)

Channel Belt Dimensions and Interpretation

The quantitative seismic geomorphology methodused for interpretation of the middle Frio fluvialchannel systems provides reasonable predictions for

Figure 19 Cross plot of channel width versus meanderwavelength of the segments of all system categories Category 1channel systems can be separated into two clusters and theyshow marked decrease of meander wavelength with decreasingchannel width Category 2 channel systems exhibit the highestchannel width and meander wavelength Category 3 channelsystems show variable narrow meander wavelengths and narrowchannel widths and they are also isolated into two clusters

EL-MOWAFY aND MARFURT 559

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 24: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

the category 1 and 2 high-amplitude sand-dominated(high netgross ratio) channel systems and higherpredictions for the category 3 low-amplitude fine-grained sandstone and siltstone-dominated crevassechannel systems The high-amplitude sand-dominatedcategory 1 and 2 channels are the main sedimenttransport conduits in the study area Identification andprediction of the sand-dominated reservoir intervals isimportant because they are typically the most prolificreservoirs and ideal targets for exploration and infilland step-out drilling in the study area

Increased attention to the dimensions of thefluvial channel belts is relevant to petroleum ex-ploration and production Based on surface fieldstudies Gouw and Berendsen (2007) indicated down-stream decrease of channel belt width along thelength of the channel belt They also found that thewidth of channel belts encased in cohesive depositsdecreases by a factor of 4 to 65 in a downstreamdirection along the length of the channel beltsHowever the width of a channel belt incised in anoncohesive substrate remains constant along theentire course These observations are related to bankerodability and stream power In the study area insouth Texas it was found that the channel beltwidths of categories 1 and 2 decrease by a factor of133 to 34 in a downstreamdirection along the lengthof the channel belts (Figures 6ndash8) and from 12 to18 of the category 3 crevasse channels encased incohesive overbank deposits (Figures 9 10) Down-stream narrowing of channels in a fluvial system is a

function of grade change and aspect ratio changesassociated with it as well as with bifurcationsSignificant narrowing of a fluvial system over a veryshort distance typically occurs after the river hasentered the backwater (T Payenberg 2015 personalcommunication) Hudson and Kesel (2000) andNittrouer et al (2012) showed the extraction ofsuspended sand fraction by net deposition mightcause channels to become narrower and deeperafter reaching the backwater length Ullah andBhattacharya (2015) identified three incised valleyfills in the downstream area in Utah that show avertical translation from fluvial to tidal facies at thetop of the valley which suggests the rivers enteredinto their backwater length at the later phase ofvalley filling To the knowledge of the authors andbased only on one core description (Kerr and Jirik1990) the basal middle Frio fluvial deposits in thesouth Texas study area do not show any tidal in-fluence However Blum et al (2013) hypothesizedthat most of the Texas coastal plain alluvial valleysare well within the range of backwater effects andthus characterized by rivers that are aggradationalavulsive and distributive in nature

Figure 20 shows an overall trend of decreasingthe width of the channel belt imaged in Figure 6 ina downstream direction The factor of decrease is29 where it decreases from 1855 to 642 m (6086 to2106 ft) Makaske et al (2007) proposed two factorsto explain the downstream changes in the channelbelt geometry the stream power and substrateerodability Decreasing stream power and or bankstability will result in decreasing the ability of theriver channel to migrate laterally The stream powerof the channel belt shown in Figure 6 may havedecreased because of the loss in the discharge inducedby the crevasse channel systems shown in Figure 9The trace or surface exposure of the AguaDulce faultmay have acted as a terrace where the channel runsbehind it This fault terrace may have been crevassedand resulted in decreasing stream power As men-tioned earlier the channel belt imaged in Figure 6 isstructurally controlled and confined in the footwallside of the Agua Dulce fault The Agua Dulce andVicksburg growth faults (Figure 5) form a half-graben structure Thus the downstream narrowingof the shallower middle Frio channel belts may berelated to the confinement experiences in this half-graben structure

Figure 20 Width of the F11 high-amplitude channel belt shownin Figure 6 plotted against downstream distance along the channelbelt axis Downstream direction is to the left The plot shows a 29factor of decrease along the meander belt length

