Architectural element analysis of turbidite systems, and selected topical problems for sand-prone deep-water systems K.T. Pickeringt,j.D. Clark2, R.D.A. Smith\ R.N. Hiscott\ F. Ried Lucchi5 and N.H. Kenyon6

1Department of Geological Sciences, University College London, Gower Street, London WCIE 6BT, UK 1Department of Geology, University of Leicester, Lekester LEI 7RH, UK

·'Koninklijke/Shell Exploratie en Produktie Laboratorium, Shell Research BV, Volmerlaan 6, 2288 GD Rijswijk ZH, The Netherlands

'Department of Earth Sciences, Memorial University of Newfoundland, Canada, AlB 3XS

'Department of Geological Sciences, University of Bologna, Via Zamhoni 67, 40127 Bologna, ltaly

''Institute of Oceanographic Sciences, Deacon Laboratory, Brook Road, Wormley, Godalming, Surrey GU28 SUB, UK

lntroduction Deep-water deposits may be defined in terms of 'architectural elements' (cf. fluvial models of Miall 1985). In fluvial sedimentology, Miall's (1985) 'architectural elements' are characterized by fades assodations and three-dimensional geometry (including orientation). The identification of depositional geometry requires the classification of a hierarchy of bounding surfaces, an approach now widely accepted for fluvial and eolian deposits. Using a hierarchy of bounding surfaces, tagether with fades analysis and geometry, it is shown how 'architectural element analysis' may be used as an aid in the interpretation of deep-water systems and their depositional history.

For fluvial sediments, architectural element analysis has proved a useful tool for describing depositional units and for modeHing andent channel deposits (e.g. Allen 1982; Friend 1983; Miall 1985). Miall 0989) suggested the use of this technique for the interpretation of turbidite systems and demonstrated its advantages over more traditional vertical sequence analysis. Miall 0985) defined an architectural element as a 'lithosome characterized by its geometry, fades composition, and scale, and represents a particular process or suite of processes occurring within a depositional system' (Fig. 1). The recognition of architectural geometries inherently involves the identification of bounding surfaces and surface hierarchy. Hierarchical schemes have been implemented by many authors working on eolian and fluvial deposits (e.g. McKee and Weir 1953; Brookfield 1977; Allen 1983; Miall 1985). Allen's 0983) hierarchy

of bounding surfaces in fluvial sediments, developed from his study of the Devonian Brownstones of the Welsh Borders, is commonly used in the description of fluvial deposits.

The concept of elements in turbidite systerns was introduced by Mutti and Normark 0987) and is equally viable for modern and andent settings: the elements recognized include large-scale features only, such as channels, overbank deposits and lobes, tagether with scours which range across several orders of magnitude in scale. Theseare identified by fades assodations and a classification of the hierarchy of events. Depositional and erosional features may also be identified at smaller scales within turbidite systems, including, for example, similar elements to those defined by Miall 0985) in fluvial sediments. The characterization of outcrops of andent deposits into their component elements on a variety of scales, will help unravel complex depositional histories and fadlitate an understanding of the development of growth stages within turbidite systems. Not all sedimentary features of andent deep­water deposits, however, can be expressed by Miall's 0985) architectural element scheme, for example sediment slides, mud mounds and scour-and-fill features (on a variety of scales), and unlike Mutti and Normark's 0987) turbidite element scheme, the fluvial scheme is restricted to the description of andent deposits. A more methodical approach to this type of analysis is required, and therefore it is proposed to describe here architectural elements, characterized by various fades and scales, from both modern and andent environments.

Element analysis scheme Depositional bodies comprise architectural elements, defined between bounding surfaces which may be erosional or conformable. Here the term 'architectural element' is reserved for interpretive characterization of a sedimentary feature defined by its geometry (including orientation), scale and fades. An appropriate measure of scale may be attained from identifying the hierarchy of the bounding surfaces

Atlas of Deep Water Environments: Architectural style in turbidite systems. Edited by K.T. Pickering, R.N. Hiscott, N.H. Kenyon, F. Ried Lucchi and R.D.A. Smith. Published in 1995 by Chapman & Hall, London. ISBN 0 412 56110 7.

defining the architectural element. The element analysis scheme rnay be used to describe depositional and erosional features of both modern and andent systems (e.g. a gravel bar element may be a characteristic element in either a modern or andent submarine channel deposit). Figure 2 shows the philosophical framework used to classify and interpret modern and andent deep-marine environments.

The approach to depositional characterization of deep-water deposits is as follows: a bounding surface hierarchy scheme as applied in fluvial sedimentology allows deposits to be delineated at various scales into their architectural elements. The architectural elements are than classified using fades and 'architectural geometry' classification schemes. Fades classifications for deep-water depositsexist in the Iiterature (e.g. Mutti and Ried Lucchi 1975; Pickering et al. 1986a, 1989; Ghibaudo 1992).

Architedural geometry Architectural geometry is defined so as to be independent of scale and fades, at least on scales greater than metres but not scale independent and to include most fades. A classification of two-dimensional architectural geometries of what are, in fact, three­dimensional sedimentary features is preferred, since the majority of geological observations in both outcrop and seafloor observations are recorded in this manner. A three-dimensional classification is possible but its applicability would be limited, since very few case studies include such data quality. However, without the third dimension many features may appear similar, e.g. a large scour may have aspect ratios and a sedimentary fill identical to that of relatively small channels: their resolution and differentiation requires three dimensions. As with the application of any

Table 1 Prindpal architectural elements documented in chapters in this atlas

Planform elemenf' CH NCH M/L IRR BF SC/SCF

Chapter 12, 13, 14, 18 7, 27, 28, 1, 2, 3, 1, 2, 4, 1, 3, 4, 15, 17, 18 31, 41, 43 4, 5, 27, 12, 13, 18, 12, 13, 14,

42 27, 28, 30 27, 28, 30 31, 42

2D Sectional element' CH SH LE SI/I IRR SC/SCF

Chapter 8, 10, 13, 15, 6, 7, 32, 20, 21, 22, 4, 18, 20, 2, 3, 5, 1, 3, 5, 16, 17, 18, 33, 34, 35, 23, 24, 25, 7, 8, 9, 7, 9, 13, 19, 20, 21, 36, 37, 38, 29, 31, 33 11, 18, 20, 14, 18, 20, 22, 23, 24, 39, 40, 44, 34, 35, 39 21, 22, 23, 21, 22, 24, 25, 26, 28, 45, 46, 47, 40 24, 26, 29, 25, 27, 28, 29, 34, 40 48, 49 31, 32, 35, 29, 30, 31,

39,42 35, 42

"CH = Channel; NCH = nested channels; M = mounds; L = lobes; NM = nested mounds; IRR = irregular; BF = depositional bedform field; SC = erosional scour; SCF = scour field.

WF

1, 2, 13, 18, 27, 28, 30

bCH = Channel; SH = sheet; LE = lens; SI = sigmoid; I = inclined; IRR = irregular; SC = scour; SCF = scour-and-fill, WF = waveform.

•

•

descriptive classification scheme, there are inherent ambiguities which should be appreciated.

Figures 3 and 4 show classifications of depositional-erosional two-dimensional architectural geometry bothin section andin plan (planform), respectively. These are, in section: channel, sheet, Jens, sigmoid/ inclined, irregular, waveform and scour/ scour­and-fill. In plan, the following elements are recognized: channel, nested channels, mound/ lobes, nested mounds, irregular, bedform field and scour/ scour field. It should be noted that the 'channel' is an architectural geometry within which all the other geometries may be found.

