Geological Society of America Bulletin › wp-content › uploads › 2019 › 10 › Helle… · Geological Society of America Bulletin, September 2003 1123 TILTING OF THE U.S. CORDILLERA

Geological Society of America Bulletin

doi: 10.1130/B25219.1 2003;115;1122-1132Geological Society of America Bulletin

Paul L. Heller, Kenneth Dueker and Margaret E. McMillan U.S. CordilleraPost-Paleozoic alluvial gravel transport as evidence of continental tilting in the

Email alerting servicescite this article

to receive free e-mail alerts when new articleswww.gsapubs.org/cgi/alertsclick

SubscribeAmerica Bulletin

to subscribe to Geological Society ofwww.gsapubs.org/subscriptions/click

Permission request to contact GSAhttp://www.geosociety.org/pubs/copyrt.htm#gsaclick

viewpoint. Opinions presented in this publication do not reflect official positions of the Society.positions by scientists worldwide, regardless of their race, citizenship, gender, religion, or politicalarticle's full citation. GSA provides this and other forums for the presentation of diverse opinions and articles on their own or their organization's Web site providing the posting includes a reference to thescience. This file may not be posted to any Web site, but authors may post the abstracts only of their unlimited copies of items in GSA's journals for noncommercial use in classrooms to further education andto use a single figure, a single table, and/or a brief paragraph of text in subsequent works and to make

GSA,employment. Individual scientists are hereby granted permission, without fees or further requests to Copyright not claimed on content prepared wholly by U.S. government employees within scope of their

Notes

Geological Society of America

on January 26, 2010gsabulletin.gsapubs.orgDownloaded from

http://gsabulletin.gsapubs.org//cgi/alerts

http://gsabulletin.gsapubs.org//subscriptions/index.ac.dtl

http://www.geosociety.org/pubs/copyrt.htm#gsa

http://gsabulletin.gsapubs.org/

For permission to copy, contact [email protected] 2003 Geological Society of America1122

GSA Bulletin; September 2003; v. 115; no. 9; p. 1122–1132; 8 figures; Data Repository item 2003121.

Post-Paleozoic alluvial gravel transport as evidence of continentaltilting in the U.S. Cordillera

Paul L. Heller†

Kenneth DuekerMargaret E. McMillan‡

Department of Geology and Geophysics, University of Wyoming, Laramie, Wyoming 82071, USA

ABSTRACT

The western United States contains threethin but remarkably widespread alluvialconglomeratic units that record episodes oflarge-scale tilting across the U.S. Cordille-ran orogen in post-Paleozoic time. Theseunits are: (1) the Shinarump Conglomerateof Late Triassic age exposed in northernArizona and adjacent parts of Utah, Neva-da, and New Mexico; (2) Lower Cretaceousgravel deposits that overlie the MorrisonFormation throughout the Rocky Mountainregion; and (3) the gravel-rich parts of theMiocene-Pliocene age Ogallala Group inwestern Nebraska and adjacent southeast-ern Wyoming. Paleoslopes of the rivers de-positing these units were in the range of1024 to 1023, based on paleohydraulic cal-culations. However, depositional thicknesstrends of these units are not sufficient tohave generated such steep paleoslopes.Thus, long wavelength tilting of the Earth’ssurface must have occurred for these grav-els to be transported. Although these unitswere deposited adjacent to large tectonicfeatures, including an evolving and migrat-ing continental arc, and, for the OgallalaGroup, the northward-propagating RioGrande Rift, the tilting occurred overwavelengths too broad to be directly gen-erated by these features. These widespreadgravel units attest to the interplay betweenthe creation of subduction-related isostaticand dynamic topography and continentalsedimentation. Hence, paleotopography, asdetermined from calculated transport gra-dients of sedimentary deposits, provides a

†E-mail: [email protected].‡Present address: University of Arkansas at Little

Rock, Department of Earth Sciences, Little Rock,Arkansas 72204, USA.

means of relating constructional landformsto mantle-driven processes.

Keywords: paleohydraulics, tectonics, dy-namic topography, U.S. Cordillera, con-glomerate, paleotopography, syntectonicsedimentation.

INTRODUCTION

A common characteristic of synorogenic al-luvial conglomerates is the relatively limiteddistance they prograde from their upliftedsource areas out into the adjacent basins—most are found within a few tens of kilometersof their associated mountain fronts (Fig. 1, A–G). The restricted distance of gravel deposi-tion reflects the long-term balance between therate of delivery of coarse sediment supply andbasin subsidence rate that acts to trap the grav-el. Since on an elastic plate both supply andsubsidence are proportional to the size of ad-jacent mountain belts, it is no surprise that ba-sin sedimentation patterns have a similarlength scale.

In contrast, a few conglomeratic units seenin the U.S. Cordillera are unusually wide-spread in distribution (Stokes, 1950; Stewartet al., 1972a; Heller and Paola, 1989), beingdeposited over many hundreds of kilometersdownstream from their source areas (Fig. 1),yet being relatively thin over their entire ex-tent. These gravels are related to the temporaland spatial pattern of subduction and late Ce-nozoic uplift of the central and southernRocky Mountains. We argue that the occur-rence of these thin, widespread gravel unitsdoes not reflect climatic controls, but must re-cord times of large-wavelength/low-amplitudetilting of the continental interior. Thus, we be-lieve they are the stratigraphic records of deepprocesses affecting the western U.S. duringpost-Paleozoic time.

THREE WIDESPREADCONGLOMERATES

The three widespread gravel units of theU.S. Cordillera are the Shinarump Conglom-erate Member of the Chinle Formation (LateTriassic) found in the southwestern USA (Fig.2A); the Lower Cretaceous conglomeratesfound over much of the U.S. Rocky Moun-tains (Fig. 2B); and gravelly parts of the Ogal-lala Group (Miocene–Pliocene) exposed alongthe western Great Plains (Fig. 2C).1

Shinarump Conglomerate

The Shinarump Conglomerate lies at thebase of the Chinle Formation, a fluvial andlacustrine unit of Late Triassic age (Carnian;Lucas, 1993) exposed in the southwesternUSA (Stewart et al., 1972a). Rivers of theChinle Formation flowed generally north fromthe ‘‘Mogollon highlands,’’ an interpreted re-gional uplands in southern Arizona and adja-cent regions (Stewart et al., 1972a; Blakey andGubitosa, 1983), eventually joining majortributaries from farther east. Together theyflowed northwest to the contemporaneousshoreline (Riggs et al., 1996). While the Shi-narump Conglomerate is dominantly sand-stone, conglomerate is common. Unit thick-ness varies, but it is typically a few metersthick. Gravel composition includes abundantquartz, quartzite, and chert grains (Blakey andGubitosa, 1983). The Shinarump Conglomer-ate forms an irregular sheet that rests erosion-ally upon older Triassic and Permian units(Blakey and Gubitosa, 1983) and is overlainby bentonitic shales and mudstones of theChinle Formation (Stewart et al., 1972a). Lo-

1Color photos and figures to accompany this pa-per can be found at: http://faculty.gg.uwyo.edu/hell-er/pubs.htm.



Geological Society of America Bulletin, September 2003 1123

TILTING OF THE U.S. CORDILLERA

Figure 1. Reported distances of gravel de-position (measured normal to orogenicfront) with respect to their orogenic sourceareas. Examples A–G are from ancientforeland basins (Heller and Paola, 1989);these are, in order, the Alps, Andes, Ap-ennines, Himalayas, Pyrenees, CatalanCoast Range, and Sevier belt. The three un-usually widespread alluvial gravels in theU.S. Cordillera are also shown.

Figure 2. Maps of three gravel units: (A) Late Triassic time showing Shinarump Con-glomerate Member of Chinle Formation (Stewart et al., 1972a; Blakey and Gubitosa, 1983;Dubiel and Brown, 1993); (B) Early Cretaceous time showing Cloverly Formation (inWyoming) and its lithostratigraphic correlatives, including Buckhorn Member of CedarMountain Formation (in Utah), Lytle and Burro Mountain Formations (in Colorado),Lakota Formation (in South Dakota), and lower member of Ephraim Conglomerate (inIdaho and western Wyoming) (Heller and Paola, 1989); and (C) Miocene time showingOgallala Formation of western Great Plains (Swinehart et al., 1985; Gustavson and Wink-ler, 1988). Maps include observed (black) and inferred (shaded) original distribution ofunits, generalized paleoflow trends of units (white arrows); line of reconstructed topog-raphy (Fig. 7) is shown as white line with round end points, and associated tectonic fea-tures are discussed in text (Baldridge et al., 1991; Armstrong and Ward, 1993; Chapinand Cather, 1994; Erslev, 2001). In Part B, heavy dashed line shows subsequent site ofthrusting in Sevier belt. Generalized extension orientation of Rio Grande Rift is shownwith double arrows and its interpreted northward extent (Tweto, 1979; Mears, 1993; Er-slev, 2001) shown with dashed lines.