560 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 25: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Applications to Geomodeling

In fluvial architecture models if channel belt di-mensions are held constant most likely they willoverestimate sand quantities and connectedness influvial successions We therefore propose changesin channel belt width to be incorporated in futuregeologic models to make more realistic estimates ofsand quantities in fluvial sequences in south Texasand elsewhere Therefore based on the limited da-tabase available to this study the shallower F-seriesmiddle Frio (Figures 2 6) reservoir models shouldhonor changes in the channel belt width in a down-stream direction instead of assuming constant widthalong the entire channel belt length This may over-estimate the predicted volume of fluvial sandstonereservoirs and as a result may overestimate thehydrocarbon volumes However the deeper F39 low-sinuosity channel belts (Figure 8 feature 3) exhibitmore or less constant width that may be caused bylimited lateral migration

This reflects the value of the quantitative mor-phometric data of the middle Frio fluvial channelsystems and the associated sand bodies mapped from3-D seismic data that are important to improve theinput parameters for subsurface modeling and res-ervoir prediction in the Texas Gulf Coast and insimilar sedimentary basins worldwide

SUMMARY AND CONCLUSIONS

The 3-D seismic horizon slices and window attributemaps revealed the dimension direction and spatiallocation of the Oligocene middle Frio fluvial archi-tectural elements in south Texas In the study areathe middle Frio category 1 and 2 channel belts trendin northeastndashsouthwest and eastndashnortheast to westndashsouthwest directions and category 3 crevasse channelsystems trend in eastndashwest and northwestndashsoutheastdirections Other category 1 abandoned channels andmeander loops trend in an eastndashwest direction

The middle Frio channel belts are highly variablein their morphology Application of quantitativeseismic geomorphologic techniques in the inter-pretation ofmiddle Frio fluvial systems improved ourunderstanding of reservoir development and dis-tribution in a growth fault depositional setting Thechannel belt systems imaged in the study area can be

divided into three categories on the basis of theirmorphometric characteristics Category 1 is mean-dering fluvial systems showing moderate to highsinuosity wide meander belts and larger meanderarc heights with point bars inside meander loopsThese systems are common in the shallower part ofthe middle Frio interval in the study area Category 2systems are straight to low-sinuosity channel beltswith wider and longer channels than categories 1 and3 Category 1 and 2 channel belts are interpreted tohave good quality sand content and they form onsubaerial unconformities during low accommodationtimes Category 3 systems are crevasse channels withhigh sinuosity narrowwidths smallmeander arc heightswhen compared with category 1 system and shallowincision when compared with categories 1 and 2

Multiple channelized reservoirs exist within thesame timewindow across the study area Categories 1and 2 channelized systems appear to have the highestsand content and better reservoir quality Category 3crevasse channels appear to be dominated by fine-grained deposits and as a result lower reservoirquality Category 2 channel systems appear to bedeposited by higher-gradient rivers compared withcategory 1 lower-gradient rivers

New morphometric data are introduced for themiddle Frio fluvial systems in south Texas Fluvialchannel architectural elements are measured across a254 km2 (98 mi2) area through the middle Friostratigraphic interval The morphometric data col-lected include CW MBW MAHML sinuosity andpoint bar width and length Category 3 crevassechannels exhibit lower CW MBW MAH andML than categories 1 and 2 high-amplitude sand-dominatedmain channel belt systems The shallowercategory 1 high-amplitude channel systems exhibithigher MBW and MAH than the deeper category 2high-amplitude channel systems Alternatively thedeeper category 2 channel systems exhibit higherCWs and lower sinuosities than the shallowerchannel systems that may be related to changesin valley gradients The morphometrics are cross-correlated with each other and relationships be-tween the different parameters could be assessed andprovide useful data for exploration risk assessmentand well planning

The morphometric data collected are comparedwith some published examples from the globalfluvial database Some morphometric parameters

EL-MOWAFY aND MARFURT 561

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 26: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

