Fine-Grained Sediments: Deep-Water Processes and Facies
Fine-Grained Sediments: Deep-Water Processes and Facies edited by D. A. V. Stow Grant Institute of Geology University of Edinburgh Scotland
and D. J. W. Piper Atlantic Geoscience Centre Geological Survey of Canada Dartmouth, Nova Scotia Canada
1984 Published for The Geological Society by Blackwell Scientific Publications Oxford London Edinburgh Boston Palo Alto Melbourne
Published by Blackwell Scientific Publications Osney Mead, Oxford OX2 0EL 8 John Street, London WC1N 2ES 9 Forrest Road, Edinburgh EH1 2QH 52 Beacon Street, Boston, Massachusetts 02108, USA 706 Cowper Street, Palo Alto, California 94301, USA 99 Barry Street, Carlton, Victoria 3053, Australia
DISTRIBUTORS USA and Canada Blackwell Scientific Publications Inc. PO Box 50009, Palo Alto California 94303 Australia Blackwell Scientific Book Distributors 31 Advantage Road, Highett Victoria 3190
First published 1984 9 1984 The Geological Society. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by The Geological Society for libraries and other users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided that a base fee of $02.00 per copy is paid directly to CCC, 21 Congress Street, Salem, MA 01970, USA 0305-8719/84 $02.00 Printed in Great Britain at The Alden Press, Oxford and bound by Butler & Tanner Frome and London
British Library Cataloguing in Publication Data Fine-grained sediments.---(Geological Society special publications, ISSN 0305-8719; v.4) 1. Sediment transport 2. Sedimentation and deposition I. Stow, D.A.V. II. Piper, D.J.W. III. Series 551.3'5 GB850 ISBN 0-632-01075-4
Contents Preface: STOW, D.A.V. & PIPER, D.J.W .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
vii
Introduction STOW, D.A.V. & PIPER, D.J.W. Deep-water fine-grained sediments: history, methodology and terminology . . . . . . . . . . . . . . . . . . . . . . . . . .
Processes GORSLINE, D.S. A review of fine-grained sediment origins, characteristics, transport and deposition . . . . . . . . . . . . . . . . . . . . . . . MCCAVE, I.N. Erosion, transport and deposition of fine-grained marine sediments . EITTREIM, S.L. Methods and observations in the study of deep-sea suspended particulate matter KRANCK, KATE Grain-size characteristics of turbidites . . . . . . . . . . .
17 35 71 83
Terrigenous turbidites and associated facies VAN WEERING,T.C.E. & VAN IPEREN, J. Fine-grained sediments of the Zaire deep-sea fan, southern Atlantic Ocean . . . . . . . . . . . . . . . . . . . . MONACO, A. & MEAR, Y. Sedimentary sequences on the north-west Mediterranean margin during the Late Quaternary: a dynamic interpretation . . . . . . . . . . . . STOW, D.A.V., ALAM, M. & PIPER, D.J.W. Sedimentology of the Halifax Formation, Nova Scotia: Lower Palaeozoic fine-grained turbidites . . . . . . . . . . . . . . KIDD, R.B. & SEARLE, R.C. Sedimentation in the southern Cape Verde Basin: regional observations by long-range sidescan sonar . . . . . . . . . . . . AUFERET, G.A., LE SUAVE, R., KERBRAT, R., SICHLER, B., ROY, S., LAJ, C. & MULLER, CI Sedimentation in the southern Cape Verde Basin: seismic and sediment facies . . . . GOT, H. Sedimentary processes on the west Hellenic Arc margin . . . . . . CHOUGH, S.K. Fine-grained turbidites and associated mass-flow deposits in the Ulleung (Tsushima) Back-arc Basin, East Sea (Sea of Japan) . . . . . . . . . . . .
95 115 127
145 153 169 185
Carbonate turbidites and associated facies HEATH, K.C. & MULLINS, H.T. Open-ocean, off-bank transport of fine-grained carbonate sediment in the Northern Bahamas . . . . . . . . . . . . . . . . . . . FAUGERES, J-C., CREMER, M., GONTHIER, E., NOEL, M. & POUTIERS, J. Late Quaternary calcareous clayey-silty muds in the Obock trough (Gulf of Aden): hemipelagites or fine-grained turbidites? . . . . . . . . . . . . . . . . . . . . . . STOW, D.A.V., WEZEL, F.C., SAVELLI,D., RAINEY, S.C.R. & ANGELL, G. Depositional model for calcilutites: Scaglia Rossa limestones, Umbro-Marchean Apennines . . . . . . .
199
209 223
Contourites STOW, D.A.V. & HOLBROOK,J.A. North Atlantic contourites: an overview . . . . . . . 245 SHOR, A.N., KENT, D.V. & FLOOD, R.D. Contourite or turbidite?: magnetic fabric of fine-grained Quaternary sediments, Nova Scotia continental rise
.
. . . . . . .