Figure 3. Example of widespread conglomeratic unit, in this case Lytle Formation (partof Lower Cretaceous gravels) north of Fort Collins, Colorado. This unit shows most typicalcharacteristics of gravel units; it is conglomerate and sandstone, which abruptly overliesmuch finer-grained sedimentary deposits, in this case fluvial rocks of Late Jurassic Mor-rison Formation (A). Unit has sharp, disconformable contact with underlying unit (B). Inmost places, unit is a single, sheet-like deposit that pinches and swells (C), but there isevidence of amalgamation of several channel-belt bodies (D), enlarged below with super-imposed line drawing. Resistant deposits typically form a cliff-capping unit that can betraced over many kilometers (E). Cars at base of cliff show scale. At D (inset), unit consistsof three amalgamated sand/gravel bodies, of which upper two interfinger with finer-grained overbank deposits.

cal relief of up to 80 m along the basal un-conformity suggests that the conglomeratefilled paleovalleys, especially in its more up-stream parts (Stewart et al., 1972a; Blakey andGubitosa, 1983). Valley development dimin-ishes downstream (northward), indicating thatpre-Shinarump valley cutting was not likelydue to sea-level change. Toward the north theunit is inferred to record deposition in a broadalluvial plain, and the preservation of mottledfacies (Stewart et al., 1972a) is interpreted torepresent soil formation in low-lying inter-fluves (Hasiotis et al., 2000).

Deposition of the Shinarump Conglomeratewas penecontemporaneous with the onset ofintrusive igneous activity in southeastern Cal-ifornia and southern Arizona (Armstrong andWard, 1993). The resulting Upper Triassic plu-tons are interpreted to represent the begin-nings of Andean-type arc magmatism in thesouthwestern USA (Armstrong and Ward,1993; Dickinson, 2000). The coincidence intiming of Shinarump deposition, at the base ofthe ash-rich Chinle Formation, and the onsetof arc magmatism suggests to us that theseevents are linked.

Lower Cretaceous Conglomerates of theRocky Mountains

Discontinuously exposed throughout theU.S. Rocky Mountains are conglomeratic flu-vial rocks of Early Cretaceous age (Figs. 2Band 3). These units are called a variety ofnames, including the Buckhorn Member of theCedar Mountain Formation in central Utah,the Burro Canyon and Lytle formations inColorado, the Cloverly Formation in Wyo-



1124 Geological Society of America Bulletin, September 2003

HELLER et al.

Figure 4. Flexural analysis of Lower Creta-ceous gravels. Shown are filled sedimentarybasin geometries required to generate back-bulge transport slopes of 1023, assuming anunbroken plate with flexural rigidities (D) of1023 to 1024 Nm and for various load geome-tries (height and length). Width and depth offilled foreland basins associated with theseback bulge slopes range from 200 km and 2.7km deep to 400 km and 5.2 km deep, de-pending upon value of D.

ming, and the Lakota Formation in the BlackHills. All of these units appear to sit discon-formably upon the Morrison Formation (Fig.3B), a widespread, mud-rich alluvial unit ofLate Jurassic age (Stokes, 1944; Kowallis andHeaton, 1987). The duration and significanceof the bounding unconformity has been con-tested (Furer, 1970; Winslow and Heller, 1987;May et al., 1995) and may represent only alithofacies change in otherwise continuous de-position. The Lower Cretaceous conglomer-ates have been correlated to each other on thebasis of composition, lithostratigraphy, grain-size trends, and general consistency of limitedage dates (Heller and Paola, 1989). Sometime-transgressive deposition of the conglom-erate has been suggested based upon strati-graphic interpretations (Currie, 1997). Gravel-clast composition is primarily chert pebbles.Lithofacies interpretations suggest a low-sinuosity, braided origin (Heller and Paola,1989; Zaleha et al., 2001).

The gravel body consists of amalgamatedchannels that locally include some finer-grained alluvial deposits (e.g., Fig. 3D; Mayet al., 1995). Overall, paleocurrent trends in-dicate flow toward the northeast (Heller andPaola, 1989), although some notable complex-ities are found locally, particularly in theBlack Hills region (Zaleha et al., 2001). Agecontrol is limited, but available biostratigraph-ic control suggests deposition during middleNeocomian to Aptian time (Tschudy et al.,1984; Heller and Paola, 1989).

Most accounts interpret the gravels as hav-ing been deposited prior to, and perhaps im-mediately prior to, the earliest deformationalong the Sevier fold-thrust belt (Heller andPaola, 1989; Yingling and Heller, 1992; Cur-rie, 1998). One interpretation suggests depo-sition in an overfilled earlier foreland basinthat formed in front of thrust faults in centralNevada (DeCelles and Currie, 1996; Currie,1998); however, the ages of thrust faults inthat region are now thought to be Early tomiddle Cretaceous (MacCready et al., 1997),too young to generate an early foreland basin.In addition, if the gravels formed in a back-bulge basin setting, as suggested by Currie(1998) and DeCelles and Currie (1996), thena thick sequence of nonmarine foreland basinfill must have existed between the farthestwest exposures of gravel and the inferredthrust belt that no longer exists. Consideringthat such a basin must have been very thick,in excess of 2 km thick, to develop a backbulge slope of 1023 that is required by paleo-hydraulic analysis (Heller and Paola, 1989;Fig. 4), its complete absence is even more dif-ficult to understand. We suggest that no such

‘‘phantom foredeep’’ existed (Royse, 1993),but rather the region was simply dipping gent-ly down toward the east.

Tectonic activity in the western U.S. duringearly Early Cretaceous time (Neocomian) wasrelatively quiescent (Dickinson, 2001). Thereis a notable paucity of sedimentary accumu-lation from that time until the initiation of theSevier orogenic belt in Aptian/Albian time(Heller et al., 1986). The U.S. Cordillerancontinental arc attained its maximum north-south extent in Jurassic time (Armstrong andWard, 1993). During Late Jurassic time, thearc migrated eastward into westernmost Utahand had begun to retract westward during Ear-ly Cretaceous time. Apparently little else ofregional tectonic significance took place, atleast within or east of the volcanic arc, duringthe Early Cretaceous. Thus, the gravel was de-posited during continued growth of the con-tinental arc.

Ogallala Group

The Ogallala Group is part of a late Tertiarybasin-fill sequence deposited within centralRocky Mountain basins and on the westernGreat Plains (Fig. 2C). In western Nebraska,geologists recognize several formations withinthe Ogallala Group, but lithologic and geo-

metric complexities make broader correlationvery difficult (Swinehart et al., 1985). Thegroup is composed of predominantly fine-grained amalgamated fluvial and eolian de-posits with dispersed interbedded conglomer-ates. Deposition began in the middle Mioceneand continued until the earliest Pliocene (19–5 Ma) based on K/Ar and fission-track dating(Izett, 1975; Naeser et al., 1980) and biostrati-graphic controls (Reeves, 1984; Swinehart etal., 1985). The Ogallala Group sits discon-formably upon the Arikaree and White Rivergroups in its northern exposure and uncon-formably on Mesozoic and Paleozoic rocks tothe south. The unit marks the regional transi-tion from a period of nearly continuous ag-gradation to one of net degradation (Diffendal,1982; Swinehart et al., 1985).

Fluvial conglomeratic deposits of the Ogal-lala Group are remarkable in that they havebeen transported over 250 km from theirsource areas. These deposits form irregularcut-and-fill sequences that are most continu-ous in the Ash Hollow Formation along theCheyenne Tablelands in eastern Wyoming andwest-central Nebraska. They were deposited byeast-northeast-flowing, low-sinuosity braidedstreams draining the Rocky Mountain front(Goodwin and Diffendal, 1987). To the south,conglomerates are less evident at the surface,but they are found in the subsurface (Gustav-son and Winkler, 1988). Gravel compositionsinclude granite and anorthosite clasts from theFront and Laramie ranges and volcanic clastsfrom flows in north-central Colorado (Black-stone, 1975). Relief on the base of the unit isup to 100 m, reflecting paleovalleys in the up-stream end and local post-depositional subsi-dence due to salt dissolution in underlyingMesozoic units (Gustavson and Winkler,1988; Oldham, 1996). The cut-and-fill alter-nations may be the result of climate fluctua-tions, autocyclic events, or tectonic uplift inthe Rocky Mountains during the lateCenozoic.

Most interpretations of the Ogallala Groupsuggest that it records uplift associated withthe propagating Rio Grande Rift (Trimble,1980; Eaton, 1986). Deposition of the Ogal-lala Group occurred at the same time as de-formation along the rift that was initiated inNew Mexico between 28 and 27 Ma, reachedcentral Colorado by 26–25 Ma (Chapin andCather, 1994), and is manifested as 10–8 Mavolcanic rocks, post-Miocene normal faulting,and high heat-flow on the Colorado–Wyomingborder (Keller and Baldridge, 1999; Mears,1998). The link between the two is further sup-ported by paleohydraulic analysis of Ogallalafluvial channels on the Cheyenne Tablelands.





These channels have been post-depositionallytilted up by an order of magnitude, most likelyby tectonic uplift in the Rocky Mountains(McMillan et al., 2002). Thus, the OgallalaGroup seems to be an indicator of long-wave-length tectonic activity both during and afterits deposition.