(eg sinuosity and channel width) are similar orfall in the range of some available global exampleswhereas other parameters such as meander beltwidthmeanderwavelength andpoint bar dimensionsare differentiated The differences in morphometricsmay be related to local bedrock geology bank resis-tance and stream power valley gradient and variationsin incision

Variable channel morphologies occur simulta-neously in the study area suggesting more influenceof local downstream controls such as syndepositionalgrowth fault activity accommodation changes in-duced by relative structure subsidence and nodalavulsions Syndepositional Agua Dulce growth faultactivity provided higher accommodation space forsediment accumulation and focusing of channels onthe hanging wall side These channels represent thedownstream part of the Gueydan fluvial system thatfed the sand into these systems

The quantitative morphometric data of themiddle Frio fluvial systems and the associated sandbodies are important to improve the database forgeologic and reservoir modeling and for petroleumexploration and production along the Texas GulfCoast Also the variations in the middle Frio fluvialchannel style and scale should be used against mak-ing simplistic assumptions about the uniformity ofthe quantitative parameters during architecturalreconstructions and reservoir modeling

REFERENCES CITED

Alexander J J S Bridge M R Leeder R E Collier andR L Gawthorpe 1994 Holocene meander-belt evolutionin an active extensional basin southwestern MontanaJournal of Sedimentary Research v B64 no 4 p 542ndash559

Blum M J Martin K Milliken and M Garvin 2013 Pa-leovalley systems Insights from Quaternary analogs andexperiments Earth-Science Reviews v 116 p 128ndash169doi101016jearscirev201209003

Brice J C 1984 Planformproperties ofmeandering rivers inC M Elliott ed River Meandering Proceedings ofRivers rsquo83 American Society of Civil Engineers NewOrleans Louisiana October 24ndash26 1983 p 843ndash856

Busch D A ed 1974 Stratigraphic traps in sandstonesmdashExploration techniques AAPG Memoir 21 174 p

Carter D C 2003 3-D seismic geomorphology Insights intofluvial reservoir deposition and performance Widurifield Java Sea AAPG Bulletin v 87 no 6 p 909ndash934

Chopra S and K J Marfurt 2007 Seismic attributes forprospect identification and reservoir characterization

Tulsa Oklahoma Society of Exploration GeophysicistsGeophysical Developments Series 11 464 p doi10119019781560801900

Coffman D K G Malstaff and F T Heitmuller 2010Characterization of geomorphic units in the alluvialvalleys and channels of Gulf Coastal Plain rivers in Texaswith examples from the Brazos Sabine and Trinityrivers US Geological Survey Scientific InvestigationsReport 2011ndash5067 42 p

Davies R J H W Posamentier L J Wood andJ A Cartwright 2007 Seismic geomorphology Appli-cations to hydrocarbon exploration and productionGeological Society London Special Publications 2007v 277 274 p

Elliott T 1976 The morphology magnitude and regime of aCarboniferous fluvial distributary channel Journal ofSedimentary Petrology v 46 no 1 p 70ndash76

El-Mowafy H Z and K J Marfurt 2008 Structural in-terpretation of the middle Frio Formation using 3-Dseismic and well logs An example from the Texas GulfCoast of the United States Leading Edge v 27p 840ndash854 doi10119012954023

EmeryD andK JMyers eds 1996 Sequence stratigraphyOxford United Kingdom Blackwell Science 297 p doi1010029781444313710

Friend P F 1983 Towards the field classification of alluvialarchitecture or sequence in J D Collinson andJ L Lewin eds Modern and ancient fluvial systemsInternational Association of Sedimentologists SpecialPublication 6 p 345ndash354

Friend P F M J Slater and R C Williams 1979 Verticaland lateral building of river channels Ebro Basin SpainJournal of the Geological Society v 136 p 39ndash46 doi101144gsjgs13610039

Galloway W E 1977 Catahoula Formation of the Texascoastal plain The University of Texas at Austin Bureauof Economic Geology Report of Investigations 100 81 p

GallowayW E 1981 Depositional architecture of CenozoicGulf Coastal Plain fluvial systems Tulsa OklahomaSEPM Special Publication 31 p 127ndash155