257
GONTHIER,E.G., FAUGERES,J-C. & STOW,D.A.V. Contourite facies of the Faro Drift, Gulf of Cadiz . . . . . . . . . . . . . . . . . . . . . . . . . . . HALEMAN, J.D. & JOHNSON, T.C. The sediment texture of contourites in Lake Superior
275 293 V
vi
Contents
Hemipelagites and associated facies of slopes and slope basins HILL, P.R. Facies and sequence analysis of Nova Scotian Slope muds: turbidite vs 'hemipelagic' deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . MCGREGOR,B.A., NELSEN,T.A., STUBBLEEIELD,W.L. & MERRILL,G.F. The role of canyons in late Quaternary deposition of the United States mid-Atlantic continental rise . . . . BALLANCE,P.F., GREGORY,M.R., Gmson, G.W., CHAPRONIERE, G.C.H., KADAR,A.P. & SAMESHIMA,Z. A late Miocene and early Pliocene upper shelf-to-slope sequence of calcareous fine sediment from the Pacific margin of New Zealand . . . . . . . . . . . . PICKERING, K. T. Facies, facies-association and sediment transport/deposition processes in a late Precambrian upper basin-slope/pro-delta, Finnmark, N. Norway . . . . . . . . KRISSEK, L.A. Continental source area contributions to fine-grained sediments on the Oregon and Washington continental slope . . . . . . . . . . . . . . . . . . . THORNTON, S.E. Basin model for hemipelagic sedimentation in a tectonically active continental margin: Santa Barbara Basin, California Continental Borderland . . . . . . . . GORSLINE, D.S., KOLPACK, R.L., KARL, H.A., DRAKE, D.E., FLEISCHER, P., THORNTON, S.E., SCHWALBACH,J.R. & SAVRDA,C.E. Studies of fine-grained sediment transport processes and products in the Californian Continental Borderland . . . . . . . . . . . . . BOURROUILH, R. & GORSLINE, D.S. Fine-grained sediments associated with fan lobes: Santa Paula Creek, California . . . . . . . . . . . . . . . . . . .
311 319
331 343 363 377
395 417
Pelagites and organic-rich sediments ROBERTSON, A.H.F. Origin of varve-type lamination, graded claystones and limestone-shale 'couplets' on the lower Cretaceous of the western North Atlantic . . . . . . . . HAYWARD, A.B. Hemipelagic chalks in a clastic submarine fan sequence: Miocene SW Turkey CREVELLO, P.D., PATTON, J.W., OESLEBY, T.W., SCHLAGER, W. & DROXLER, A. Source rock potential of Bahamian Trough carbonates . . . . . . . . . . . . . . . . ISSACS, C.M. Hemipelagic deposits in a Miocene basin, California: toward a model of lithologic variation and sequence . . . . . . . . . . . . . . . . . . . . . ANASTASAKIS,GEORGEC. & STANLEY, DANIELJEAN. Sapropels and organic-rich variants in the Mediterranean: sequence development and classification . . . . . . . . . . THICKPENNY, A. The sedimentology of the Swedish Alum Shales . . . . . . . . . . ARTHUR, M.A., DEAN, W.E. & STOW, D.A.V. Models for the deposition of Mesozoic-Cenozoic fine-grained organic-carbon-rich sediment in the deep sea . . . . . . . . . .
437 453 469 481 497 511 527
Internal characteristics FAAS,R.W. Plasticity and compaction characteristics of the Quaternary sediments penetrated on the Guatemalan Transect--DSDP Leg 67 . . . . . . . . . . . . . . . . . MOON, C.F. & HURST, C.W. Fabrics of muds and shales: an overview . . . . . . . . WETZEL, A. Bioturbation in deep-sea fine-grained sediments: influence of sediment texture, turbidite frequency and rates of environmental change . . . . . . . . . . . .
563 579 595
Facies models: synthesis STOW, D.A.V. & PIPER, D.J.W. Deep-water fine-grained sediments: facies models
611
Preface This volume has been edited from the proceedings and discussion at an International Workshop on Fine-Grained Sediments held at Dalhousie University, Halifax, Canada, in August 1982, and co-sponsored by the Geological Society of London and the Geological Association of Canada. Additional contributions have been solicited to provide a more comprehensive publication, although we have retained the original narrow remit of the research meeting. This focused on modern and ancient, siliciclastic and biogenic, fine-grained sediments from a range of deep-water environments. We aimed to limit the scope of our discussions so that advances in understanding resulting in a useful synthesis might be achieved, but to leave the scope sufficiently broad that valuable cross-fertilization of ideas could take place. We hope that the result will be helpful to many research sedimentologists as a state-of-theart summary of a large and important class of rocks. The volume contains thirty-eight separate papers, including short regional contributions and longer review papers, organized into nine sections. There is a short introduction to the history, methodology and terminology of finegrained sediment studies. This is followed by a more substantive section dealing with processes of erosion, transport and deposition in deepwater. There are then sections on each of the main facies types including siliciclastic turbidites, carbonate turbidites, contourites, mixed (mainly hemipelagic) facies of slopes and slope basins,
pelagites and organic-carbon-rich sediments. A section on internal characteristics covers plasticity and compaction, mud and shale fabric, and bioturbation. Our final synthesis paper on facies models attempts to relate fine-grained sediment facies to depositional processes and deep-water environments. Each paper has been reviewed by at least two external referees as well as by the editors. In editing, we have not aimed for rigorous uniformity of style or terminology, as this was clearly not possible given the different approaches and persuasions of the seventy-five authors and coauthors who have contributed. Finally, our thanks to the many people who helped to make possible this publication: to all the participants at the meeting and to the contributors who were unable to attend; to Phil Hill, Tony Bowen and Martin Gibling for their help with the organization and running of the Halifax workshop; to those who refereed papers; to our respective institutions for secretarial, drafting and technical support; to the Natural Environment Research Council, UK, the Canadian Natural Sciences and Engineering Research Council, Dalhousie University, the Geological Survey of Canada, and to our sponsoring organizations for support; and to Nigel Palmer and Jane Grisdale of Blackwell Scientific Publications for their patient and diligent work. Dorrik A.V. Stow, David J.W. Piper
August 1983
Deep-water fine-grained sediments; history, methodology and terminology D.A.V. Stow and D.J.W. Piper SUMMARY: To introduce this collection of papers presented at an international research workshop in Halifax, Canada (1982), we highlight briefly three aspects of deep-water fine-grained sediments that are alluded to throughout the volume but never discussed specifically. These are: (a) an historical outline of the research that has made both possible and necessary the workshop and the volume; (b) a review of the methodology currently used in the study of fine-grained sediments; and (c) an assessment of the state of terminology as applied to this class of rocks.