Shared Characteristics

These three fluvial conglomerate units shareseveral key features in common. While ex-posures today are discontinuous, paleogeo-graphic reconstructions augmented by welldata suggest that these units were far morecontinuous in the past and represent through-going alluvial systems. The Shinarump Con-glomerate and Ogallala Group are dominantlysandstone, but gravels are found in locallyabundant patches over long down-flow dis-tances. The Lower Cretaceous units are morestrongly conglomeratic. We think of these asdiscontinuous gravel sheets; however, we notethat: (1) the dispersed presence of low-reliefinterfluves and occasional intercalated over-bank deposits demonstrate that they formed asan amalgam of individual channel belts (Fig.3D), and (2) there are lateral variations ingrain size, particularly for the ShinarumpConglomerate and Ogallala Group, that dem-onstrate that a variety of individual river chan-nel systems were active intermittently in spaceand time across the region. Most remarkably,the units are thin, typically a few meters thick,over nearly their entire lateral extent (Fig. 3).However, in their upstream areas, they cut andthen fill paleovalleys into underlying deposits.The Shinarump Conglomerate contains manysuch paleovalleys toward the south that cutdown into the Moenkopi Formation (Triassic)and the DeChelley Sandstone (Permian).These range up to a few tens of meters deepand up to 5 km wide, but northward the unitis more continuous, typically 10 m thick (upto 20 m in places) (Stewart et al., 1972a). TheLower Cretaceous gravels in Utah reach up to35 m thick along trunk stream paleovalleyscut into the underlying Morrison Formationbut are typically more consistent (;10 m) inthickness (Heller and Paola, 1989; Yinglingand Heller, 1992; Currie, 1998). The entireOgallala Group, deposited between 19 and 5Ma (Swinehart et al., 1985), ranges from 30to 240 m thick, however much of this fill issandstone and the greatest thickness is con-fined to paleovalleys (Swinehart et al., 1985;Gustavson and Winkler, 1988; Gustavson,1996).

The deposits contain clasts of resistant li-thologies, such as quartz, quartzite, chert, ig-

neous, and metamorphic pebbles, ranging upto cobble size, that were deposited by low-sinuosity (braided?) streams (Stokes, 1950;Blakey and Gubitosa, 1983; Swinehart et al.,1985; Gustavson and Winkler, 1988; Hellerand Paola, 1989; Currie, 1998). Gravel bodiesare widespread, laterally persistent, albeit notalways continuous, and contain coarse-tail up-ward fining and/or nested channel cut-and-fillstructures. Occasionally, the gravel units containbar accretion surfaces, trough cross-bedding inthe sandier fractions, and clast imbrication.

The units rest on underlying deposits withsharp erosional contact (Fig. 3B), which sug-gests that they are not parts of a continuouscoarsening-upward progression but eitherflowed out upon a preexisting erosion surfaceof low relief or that they first beveled a low-angle scour surface (Stokes, 1950; Swinehartet al., 1985). The underlying units are domi-nantly fine-grained fluvial to nearshore marinedeposits. Transport directions for units be-neath the Lower Cretaceous gravels (the Mor-rison Formation) and Ogallala Group (Arika-ree Group) were subparallel to the sheetgravels (Peterson, 1984, 1987; Winslow andHeller, 1987; Yingling and Heller, 1992).These underlying units are much finer grainedthan the gravel sheets, suggesting that theywere transported by rivers of comparativelylow slope. In the case of the Shinarump Con-glomerate, the underlying Moenkopi Forma-tion was not only a finer-grained, hence low-gradient, coastal plain to deltaic system, butpaleoflow was at acute angles to that of theoverlying gravel deposit (Stewart et al.,1972b). Thus, in all three cases, the onset ofgravel deposition coincided with, or soon fol-lowed, a change to steeper transport gradients.

METHODOLOGY

Our approach combines two types of infor-mation. The first is a determination of thetransport slope needed to move gravels out theobserved distances from the source areas. Riv-er transport can be achieved by either buildinga sufficient slope or increasing flow depth togenerate sufficient shear stresses to carrymaterial. If we can constrain the flow depthsfrom field observations, we can calculate thepaleoslopes, which then record regionalpaleotopography.

The second type of information is unitthickness. The preserved thickness of graveldeposits and related units constrains howmuch aggradation can account for building thenecessary topography to reach the neededtransport slopes. The way these slopes arebuilt comes from a combination of aggrada-

tion in the proximal part of basins, degrada-tion in the distal parts of basins (not importantin the case of net deposition), and tilting oftopography by isostatic processes.

Paleoslope Estimates

The key to determining tilt histories is es-timating paleoslopes from these widespreadalluvial gravels. Various quantitative paleo-hydraulic analyses exist (e.g., Cotter, 1971;Ethridge and Schumm, 1978; Gardner, 1983);however, the most appropriate approach to de-termining paleoslopes for these types ofcoarse-grained deposits is that described byPaola and Mohrig (1996). This method utiliz-es the observed relationship between calculat-ed basal shear stress and characteristic grainsize of the mobile bed in gravelly braidedstreams. These streams tend to adjust to anequilibrium condition where basal shear stressis slightly above that needed to move a rep-resentative grain size. Assuming grain densityof 2700 kg m23, slope (S) is related to grainsize (D) and flow depth (H) by:

21S 5 c D Hx

(see Paola and Mohrig (1996) for derivation),where c 5 9.4 3 1022 when D50 (median di-ameter) is used (Paola and Mohrig, 1996), andc 5 2.38 3 1022 when D90 (90th percentilediameter) is used (Heller and Paola, 1989).

The slope calculation requires a measure-ment of characteristic flow depth (H) alongreaches in which the measured gravels weredeposited. We have chosen thickness of thefining-upward sequence in which gravels weremeasured as a proxy for flow depth. We re-alize that a preserved fining-upward sequencemay vary slightly from true flow depth due tochannel perching during filling, which wouldlengthen the fining-upward sequence, or sub-sequent scouring which might remove the up-per part of the sequence. Although these ef-fects tend to counter each other, we made aneffort to reduce the second problem by onlychoosing sites where sequences include thefine-grained sandy to silty capping deposits.

Application of this method to modern grav-elly rivers indicates uncertainties (2s) of afactor of two or less (Paola and Mohrig,1996). The four key assumptions in using thismethod are that flows were quasi-steady state,channel banks were noncohesive, bedformswere absent on the channel bed, and transportwas dominantly as bedload. Thus, applicationis best applied to deposits that do not showsigns of highly variable, short-lived, rapid de-position, where there is no evidence of plant




HELLER et al.

Figure 5. Conceptual model showing meth-odology used in this study. If paleotrans-port slope (a) and aggradational thicknessof deposits (stippled pattern) are known,then tilt of underlying basement (patterned)can be deduced. (A) Required slope is builtup atop horizontal basement. Slope is main-tained during aggradation, leading to bas-inward progradation of system. (B) As inA, except transport slope steepens overtime. This may result in a coarsening-upwardsequence. (C) If thickness of unit increasesupstream beyond that needed to maintaintransport slope, then back-tilted subsidencemust have occurred. This would be true ina typical foreland basin setting (Fig. 6). (D)If unit does not thicken upstream sufficient-ly to generate needed transport slope, thenbasement tilting must have occurred toreach transport slope a.

Figure 6. Interpreted cross section showing Early to middle Cretaceous units of centralUtah that formed during initial growth of Sevier orogenic belt (modified from Yinglingand Heller, 1992). Note strong asymmetry of basin fill and limited transport distance ofconglomeratic deposits of ‘‘Cedar Mountain’’ and San Pete formations. Pre-foreland de-posits include widespread gravel unit (in black)—Buckhorn Conglomerate Member ofCedar Mountain Formation of Early Cretaceous age.

roots or fine-grained cohesive sediment, andwhere channel-filling gravels are either mas-sive or at most contain crude, parallel bed-ding. Field criterion for determining the suit-ability of deposits for analysis are more fullydiscussed in Paola and Mohrig (1996).

Table DR1 shows our calculations for pa-leoslopes at various sites for each unit.2 Shi-narump Conglomerate was collected at two

2GSA Data Repository item 2003121, data usedin paleoslope reconstructions, is available on theWeb at http://www.geosociety.org/pubs/ft2003.htm.Requests may also be sent to [email protected].

sites in southern Nevada, along the east sideof the Spring Mountains and at Firehole Val-ley (locations N-1 and N-3 of Stewart et al.[1972a]). Slopes for the Lower Cretaceousgravels were determined by Heller and Paola(1989) using D90 and come from eight sitesbetween central Utah and southeastern Wyo-ming. Data from the Ogallala Group comefrom 10 sites across the Cheyenne Tablelands,which run east from southern Wyoming intowestern Nebraska (McMillan et al., 2002).

Our results indicate that paleoslopes forthese transport systems span the range of 1023

to 1024, not unlike modern river gradients forgravelly braided streams (Church and Rood,1983; Paola and Mohrig, 1996). Estimated un-certainties on these values are about a factorof 2.

Determining Tilt

For the streams to reach and maintain thecalculated paleoslopes, sufficient topographymust have developed prior to or during de-position. One way topography can be devel-oped is by higher aggradation rates in the up-stream reaches, such as occurs during alluvialfan growth (Fig. 5A). In this case, early allu-vial deposition takes place preferentially up-stream and builds out over time; otherwise,deposition occurs basinwide but faster up-stream so that a sufficiently steep transportslope forms. In the former case, sedimentationwould downlap basinward over time. In thelatter case, slopes would gradually increaseupsection as relief increased. In either case,

deposition must be greater upstream for reliefto develop to the point where slopes could car-ry the observed gravels far into the basins(Fig. 5, A and B).