Galloway W E 1989 Genetic stratigraphic sequences inbasin analysis II Application to northwestGulf ofMexicoCenozoic basin AAPGBulletin v 73 no 2 p 143ndash154

GallowayW E andD KHobday 1996 Terrigenous clasticdepositional systems Applications to fossil fuel andgroundwater resources New York Springer-Verlag 489p doi101007978-3-642-61018-9

Galloway W E D K Hobday and K Magara 1982a FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon dis-tribution AAPG Bulletin v 6 no 6 p 649ndash688

Galloway W E D K Hobday and K Magara 1982b FrioFormation of the Texas coastal plain Depositional sys-tems structural framework and hydrocarbon originmigration distribution and exploration potential TheUniversity of Texas at Austin Bureau of EconomicGeology Report of Investigation 122 78 p

Gibling M R 2006 Width and thickness of fluvial channelbodies and valley fills in the geological record A literature

562 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 27: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

compilation and classification Journal of SedimentaryResearch v 76 p 731ndash770 doi102110jsr2006060

Gouw M J and H J Berendsen 2007 Variability ofchannel-belt dimensions and the consequences for allu-vial architectureObservations from theHoloceneRhine-Meuse Delta (The Netherlands) and Lower MississippiValley (USA) Journal of Sedimentary Research v 77p 124ndash138 doi102110jsr2007013

Hammes U H Zeng L F Brown R Loucks andP Montoya 2005 Seismic geomorphology of OligoceneFrio lowstand slope and basin floor sedimentary bodies ingrowth-faulted subbasins in South Texas Gulf CoastAssociation of Geological Societies Transactions v 55p 278ndash282

Hardage B A R Edson R A Levey V Pendelton andJ Simmons 1994 A 3D seismic case history evaluatingfluvially deposited thin-bed reservoirs in a gas-producingproperty Geophysics v 59 p 1650ndash1665 doi10119011443554

Holbrook J R W Scott and F E Oboh-Ikuenobe 2006Base-level buffers and buttresses A model for upstreamversus downstream control on fluvial geometry and ar-chitecture within sequences Journal of SedimentaryResearch v 76 p 162ndash174 doi102110jsr200510

Hubbard S M D G Smith H Nielsen D A LeckieM Fustic R J Spencer and L Bloom 2011 Seismicgeomorphology and sedimentology of a tidally influencedriver deposit Lower Cretaceous Athabasca oil sandsAlbertaCanadaAAPGBulletin v 95 no 7 p 1123ndash1145doi10130612131010111

Hudson P F and R H Kesel 2000 Channel migration andmeander-bend curvature in the Mississippi River prior tomajor humanmodificationGeology v 28 p 531ndash534 doi1011300091-7613(2000)28lt531CMAMCIgt20CO2

Kerr D R 1990 Reservoir heterogeneity in the middle FrioFormation Case studies in Stratton and Agua Dulcefields Nueces County Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 363ndash372

Kerr D R and L A Jirik 1990 Fluvial architecture andreservoir compartmentalization of the Oligocene middleFrio Formation south Texas Gulf Coast Association ofGeological Societies Transactions v 40 p 373ndash380

Kosters E C D G Bebout L F Brown S P DuttonR J Finley C M Garrett H S Hamlin S C RuppelS J Seni and N Tyler 1989 Atlas of major Texas gasreservoirs Austin Texas The University of Texas atAustin Bureau of Economic Geology Special Pub-lication 161 p

Kukulski R B S M Hubbard T F Moslow andM K Raines 2013 Basin-scale stratigraphic architectureof upstream fluvial deposits Jurassic-Cretaceous fore-deep Alberta Basin Canada Journal of SedimentaryResearch v 83 p 704ndash722 doi102110jsr201353

Labrecque P A SM Hubbard J L Jensen andH Nielsen2011 Sedimentology and stratigraphic architecture of apoint bar deposit Lower Cretaceous McMurray For-mation Alberta Canada Bulletin ofCanadian PetroleumGeology v 59 no 2 p 147ndash171 doi102113gscpgbull592147