History The development of knowledge concerning finegrained sediments in deep water has been closely related to the history of both sedimentology and oceanography, and their interpretation has evolved along with the science of geology as a whole. Major advances within each of these disciplines have, on the one hand, improved our understanding of fine-grained sediments, while on the other hand have acted as significant obstacles to progress in the years immediately following each breakthrough. It is as Kuhn (1970) suggested in his theory of scientific revolutions: the establishment of a paradigm both advances an area of study and leads to a period of blinkered normal science that serves to elaborate the paradigm but to digress little from it. We can illustrate this with several examples (Fig. 1).
Pelagic sediments The systematic study of deep-sea sediments and the birth of modern oceanography began with the voyage of HMS Challenger (1872-1876) which established the general morphology of the oceans and the types of sediments they contained. The report on Deep-Sea Deposits by Murray & Renard (1891), which documented the calcareous and siliceous pelagic oozes, their microflora and microfauna, the pelagic red clays, and deep marine ferromanganese deposits became the cornerstone of deep-sea sedimentology for over half a century. Although a major advance over previous knowledge, this new paradigm held that only pelagic clays and biogenic oozes were found in the deep sea and that all coarser-grained clastics were restricted to shallow water or continental environments. It also dismissed any suggestion that rocks now exposed on land were comparable with deep-ocean pelagic sediments. This, therefore,
had two main restricting effects on the study of ancient rocks: (a) clastic sediments, other than red clays, were not considered as deep-water; and (b) the interpretation of chalks, cherts and limestones was thrown into confusion. The growing body of evidence that invoked a deep water origin for thick flysch successions (e.g. Lesley 1892; Natland 1933; Bailey 1936; Pettijohn 1943) was held at bay for many years. Similarly, the Alpine geologists who claimed deep-water pelagic successions (e.g. Suess 1875; Fuchs 1877; Steinmann 1905; Heim 1924) and equivalent interpretation in other parts of the world (e.g. Hinde 1890; Molengraaf 1922), had to bide their time before being more generally accepted.
Systematic sedimentology If Murray & Renard were pioneers of modern oceanography, so Sorby was the father of sedimentology as we know it today. As early as 1861, even prior to the Challenger expedition, he had equated the English Chalk with deep-sea coccolith ooze based on the first reports of coccoliths from Atlantic sediments by Huxley (1858). His paper early in the 20th century (Sorby 1908), On
the application of quantitative methods to the study of the structure and history of rocks, was equally far ahead of his time and extremely perceptive about fine-grained sediments. 'Possibly many may think that the deposition and consolidation of fine-grained mud must be a very simple matter, and the results of little interest. However, when carefully s t u d i e d . . , it is soon found to be so complex a question, and the results dependent on so many variable conditions, that one might feel inclined to abandon the inquiry, were it not that so much of the history of our rocks appears to be written in this language.' H.C. Sorby (1908). It is not entirely clear why his plea for systematic sedimentology was not taken up by students of
4
D . A . V . S t o w and D . J . W . Piper /!
80-'
/
Information explosion
t/ / I
70"
968 Glomar Challenger sails (DSDP) #
966 Contourites (Heezen + others) I
6 0 - 1960s Plate tectonics revolution I I I ! I
shales for so long. Perhaps the rigorous scientific approach was difficult to apply to the finergrained rocks, or perhaps the methodology was lacking. There were, however, some notable exceptions: Helm's (1924) classic work on Alpine sediments; Archanquelsky's (1927) little known study of Black Sea sediments which he likened to the argillaceous fill of geosynclines; and Ruedemann's (1935) discussion of the ecology and sedimentology of Palaeozoic black shales in the Appalachians. Turbidite revolution
Walker (1973) has documented the four lines of research that effected the mid-20th century revolution in our thinking about deep-sea sediments. I These were: (a) the recovery of varied sediment i types other than pelagites on a number of Euroi 40I pean and American oceanographic expeditions in I I the first half of the century (e.g. Bdggild 1916; Long period of 'normal' science Andree 1920; Shepard 1932, 1948; Stocks 1933; data accumulation Bramlette & Bradley 1940; Arrhenius 1950); (b) 5 0 - no new paradigms the persuasive evidence presented by Daly (1936 ) and Johnson (1938) suggesting that sedimentladen density currents had excavated submarine canyons as they flowed downslope; (c) a series of flume experiments on both dilute and high-den20" sity flows carried out by Kuenen (1937, 1950); and (d) numerous observations of graded sand beds in probable deep-water successions on land (e.g. Natland 1933; Migliorini, 1946). The classic IO,~ paper Turbidity currents as a cause of graded 1908 Systematic sedimentology bedding (Kuenen & Migliorini 1950) became the I I (Sorby) manifesto of the revolution, and stimulated an intense period of systematic field, laboratory and II 1900oceanographic studies. I I However, attention became focused on the I sandstones and coarser-grained sediments to the I detriment of the finer-grained material. Much of 1891 Pelagic sediments the interbedded mudstone was dismissed as 90( Murray + Renard) 'background', 'ubiquitous' hemipelagic sediment. This has been more true of carbonate turbidites and the interbedded calcilutites which are still largely considered pelagic. The Bouma 80 (1962) standard structural sequence for sandy turbidites has stood for a long time as the sequence for turbidites, although it is clearly ~ 8 7 2 _ 7 6 HMS Challenger unsatisfactory for resedimented muds and silts or expedition for fine-grained biogenics. This neglect of fine70.grained turbidites was largely a consequence of the 1950s concepts of turbidity current dynamics. FIG. 1. Historical sketch showing the main concepts Although evidence for deposition of mud beds that have advanced our understanding of deep-sea from turbidity currents was recognized (Dzufine-grained sediments. The dashed line indicates the lynski &Kinle 1957) it was thought that little clay stepped increase in our level of understanding, as each new concept (or paradigm) leads both to a could settle out except at velocities that would significant advance and to period or relative give only very dilute turbidity currents, so that the standstill. deposit would be very thin (Dzulynski et al. 5 0 " 1950 Turbiditesl (Kuenen + Migliorini) I I I
Deep-water fine grained sediments," history, methodology and terminology 1959). Clay deposition was believed to require velocities of less than 1 cm/sec. The observational evidence for deposition of clay from typical turbidity currents could be explained by deposition as faecal pellets or shale chips, ponding, or repeated small turbidity currents (Dzulynski et al. 1959). In the 1960s, it became clear that clay could be deposited from suspension at velocities at which silt and fine sand could be transported (Einstein & Krone 1962), and increasing investigation of core samples clearly demonstrated that thick turbidite muds were a common deep-sea facies (van Straaten 1967; Hesse 1975).