If more sedimentation takes place upstreamthan is needed to achieve the necessary slopesto carry the gravel forth, there must be sub-sidence in the proximal part of the basin totrap the extra sediment. This situation can beobserved by hanging the available sedimen-tary thickness from the calculated paleoslopeand seeing if the basement surface on whichthese deposits are laid had a back-tilting ge-ometry (Fig. 5C). This case exists in mostsynorogenic gravel deposits (e.g. Fig. 6),where flexural or structural back tilting ac-companies source-area uplift.

The converse is true when the aggradationalthickness upstream is not sufficient to generatethe needed slope to deliver gravels far out intothe basin (Fig. 5D). In this case, the alterna-tive way of generating enough topography toreach the needed transport slope is by tiltingthe basement down toward the distal part ofthe basin. The amount of tilting required is thedifference between the maximum preservedthickness and the required slope. This tiltingcan be preexisting or can occur early duringdeposition of the gravel deposits.

A more complex alternative case is wheretransport slopes are formed by aggradation inthe proximal basin followed by tilting and ero-sion in the same area after deposition of thegravels. Such is the case for the post-orogenicrebound model of Heller et al. (1988) andFlemings and Jordan (1989). However, in this





Figure 7. Results for three units: (A) Shi-narump Conglomerate, with average slopeof 6 3 1024; (B) Lower Cretaceous con-glomerate, with average slope of 1.6 3 1023

to west flattening to 2 3 1024 eastward; (C)Ogallala Group (Ash Hollow Member) onCheyenne Tablelands with average slope of1.5 3 1023 to west and 1 x 1023 eastward.Generalized thicknesses of units are hungfrom this surface. Thickness is smoothedfrom measured sections and isopach mapsreported in Stewart et al. (1972a) for Shi-narump Conglomerate, Heller and Paola(1989) for Lower Cretaceous gravels, andNebraska Geological Survey (1969) forOgallala Group. Two transport slopes areshown. Best-fit slope (solid line) and mini-mum transport slope (dashed line; a factorof 2 less than best-fit slope). Lines of crosssections onto which data are projected areshown in Figure 2.

case, a significant unconformity would formatop the gravel units in the proximal part ofthe basin (Heller et al., 1988; Flemings andJordan, 1989), which does not seem to be thecase in our examples.

RESULTS

Figure 7 shows the results of our analysis.For each unit studied, the calculated topogra-

phy of the river systems is shown as reliefabove the most downstream known extent ofthe gravel units. Topography is determined byprojecting slopes upstream from these down-stream points. Since we take topographic gra-dient to represent depositional slopes existingat the end of aggradation, we hang the pre-served stratigraphic thickness of the gravellyunits from this datum. In this figure, we showthe generalized maximum stratigraphic thick-ness using results from published measuredsections (Stewart et al., 1972a; Heller andPaola, 1989; McMillan et al., 2002). Althoughthe conglomerate units are locally erodedaway at present, we show the deposits as con-tinuous, which they likely were at the end ofdeposition. Generalizing thickness in this waytreats the stratigraphy as a continuous sheetalong the cross sections, while preservation ispatchy in truth. In addition, since we are draw-ing an envelope larger than the reported thick-nesses, we err on the side of overestimatingstratigraphic thickness for these units. This isa conservative approach (i.e., provides a min-imum estimate of tilt), since the more aggra-dation can account for building transportslopes, the less tectonic tilting is needed.

The results for all three units are the same:the preserved stratigraphic thickness of thegravelly units is nowhere near enough to de-velop a sufficient transport slope by aggrada-tion alone (Fig. 7). This is true even if mini-mum paleoslopes are used (Fig. 7, dashedlines). The amount of topography required be-yond that of the aggradational buildup by thestreams themselves is several hundred metersvertically over distances of several hundredkilometers (i.e. a slope of 1023). To developthis relief without evidence of significant dif-ferential aggradation in the upstream side ordegradation in the downstream parts of thesystem requires tilting of the substratum.

DISCUSSION

Our results suggest that topography musthave formed prior to, or during, deposition ofthe gravel units for the slopes to have beenbuilt steeply enough to deliver gravel far intothe basins. Whatever mechanism generatessuch slopes, it is apparently not a typical oc-currence as evidenced by the rarity of suchwidespread gravelly deposits.

The misfit between required slopes andavailable sediment thickness is so large thatsecond-order effects—such as differentialcompaction, modest erosion of the proximal ba-sin deposits or how the sections are hung—arenot contenders. If significant post-depositionalerosion of the proximal basin occurred, then

to explain the results, (1) there must have beenseveral hundred meters of erosion in the prox-imal basin to remove enough gravel to negatethe requirement of tilting, and there is no ev-idence of significant, or any, erosion for theseunits (Witkind, 1956; Yingling and Heller,1992); and (2) the amount of erosion wouldhave to be carefully adjusted across the basinso the resultant deposit was of uniform thick-ness to yield the present sheet-like thicknessof the units.

While the thickness of the gravel depositsis insufficient to generate enough relief to al-low such extensive progradation, it could bethat steep slopes were generated by differen-tial aggradation and/or erosion of the under-lying sedimentary units. To generate topog-raphy in this way, aggradation would have tobe concentrated in the proximal part of thebasin and erosion in the distal part. This isopposite to the observed patterns of erosioninto underlying deposits. Erosional paleoval-leys are primarily found beneath the upstreamlimits of the gravel units and are minor or ab-sent farther downstream.

Underlying units are fluvial to fluvio-deltaic,presumably deposited at lower slopes than theoverlying gravel. This need not be true forfine-grained fluvial deposits, as transport slopecan be set by sediment flux and not just grainsize. In the absence of an estimate of gradientsfor underlying deposits, we suspect they weremuch lower than the overlying gravels, basedupon: (1) their fine-grained nature, which doesnot require steep gradient for transport, (2) theabsence of gravel lenses forming an overallcoarsening-upward sequence, which wouldshow that gradients were building over timeto a point where gravel could be continuouslytransported by the evolving stream system,and (3) the presence of disconformity and par-aconformity below the gravel units, indicatingthat gravel deposition commenced after a dis-crete, indeterminate period of erosion.

Thus, we interpret these results to indicatethat sufficient slopes existed to transport thewidespread gravels and that the most plausibleexplanation is tilting of the Earth’s surface overlong (hundreds of kilometers) wavelengths.

Role of Climate

Given the thin, abrupt nature of these de-posits, it might seem like rapid climate changecould play a role in their origin. The primaryinfluence of climate on the origin of alluvialstratigraphy is through four mechanisms: (1)an increase in precipitation, and therefore wa-ter flux in river systems; (2) changing thegrain-size distribution derived from the source




HELLER et al.

area by altering the mechanisms of weather-ing; (3) changing vegetation patterns, whichcan affect bank strength and hydrographshape; and (4) by forcing increased regionalerosion that leads to isostatic rebound andtilting.

Changing water flux in rivers affects rivercapacity and competence, which can generateat least transient gravel transport events (Paolaet al., 1992). However, increased water fluxdoes not play a role in the origin of thesewidespread gravels. Flow depth is a key ob-servation made in reconstructing paleoslope;therefore, any increase in water depth has al-ready been accounted for in the paleoslope de-termination. In addition, in truly braided riversystems, changes in discharge are compensat-ed for by changing the number of active chan-nels and, therefore, the width of the activebraid plain. Such lateral variations are unim-portant in generating the shear stress neces-sary to move gravel at any single point. If thesheet-like distribution of the units requiredmultiple channel threads active at the sametime, then high water discharge would be re-quired. However, it is not clear whether theseunits formed by multiple channels active atthe same time or by the amalgamation of de-posits formed by fewer channels active over asustained period.

The second possible influence of climate ison the grain-size distribution delivered to riv-ers from hill slopes. A change in weatheringprocess could change the relative abundanceof gravel derived from source areas. A signif-icant increase in the input material grain sizewould change the sorting of resultant deposits.While this influences the availability of grav-el, it does not mitigate the need for tilt.

The third possible influence of climate is onchannel pattern. If channel braiding is primar-ily a result of reduction of bank cohesiveness(Parker, 1978), then climate can come intoplay by both decreasing the abundance ofclay-sized material in the water load and byaltering the type and abundance of vegetationalong the channel banks. The cohesiveness ofclay exerts primary influence on bank stabilityand, therefore, channel form (Schumm andThorne, 1989). Climate can also influencevegetation types along the flood plain, whichcan influence bank stability through rootingbehavior. None of this negates the necessity togenerate sufficient slopes to transport the ma-terials. The only way for this scenario to workis if steep slopes formed earlier than the graveldeposits and the transition into gravels record-ed a grain-size change from unroofed sources,not the onset of steeper gradients. However,evaluating this possibility would require re-

constructing the history of slope changes inunderlying, finer-grained rocks, which was notdone in this study and would require a differ-ent methodology.