Leeder M R 1993 Tectonic controls upon drainage basindevelopment river channel migration and alluvial ar-chitecture Implications for hydrocarbon reservoir de-velopment and characterization in C P North andD J Prosser eds Characterization of fluvial and aeolianreservoirs Geological Society London Special Pub-lications 1993 v 73 p 7ndash22 doi101144GSLSP19930730102

Leeder M R and J Alexander 1987 The origin and tectonicsignificance of asymmetric meander belts Sedimentologyv34p217ndash226doi101111j1365-30911987tb00772x

Leopold L B and W G Wolman 1960 River meandersGeological Society of America Bulletin v 71 p769ndash794 doi1011300016-7606(1960)71[769RM]20CO2

Lorenz J C D M Heinze J A Clark and C A Searls1985 Determination of widths of meander-belt sand-stone reservoirs from vertical downhole data MesaverdeGroup Pieceance Creek Basin Colorado AAPG Bulle-tin v 69 no 5 p 710ndash721

Makaske B H J A Berendsen andMHMVanRee 2007Middle Holocene avulsion-belt deposits in the centralRhinendashMeuse Delta The Netherlands Journal of Sedi-mentary Research v 77 p 110ndash123 doi102110jsr2007004

Miall A D 1996 The geology of fluvial deposits Sedi-mentary facies basin analysis and petroleum geologyHeidelberg Germany Springer-Verlag 582 p

Miall A D 2002 Architecture and sequence stratigraphy ofPleistocene fluvial systems in the Malay Basin based onseismic time-slice analysis AAPG Bulletin v 86 no 7p 1201ndash1216

MiallAD 2014 Fluvial depositional systems Berlin Springer-Verlag 316 p doi101007978-3-319-00666-6

Miall A D 2015 Modern chronostratigraphic data dem-onstrate that currently popular sequence models forfluvial systems donrsquot work Canadian Society of Petro-leum Geologists Geoconvention 2015 New HorizonsCalgary Alberta Canada May 4 2015 5 p

Milliken K M Blum and J Martin 2012 Scaling rela-tionships in fluvial depositional systems Search andDiscovery article 30245 accessed December 2 2015httpwwwsearchanddiscoverycomdocuments201230245millikenndx_millikenpdf

Nittrouer J A J Shaw M P Lamb and D Mohrig 2012Spatial and temporal trends for water-flow velocity andbed-material sediment transport in the lower MississippiRiver Geological Society of America Bulletin v 124 no3ndash4 p 400ndash414 doi101130B304971

Nuse B D Pyles and K Kirschbaum 2015 Associatedsedimentation styles in a fluvial channel belt Three-dimensional outcrop study of the Cedar Mountain For-mation Utah (abs) AAPG Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed March 14 2016 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102706html

Posamentier H W 2002 Ancient shelf ridgesmdashA poten-tially significant component of transgressive systems tract

EL-MOWAFY aND MARFURT 563

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems

Page 28: Quantitative seismic geomorphologyofthemiddleFrio ...mcee.ou.edu/aaspi/publications/2016/Hamed_M2016.pdf · Quantitative seismic geomorphologyofthemiddleFrio fluvial systems, south

Case study from offshore northwest Java AAPG Bul-letin v 86 no 1 p 75ndash96

Posamentier HW R Davies L J Wood and J Cartwright2007 Seismic geomorphologymdashAnoverview inRDaviesH W Posamentier L J Wood and J Cartwright edsSeismic geomorphology Application to hydrocarbon ex-ploration and production Geological Society LondonSpecial Publications 2007 vol 277 p 1ndash20

Reynolds A D 1999 Dimensions of paralic sandstonebodies AAPG Bulletin v 83 no 2 p 211ndash229

Rust B R 1977 A classification of alluvial channel systemsFluvial sedimentology DallasGeological SocietyMemoir5 p 187ndash198

Ryseth A H Fjellbirkeland I K Osmundsen A Skalnesand E Zachariassen 1998 High-resolution stratigraphyand seismic attribute mapping of a fluvial reservoirMiddle Jurassic Ness Formation Oseberg Field AAPGBulletin v 82 no 9 p 1627ndash1651