Modern sedimentology The turbidite revolution also heralded the development of the modern science of sedimentology. Sedimentary structures were recognized as indicators not just of way-up in folded strata, but more importantly of processes (which could be quantified) and hence sedimentation facies and environments (Middleton 1965). Analysis of the sequence of sedimentation facies in a stratigraphic succession, enunciated in the last part of the nineteenth century by Walther (1893), was developed into a standard tool (de Raaf et al. 1965). These two techniques lead to the development of the facies model approach (Walker 1976). At the same time, the principles of carbonate rock classification developed by Folk (1959) provided the foundation for a major expansion of understanding of carbonate rock sedimentology accompanied by major investigations of modern shallow-water carbonate environments of Florida, the Bahamas and the Persian Gulf (Folk 1973).
Plate tectonics The major upheaval in geology that culminated in the early 1960s ideas of sea floor spreading and plate tectonics (Hess 1960; Dietz 1961; Runcorn 1962; Vine & Matthews 1963), had repercussions throughout the discipline. It finally sanctioned the idea that deep oceanic pelagic sediments may be found high up in mountain ranges, and threw new light on our interpretations of the very thick and largely fine-grained fill ofgeosynclines. It also spawned new major scientific investigation of the ocean basins and their sediments. However, even these undoubtedly profound advances had their drawbacks. In particular, scientific attention and effort was directed towards the large-scale picture, the placing of giant pieces in a new jigsaw puzzle, and this necessarily detracted from detailed sedimentology.
5
Contourites Only a decade after the explosion of research on turbidites many sedimentologists were beginning to see flaws in this undoubtedly elegant paradigm. Land geologists repeatedly described examples of 'turbidite' sequences having orthogonal palaeocurrent directions, even from bottom to top of the same bed (e.g. Kelling 1958; Craig & Walton 1962; Ballance 1964; Klein 1966). Marine geologists were documenting many other kinds of evidence that indicated an important role for bottom currents flowing alongslope in deep water (e.g. Heezen 1959; Heezen & Johnson 1963; Hubert 1964). Finally, it was the compilation of evidence by Heezen et al. (1966) on Shaping of the
continental rise by deep geostrophic contour currents that caught the imagination of many more geologists. Stow & Lovell (1979) suggested that this paper marked the beginning of a revolution in sedimentology comparable to the turbidite revolution. Important though this paradigm has been in adding a new dimension to our understanding of deep-sea processes and fine-grained facies, it has nevertheless proved an obstacle to progress in two respects: (a) research focused initially on sandy contourites and largely ignores muddy contourites, although the latter now appear volumetrically more significant; and (b) many authors apparently jumped onto the contourite bandwagon without a clear appreciation of just how to recognize a contourite, modern or ancient, and hence many erroneous interpretations have been published.
Information explosion In 1968, nearly 100 years after HMS Challenger set sail, the Glomar Challenger drillship put to sea at the beginning of what was to become the modern equivalent of that historic voyage. By the end of this international scientific mission (late 1983) nearly 100 separate legs of the expedition had been completed and over 600 holes drilled into the deep ocean floor. At the same time, many extensive national marine programmes have ventured into the deep sea; wide-ranging international, national and individual work has been carried out on land; and an enormous amount of geological data has been obtained in the search for hydrocarbons. Moreover, fine-grained sedimentology no longer takes the back seat. Research has included the muddy turbidites and contourites, hemipelagites, and fine biogenic oozes, red clays and black shales; the processes and controls on the dispersion and deposition of these facies; their micro-
6
D.A.V. Stow and D.J. IV. Piper
characteristics and physical properties; the compaction, disgenesis and association with both hydrocarbon generation and metallic ores. Clearly, it is difficult to assess the historical significance of such research and, similarly, difficult to see where we are being short-sighted or blinkered by new paradigms. One factor that mitigates against progress is the sheer volume of data now available, mostly undigested and not assimilated into the mainstream of knowledge. It is this information explosion that has made it necessary to focus our attention so narrowly for this workshop and volume in the hope that it may thus be able to contribute a little towards the synthesis and understanding of one small class of rocks.