Fourth, while climate-induced erosion andresultant isostatic rebound provides a possiblelink between climate and tilting (e.g., Molnarand England, 1990), in the case of the wide-spread gravels, the area of deposition was overwavelengths far larger than flexural responsedistances of earth lithosphere (Watts, 1992;McKenzie and Fairhead, 1997; Stewart andWatts, 1997). Thus, for this mechanism toplay a role would require broadly distributed,but asymmetric, erosion along most of thelength of the depositional basin. Tilting at dis-tances far from the eroded source areas re-quires an unusual distribution of erosionacross the basin such that some areas are erod-ed deeply enough to generate significant iso-static rebound, while immediately adjacent ar-eas continue to transport and deposit sandsand gravels that record this tilt. Such a sce-nario is not only unlikely and oxymoronic foraggradational settings, but this hypothesis hasbeen tested and negated for the OgallalaGroup (McMillan et al., 2002).

Finally, if these deposits were related to cli-mate, it would be curious for such events tobe found only at these times and in these lo-cations and not more broadly in North Amer-ica or the world. It seems too geographicallylimited.

Role of Source Area Lithology

If source areas contained a change in li-thology either in depth or over an area, thena change in abundance of gravel productioncould reflect, respectively, simple unroofing orintegration of streams into larger catchments.Either way, lithologies capable of generatingmore resistant gravel-sized clasts could betapped over time. This is similar to the secondpossible climatic influence discussed above,and, as with that case, it has no bearing on theneed to generate a tilt to allow these gravelsto reach points farther out across the deposi-tional basin. However, it does provide an in-triguing possibility that tilting began earlierthan we assume and that the occurrence ofgravel reflects unroofing sometime after tiltinghad started. We suspect that such may be thecase for the Lower Cretaceous gravels in theRocky Mountains. The underlying unit, theMorrison Formation of Late Jurassic age, isfine-grained overall, but it does contain gravelalong its most upstream, western, exposures(Peterson, 1980). In particular, the Salt WashMember of the Morrison Formation is rela-

tively coarse grained. Based on detrital fossilspreserved within clasts of the Lower Creta-ceous gravel (Winslow and Heller, 1987; Hell-er and Paola, 1989), the source area includesupper Paleozoic rocks of Nevada. These areashave a suitable stratigraphy for unroofing toplay a major role in the evolution of the LateJurassic through Early Cretaceous stratigraphyin the Rocky Mountains. Triassic rocks of Ne-vada contain many marine mud-rich units, in-cluding the Auld Lang Syne Group, often re-ferred to informally as the ‘‘mud-pile.’’ Thesedeposits sit atop Pennsylvanian-Permian ageshelf carbonates that include secondary chertnodules. Erosion of this source area could leadto the dominantly mud-rich fluvial rocks ofthe Morrison Formation. As deeper levelswere unroofed in the source area, the Paleo-zoic chert-bearing carbonates could be tapped,providing clasts found in the Lower Creta-ceous gravels. If this is the case, it may wellbe that tilting began in Late Jurassic time inthis area and helped generate slopes in theMorrison Formation. This possibility could betested once a methodology for reconstructingslopes in finer-grained fluvial systems isestablished.

Tectonic Timing of Continental Tilting

In search of a causal mechanism for large-scale continental tilting, we note that each ofthe gravel sheets is associated with changes inmantle density structure associated either withsubduction or late Cenozoic arrival of warmer-than-normal mantle beneath the southernRocky Mountains. The Shinarump Conglom-erate was deposited during Late Triassic time(Stewart et al., 1972a; Blakey and Gubitosa,1983), coincident with a transition from a pas-sive (and/or transform) margin to initiation ofan Andean-type convergent margin in thesouthwestern USA (Walker, 1988; Dickinson,2000). Continental arc plutonism began afterdeposition of the Moenkopi Formation andwas commensurate with deposition of theChinle Formation, of which the Shinarump isthe basal member (Stewart et al., 1972a, b;Armstrong and Ward, 1993; Dickinson, 2000).

Lower Cretaceous gravels were depositedsometime during Neocomian-Aptian time,close to, but arguably immediately prior to,earliest thrusting in the Sevier belt (Fig. 2B)(Heller et al., 1986; Yingling and Heller, 1992;Currie, 1998). The onset of deformation in theSevier belt post-dated deposition of the grav-els, as determined from source area prove-nance of gravel composition (Heller and Pao-la, 1989; Yingling and Heller, 1992), lack ofcontemporaneous foreland basin subsidence





(Heller et al., 1986), and the known chronol-ogy of tectonic activity in adjacent parts ofNevada (Dickinson, 2001). Also, there is nogeologically reasonable rendition of thrust beltloading that can result in basinward tilts ofsufficient steepness or width to generate theobserved distribution of Lower Cretaceousgravels (Fig. 4; c.f. Heller et al., 1986; De-Celles and Giles, 1996; Currie, 1998). Morelikely, basin partitioning by thrust-belt defor-mation led to local trapping of gravel deposits,such as seen in the early foreland basin fill ofUtah (Fig. 6). Deposition of the Lower Cre-taceous gravels instead took place following aperiod of continued eastward expansion of theMesozoic continental arc in the western USA(Armstrong and Ward, 1993; Dickinson,2001).

The Ogallala Group was deposited eastwardacross the western Great Plains following in-creased magmatic activity in Colorado andNew Mexico (Chapin and Cather, 1994), up-lift, and northward-propagation of the RioGrand Rift. In New Mexico and southern Col-orado, rifting began in Oligocene time (Chap-in and Cather, 1994). Farther north, the sur-face manifestation of rifting is discontinuous,and available age data suggest a more recentinitiation (Leat et al., 1991; Mears, 1993; Er-slev, 2001). Young uplift of the western GreatPlains, inferred to relate to propagation of therift (Angevine and Flanagan, 1987), is coin-cident with deposition of the Ogallala Group(Miocene) and younger fluvial units in thatarea (McMillan et al., 2002). Symmetric de-position to the west of the rift was neither con-tinuous nor widespread, presumably due to in-herited topography from earlier orogeny thatforced deposition into more isolated basins.

Tilt Origins

In each case, gravel sheet deposition mostlytook place inboard of the locus of tectonic ac-tivity and was not directly affected by asso-ciated structural deformation; therefore, crustal-scale structure played no role in regionaltilting. Tilting at wavelengths greater than afew hundred kilometers can be accomplishedin a couple of ways. Isostatic topography re-sults primarily from lithospheric density var-iations. A defining characteristic of isostatictopography is that it generally changes slowlyin response to erosion and major orogenic/magmatic episodes. Dynamic topography, onthe other hand, is the deflection of the Earth’ssurface that balances the stresses applied at thebase of the elastic lithosphere by mantle con-vection (Burgess and Moresi, 1999). In con-trast to isostatic topography, dynamic topog-

raphy is transient because it results from thestresses created by the evolving pattern ofmantle convection. Dynamic topography iscalculated by solving the equations for mantleconvection with a specified distribution ofdensity and viscosity throughout the Earth’smantle. The effect of the 660 km phase tran-sition on mass transport across this boundarycan also significantly modulate the time evo-lution of dynamic topography, especially onshort time scales (Pysklywec and Mitrovica,2000).

While the magnitude of dynamic topogra-phy is model-dependent, it is now widely ac-cepted that most lateral variations in mantledensity structure can be reconstructed by in-tegrating the last 180 m.y. of plate subduction(Richards and Engebretson, 1992). The re-maining dynamic topography probably resultsfrom warm upwellings from the deep mantle(Lithgow-Bertelloni and Silver, 1998). In thecase of the western USA, the subduction of7000 km of oceanic plate over the last 180Ma is well documented (Engebretson et al.,1992) and apparently has exerted a first-ordercontrol on the dynamic topography of the re-gion as documented by the sedimentary record(Mitrovica et al., 1989; Coakley and Gurnis,1995; Burgess et al., 1997; Pysklywec andMitrovica, 2000). More controversial are theroles of tectonic uplift (due to increased man-tle buoyancy) versus climate change in thesouthern Rocky Mountains as the cause of de-position of the post-Laramide Ogallala Group(McMillan et al., 2002).

Our observations suggest that 400–800 mof differential uplift occurred over areas 400–800 km wide on time scales of 1–10 m.y.These tilts are consistent with the predictionof dynamic topography from time-varyingsubduction models (Mitrovica et al., 1989;Lithgow-Bertelloni and Silver, 1998; Burgessand Moresi, 1999; Pysklywec and Mitrovica,2000). While it is true that the dynamic to-pography is filtered by the lithosphere’s elasticresponse at wavelengths shorter than 500–1000 km, we do not consider its effect im-portant to our first-order conclusions.

Shinarump ConglomerateWhile little is known about plate kinematics

during Late Triassic time, we assume the Shi-narump Conglomerate is related to the pene-contemporaneous onset of subduction alongthe southwestern U.S. continental margin (Fig.8A). Certainly substantial variations in iso-static and dynamic topography are predictedduring the early evolution of any subductionzone as a result of initial slab emplacement,initiation of arc magmatism, and attendant

changes in extensional-compressional re-gimes. Of the cases studied, the ShinarumpConglomerate is the least compelling for dy-namic topography because the wavelength ofdeposition is smaller and the variability of pa-leocurrents is greater than in the other cases.However, the correlation of timing of deposi-tion of the Shinarump following the onset ofsubduction is consistent with isostatic adjust-ments associated with arc creation playing amajor role in regional tilting. More data on thetiming of Late Triassic volcanism and tecto-nism in the southwestern USA and/or a betterestimate of plate kinematics at the time areneeded to constrain dynamical models.