II"' I UV141 ~ M " 1965

SB Sand bedf01111

LS L.am 11Kl sand

[ 02-20m

OF Ov n 11es

Although this classification of geometric elements is largely applicable over a !arge range of scales (but not scale independent) the limitations of remote sensing of modern systems and subsurface ancient systems will introduce an unavoidable scale dependency into any geometrical classification (Normark et al. 1979).

Bounding surfaces Bounding surfaces can be classified by type and order, both in cross-section and in plan. To date, geologists working in the ancient rock record have focused on the two-dimensional, cross-sectional form of bounding

surfaces. Allen (1983) recognized four types of bounding surface in the fluvial sediments of the Devonian Brownstones of the Welsh Borders: concordant non-erosional (normal bedding), discordant non-erosional, concordant erosional and discordant erosional contacts. All such contacts can be found in turbidite systems.

Bounding surface hierarchy classifications can also be readily applied to the submarine environment. Allen's (1983) classification is chosen here because of its simplicity, although fourth, fifth and sixth orders have been added to include basin-scale heterogeneity. Figure 5 shows schematic sections of deep-water

Hierarchy of

c: Bounding Surfaces

0 ...... c.. ·;:::: ACOUSIIC I Sonic (.) CJ) Sedimentary I Facies Q) I EchoType 0 Facies

~ - 1 ~ r

c: 0

:;::; ro (.) - Deposit Geometry Factes Associattons "(i:j I CJ) ro ü "'T -

~ Archttectural Elements I c: -'------

0 :;::; ro ...... Q) .... c.. .... Q)

Depositional ...... c: History -

tJ Depositional Model I

Fig. 1. Architectural elements in fluvial, deposits afterMiaU 0985). Fig. 2. Architectural characterization of modern and ancient depositional bodies in deep-water environments.

deposits, including a submarine channel, showing the bounding surface hierarchies and how the classification of these surfaces can easily be applied to sectional geometries of deep-water deposits. The normal concordant bedding contacts between strata and laminae form the zeroth-order of bounding surface. First-order surfaces bound, for example, packages of cross-bedding sets or concordant bedding. Typically, these surfaces are erosional (concordant or discordant). Second-order surfaces bound units delineated by first-order surfaces to form distinct sedimentary complexes of genetically related facies and/ or palaeocurrent directions. These complexes are equivalent to the 'storeys' of Friend et al. (1979) . Third­order surfaces are major erosional features dividing groupings of complexes (as delineated by second­order bounding surfaces). These units are commonly informally referred to in the Iiterature as depositional bodies. Fourth-order surfaces have been added to express erosional contacts which can range up to a basin-wide scale and define, for example , groups of channels and palaeovalleys; they are equivalent to Miall's (1985) sixth-order surfaces and separate Mutti and Normark's (1987) 'stages of growth' within an individual deep-marine system (their third-order of physical scale). Mappable Stratigraphie units, such as members or sub-members, are bounded by these fourth-order surfaces (MiaU 1989). Fifth-order surfaces define individual fan systems and are equivalent to surfaces defining Mutti and Normark's (1987) second­order of physical scale . Finally, sixth-order surfaces delineate basin-fill sequences and supergroups. These bodies are equivalent to Mutti and Normark's (1987) first-order of physical scale.

It should be noted that the appropriate hierarchy may differ between systems, simply because some turbidite systems are more punctuated into stages and sub-stages than others. In this respect, any hierarchy that is developed for one system should not be assumed a priori to be comparable to another .

Bounding surfaces seen in plan view are less easily defined by current classification schemes, but should nevertheless be used to help characterize architectural elements. Identification of such bounding surfaces , and bounding surface hierarchy, requires some interpretation of the element that is delineated by the bounding surface. For example , a major channel element seen on sidescan images will be defined by third-order bounding surfaces, using the similar diagnostic criteria as adopted for defining sectional bounding surfaces (see above).

Sectional classification of deep-water architectural element geometry (applicable over a wide range of scales)

Sigmold (SI) lncllned (I)

Waveform (WF)

Channel (CH)

lrregular (IRR}

Sheet (SH)

Lens (LE}

lrregular (IRR) (lncluding fnjection features)

Scour(SC) ~ Scour-and-fill (SCF} /////////// //

Planform classification of deep-water architectural geometry (applicable over a wide range of scales)

Nested channels (NCH) Mounds I Lobes (M I L) Nested mounds I Lobes (N M}

Deposltional Badform Reld (BF}

Facies Various facies schemes for deep-water sediments exist in the Iiterature (e.g. Mutti and Ricci Lucchi 1972, 1975; Mutti 1977, 1992; Pickering et al. 1986a, 1989; Ghibaudo 1992). In this paper the scheme of Pickering et al. (1989) (redrawn in Fig. 6) is adopted. In modern systems, core samples do not readily permit the mapping of sedimentary features on the seafloor, but where such information exists the sedimentary facies can be described as mentioned above. In the absence of such information, as in many modern turbidite systems, sonar facies can be characterized, but with equivocal interpretations as to their true sedimentological character. For planform analysis, acoustic facies from sidescan sonar images may simply be classified as high, moderate or low backscatter facies, with classifications of seismic facies in seismic profiles for sectional analysis already available (e.g. Damuth 1975, 1980; Droz and Bellaiche 1985; Pratson and Laine 1989).

Application of scheme Once the bounding surface hierarchy of modern and ancient deposits is established, the geometry can be classified using the scheme proposed here. Architectural geometries bounded by at least third­order surfaces are defined as third-order elements, second-order bounding surfaces define second-order elements and so on. Sedimentary or acoustic facies can be assigned to the various orders of architectural geometries. These steps Iead to a descriptive classification of the sedimentary feature. Using this description facilitates the interpretative characterization of the feature by assigning to it an 'architectural element'. For example, a pebbly sandstone and clast-supported conglomerate second-order lens element within a submarine canyon, may be interpreted as a second-order gravel bar architectural element. Architectural elements are defined, therefore, by geometry (including orientation), order of bounding surface and facies. Table 1 summarizes the principal architectural elements described in this atlas.

Comparing modern and ancient systems The description of deep-water sedimentary environments inevitably Ieads to camparisans being made with other ostensibly similar systems. Careful

consideration should be given to the appropriateness of selected camparisans between modern and ancient deep-water systems. In the past, classification of deep­water systems and their tectono-sedimentary setting have been, at least, implicitly assumed to compare similar features from modern and ancient systems (e.g. Stow 1986; Mutti and Normark 1987, 1991; Shanmugam and Moiola 1988). From the following discussion, the authors are sceptical that many of these modern and ancient camparisans are valid. The philosophy adopted for the schemes presented in this chapter is not concerned with comparing fan systems, type of basin, etc., but rather concentrates on describing the architectural elements within deep­water systems. Camparisans can then be made between similar architectural elements, or suites of elements.

Camparisans between geological features in deep­water systems must be made at similar spatial and temporal scales (Mutti and Normark 1987). In the scheme outlined in this paper, the importance of an enveloping bounding surface, or its hierarchy with respect to other surfaces, is used as an appropriate measure of scale. However, some care must still be exercised when comparing architectural elements of the same order, e .g. third-order channel elements may range considerably in size, but the importance of the bounding surface is only that it represents changes in depositional pattern; the interpretation that similar processes caused these changes is not justified without additional evidence.