Methodology Part of the reason that the study of fine-grained sediments has for so long lagged behind that of coarser-grained rocks is related to methodology. Although the approach and techniques are largely the same, there are certain problems unique to the fine end of the grain-size spectrum. These include: (a) the resolution of the individual particles, often less than 5 /~m in diameter, requires high-powered instrumentation and routine use of electron microscopy; (b) fine-grained rocks cannot be adequately characterized in the field or from cores: laboratory-based analyses are imperative; (c) because clays flocculate, textural analysis cannot be easily used to infer hydrodynamic processes (in contrast to sands); and (d) the dramatic post-depositional changes suffered by clays and fine biogenic material often make it extremely difficult to reconstruct the sediment characteristics at the time of deposition. However, these problems are by no means insurmountable. Many standard techniques are now successfully applied to fine-grained sediments, and new methods or refinements continually being developed. Coulter counters are routinely used for grain-size analysis, scanning electron microscopy for fabric, structure and composition, X-ray diffraction and X-ray fluorescence for mineralogy and geochemistry, CHN analysis for organic carbon determination, induction magnetometers for measuring the anisotropy of magnetic susceptibility, and X-radiography for study of sedimentary structures. In order to fully understand and accurately interpret a given sedimentary succession, it is important that the study be approached from many directions, at different scales of operation and employing a wide range of methods. For fine-grained sediments in particu-
lar, it is crucial that field examination is combined with laboratory analysis. Perhaps the best general approach to use is that described by Potter et al. (1980) as the 'Question Set Approach'. They developed a series of questions for the study of shale (mudrock) successions, that help to lead the investigator through the required series of techniques and analyses at the micro, meso and macro scales. We have reproduced their Question Set below (Table 1). Clearly, each particular study might require a modified question set, especially studies of finegrained biogenic sediments, for example. In addition, many investigators might choose to study in greater detail just one aspect of the whole (e.g. clay fabric, sediment grain-size, etc) but it is nevertheless useful to see where this micro-study fits into the overall picture. Clearly, we do not intend to describe all the variety of methods used in this short introduction. Instead we list below (Table 2) the main source books on methodology with a brief comment on their relevance to the study of finegrained sediments.
Terminology Fine-grained sediments are those rocks, both hard and soft, biogenic and clastic, that have a dominant grain size in the clay or silt grades (i.e. over 50% < 63 #m). Deep-water implies anything below wave base (say about 50m) and so includes the deeper shelf areas, shelf basins and deep lakes as well as the open ocean basins and marginal seas. Most of the papers in this volume, however, are concerned with sediments deposited at oce~/nic depths greater than about 200m. As a discipline grows, so the terminology within that discipline proliferates and confusion is inevitable. However, it is clearly important for improved communication within that discipline, as well as within the earth sciences as a whole (e.g. between sedimentologists and engineers), that a generally-accepted standard terminology is widely used. Fortunately, the state of confusion in the terminology applied to fine-grained sediments is not yet too severe, and so we are able here to propose a degree of standardization. We outline first a purely descriptive terminology for fine-grained sediment classification and description and, second, some of the terms used in the interpretation of deep-water facies.
Descriptive terminology Classifications of fine-grained sediments have been based most commonly on texture and
TABLE 1.
Question set for study of shale (mudrock) sequence (from Potter et al. 1980)
Describing outcrops and cores and using wire-line logs 1. Where is the section? 2, What are the major units? 3. What is to be done next? 4. What should be described? 5. What terms should be used in the field for the description of the major lithologic types of shales? 6. How should the observations be recorded? 7. What palaeontologic observations can and should be made in the field? a. Relative abundances of different macrofossils? b. Is the distribution ofmacrofossils patchy, uniform, or random? Are they concentrated within beds or on bedding planes? c. Are the macrofossils intact and well preserved or fragments and worn? Molds or casts? Recrystallized or replaced? d. What functional types of organisms are present? Encrusters, sediment trappers or binders, epifauna, or infauna? Mobile or sessile? Suspension feeders, deposit feeders, scavengers, or predators? 8. What is the gamma-ray profile--what is it good for and how is it obtained? 9. Wire-line logs--what are they and how can they be best used? 10. Why and how to describe cuttings? Laboratory studies 1. What samples should be selected? 2. What tools and techniques should be used? 3. What sequence of study should be followed? 4. How should the observations be recorded? 5. What components are present, what is their abundance, and what do they all mean (fundamental to the understanding of every rock, the key questions are always the same)? a. Large detrital grains, such as quartz, feldspar micas, heavy minerals, and carbonate grains? b. Detrital clays? c. Authigenic grains, including carbonates (calcite, dolomite, and siderite), quartz, feldspar, zeolites, and the authigenic clays, glauconite, sepiolite, etc.? i. How are authigenic minerals recognized in a mudstone or shale? Those formed after deposition either by precipitation in pores or by transformation of original detrital minerals (by solution and replacement). it. Significance? d. Mineral cements, such as silica, carbonate, or zeolites? e. Floccules? f. Pellets and pelaggregates? g. Organic particles? h. What can be learned from micropalaeontology and palynology? 6. What textural parameters should be measured and what do they mean? a. What proportions of clay, silt, and sand? b. Percentage of large micas? c. Size ranges and modes of diverse detrital grains? d. How well oriented are the different components of the shale sediment--the clay minerals, the silt and sand grains commonly found in t h e m - - a n d what significance, if any, does their orientation have for palaeocurrents? ~. Perfection of framework orientation? it. Silt and sand grains? e. Burrowed and/or mottled textures? f. Pore geometry--kinds and amounts? g. Is there any significance to the shape of shale cuttings? h. What can be seen by radiography? 7. What name is to be used and how should the shale be classified now that we know so much about it? 8. The petrographic report: What is the best way to organize the foregoing petrology and texture into a useful, concise, and coherent petrographic report? 9. What can be learned from the study of inorganic geochemistry? a. The major elements? b. The trace elements? c. Exchangeable cations? d. Pore-water chemistry? e. Stable isotope geochemistry? 10. Organic chemistry? a. What are the best indicators of thermal history? b. What does the study of palaeobiochemical indicators tell us? Making a synthesis and basin analysis 1. Where did the mud come from? 2. How was it transported to its final depositional site? 3. At what water depth, sedimentation rate, oxygenation, and toxicity to life was the mud deposited? 4. What has happened to the mud since deposition?