Schumm S A 1960 The effect of sediment type on theshape and stratification of some modern fluvial depositsAmerican Journal of Science v 258 p 177ndash184 doi102475ajs2583177

Schumm S A 1968 Speculations concerning paleohydrauliccontrols on terrestrial sedimentation Geological Society ofAmerica Bulletin v 79 p 1573ndash1588 doi1011300016-7606(1968)79[1573SCPCOT]20CO2

Schumm S A 1981 Experimental fluvial geomorphologyNew York Wiley and Sons 376 p

Shanley KW and P J McCabe 1993 Alluvial architecturein a sequence stratigraphic framework A case historyfrom the Upper Cretaceous of southern Utah USA inS S Flint and I D Bryant eds The geological modelingof hydrocarbon reservoirs and outcrop analogues Inter-national Association of Sedimentologists Special Pub-lication 15 p 21ndash56

Stanistreet I G B Cairncross and T S McCarthy 1993Low sinuosity and meandering bedload rivers of theOkavango Fan Channel confinement by vegetated leveeswithout fine sediment Sedimentary Geology v 85p 135ndash156 doi1010160037-0738(93)90079-K

Thies K J B E Brown R N Rosen B L Shaffer andJ A Adamick 1993 Sequence stratigraphy of the upperand middle Frio Orange County Texas Gulf CoastAssociation of Geological Societies Transactions v 43p 413ndash419

Tye R S 1991 Fluvial sandstone reservoirs of the TravisPeak Formation East Texas basin in A D Miall andN Tyler eds The three-dimensional facies architecture

of terrigenous clastic sediments and its implications forhydrocarbon discovery and recovery SEPM Concepts inSedimentology and Paleontology v 3 p 172ndash188

Tye R S J P Bhattacharya J A Lorsong S T SindelarDGKnockDDPuls andRALevinson 1999Geologyand stratigraphy of fluvio-deltaic deposits in the IvishakFormation applications for development of PrudhoeBay Field Alaska AAPG Bulletin v 83 no 10p 1588ndash1623

Tyler N and F G Ethridge 1983 Fluvial architecture ofJurassic uranium-bearing sandstones Colorado Plateauwestern United States in J D Collinson and J Lewineds Modern and ancient fluvial systems InternationalAssociation of Sedimentology Special Publication 6p 533ndash547 doi1010029781444303773ch42

Ullah M S and J P Bhattacharya 2015 Interpretingbackwater effects on fluvial style and architecture in ahigh-gradient compound incised-valley deposits Exam-ple from Cretaceous Ferron Notom Delta southeasternUtah (abs) AAPG 2015 Annual Convention andExhibition Denver Colorado May 31ndashJune 3 2015accessed December 2 2015 httpwwwsearchanddiscoverycomabstractshtml201590216aceabstracts2102362html

Wilcox D B 1993 An aquatic habitat classification systemfor the Upper Mississippi River system US Fish andWildlife Service Long Term Resource Monitoring Pro-gram Technical Report 93-T003 31 p

Wood L J 2007 Quantitative seismic geomorphology ofPliocene and Miocene fluvial systems in the northernGulf of Mexico Journal of Sedimentary Research v 77p 713ndash730 doi102110jsr2007068

Wood L J and K L Mize-Spansky 2009 Quantitativeseismic geomorphology of a Quaternary leveed-channelsystem offshore eastern Trinidad and Tobago north-eastern South America AAPG Bulletin v 93 no 1p 101ndash125 doi10130608140807094

Wright V P and S B Marriott 1993 The sequencestratigraphy of fluvial depositional systems The role offloodplain sediment storage SedimentaryGeology v 86p 203ndash210 doi1010160037-0738(93)90022-W

Zaleha M J J W Nathan and L J Suttner 2001 Effects ofsyndepositional faulting and folding on early Cretaceousrivers and alluvial architecture (Lakota and CloverlyFormations Wyoming USA) Journal of SedimentaryResearch Section B Stratigraphy and Global Studiesv 71 no 6 p 880ndash894

564 Quantitative Seismic Geomorphology of the Middle Frio Fluvial Systems