The way in which geological and oceanographic observations are made for ancient and modern deep­water systems usually results in a Iack of common data sets. Field observations of ancient rocks provide detailed description of facies, facies-associations, vertical sequences and identification of meso- and microscale sedimentary structures. These observations are not commonly available in any synthesized form to workers on modern deep-water systems. The mapping of modern deep-water environments is commonly restricted to the identification of larger-scale features. Long-range sidescan sonar instruments (e.g. GLORIA) enable oceanographers and sedimentologists to map out in plan extensive areas of seafloor Sedimentation. The best high-resolution of GLORIA, however, is >100m: given that a typical field outcrop is commonly of this order of scale or smaller, the problems involved in identifying similar sized modern and ancient features become apparent. Detailed photogeological mapping of ancient systems may go some way to bridging this gap in observations (e.g. Sgavetti 1991).

Deep-towed sidescan instruments (e.g. TOBI,

•

•

Bounding surface hierarchy in outcrops of ancient deep-water deposits

B. ---

Fig. 5. Sehemarie illustrations of bounding surface hierarchy in outcrops of ancient deep-water deposits. A submarine channel is shown to demoostrate how this scheme may be applied to sectional geometries of deep-water deposits. The normal concordant bedding contacts between strata and laminae form the zeroth­order of bounding surface. First-order surfaces bound, for example, packages of cross-bedding sets or concordant bedding. Typically, these surfaces are erosional (concordant or disconcordant). Second-order surfaces bound units delineated by first-order surfaces to form distinct sedimentary complexes of genetically related facies and/or palaeocurrent directions. Third-order surfaces are major erosional features dividing groupings of complexes (as delineated by second-order bounding

surfaces). These units are often informally referred to in the Iiterature as depositional bodies. Fourth-order surfaces have been added to express erosional contacts which can range up to a basin-wide scale, and define, for example, groups of channels and palaeovalleys and are equivalent to Miall's (1985) sixth­order surfaces, and separate Mutti and Normark's (1987) 'stages of growth' wirhin an individual deep-marine system (their third-order of physical scale) . Fifth-order surfaces define individual fan systems, and are equivalent to surfaces defining Mutti and Normark's (1987) second-order of physical scale. Finally, sixth-order surfaces delineate basin-fill sequences and supergroups, and are equivalent to Mutti and Normark's (1987) first-order of physical scale.

SEAMARC) give a more detailed view of the seafloor (resolution <20m) and, although these instruments are not generally used for the mapping of extensive areas of deep-water systems, they can show architectural elements of the order of scale that could be identified from outcrop studies. Sidescan sonographs, most useful in modern marine studies, show plan-view sections of the sedimentation surface rarely obtainable in ancient outcrops at such scales. Good secrional data, however, can be obtained from outcrop studies and the standard of sectional correlations between outcrops cannot be matched using high-frequency seismic profilers, commonly used in the study of modern systems.

The importance of this scheme lies in the assumption that an architectural element defined from outcrop by facies, geometry and bounding surface hierarchy (which largely involves sectional analysis), can be compared with a similar architectural geometry defined by the same parameters in a modern deep­water system (from either plan or sectional analysis). For example, high-backscatter (or sidescan sonar), low-penetration (on high-frequency seismic profile), second-order bedform elements from a modern system, may be interpreted as relatively coarser grained second-order bedforms and used for camparalive studies and analogies with similar second­order ancient coarse-grained wave or bedform architectural geometry. Such a comparison does not necessarily Iead to the interpretation that the two elements formed from similar processes.

The channel types of Normark 0970), erosional, depositional and erosional-depositional or mixed, and their related fill deposits can be identified in both modern and ancient settings (Fig. 7). This submarine channel classification scheme is useful for modeHing sediment supply, controlled by intra-basin, tectonic and eustatic sea-level controls. The Eocene Ainsa I and Ainsa II channel complexes in the southern Centrat Pyrenees, provide good examples of mixed erosional-depositional channels (Mutti and Normark 1991). In the northern North Sea Paleogene hydrocarbon province, good examples of slope and lower slope erosional-depositional channel systems are documented from the Paleocene Baider Formation in UK Quadrant 9 by Timbrell 0 993), the Paleocene-Eocene Gryphon Field by Newman et al. 0 993) and the Eocene Alba Field in UK Block 16/ 26 (Newton and Flanagan 1993).

CLASS

(/) "0 c Cl3

Cf)

m

0 LlJ ~ z ~ a: 0 Cl)

Ci <C

0 LlJ Cl)

z ~ a: 0 Cl)

Ci CD

u

0 LlJ ~ z ~ a: ~ Ci w

A1.1

81.1 81 .2

C1 .2

01.1 01.2

E1.1 E1.2

~ o_ o -::."1 ooo.

~ A1 .3

01 .3

E1.3

GROUP

0 LlJ Cl)

z <C (!) a:

A2.1

0

~m A2.5

0

A2.2

F!JJ A2.6

II I 0 ~ 82.1 82.2

rs c2.1 C2.2

~ 02.1 02.2

0

~ ~ ~ 0 fij E2.1 E2.2

A2.3 A2.4

~ A2.7 A2.8

C2.3 C2.4

02.3

F1 EXOTIC CLASTS F2 GONTORTED & DISTURBED STRATA

.~ (/) --o ·-Cll(/) .s=o üaf LL"'

(/)

u "0 -·- ::I Cl3 (/) c -Eu_ <D (/) E·-e»<Du (/) 0 N ·- <D 0 ·- 0 c .s= a. !ll 0 <Dü <D '"' 0> "0 "' 0~

jj)

F1.1 F1 .2 F2.1

G1 BIOGENIC OOZES G2 BIOGENIC MUDS

G1 .1 G1.2 G2

Fig. 6. Classification scheme for deep-water sedimentary facies after Pickering et al. (1986a, 1989).

~ F2.2

G3 CHEMICAL DEPOSITS

r---':tJ .. 'f/1

• 'f/1. 'f/1 ..

~ G3

A.

B.

C.

Fig. 7. Three channel types: A, erosional; B, erosional-depositional (or mixed type); C, depositional (cf. Normark 1970; Mutti and Normark 1987).

Channel development and infill In many documented ancient turbidite exposures it is not possible to 'walk-out' beds from within a channel into the levee or overbank deposits. The main reason for this paradox is that many ancient channels, although comparable in scale to modern middle-fan channels, are erosional and erosional-depositional channel complexes developed within basin-slope sediments, rather than the aggradational (essentially depositional) channel-levee-overbank complexes common to many large-radius, very low-gradient modern fans. There are good reasons for this discrepancy between modern and ancient systems. Apart from accretionary prisms, most of the ancient rock record in deep-marine/ deep-water systems represent the vestiges of upper continental slope, intra­sbelf or aulacogen-related deposits, where basins are up to orders of magnitude smaller than continental margins and ocean basins, with few, if any, essentially flat basin floors and where slope gradients are commonly high. A corollary of this is that erosional and erosional-depositional, low- to moderate-sinuosity channel systems should prevail.