Cretaceous ConglomerateAs the western U.S. subduction zone

lengthened over the next few tens of millionsof years, so too did the associated continentalarc grow along the North American Cordillera(Armstrong and Ward, 1993). By Cretaceoustime, the North American continent had drift-ed over the large mass of negatively buoyantsubducted slab that had accumulated sinceLate Triassic time. The sinking of this masscreates lithospheric subsidence with relief ofhundreds of meters over wavelengths of hun-dreds of kilometers (Mitrovica et al., 1989;Gurnis, 1991, 1992). These tilts, approaching1023, are sufficient to promote the observedwidespread distribution of Lower Cretaceousgravels (Fig. 8B). With continued westwardmigration of the continent over the moment ofaccumulated slab mass in the middle, mantlemigrates beneath the central North Americancontinent by Late Cretaceous time. This, co-incident with a time of global sea level high-stand (Haq et al., 1987), produces the subsi-dence needed to create the Western InteriorSeaway (Fig. 8C) (Gurnis, 1992; Mitrovica etal., 1989). It is notable that topography in-ferred for the Western Interior Seaway (a fewhundred meters of relief over a few hundredkilometers in length [Martinson et al., 1998])is the same scale as is needed immediatelypreceding deposition of the Lower Cretaceousgravels somewhat farther west. While we sus-pect that other widespread fluvial units depos-ited between Triassic and Cretaceous time,specifically the Kayenta and Morrison for-mations, may be related to the same phenom-ena (e.g., Lawton, 1994), paleoslope recon-structions of these finer-grained units are notyet available.

Ogallala GroupFinally, we relate continental tilting asso-

ciated with the Ogallala Group with the oc-currence of the Neogene uplift of the southern




HELLER et al.

Figure 8. Interpreted relationship of mantle features to regional tilt and deposition ofwidespread units. (A) Subsidence associated with initial underthrusting of subducted slabis correlated with tilt and deposition of Shinarump Conglomerate in Late Triassic time.Penetration of slab may have caused subsidence either by inducing asthenospheric coun-terflow above subducted slab or by disrupting vertical heat loss from mantle. (B) Contin-ued subduction, growth, and migration of continental arc is related to northeast-migratingtilt and depocenter recorded by Lower Cretaceous gravel units in Rocky Mountains. (C)Continued eastward migration of tilt and Western Interior Seaway depocenter of LateCretaceous age is tied to gradual eastward migration of low-angle subducted slab. SL—sea level during Cretaceous eustatic highstand. (D) Tilting associated with progradationof Ogallala Group is interpreted to relate to regional uplift and active extension associatedwith evolving Rio Grande Rift in Miocene time.

and central Rockies (Fig. 8D) (Eaton, 1987;McMillan et al., 2002). This uplifted regionhas been correlated with low-velocity uppermantle currently imaged beneath this regionthat is interpreted to be mantle at or near thedry peridotite solidus (Goes and van der Lee,2002). The North American upper mantle den-sity model of Goes and van der Lee (2002)predicts ;1 km of isostatic elevation differ-ence between the southern Rocky Mountainsand Kansas. This effective tilt on the order of1023 is sufficient to enable transport of thegravel deposits of the Ogallala Group. Wewould suggest that a peak in volcanism inColorado and New Mexico ca. 35 Ma (Mc-Millan Nancy et al., 2000), coincident with theonset of extension along the Rio Grande Rift(Chapin and Cather, 1994), marks the onset ofa lithospheric warming event associated withregional uplift. At longer wavelengths, the dy-namical rebound of the western USA is pre-dicted as the North American continent driftedwell west of the Farallon slab, which now re-sides in the middle mantle off the east coastof North America (Grand et al., 1997). Thus,a combination of dynamic rebound and the re-gional lithospheric warming event are respon-sible for the origin of the Ogallala Group, asopposed to subsidence, which led to the for-mation of the other two gravel sheets. Thepreservation potential of the Ogalla Group istherefore low; continued uplift and tilting ofthe deposits as the rift continues to grow leadsto erosion and reworking of the previously de-posited materials (Heller et al., 1988).

CONCLUSIONS

By combining paleohydraulic requirementsof gravel transport with preserved isopach ge-ometries, we analyzed the tilt history of thewestern USA at various times in its evolution.The presence of widespread gravels, such asthose described here, indicates large-scale de-flections of continental crust in areas with lit-tle preexisting topography. Such regional con-tinental tilts result from dynamic topographycreated by mantle convection. The rarity ofthese widespread units suggests that either pe-riods of regional continental tilting are unusu-al or that preservation potential of these de-posits is low. However, it is possible thatsimilar thin, widespread conglomeratic unitsmay have been recognized elsewhere but werelumped in with surrounding alluvial depositsfor the purposes of geologic mapping. Iden-tification of such units is important becausethey provide at least semiquantitative evi-dence of broad wavelength topography thatcan help constrain the time-integrated effect of

slab subduction and mantle convection on sur-face processes. As such, paleotopography, asdetermined from calculated transport gradientsof sedimentary deposits, provides a means ofrelating constructional landforms to mantle-driven geodynamics.

ACKNOWLEDGMENTS

Supported in part by National Science Foundationgrant EAR-0106029. Helpful reviews were providedby R. Dorsey, E. Humphreys and D. Mohrig. Weare grateful for discussions and/or field assistanceprovided by R.C. Blakey, M. Cheadle, W.R. Dick-inson, R. Dunne, S. Liu, H. Luo, R.M. Lynds, C.Paola, D. Saffer, A.W. Snoke, A.K. Souders, and K.Trujillo.

REFERENCES CITED

Angevine, C.L., and Flanagan, K.M., 1987, Buoyant sub-surface loading of the lithosphere in the Great Plainsforeland basin: Nature, v. 327, p. 137–139.

Armstrong, R.L., and Ward, P.L., 1993, Late Triassic toEarliest Eocene magmatism in the North AmericanCordillera: Implications for the Western Interior Basin,in Caldwell, W.G.E., and Kauffman, E.G., eds., Evo-lution of the Western Interior Basin: Geological As-sociation of Canada Special Paper 39, p. 49–72.

Baldridge, W.S., Perry, F.V., Vaniman, D.T., Nealey, L.D.,Leavy, B.D., Laughlin, A.W., Kyle, P., Bartov, Y.,Steinitz, G., and Gladney, E.S., 1991, Middle to lateCenozoic magmatism of the southeastern ColoradoPlateau and central Rio Grande rift (New Mexico andArizona, U.S.A.): A model for continental rifting: Tec-tonophysics, v. 197, p. 327–354.

Blackstone, D.L., Jr., 1975, Late Cretaceous and Cenozoic his-tory of Laramie Basin region, southeast Wyoming: Geo-logical Society of America Memoir 144, p. 249–279.





Blakey, R.C., and Gubitosa, R., 1983, Late Triassic paleo-geography and depositional history of the Chinle For-mation, southern Utah and northern Arizona, in Reyn-olds, M.W., and Dolly, E.D., eds., Mesozoicpaleogeography of west-central United States: Denver,Colorado, Society of Economic Paleontologists andMineralogists (SEPM) Rocky Mountain Paleogeogra-phy Symposium, v. 2, p. 57–76.

Burgess, P.M., and Moresi, L.N., 1999, Modelling rates anddistribution of subsidence due to dynamic topographyover subducting slabs: Is it possible to identify dynam-ic topography from ancient strata?: Basin Research,v. 11, p. 305–314.

Burgess, P.M., Gurnis, M., and Moresi, L., 1997, Formationof sequences in the cratonic interior of North Americaby interaction between mantle, eustatic, and strati-graphic processes: Geological Society of AmericaBulletin, v. 109, p. 1515–1535.

Chapin, C.E., and Cather, S.M., 1994, Tectonic setting ofthe axial basins of the northern and central Rio GrandeRift, in Keller, G.R., and Cather, S.M., eds., Basins ofthe Rio Grande Rift; structure, stratigraphy, and tec-tonic setting: Geological Society of America SpecialPaper 291, p. 5–25.

Church, M., and Rood, K., 1983, Catalogue of alluvial riverchannel regime data: Vancouver, University BritishColumbia, Department of Geography, 99 p.

Coakley, B., and Gurnis, M., 1995, Far-field tilting of Lau-rentia during the Ordovician and constraints on theevolution of a slab under an ancient continent: Journalof Geophysical Research, v. 100, p. 6313–6327.

Cotter, E., 1971, Paleoflow characteristics of a Late Creta-ceous river in Utah from analysis of sedimentarystructures in the Ferron Sandstone: Journal of Sedi-mentary Petrology, v. 41, p. 129–138.

Currie, B.S., 1997, Sequence stratigraphy of nonmarine Ju-rassic–Cretaceous rocks, central Cordilleran foreland-basin system: Geological Society of America Bulletin,v. 109, p. 1206–1222.