8
D.A.V. Stow and D.J.W. Piper
TABLE 2. Principal texts on methods in sedimentary geology BOUMA,A.H. 1969. Methods for the Study of Sedimentary Structures. Wiley-Interscience, New York. 446 PP. Comprehensive for sedimentary structures. Good on X-radiography, graphical presentation, impregnation. BRINDLEY, G.W. & BROWN, G. (eds) 1980. Crystal structures of clay minerals and their X-ray identification. Mineral. Soc. Monograph. No 5. The most thorough and up-to-date reference for clay mineralogy and methods. CARVER,R.E. 1971. Procedures in Sedimentary Petrology, Wiley-Interscience, New York. 653 pp. One of the most comprehensive books on laboratory techniques. Covers sedimentary structures, grain size, grain attributes, and textural, mineralogical and chemical analyses. Lacks many techniques developed over the past 10-15 years. COMPTON, R.R. 1962. Manual of Field Geology. John Wiley, New York. 378 pp. Early but thorough treatment of what to do in the field. CONYBEARE,C.E.B. & CROOK,K.A.W. 1968. Manual of Sedimentary Structures. Australian Dept. Natl. Development. Bull. Bur. Min. Res. Geol. & Geophys. 102, 327 pp. A useful guide to structures in all types of sediments. FLOGEL, E. 1982 (english edition), Microfacies Analysis of Limestones. Springer-Verlag, New York, Berlin, 633 pp. Does not detail methods used in the study of limestones at any great length, but has wide-ranging coverage and extensive references. FOLK, R.E. 1974. Petrology of Sedimentary Rocks.
Hemphill Pub. Co., Austin, Texas. 159 pp. General sedimentology text with useful sections and references on methodology. FREe, R.W., (ed.) 1975. The Study of Trace Fossils. Springer-Verlag, New York. Useful illustrations of trace fossils in sediments. GR1FFITHS,J.C. 1967. Scientific Method in the Analysis of Sediments. McGraw-Hill, New York. 508 pp. Good on texture, fabric, composition, data processing and statistics. KUMMEL, B. & RAUP, B. (eds) 1965. Handbook of Paleontological Techniques. Freeman & Co., San Francisco. 852 pp. Still one of the most thorough books on palaeontological techniques. MULLER, B. 1967. Methods in Sedimentary Petrology. Hafner. Fairly comprehensive, but slightly dated. PETTIJOHN, F.J. & POTTER,P.E. 1964. Atlas and Glossary of Primary Sedimentary Structures, SpringerVerlag, New York. 370 pp. Well-illustrated guide to structures in all rock types. POTTER, P.E., MAYNARD, J.B. & PRYOR, W.A. 1980. Sedimentology of Shale, Springer-Verlag, New York. 303 pp. Informative study guide and reference source for shales, including guide and references to methodology. TUCKER,M.E. (1982). The FieMDescription of Sedimentary Rocks. Open University Press, Milton Keynes, UK and Halsted Press, New York. 112 pp. A useful field manual covering all sedimentary rock types.
composition and secondarily on fissility, colour, degree of metamorphism and depositional environment. We propose here a twofold division of fine-grained sediments into mudrocks, with an implied siliciclastic composition, and, biogenic mudrocks, having either a calcareous or siliceous composition (Tables 3 and 4). The term mudrock seems more appropriate than shale although both are widely used as general class names (e.g. Potter et al. 1980). In subdividing the two groups we have tried to use generally-accepted simple terms that can be readily applied in the field and subsequently modified by the appropriate descriptors after laboratory analysis. The terminology presented in Table 2 is broadly in line with that proposed by many previous authors (e.g. Wentworth 1922; Ingrain 1953; Shepard 1954; Dunbar & Rogers 1957; Folk 1968; Picard 1971; Weser 1974; Pettijohn 1975; Blatt et al. 1980; Potter et al. 1980). It is also commonly used by sedimentologists, marine geologists, engineers and soil scientists. Minor differences that exist between classification systems are
mainly concerned with the exact percentage of sand that a mud must contain before it becomes a sandy mud, and so on. We suggest it is better to keep the terms more general in application, so that the textural or compositional contents are estimates rather than strict definitions. Mixtures of sand, silt and clay are commonly displayed graphically on triangular diagrams (Fig. 2). Colour, chemical composition or genetic terms can be used as additional descriptors to the basic terms as appropriate (e.g. black shale, uraniferous mudstone). These, and mineralogical terms such as chlorite illite mudstone, are mainly applicable only after detailed laboratory investigations. Spears (1980) notes that the percentage of quartz in mudrocks is generally proportional to the grain-size or siltiness. In ancient well-lithifled rocks it is often simpler to estimate quartzcontent than grain size directly, particularly by laboratory methods, but the same basic terminology should still be applied. Lewan's (1979) attempt to erect a laboratory classification of very fine-grained sediments is also based on textural
D e e p - w a t e r f i n e g r a i n e d sediments," history, m e t h o d o l o g y a n d t e r m i n o l o g y TABLE 3. Mudrock terminology (after Stow 1981) Mudrock (> 50% less than 63 llm)
Basic terms Unlithified
Lithified/non-fissile
Silt Mud Clay
Lith!fied/fissile
Siltstone Mudstone Claystone
Silt-shale Mud-shale Clay-shale
Approx. proportions/grain-size > ~ silt-sized (4-63/~m) silt and clay mixture (< 63/~m) > 2 clay-sized ( < 4/~m)
Metamorphic terms Argillite Slate
slightly metamorphosed/non-fissile metamorphosed/fissile
Textural descriptors
Approx. proportions
Silty Muddy Clayey Sandy, pebbly, etc
> > > >
Compositional descriptors
10% silt-size 10% silt- or clay-size (applied to non-mudrock sediments) 10% clay size 10% sand-size, pebble-size, etc.