In ancient outcrops, one of the main reasons for the difficulty of tracing beds from intrachannel to extrachannel sites results from the recognition of only

the lower parts of channel-fills, and possibly even misidentifying parts of canyon-fills as middle-fan channels. For example, within the Guaso channel complex in the Eocene Hecho Group, Spanish Pyrenees, there are mid-fan-scale channel-like sandstone bodies within canyon systems (Stocchi 1992 in Mutti 1992). Channel depth estimates rarely include much fine-grained fill , contrary to the observations in some modern fan channels (e.g. Pickering et al. 1986b). In ancient outcrops, the inter-sandstonepacket fine-grained lithologies (muds and silts) are commonly poorly exposed at critical outcrops, and may have undergone severe post-depositional, early wet­sediment and tectonic deformation to obscure original bedding relationships. Such severe limitations frequently conspire to encourage under-estimation of channel dimensions and the interpretation of packets of fine-grained sediments more than metres in thickness as 'levee', 'interchannel' or 'overbank' deposits. This means that the lower parts of many ancient channel fills, commonly recognized by large­scale erosional surfaces overlain by coarse-grained facies (Facies Classes A-C, Pickering et al. 1986a, 1989), may represent only part of much !arger erosional-depositional sediment conduits, for example, as exemplified in the Ainsa I channel margin, Eocene Hecho Group, Spanish Pyrenees, d assie onlap relationships. In these, and many other ancient examples, much of the fine-grained facies probably represents mainly slope sediments sensu stricto into which the channels were incised, and possibly much of the channel-fill may be muds and silts, not coarser sediments.

With the above caveats in mind, and assuming that most ancient documented channel complexes are erosional and erosional-depositional (sensu Mutti 1985), the following general channel-fill model is proposed (Fig. 8 (a and b)).

Phase I - Erosion, through-flow and non-deposition

The initial phase of channel development involves the excavation, commonly by multiple discrete erosional events, of a complex, often stepped or benched, erosional surface. During this phase, small relatively straight slope gullies may form on the slope and grow both by headward erosion and an increase in sinuosity (cf. canyon development, Twichell and Roberts 1982; McGregor et al. 1982). The initial position of channels will be determined by many factors, both extra- and intra-basinal, the specifics of which must be determined individually for each channel complex. As

•

/ residuaVIag deposits

inner bank fines c-1 ("levee" facies) cross-bedded I

Erosional Phase sands ""- pebbly sands

• E rosional-depositional Phase

fining-upward sequence r/

Backfilling Phase ~

abandonment muds/silts

----- I ~~~~~~:::::::::::::::::~~~~~

""111!!::::

Abandonment Phase ~~

-::~ ~

......_ Onlap direction - lsochronous surface emphasized

residual/lag deposits I c-1 levee development

cross-bedded I Erosional Phase sands ~ pebbly sands

~> ::::::s J2& --~

Erosional-depositional Phase ~ --

fining-upward

~~ ~~ sequence

A/ ""'~~!@!::::

Backfilling Phase ~ ~ ::::

--------- I abandonment muds/silts

~---:-·::::.~-::.-~ ~~~:-::~-::-::==-::===: - ::---- .

~I Abandonment Phase ~ -......_ Onlap direction - lsochronous surface emphasized

Mainly bypass through the channel to distal part of system. "Levee" facies on benches and highs.

Many flows still bypass channel, but deposition within channel becoming significant. Thaiweg channel infill and switching.

Main phase of channel infill. Onlap of channel sands onto inner banks of channel. Channel as conduit for sand-laden flows reduced.

Channelabandoned. Residual topography infilled by silts and muds, incl. bank collapse

(a)

Rapid development of levees by flow-stripping from flows that mainly bypass the channel to distal part of system

Many flows still bypass channel, but deposition within channel becoming significant. Thaiweg channel infill. Levee growth slowed.

Main phase of channel infill. Onlap of channel sands onto inner banks of channel. Levee growth very slow to insignificant. Channel as conduit for sand-laden flows reduced.

Channelabandoned. Residual topography infilled by silts and muds, incl. bank collapse

(b)

the channels develop and extend in both a headward and basinward direction, the relatively straight courses will tend to become more sinuous, particularly where gradients are lower (Clark et al. 1992).

During Phase I, the channel acts essentially as a conduit for the through-put of channelized sediment gravity flows and is, therefore, the time of maximum development of channellag deposits (e.g. as pebble stringers, etc.), reworking of previously deposited sands into bedform fields (e.g. cross-stratified sands) and irregular and localized scouring of the channel floor. Additionally, channel overspill and flow-stripping are most common during Phase I times, leading to maximum growth of the levees, because through­channel flows have their greatest competence and capacity during Phase I compared to other times of channel development. Phase I is associated with the most irregular bed geometries, irrespective of the sediment fades.

Phase II - Deposition This is the main phase of channel infill by quintessentially coarse-grained fades, in which back­filling, sand-bar aggradation, point-bar growth (if relevant) and other depositional processes occur. It is during this phase that the erosional channel walls are typically onlapped by more sheet-like beds compared to Phase I deposition. Phase II is associated with the most commonly described relatively coarse-grained intrachannel fades.

Phase 111 - Abandonment The transition from Phase II to Phase III deposition has the greatest propensity to generate thinning-and-fining­upward sequences, as the channel becomes abandoned as a conduit for sandy flows and their deposits. Channel abandonment may be caused by many processes, e.g. channel avulsion further up­system and the progressive plugging of the newly constructed channel margin-levee, leading to systematically less overspill of coarser sediments along

Fig. 8. (Left) Channel-fill models to show three principal phases in the history of a channel. Changes in the channel phases are related to changes in relative base Ievel, i.e. variations in accommodation space, sediment supply and eustasy. (a) Canyon and non­leveed channels; (b) channel-levee-overbank complexes.

the older conduit, or an extra-basinal change in base Ievel, such as a relative rise in sea Ievel or decreased tectonic activity in the source region. During Phase III the channel may be partially or completely filled by fine-grained fades.

Repetition of Phases 1-111

It should be noted that any channel history may involve the repetition of any of the above phases, e.g. if the channel is reactivated because of changes in base Ievel, it may go from Phase III to Phase I or Phase II. Consequently, the vertical and lateral fades relationships observed within channel-fills are commonly both complex and unpredictable. For example, any vertical sequence through a channel-fill may show several phases of deep scouring overlain by channel lag deposits, and it is at such times of channel reactivation that a channel may dramatically shift or migrate sideways. A good example of this process is seen in the Ainsa II channel complex, Eocene Hecho Group, Spain, or the late Precambian Kongsfjord Formation, N. Norway.

Stacking architecture of channel sandbodies appears very variable (Fig. 9). Channels may show varying combinations of: 1. gradual lateral migration (.±. point­bar development); 2. offset stacking; 3. vertical aggradation; 4. chaotic stacking. The interplay between tectonics, sediment supply and eustasy naturally control the actual development of channels and their infill history. Third-order channel elements showing sequential offset stacking relationships, such as the eight nested channels of the Capistrano Formation (Walker 1975), indicate progressive lateral shifting of the channel axis, due to either channel migration or avulsion (Fig. 10). The well-studied Ainsa II channel of the Eocene Hecho Group, in the South Central Pyrenees, is another example of a channel showing third-order offset stacked channel bodies (Clark, Ch. 20). Second-order offset stacked channel elements are seen in turbidite channels in the late Precambian Kongsfjord Formation, N. Norway (Pickering 1983). Purely autocyclic processes that result in a shift of channel axis can be modelled to demonstrate the stacking architecture and channel body connectivity. Reservoir models exist for fluvial channels demonstrating that the interconnectivity and distribution of channels within the flood plain deposits is related to: (a) laterally variable aggradation; (b) compaction of fine sediment; (c) tectonic movement at flood plain margins and (d) channel avulsion (Bridge and Leeder 1979). Thesemodels do not account for fault-controlled channel axis location. In ancient

sequences deposited in tectonically active settings, submarine channel courses may be strongly controlled by syn-sedimentary faulting, and such ancient deposits will show vertically stacked channel architecture. Detailed correlations of the Llandovery Caban Conglomerate Formation, South Central Wales, constrained by good graptolite biostatigraphic subdivisions (Smith et al. 1991), demonstrate vertical stacked architecture of the channel conglomerate elements controlled by syn-sedimentary faulting (Davies and Waters, Ch. 26).