Currie, B.S., 1998, Upper Jurassic-Lower Cretaceous Mor-rison and Cedar Mountain Formations, NE Utah-NWColorado: Relationships between nonmarine deposi-tion and early Cordilleran foreland-basin develop-ment: Journal of Sedimentary Research, v. 68,p. 632–652.

DeCelles, P.G., and Currie, B.S., 1996, Long-term sedimentaccumulation in the Middle Jurassic-early EoceneCordilleran retroarc foreland-basin system: Geology,v. 24, p. 591–594.

DeCelles, P.G., and Giles, K.A., 1996, Foreland basin sys-tems: Basin Research, v. 8, p. 105–123.

Dickinson, W.R., 2000, Geodynamic interpretation of Pa-leozoic tectonic trends oriented oblique to the Meso-zoic Klamath-Sierran continental margin in California,in Soreghan, M.J., and Gehrels, G.E., eds., Paleozoicand Triassic paleogeography and tectonics of westernNevada and northern California: Geological Societyof America Special Paper 347, p. 209–245.

Dickinson, W.R., 2001, Tectonic setting of the Great Basinthrough geologic time: Implications for metallogeny,in Shaddrick, D.R., Zbinden, E.A., Mathewson, D.C.,and Prenn, C., eds., Regional tectonics and structuralcontrol of ore: The major gold trends of northern Ne-vada: Geological Society of Nevada Special Publica-tion 33, p. 27–53.

Diffendal, R.F., Jr., 1982, Regional implications of the ge-ology of the Ogallala Group (upper Tertiary) of south-western Morrill County, Nebraska and adjacent areas:Geological Society of America Bulletin, v. 93,p. 964–976.

Dubiel, R.F., and Brown, J.L., 1993, Sedimentologic anal-ysis of cores from the Upper Triassic Chinle Forma-tion and Lower Permian Cutler Formation, LisbonValley, Utah: U.S. Geological Survey Bulletin,v. 2000-E, p. E1–E40.

Eaton, G.P., 1986, A tectonic redefinition of the southern RockyMountains: Tectonophysics, v. 132, p. 163–193.

Eaton, G.P., 1987, Topography and origin of the southernRocky Mountains and Alvarado Ridge, in Coward,M.P., Dewey, J.F., and Hancock, P.L., eds., Continentalextensional tectonics: Geological Society [London]Special Paper 28, p. 355–369.

Engebretson, D.C., Kelley, K.P., Cashman, H.J., and Rich-ards, M.A., 1992, 180 million years of subduction:Geological Society of America Today, v. 2,p. 93–100.

Erslev, E.A., 2001, Multistage, multidirectional Tertiaryshortening and compression in north-central NewMexico: Geological Society of America Bulletin,v. 113, p. 63–74.

Ethridge, F.G., and Schumm, S.A., 1978, Reconstructing paleo-channel morphologic and flow characteristics: Method-ology, limitations and assessment, in Miall, A.D., ed., Flu-vial Sedimentology: Calgary, Canadian Society ofPetroleum Geologists Memoir 5, p. 703–721.

Flemings, P.B., and Jordan, T.E., 1989, A synthetic strati-graphic model of foreland basin development: Journalof Geophysical Research, v. 94, p. 3851–3866.

Furer, L.C., 1970, Petrology and stratigraphy of nonmarineUpper Jurassic-Lower Cretaceous rocks of westernWyoming and southeastern Idaho: American Assoco-ciation of Petrology Geology Bulletin, v. 54,p. 2282–2302.

Gardner, T.W., 1983, Paleohydrology and paleomorphologyof a Carboniferous, meandering, fluvial sandstone:Journal of Sedimentary Petrology, v. 53, p. 991–1005.

Goes, S., and van der Lee, S., 2002, Thermal structure ofthe North American uppermost mantle inferred fromseismic tomography: Journal of Geophysical Re-search, v. 107, p. 2000JB000049.

Goodwin, R.G., and Diffendal, R.F., Jr., 1987, Paleohy-drology of some Ogallala (Neogene) streams in thesouthern panhandle of Nebraska, in Ethridge, F.G., etal., ed., Recent developments in fluvial sedimentology,Volume 39: Society of Economic Paleontologists andMineralogists Special Publication: Tulsa, Society ofEconomic Paleontologists and Mineralogists (SEPM),p. 149–157.

Grand, S.P., van der Hilst, R.D., and Widiyantoro, S., 1997,Global seismic tomography: A snapshot of convectionin the Earth: Geological Society of America Today,v. 7, p. 1–7.

Gurnis, M., 1991, Continental flooding and mantle-lithospheredynamics, in Sabadini, R., Lambeck, K., and Boschi, E.,eds., Glacial isostasy, sea-level and mantle rheology: Dor-drect, Kluwer Academic Publishers, p. 445–491.

Gurnis, M., 1992, Rapid continental subsidence followingthe initiation and evolution of subduction: Science,v. 255, p. 1556–1558.

Gustavson, T.C., 1996, Fluvial and eolian depositional sys-tems, paleosols, and paleoclimate of the Upper Ce-nozoic Ogallala and Blackwater Draw Formations,southern high plains, Texas and New Mexico: Austin,The University of Texas, 62 p.

Gustavson, T.C., and Winkler, D.A., 1988, Depositional fa-cies of the Miocene-Pliocene Ogallala Formation,northwestern Texas and eastern New Mexico: Geolo-gy, v. 16, p. 203–206.

Haq, B.U., Hardenbol, J., and Vail, P.R., 1987, Chronologyof fluctuating sea levels since the Triassic: Science,v. 235, p. 1156–1167.

Hasiotis, S.T., Demko, T.M., and Dubiel, R.F., 2000, Shi-narump incised valley cut-and-fill deposits, paleosols,and ichnofossils in the lower part of the Upper TriassicChinle Formation, Petrified Forest National Park, Ar-izona: It really does exist, and it has regional strati-graphic implications: Geological Society of AmericaAbstracts with Programs, v. 33, p. 23.

Heller, P.L., and Paola, C., 1989, The paradox of LowerCretaceous gravels and the initiation of thrusting inthe Sevier orogenic belt, United States Western Inte-rior: Geological Society of America Bulletin, v. 101,p. 864–875.

Heller, P.L., Angevine, C.L., Winslow, N.S., and Paola, C.,1988, Two-phase stratigraphic model of foreland-basin sequences: Geology, v. 16, p. 501–504.

Heller, P.L., Bowdler, S.S., Chambers, H.P., Coogan, J.C.,Hagen, E.S., Shuster, M.W., Winslow, N.S., and Law-ton, T.F., 1986, Time of initial thrusting in the Sevierorogenic belt, Idaho-Wyoming and Utah: Geology,v. 14, p. 388–391.

Izett, G.A., 1975, Late Cenozoic sedimentation and defor-mation in northern Colorado and adjoining areas, inCurtis, B.F., ed., Cenozoic history of the Southern

Rocky Mountains: Geological Society of AmericaMemoir 144: Boulder, Geological Society of America,p. 179–209.

Keller, G.R., and Baldridge, W.S., 1999, Lithospheric struc-ture and evolution of the Rocky Mountains, Part II:Rocky Mountain Geology, v. 34, p. 121–130.

Kowallis, B.J., and Heaton, J.S., 1987, Fission-track datingof bentonites and bentonitic mudstones from the Mor-rison Formation in central Utah: Geology, v. 15,p. 1138–1142.

Lawton, T.F., 1994, Tectonic setting of Mesozoic sedimen-tary basins, Rocky Mountain Region, United States,in Caputo, M.V., Peterson, J.A., and Franczyk, K.J.,eds., Mesozoic systems of the Rocky Mountain Re-gion, Society for Sedimentary Geology (SEPM)Rocky Mountain Section, p. 1–25.

Leat, P.T., Thompson, R.N., Morrison, M.A., Hendry, G.L.,and Dickin, A.P., 1991, Alkaline hybrid mafic magmasof the Yampa area, NW Colorado, and their relation-ship to the Yellowstone mantle plume and lithosphericmantle domains: Contributions to Mineralogy and Pe-trology, v. 107, p. 310–327.

Lithgow-Bertelloni, C., and Silver, P.G., 1998, Dynamic to-pography, plate driving forces and the African super-swell: Nature, v. 395, p. 269–272.

Lucas, S.G., 1993, The Chinle Group: Revised stratigraphyand biochronology of Upper Triassic nonmarine stratain the western United States: Museum of Northern Ar-izona Bulletin, v. 59, p. 27–50.

MacCready, T., Snoke, A.W., Wright, J.E., and Howard,K.A., 1997, Mid-crustal flow during Tertiary exten-sion in the Ruby Mountains core complex, Nevada:Geological Society of America Bulletin, v. 109,p. 1576–1594.

Martinson, V.S., Heller, P.L., and Frerichs, W.E., 1998, Dis-tinguishing middle Late Cretaceous tectonic eventsfrom regional sea-level change using foraminiferaldata from the U.S. Western Interior: Geological So-ciety of America Bulletin, v. 110, p. 259–268.

May, M.T., Furer, L.C., Kvale, E.P., Suttner, L.J., Johnson,G.D., and Meyers, J.H., 1995, Chronostratigraphy andtectonic significance of Lower Cretaceous conglom-erates in the foreland of central Wyoming, in Dorobek,S.L., and Ross, G.M., eds., Stratigraphic evolution offoreland basins: Tulsa, Oklahoma, Society for Sedi-mentary Geology (SEPM) Special Publication 52,p. 97–110.