Approx. proportions
Calcareous Siliceous Carbonaceous Pyritiferous t Ferruginous Micaceous and others
> 10~ CaCO3 (foraminiferal, nannofossil, etc) > 10~ SiO2 (diatomaceous, radiolarian, etc) > 1~ Organic carbon Commonly used for contents greater than about 1-5%
criteria and compositional modifiers. However, his redefinition ofmudstone, shale, claystone and marlstone are not very helpful. Pelite and lutite (pelitic and lutaceous) are synonymous with mudstone (muddy) and are not recommended terms, particularly since pelite is widely used to indicate a metamorphic rock. Siltite for siltstone is also a redundant term. Argillaceous sedimentary rock is more conveniently replaced by mudrock, but argillaceous as a strictly compositional term (meaning rich in clay minerals) is still useful in certain cases. C LAY
..........
~MUD
: s.~!t..... !i:-i\
SAND
80
silt and clay mixture silt and clay mixture
50%
80
SILT
FIG. 2. Terminology for sand-silt-clay sediment admixtures (after Shepard 1954). The shaded area denotes mud, a term much used for silt-clay mixtures following Folk (1968).
The terminology for biogenic mudrocks shown in Table 4 follows closely the various systems in common use (e.g. Gealy et al. 1971; Weser 1974; Berger 1974; Davies et al. 1977). Dean et al. (1984) have proposed that a more formal intermediate category is needed between ooze and mud for mixtures of biogenic material and mud (Fig. 3). They recognized marl as a very common and useful term for calcareous biogenic muds and introduced sarl for siliceous biogenic muds and smarl for mixed calcareous and siliceous biogenic muds. The general term for sediments in this category therefore becomes arl. Descriptors of all kinds (composition, texture, colour, etc.) can then be applied to the basic terms as for the siliciclastic sediments. A single descriptor indicates the dominant component, and a second descriptor (added at the beginning) indicates the second most important component when applied to oozes and muddy oozes (arls). Further classifications of lithified carbonate rocks, commonly after laboratory analyses, should follow the systems proposed by Folk (1959) or Dunham (1962). The former is based on the type of carbonate particle (ooliths, skeletal grains, pellets) and the type of cement (microcrystalline micrite or macrocrystalline sparry calcite). The latter is concerned more with the relative proportions of matrix and discrete particles. The terms calcilutite and calcisiltite, together with calcarenite and calcirudite for coarser-grained carbonates, can be useful when the grain-size of
Io
D.A. V. S t o w and D.J. W. Piper TABLE 4. Biogenic mudrock terminology Biogenic mudrock (50% biogenic, 50% less than 63 pm)
Basic terms Unlithified Ooze Muddy ooze* Mud
Descriptors Calcareous t Siliceous Foraminiferal Diatomaceous Radiolarian | etc. j Carbonaceous
Lithified
Approx. proportions
chalk, limestone, diatomite, chert, etc. Argillaceous chert, chalk etc. mudstone
> 2/3 biogenic biogenic/terrigenous mix > 2/3 terrigenous
Approx. proportions e.g. siliceous ooze, biogenic SiO2 > 50% foraminiferal ooze, forams > 50% foraminiferal-nannofossilooze, forams & nannos > 50% with nannos dominant diatomaceous muddy ooze, diatoms dominant biogenic component calcareous mud, 300,/0> CO3 > 10,~ organic carbon > 1%
* Dean et al (1984) suggest Arl=muddy ooze hence marlstone, sarlstone, smarlstone. the rock is considered its most important feature as, for example, with resedimented carbonates. Another area of descriptive terminology in which there is often confusion is that applied to sedimentary structures, and in particular to lamination and bedding. The most widely accepted system for stratification thickness is that of Ingram (1954) shown below (Table 5). Grading can be either positive (normal) in which the grain-size decreases upwards, or negative (reverse) in which the grain-size increases upwards. This is commonly applied to a single bed or discrete unit. Series of beds that show an overall grain-size change are termed finingupward or coarsening-upward sequences, although this usually refers only to the coarser (silt or sand) beds within the sequence. Similarly,
a graded laminated unit is an interval (commonly 2-20 cm thick) through which discrete silt laminae, separated by a finer-grained mud, show an overall upward decrease in grain-size. The terms massive, homogeneous and structureless are used synonymously to mean an ungraded and featureless sediment. In Fig. 4 we list the common sedimentary structures encountered in fine-grained sediments and show the symbols most often to represent them graphically. Standardization in this aspect of description and terminology would be very helpful. TABLE 5. Stratification types and thicknesses
(after Ingram 1954) Very thickly bedded Thickly bedded Medium bedded Thinly bedded Very thinly bedded Thickly laminated Thinly laminated Very thinly laminated
MUD
> 1m 30-100 cm 10-30 cm 3-10 cm 1-3 cm 3-10 mm 1-3 mm < 1 mm
Interpretative terminology
BIOGENIC CARBONATE
2/3
approx 2/3
BIOGENIC SILICA
FIG. 3. Terminology for biogenic carbonage, biogenic silica and terrigenous mud mixtures (as proposed by Dean et al. 1984).