In the Paleogene of the northern North Sea, the Lower Eocene Baider Sands and Frigg Sand, above the Baider Clays, show an eastward to north-eastward lateral offset stacking of submarine channels (Timbrell

Embedded (e.g., within canyon)

1993), which compares well in scale with the Eocene Ainsa II channel growth pattern from the Spanish Pyrenees. Also, as with the tectonic control on the offset in the Ainsa II channels, the lateral offset in the Baider Sands channels appears to have been controlled by faulting. Offset-stacked channels appear particularly common where advancing thrust stacks cause a consistent, basinward relocation of depocentres and associated channel systems, as in submarine trenches and foreland basins, e.g. the progressive southward shift with time in the stacking of the channel complexes of the Eocene Hecho Group, Spanish Pyrenees.

lsolated stacked

Vertically stacked

Offset stacked

Offset stacked lsolated single

Fig. 9. (Above) Stacking patterns in submarine channels.

Offset stacked (nested)

Fig. 10. (Right) Architectural element interpretation of San Clemente beach section in Upper Miocene Capistrano Formation, California (redrawn from Walker 1975) . Note the gradual southward shift in palaeocurrents between successive channel-fills.

0 40m

Post-depositional modification features Post-depositional modification of channel deposits exerts a primary influence on the observable features and characteristics of ancient channels and can control the location of subsequent channels in active deep­marine systems. Post-depositional modification features include clastic dykes and sills, growth faults and differential compaction.

Fluidization and liquefaction features, such as clastic dykes and sills, are documented from many ancient deep-water systems at outcrop, e.g. associated with slope sandstone gully fills (Surlyk 1987), inferred submarine channel-fills (Hiscott 1979; Pickering 1981,

CH 8

CH 7

CH 6

r.(" Channel lag elemenl

1983; Smith and Spalleti, Ch. 24). In the subsurface, good examples of large-scale sand-rich injection structures in the Paleogene of the northern North Sea include sand dykes up to many metres wide intruding vertically up through tens of metres of sediments in the Baider Formation, as described by Jenssen et al. (1993), and on a decimetre scale in core from the Gryphon Field (Newman et al. 1993).

A common feature recorded in many industry subsurface cores, e.g. from the Paleogene of the northern North Sea, is the presence of apparently chaotic mud-flake breccias and conglomerates with a sandy matrix. Although, on the basis of micropalaeontology and composition, including colour, some of these deposits are interpreted as sediment

~ QZ]

I

CH4

Mainly classic turbidites with interbedded shales.

Thick-bedded pebbly sandstones

Sillstonesand mudstones

Palaeocurrent directions (shown oriented with respect to North)

CH 3

CH 2

CH 1 .· Wi

Eight stacked (nested) channels, Capistrano Formation, San Clemente Beach, California

•

•

slides and debris flow deposits (with extra- and intra­formational clasts), there are many cases where it appears that the deposit was formed by the pervasive injection of turbiditie and hernipelagie muds (see below).

Growth faulting, partieularly at channel margins associated with differential compaction, may encourage the offset-stacking of channel sandbodies, even very early after deposition when only perhaps tens of metres of additional stratigraphy have accumulated. Growth faults also may act as surfaces and zones of weakness along whieh large-scale wet­sediment intrusions occur.

Differential compaction of ribbon-like channel sandbodies and their associated overbank-levee deposits can Iead to inversion of the primary topographie relief as the sands compact less than the enveloping muds. Such post-depositional processes may be associated with the development of growth faults along the margins of the buried channels and even the injection of sandy clastie dykes along such faults.

Sequence stratigraphy and architedural elements The 1980s witnessed the development of the hypothesis of sequence stratigraphy, the progeny of seismie stratigraphy from the 1970s (Mitchum 1977; Mitchum et al. 1977). Useful definitions include 'Sequence stratigraphy is the study of rock relationships within a chronostratigraphie framework of repetitive, genetieally-linked strata bounded by surfaces of erosion or non-deposition, or their correlative conformities' (van Wagoner et al. 1988), and, 'Sequence stratigraphy is the study of genetieally related fades within a framework of chronostratigraphieally signifieant surfaces. The sequence is the fundamental stratal unit for sequence Stratigraphie analysis' (van Wagoner et al. 1990). Sequence stratigraphy provides a useful framework for understanding the interplay between accommodation space, subsidence and/or uplift, and rates of sea-level change in many sedimentary environments (e.g. van Wagoner et al. 1988, 1990).

Although philosophieally similar in many respects to the Exxon approach, Galloway 0989a, b) put forward the concept of a genetic Stratigraphie sequence. A genetic Stratigraphie sequence is the sedimentary product of a depositional episode, each sequence comprising: 1. a progradational facies-association; 2. an aggradational facies-association; 3. a retrogradational or transgressive facies-association. Genetie sequences

are bounded by a sedimentary veneer or surface that records the depositional hiatus that occurs over much of the transgressed shelf and adjacent slope during maximum marine flooding.

Einseie 0985) considered the response of sediments to sea-level changes in differing storm-dominated margins and epeiric seas, partieularly the Mesozoie epieontinental, mud-dominated, seas in the slowly subsiding basins of Germany. Such basins cantrast markedly with the rapidly subsiding shelf-margin seas subject to rapid changes in sea Ievel that are glacio-eustatieally driven, and from whieh much of the Exxon philosophy is based. He pointed out that the base Ievel to whieh sediments aggrade is the storm wave baserather than sea Ievel sensu stricto. Einseie 0985) believes that sediment accumulation patterns are partieular to depositional sites and conditions.

Perhaps, the principal difference between the Exxon model and Einsele's 0985) perspective is that: 1. regressions are perceived as gradual rather than abrupt; 2. purely aggradation depositional units reach their maximum thiekness where basin subsidence is most rapid, for example in the centre of basins; 3. the base Ievel to which sediments may aggrade is the storm wave base.

Although the global synchroneity, or correlatability, of events in relation to the proposed eustatie sea-level curves presented by Haq et al. (1977, 1988) has been undermined (see discussion by MiaU 1992), the principles of sequence stratigraphy, the associated ways of dividing up Stratigraphie units and interpreting many of the causal mechanisms, remain valid. For these reasons, a thorough understanding of the basie principles of sequence stratigraphy are important, but at the present time their applicability in deep-marine environments remains more problematical than is the case in shallow marine settings. Fora useful review of the nature and causes of cyclicity in turbidite and related deep water systems, see Cycles and Events in Stratigraphy, edited by Einseie et al. 0991).