McKenzie, D., and Fairhead, D., 1997, Estimates of theeffective elastic thickness of the continental litho-sphere from bouguer and free air gravity anomalies:Journal of Geophysical Research, v. 102,p. 27,523–27,552.

McMillan, M.E., Angevine, C.L., and Heller, P.L., 2002,Postdepositional tilt of the Miocene-Pliocene OgallalaGroup in the western Great Plains: Evidence of lateCenozoic uplift of the Rocky Mountains: Geology,v. 30, p. 63–66.

McMillan, Nancy, J., Alan, P.D., and Haag, D., 2000, Evo-lution of magma source regions in the Rio GrandeRift, southern New Mexico: Geological Society ofAmerica Bulletin, v. 112, p. 1582–1593.

Mears, B., Jr., 1993, Geomorphic history of Wyoming andhigh-level erosion surfaces, in Snoke, A.W., Steidt-mann, J.R., and Roberts, S.M., eds., Geology of Wy-oming: Laramie, Geological Survey of WyomingMemoir 5, p. 698–626.

Mears, B., Jr., 1998, Neogene normal faulting superposedon a Laramide uplift: Medicine Bow Mountains, Si-erra Madre, and intervening Saratoga Valley, Wyo-ming and Colorado: Contributions to Geology, v. 32,p. 181–185.

Mitrovica, J.X., Beaumont, C., and Jarvis, G.T., 1989, Tilt-ing of continental interiors by the dynamical effectsof subduction: Tectonics, v. 8, p. 1079–1094.

Molnar, P., and England, P., 1990, Late Cenozoic uplift ofmountain ranges and global climate change: Chickenor egg?: Nature, v. 346, p. 29–34.

Naeser, C.W., Izett, G.A., and Obradovich, J.D., 1980, Fission-track and K-Ar ages of natural glasses: U.S. Geolog-ical Survey Bulletin 1489, 31 p.

Nebraska Geological Survey, 1969, Geologic bedrock map




HELLER et al.

of Nebraska: Lincoln, Nebraska Geological Survey,scale 1:2,900,000, 1 sheet.

Oldham, D.W., 1996, Permian salt in the northern DenverBasin; controls on occurrence and relationship to oiland gas production from Cretaceous reservoirs, inLongman, M.W., and Sonnenfeld, M.D., eds., Paleo-zoic systems of the Rocky Mountain region: Denver,Society for Sedimentary Geology (SEPM), RockyMountain Section, p. 335–354.

Paola, C., and Mohrig, D., 1996, Paleohydraulics revisited:Paleoslope estimation in coarse-grained braided rivers:Basin Research, v. 8, p. 243–254.

Paola, C., Heller, P.L., and Angevine, C.L., 1992, The large-scale dynamics of grain-size variation in alluvial ba-sins, 1: Theory: Basin Research, v. 4, p. 73–90.

Parker, G., 1978, Self-formed rivers with equilibrium banksand mobile bed. Part 2: The gravel river: Journal ofFluid Mechanics, v. 89, p. 127–146.

Peterson, F., 1980, Sedimentology of the uranium-bearingSalt Wash Member and Tidwell unit of the MorrisonFormation in the Henry and Kaiparowits Basins, Utah,in Picard, M.D., ed., Henry Mountains Symposium:Salt Lake City, Utah Geological Association Publi-cation, v. 8, p. 305–322.

Peterson, F., 1984, Fluvial sedimentation on a quivering cra-ton: Influence of slight crustal movements on fluvialprocesses, Upper Jurassic Morrison Formation, west-ern Colorado Plateau: Sedimentary Geology, v. 38,p. 21–49.

Peterson, F., 1987, The search for source areas of Morrison(Upper Jurassic) clastics on the Colorado Plateau:Geological Society of America Abstracts with Pro-grams, v. 19, p. 804.

Pysklywec, R.N., and Mitrovica, J.X., 2000, Mantle flowmechanisms of epeirogeny and their possible role inthe evolution of the western Canada sedimentary ba-sin: Canadian Journal of Earth Sciences, v. 37,p. 1535–1548.

Reeves, C.C., Jr., 1984, The Ogallala depositional mystery,in Whetstone, G.A., ed., Proceedings, Ogallala Aqui-fer symposium II: Lubbock, Texas Tech UniversityWater Resources Research Center, p. 129–156.

Richards, M.A., and Engebretson, D.C., 1992, Large-scalemantle convection and the history of subduction: Na-ture, v. 355, p. 437–440.

Riggs, N.R., Lehman, T.M., Gehrels, G.E., and Dickinson,W.R., 1996, Detrital zircon link between headwatersand terminus of the Upper Triassic Chinle-Dockumpaleoriver system: Science, v. 273, p. 97–100.

Royse, F., Jr., 1993, Case of the phantom foredeep: EarlyCretaceous in west-central Utah: Geology, v. 21,p. 133–136.

Schumm, S.A., and Thorne, C.R., 1989, Geologic andgeomorphic controls on bank erosion, in Ports-Mi-chael, A., ed., Proceedings of the 1989 national con-ference on Hydraulic engineering: New York, Amer-ican Society of Civil Engineers, HydraulicsDivision, p. 106–111.

Stewart, J., and Watts, A.B., 1997, Gravity anomalies andspatial variations of flexural rigidity at mountainranges: Journal of Geophysical Research, v. 102,p. 5327–5352.

Stewart, J.H., Poole, F.G., and Wilson, R.F., 1972a, Stratig-raphy and origin of the Chinle Formation and relatedUpper Triassic strata in the Colorado Plateau Region:U.S. Geological Survey Professional Paper 690,336 p.

Stewart, J.H., Poole, F.G., and Wilson, R.F., 1972b, Stra-tigraphy and origin of the Triassic Moenkopi For-mation and related strata in the Colorado Plateau re-gion: U.S. Geological Survey Professional Paper691, 195 p.

Stokes, W.L., 1944, Morrison Formation and related de-posits in and adjacent to the Colorado Plateau: Geo-logical Society of America Bulletin, v. 55,p. 951–992.

Stokes, W.L., 1950, Pediment concept applied to Shinarumpand similar conglomerates: Geological Society ofAmerica Bulletin, v. 61, p. 91–98.

Swinehart, J.B., Souders, V.L., DeGraw, H.M., and Diffen-dal, R.F., Jr., 1985, Cenozoic paleogeography of west-ern Nebraska, in Flores, R.M., and Kaplan, S.S., eds.,Cenozoic paleogeography of west-central UnitedStates: Denver, Colorado, Society of Economic Pale-

ontologists and Mineralogists (SEPM) Rocky Moun-tain Section, Symposium, Volume 3, p. 209–229.

Trimble, D.E., 1980, Cenozoic history of the Great Plainscontrasted with that of the southern Rocky Mountains:A synthesis: The Mountain Geologist, v. 17, p. 59–69.

Tschudy, R.H., Tschudy, B.D., and Craig, L.C., 1984, Pal-ynological evaluation of the Cedar Mountain and Bur-ro Canyon Formations, Colorado Plateau: U.S. Geo-logical Survey Professional Paper 1281, p. 24.

Tweto, O., 1979, The Rio Grande rift system in Colorado,in Riecker, R.E., ed., Rio Grande rift; tectonics andmagmatism: Washington, D.C., American Geophysi-cal Union, p. 33–56.

Walker, J.D., 1988, Permian and Triassic rocks of the Mo-jave Desert and their implicatons for timing and mech-anisms of continental truncation: Tectonics, v. 7,p. 685–709.

Watts, A.B., 1992, The effective elastic thickness of thelithosphere and the evolution of foreland basins: BasinResearch, v. 4, p. 169–178.

Winslow, N.S., and Heller, P.L., 1987, Evaluation of un-conformities in Upper Jurassic and Lower Cretaceousnonmarine deposits, Bighorn Basin, Wyoming andMontana: Sedimentary Geology, v. 53, p. 181–202.

Witkind, I.J., 1956, Uranium deposits at base of the Shi-narump Conglomerate, Monument Valley, Arizona:Washington, U.S. Geological Survey Bulletin,v. 1030-C, p. 99–130.

Yingling, V.L., and Heller, P.L., 1992, Timing and recordof foreland sedimentation during the initiation of theSevier orogenic belt in central Utah: Basin Research,v. 4, p. 279–290.

Zaleha, M.J., Way, J.N., and Suttner, L.J., 2001, Effects ofsyndepositional faulting and folding on Early Creta-ceous rivers and alluvial architecture (Lakota andCloverly Formations, Wyoming, U.S.A.): Journal ofSedimentary Research, v. 71, p. 880–894.

MANUSCRIPT RECEIVED BY THE SOCIETY 20 JULY 2002REVISED MANUSCRIPT RECEIVED 24 FEBRUARY 2003MANUSCRIPT ACCEPTED 18 MARCH 2003

Printed in the USA



Documents

Geological Society of America Bulletin › wp-content › uploads › 2019 › 10 › Helle… · Geological Society of America Bulletin, September 2003 1123 TILTING OF THE U.S. CORDILLERA