Relative lowstands in sea Ievel tend to be associated with lowstand turbidite currents as basin-floor and/or slope fans, for example, as has been demonstrated for the Paleogene of the Central and northern North Sea (Armentrout et al. 1993; Den Hartog Jeger et al. 1993; Galloway et al. 1993; Vining et al. 1993). This is because a fall in relative sea Ievel, which may be eustatically driven, Ieads to the effective narrowing of shallow-marine shelves and the propensity for fluvial/coastal systems to prograde towards, even reaching, the shelf-slope break. Additionally, under conditions of lowered sea Ievel, shallow-marine processes, such as severe storm wave pounding, are

more likely to enhance sediment failure along basin slopes and, therefore, further increase sediment delivery rates into deep-marine environments.

Teetonic and/or climatically driven changes in the source area, even in the absence of any significant changes in relative sea Ievel relative to the coastline, can increase sediment delivery rates to the deep sea and, therefore, favour the growth of deep-marine clastie systems (e.g. Kolla and Macurda 1988). For example, Mississippi Fan sandy turbidite Sedimentation, initiated during the falling and maximum relative lowstand stages of sea Ievel during the last glacio-eustatie cycle, continued into the Holocene mid to late sea-level rise at about 12,000-11,000 years ago, due to: (i) landward extension of the Mississippi Canyon into the mid-shelf water depths as sea Ievel rase; (ii) increased glacial meltwater discharge and pebble to clay size sediment Ioads delivered directly to the canyon head during rising sea Ievel; (iii) persistent interception of Iangshare drift by the canyon as it eroded headward; (iv) steep gradients at the canyon head favoured sediment failure as sediment slides, debris flows and turbidity currents, and (v) sediment bypass of coarse­grained material through the canyon into deep water with the absence of deltaic stratal patterns within the canyon (Kolla and Perlmutter 1993). Kolla and Perlmutter 0993) note that the late sand-prone turbidite sedimentation in the Mississippi Fan is compatible with the occurrence of sandy turbidites in the middle Amazon Fan subsequent to 13,285±650 years ago. However, in both the Mississippi and Amazon Fan these late sand-prone phases of fan growth may coincide with at least the earliest part of the Younger Dryas global cooling and, therefore, represent renewed lowstand sand accumulation, an aspect requiring further evaluation. Furthermore, in a discussion of lowstand deep-water siliciclastic depositional systems, Kolla 0993) has shown that the terms 'basin-floor fan' and 'slope fan' are too restrictive both literally and conceptually to represent the many aspects of such systems. We therefore prefer the term 'active fan growth phase' to lowstand system.

In order to better constrain the sediment types, grain sizes and rates of sediment accumulation, and the fundamental controls on deposition, future research must focus on providing high-resolution stratigraphy in areas where chronostratigraphy surfaces can be correlated from basin to shelf. Chemical methods, including isotopie analysis, provide a major way forward, partieularly where global and/or regional climate change has exerted a major influence on deposition.

Problem of structureless {massive) sands One of the current areas of considerable intellectual curiosity concerns the origin of deep-water structureless (massive) sands, not least because it is fuelled by the economic importance of such sands as hydrocarbon reservoirs. The origin of massive ( very thick and essentially structureless) sands is due to a variety of processes, involving both steady and unsteady flows, principally:

1. Deposition from high-concentration turbidity currents where grain concentration at the depositional surface is sufficiently high so as to cause hindered settling. Some sands containing scattered outsize clasts may originate from this process.

2. Deposition from sandy debris flows, with or without outsize clasts.

3. Syn- to post-depositionalliquefaction and fluidization to destroy any primary sedimentary structures. This process probably accounts for beds where the only sedimentary structures are fluidization pipes (pipe structure) and dish structure.

Problem of outsize mud clasts in sandstones Outsize semi-lithified mud-silt clasts and/ or clasts comprising very thin-bedded/laminated sand, silt-mud couplets, are commonly documented from many sandstones in ancient outcrop and subsurface deep­water systems. Amongst the processes responsible for their genesis, are:

1. Long-distance transport by turbulent suspension followed by flow transformation (see Fisher 1983) to high-concentration, cohesion-dominated, debris flow during deposition.

2. Long-distance transport by turbulent suspension followed by flow transformation to high­concentration, bindered settling dominated flow during deposition.

3. Deposition from sandy debris flows. 4. Entrainment of clasts immediately prior to

deposition, leading to rapid flow deceleration and flow transformation.

5. Two-layer flow with high-density, relatively high­viscosity, slow, lower layer and faster lower­density upper turbulent layer, in which the clasts

concentrate at the interface between layers (Postma et al. 1988).

6. Two-layer flow with upper layer of plug-flow

(containing outsize clasts) moving above a lower,

high shear stress layer, where the outsize clasts

may be sheared and progressively deformed in

the plug (Clayton 1992).

7. Amalgamation of two sands beds and liquefaction

welding to destroy local evidence for multiple

flow events.

8. Pervasive sand injection (by fluidization

processes) into neighbouring, commonly

overlying, muds, silty mudstones and siltstones.

Where the sand injection is into the upper, mud­

rieb part of the same bed the process is referred

to as 'autobrecdation'.

9. Sediment slides, leading to the mixing of

unconsolidated sand and cohesive, semi-lithified

muds.

Endpiece The advantages of the architectural element scheme

presented here over other analytical techniques is

twofold. First, for individual systems, architectural

element analysis provides a useful tool in unravelling

the complexities of different sedimentological stages of

development, or characterizing the sedimentology of a

particular development stage in the deep-water

environment. In the examples shown here, simply

applying the architectural element scheme at a purely

descriptive level does not provide any new

interpretations of the sedimentology over the work

done by the original investigators. The method of

architectural analysis, however, provides a unifying

classification scheme for much less ambiguous

communication between researchers working on

modern and andent channel complexes, or deep-water

systems in general. For example, the scheme must be

used flexibly in order to accommodate different styles

of system organization.

The scheme presented here is also an improvement

on the fluvial scheme of Miall (1985) and its extension

to the description of andent turbidite deposits (Miall

1989). The classification, adopted here, of architectural

geometries (applicable over a wide range of scales and

fades, but not scale independent), permits deep-water

deposits to be broken down into a greater range of

architectural elements based on their geometry and

fades.

Heferences Allen, J.R.L. 1982. Sedimentary structures: their character and

physieal basis. Developments in Sedimentology 30 (Parts I and II). Elsevier, Arnsterdam, 593 and 663 pp., respectively.

Allen, J.R.L. 1983. Sturlies in fluviatile sedimentation: bars, bar-complexes and sandstone sheets (low-sinuosity braided streams) in the Brownstones (L. Devonian), Welsh Borders. Sedimentary Geology, 33, 237-293.

Armentrout, J.M., Malecek, S.J., Fearn, L.B., Sheppard, C.E., Naylor, P.H., Miles, A.W., Desmarais, R.J. and Dunay, R.E. 1993. Log-motif analysis of Paleogene depositional systems tracts, Central and Northern North Sea: defined by sequence Stratigraphie analysis. In: Parker, J.R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conjerence. The Geologieal Society, London, pp. 45-57.

Brookfield, M.E. 1977. The origin of bounding surfaces in ancient aeolian sandstones. Sedimentology, 24, 303-332.

Bridge, J.S. and Leeder, M.R. 1979. A simulation model of alluvial stratigraphy. International Association of Sedimentologists, 26, 617-644.

Clark, J.D., Kenyon, N.H. and Piekering, K.T. 1992. Quantitative analysis of the geometry of submarine channels: Implications for the classifieation of submarine fans. Geology, 20, 633-636.

Clayton, C.J. 1992. Contrasting sediment gravity flow processes in the Late Llandovery, Rhuddnant Grits turbidite systems, Welsh Basin. Geological Magazine.

Damuth, J.E. 1975. Echo-character of the Western Equatorial Atlantie floor and its relationship to the dispersal and distribution of terrigenaus sediments. Marine Geology, 18,

17-45. Damuth, J.E. 1980. Use of high-frequency (3.5-12kHz)

echograms in the study of deep-bottom Sedimentation processes in the deep-sea: A review. Marine Geology, 38,

51-75. Den Hartog Jager, D., Giles, M.R. and Griffiths, G.R. 1993.

Evolution of Paleogene submarine fans of the North Sea in space and time. In: Parker, J.R. (ed.) Petroleum Geology oj Northwest Europe: Proceedings oj the 4th Conjerence. The Geologieal Society, London, pp. 59-71.

Droz, L. and Bellaiehe, G. 1985. Rhone deep-sea fan: morphostructure and growth pattern. 1be American Association of Petroleum Geologists Bulletin, 69, 460-479.

Einsele, G. 1985. Responses of sediments to sea-level changes in differing subsiding storrn-dorninated marginal and epeirie basins. In: Bayer, U. and Seliacher, A. (eds) Sedimentary and Evolutionary Cycles. Springer-Verlag, Berlin, pp. 68-112.

Einsele, G., Ricken, W. and Seilacher, A. (eds) 1991. Cycles and Events in Stratigraphy. Springer-Verlag, Berlin, 955 pp.

Fisher, R.V. 1983. Flow transforrnations in sediment gravity flows. Geology, 11, 273-274.

Friend, P.F. 1983. Towards the field classification of alluvial

architecture or sequence. In: Collinson, ].D. and Lewin,]. (eds) Modern and Ancient Pluvial Systems. International Association of Sedimentologists Special Publication, 6, 345-354.

Friend, P.F., Slater, M.J. and Williams, R.C. 1979. Vertieal and lateral building of river sandstone bodies, Ebro Basin, Spain. Journal of the Geological Society, London, 136, 39-46.

Galloway, W.E. 1989a. Genetic Stratigraphie sequences in basin analysis 1: Architecture and genesis of flooding­surface bounded depositional units. 1be American Association ojPetroleum Geologists Bulletin, 73, 125-142.

Galloway, W.E. 1989b. Genetic Stratigraphie sequences in basin analysis II: Application to northwest Gulf of Mexico Cenozoie basin. 1be American Association of Petroleum Geologists Bulletin, 73, 143-154.

Galloway, W.E., Garber, J.L., Xijin Liu and Sloan, B.J. 1993. Sequence Stratigraphie and depositional framework of the Cenozoie fill, Central and Northern North Sea Basin. In: Parker, J.R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conjerence. The Geologieal Society, London, pp. 33-43.

Ghibaudo, G. 1992. Subaqueous sediment gravity flow deposits: Practieal criteria for their field description and classification. Sedimentology, 39, 423-454.

Haq, B.U., Hardenbol,]. and Vail, P.R. 1987. Chronology of fluctuating sea Ievels since the Triassic. Science, 235, 1156-1167.

Haq, B.U., Hardenbol,]. and Vail, P.R. 1988. Mesozoie and Cenozoic chronostratigraphy and eustatic cycles. In: Wilgus, C.K. et al. (eds) Sea-level Research: An integrated approach. Society of Econornic Paleontologists and Mineralogists Special Publication, 42, 71-108.

Hiscott, R.N. 1979. Clastic sills and dikes associated with deep-water sandstones, Tourelle Formation, Ordovician, Quebec. Journal ojSedimentary Petrology, 49, 1-10.

Jenssen, A.l., Bergslien, D., Rye-Larsen, M. and Lindholm, R.M. 1993. Origin of complex mound geometry of Paleocene submarine-fan reservoirs, Baider Field, Norway. In: Parker, J.R. (ed.) Petroleum Geology oj Northwest Europe: Proceedings ofthe 4th Conjerence. The Geological Society, London, pp. 135-143.

Kolla, V. 1993. Lowstand deep-water siliciclastic depositional systems: characteristics and terrninologies in sequence stratigraphy and sedimentology. Bulletin Centres Recherehes Exploration-Production Elf Aquitaine, 17, 67-78.

Kolla, V. and Macurda, D.B. Jr. 1988. Sea-level changes and timing of turbidity-current events in deep-sea fan systems. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G. St. C., Posamentier, H.W., Ross, C.A. and Van Wagoner, J.C. (eds) Sea-Level Changes-An Integrated Approach. Society of Econornic Paleontologists and Mineralogists Special Publication 42. Tulsa, Oklahoma: The Society of Economic Paleontologists and Mineralogists, pp. 381-392.

Kolla, V. and Perlmutter, M.A. 1993. Timing of turbidite sedimentation on the Mississippi Fan. 1be American

Association of Petroleum Geologists Bulletin, 77, 1129-1141.

McGregor, B.A., Stubblefield, W.L., Ryan, W.B.F. and Twichell, D.C. 1982. Wilrnington submarine canyon: a marine fluvial-like system. Geology, 10, 27-30.

McKee, E.D. and Weir, G.W. 1953. Terrninology of stratification and cross-stratification. Geological Society oj America Bulletin, 64, 381-390.

Miall, A.D. 1985. Architectural-element analysis: A new method of fades analysis applied to fluvial deposits. Earth Science Reviews, 22, 261-308.

Miall, A.D. 1989. Architectural elements and bounding surfaces in channelized clastic deposits: notes on camparisans between fluvial and turbidite systems. In: Taira, A. and Masuda, F. (eds) Sedimentary Facies in the Active Plate Margin. Terra Scientific Publishing Company (TERRAPUB), Tokyo, pp. 3-15.

Miall, A.D. 1992. Exxon global cycle chart: An event for every occasion? Geology, 20, 787-790.

Mitchum, R.M. 1977. Seismic stratigraphy and global changes of sea Ievel, Part 1: Glossary of terrns used in seisrnic stratigraphy. In: Payton, C.E. (ed.) Seismic Stratigraphy­Applications to Hydrocarbon Exploration. American Association of Petroleum Geologists Memoir (Tulsa), 26,

205-212. Mitchum, R.M., Vail, P.R. and Thompson III, S. 1977. Seisrnic

stratigraphy and global changes of sea Ievel, Part 2: The depositional sequence as a basic unit for Stratigraphie analysis. In: Payton, C.E. (ed.) Seismic Stratigraphy­Applications to Hydrocarbon Exploration. American Association of Petroleum Geologists Memoir (Tulsa), 26,

53-62. Mutti, E. 1977. Distinctive thin-bedded turbidite fades and

related depositional environments in the Eocene Hecho Group (south-central Pyrenees, Spain). Sedimentology, 24, 107-131.

Mutti, E. 1985. Turbidite systems and their relations to depositional sequences. In: Zuffa, G.G. (ed.) Provenance oj Arenites. NATO Advanced Scientific Institute Series. D. Reidel, Holland, pp. 65-93.

Mutti, E. 1992. Turbidite Sandstones. Agip, Milan, 275 pp. Mutti, E. and Ried Lucchi, F. 1972. Le torbidit dell'Apennino

settentrionale: introduzione all'analisi di fades. Memoir Society ojGeology Italy, 11, 161-199. (English translation by Nilsen, T.H. 1978. International Geological Review, 20,

125-166). Mutti, E. and Ried Lucchi, F. 1975. Turbidite fadesandfades

associations. In: Examples oj Turbidite Facies and Facies Associations jrom Selected Formations of the Northern Apennines. Field trip guidebook A-11, IX International Congress Sedimentologists, Nice, France. International Association Sedimentologists, 21-36.

Mutti, E. and Norrnark, W.R. 1987. Comparing examples of modern and ancient turbidite systems: Problems and concepts. In: Leggett, J.K. and Zuffa, G.G. (eds) Marine Clastic Sedimentology. Graham and Trotrnan, London, pp. 1-38.

•

Mutti, E. and Normark, W.R. 1991. An integrated approaeh to the study of turbidite systems. In: Link, M.H. and Weimer, P. (eds) Seismic Facies and Sedimentary Processes of Submarine Fansand Turbidite Systems. Springer-Verlag, pp. 75-106.

Newman, M. St. )., Reeder, M.L., Woodruff, A.H.W. and Hatton, I.R. 1993. The geology of the Gryphon Field. In: Parker, J.R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference. The Geologieal Soeiety, London, pp. 123-133.

Newton, S.K. and Flanagan, K.P. 1993. The Alba Field: Evolution of the depositional model. In: Parker, J.R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference. The Geological Society, London, pp. 161-171.

Normark, W.R. 1970. Growth patterns of deep-sea fans. Tbe American Association of Petroleum Geologists Bulletin, 54, 2170-2195.

Normark, W.R., Piper, D.W.J. and Hess, G.R. 1979. Distributary ehannels, sand Iobes, and mesotopography of Navy Submarine Fan, California Borderland, with applieations to aneient fan sediments. Sedimentology, 26,

749-774. Phillips, S. 1987. Dipmeterinterpretation of turbidite-ehannel

reservoir sandstones, Indian Draw Field, ;\Iew Mexieo. In: Tillman, R.W. and Weber, K.J. (eds) Reserooir Sedimentology. Soeiety of Eeonomie Paleontologists and Mineralogists Special Publieation, 40, 113-128.

Pickering, K.T. 1981. The Koogsfjord Formation- a late Preeambrian submarine fan in north-east Finnmark, north Norway. Norges Geologiske Undersogelse, 367, 77-110.

Piekering, K.T. 1982a. The shape of deep-water silieiclastie systems- a diseussion. Geo-Marine Letters, 2, 41-46.

Piekering, K.T, 1982b. Middle-fan deposits from the late Preeambrian Koogsfjord Formation Submarine Fan, northeast Finnmark, northern Norway. Sedimentary Geology, 33, 79-110.

Pickering, K.T. 1983. Transitional submarine fan deposits from the late Preeambrian Koogsfjord Formation submarine fan, NE Finnmark, N. Norway. Sedimentology, 30, 181-189.

Pickering, K.T., Stow, D.A.V., Watson, M.P. and Hiseott, R.N. 1986a. Deep-water facies, proeesses and models: A review and classification seheme for modern and aneient sediments. Earth Science Reviews, 23, 75-174.

Piekering, K.T., Coleman, )., Cremer, M., Droz, L., Kohl, B., Normark, W., O'Connell, S., Stow, D. and Meyer-Wright, A. 1986b. A high-sinuosity, laterally-migrating submarine fan ehannel-levee-overbank: results from DSDP Leg 96 on the Mississippi Fan, Gulf of Mexieo. Marine Petroleum Geology, 3, 3-18.

Pickering, K.T., Hiseott, R.N. and Hein, F.J. 1989. Deep Marine Environments. Harper-Collins, London, 416 pp.

Postma, G., Nemee, W. and Kleinspehn, K.L. 1988. Large floating clasts in turbidites, a meehanism for their emplaeement. Sedimentary Geology, 58, 47--61.

Pratson, L.F. and Laine, E.P. 1989. The relative importanee of

gravity-indueed versus eurrent-eontrolled Sedimentation during the Quaternary along the Mideast U.S. outer eontinental margin revealed by 3.5 kHz eeho eharaeter. Marine Geology, 89, 87-126.

Sgavetti, M. 1991. Photostratigraphy of Aneient Turbidite Systems. In: Link, M.H. and Weimer, P. (eds) Seismic Facies and Sedimentary Processes of Submarine Fans and Turbidite Systems. Springer-Verlag, New York, pp. 107-125 .

Shanmugam, G. and Moiola, R.J. 1988. Submarine fans: Charaeteristies, models, classifieation and reservoir potential. Earth Science Reviews, 24, 383-428.

Smith, R.D.A., Waters, R. and Davies,). 1991. Upper Ordovician and Lower Silurian turbidite systems in the Welsh Basin. International Sedimentological Congress, Nottingham, Field Trip Guide No. 20.

Stow, D.A.V. 1986. Deep clastie seas. In: Reading, H.G. (ed.) Sedimentary Environments and Facies, 2nd edn. Blaekwell Seientifie, Oxford, pp. 399-444.

Surlyk, F. 1987. Slope and deep shelf gully sandstones, Upper Jurassic, East Greenland. Tbe American Association of Petroleum Geologists Bulletin, 71, 464-475.

Timbrell, G. 1993. Sandstone arehiteeture of the Baider Formation depositional system, UK Quadrant 9 and adjaeent areas. In: Parker, J.R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference. The Geologieal Soeiety, London, pp. 107-121.

Twitehell, D.C. and Roberts, D.G. 1982. Morphology, distribution, and development of submarine eanyons on the United States Atlantie eontinental slope between Hudson and Baltimore Canyons. Geology, 10, 408-412.

Van Wagoner, J.C., Posamentier, H.W., Mitehum, R.M., Vail, P.R., Sarg, J.F., Loutit, T.S. and Hardenbol,). 1988. An overview of sequenee stratigraphy and key definitions. In: Wilgus, C.W., Hastings, B.S., Kendall, C.G. St., Posamentier, H.W., Ross, C.A. and Van Wagoner, J.C. (eds). Sea Level Changes: An integrated approach. Society of Eeonomic Paleontologists and Mineralogists, Special Publieation No. 42, pp. 39-45.

Van Wagoner, J.C., Mitehum, R.M., Campion, K.M. and Rahmanian, V.D. 1990. Siliciclastic Sequence Stratigraphy in Welt Logs, Cores and Outcrops: Concepts jor high­resolution correlation oftime andfacies. Ameriean Association of Petroleum Geologists Methods in Exploration Series, Tulsa, No. 7, 55 pp.

Vining, B.A., lonnides, N.S. and Piekering, K.T. 1993. Stratigraphie relationships of some Tertiary lowstand depositional systems in the Central North Sea. In: Parker, J.R. (ed.) Petroleum Geology of Northwest Europe: Proceedings oj the 4th Conjerence. The Geologieal Society, London, pp. 17-29.

Walker, R.G. 1975. Nested submarine-fan ehannels in the Capistrano Formation, San Clemente, California. Geological Society of America Bulletin, 86, 915-924.

Walker, R.G. 1978. Deep water sandstone facies and aneient submarine fans: models for exploration for Stratigraphie traps. Tbe American Association of Petroleum Geologists Bulletin, 62, 932-966.