The first phase of a study is, clearly, to describe the sediments and their characteristics, as far as possible objectively and using a standard format. The second phase involves interpretation of these primary data in terms of depositional processes, environmental setting, stratigraphic position and so on. International stratigraphic terminology is
Deep-water fine grained sediments," history, methodology and terminology SYMBOL
///
DESCRIPTION
SYMBOL
BEDDING PLANE STRUCTURES
massive, structureless
surface lineation
parallel bedding
flutes
parallel lamination
grooves
inclined bedding/lamination
load casts
cross-bedding/lamination
scour and fill
flaser bedding, fading ripples
flame (injection) structure
convolute bedding/lamination
mud cracks
slumped
INTERNAL STRUCTURES
oo@
lenticular bedding/lamination
~176
o~176
negative (reverse) grading
water escape pipes
II
dish structures
u~,s
positive (normal) grading
filled fracture
l
LAYER BOUNDARIES
microfault
sharp contact
concretion
sharp irregular (erosive) contact
OTHER
disturbed contact
disturbed section
gradational contact
fining-upward sequence
A V
BIOGENIC STRUCTURES
coarsening-upward sequence
bioturbation minor (0-50%)
tli 0
imbrication mud clast
wedge-shaped layer m
DESCRIPTION
STRATIFICATION
wavy bedding/lamination
II
interval over which structure occurs
bioturbation moderate (50-60~
(
bioturbation intense ( > 6 0 % )
I())
)
structure indistinct structure very indistinct
burrows
FIG. 4. Typical sedimentary structures and recommended graphic symbols (modified from Bouma 1962, and DSDP schemes). well established (Hedberg 1976; Holland et al. 1978). Environmental terms present more of a problem, particularly with regard to interpretation of palaeoenvironments. Deep-water morphological environments in which fine-grained sediments accumulate have been initially defined in large oceans (Heezen et al. 1959) as continental shelf (especially outer shelf), slope, rise and abyssal plain. The terms basin slope, basin rise and basin plain are more applicable to smaller basins (such as in the Mediterranean sea or California Borderland). Deep-sea fans are large morphogenetic environments that include both rise and plain settings. Many terms have been applied to particular
subenvironments or morphological elements within these larger-scale settings (e.g. Stow, in press). The most common include canyons, channels, gullies, levees, interchannel areas, lobes, slump masses, debris-flow masses, slump scars, sediment drifts, upper, middle and lower fans, and so on. Such present day morphological features are easily recognized by acoustic profiling in the oceans, but their recognition in ancient rock sequences requires a detailed knowledge of stratigraphic relationships and large-scale facies associations and sequences. This is particularly difficult in fine-grained rocks because of their paucity of diagnostic facies indicators and their common high degree of tectonic deformation. Extreme caution should therefore be used in
D.A.V. Stow and D.J.W. Piper
I2
TABLE 6.
Deep-waterfine-grained sediment facies (from Stow & Piper, this volume) Depositional process
Facies
Characteristics
Silt turbidite Mud turbidite Biogenic turbidite Disorganized turbidite
> 50,~ silt "~ > 50~ mud > 500/;,biogenics mud or biogenics or mixture
Silty (sandy) contourite
) > 50~ silt or fine "} commonly mixed sand size I> biogenic-terrigenous, k, bottom irregular "sequence', ( ( c o n t o u r ) > 50~o mud size bioturbated. ) current
Muddy contourite Pelagic ooze Hemipelagite Pelagic clay
> 2/3 biogenic "~ (planktonics) biogenick;. terrigenous m i x [ > 2/3 terrigenous clay (mud). ,,'
standard structural sequence no sequence
no structural sequence, rhythmic interbedding common, bioturbated
t
t
turbidity current
pelagic settling
Fine-grained sediment facies in deep water can making palaeoenvironmental interpretations for mostly be interpreted in terms of these depositancient fine-grained rock successions. There are three main groups of processes by ional processes, although it must be recognized which fine-grained sediments are deposited in the that the processes are in fact part of a continuum, deep-sea (Gorsline, this volume; McCave, this and that a similar continuum of facies therefore volume). (a) Resedimentation processes (synony- exists. Stow & Piper (this volume) have identified mous with mass gravity transport) are all those nine separate facies models (Table 6), each related processes that move sediment downslope over the to a specific depositional process and each characsea floor from shallower to deeper water and that terized by a standard sequence of structures or a are driven by gravitational forces. For fine- 9standard suite of sedimentary features. We suggest that these facies models, although grained sediments the most important of these are slides and slumps, sediment creep, debris-flows not exhaustive of the possible processes and facies and turbidity currents. (b) Normal bottom cur- in the deep sea, are all relatively well documented rents are all those deep currents that erode, and understood. Until similar evidence exists for transport and deposit sediment on the sea floor other facies (e.g. the 'nepheloidites' or 'suspenand that are driven by normal thermohaline or sion cascadites' invoked by some authors), it wind-driven circulation within the oceans. They seems better not to use the terms. Similarly, include, internal tides and waves, canyon cur- purely descriptive facies terms such as 'laminites' rents, bottom (contour) currents and deep surface and 'unifites' would be better replaced by the currents. (c) Pelagic settling through the water standard descriptive terminology outlined precolumn is an ubiquitous, slow and predominantly viously. vertical process under the influence of gravity. Much of the sediment settles as flocs and faecal pellets rather than individual particles.
References ANDREE, K. 1920. Geologic des Meeresbodens. Borntraeger (PUN.), Leipzig. 2, 689 pp. ARCHANQUELSKY, A.D. 1927. Ob Osadkak Chernov Morya i ik Znachenii v Pozanii Osadochnik Gornik Porod. Bull. Soc. Naturalistes Moscou Sec. Geol., N.S. 5, 261-89. ARRHENIUS, G. 1950. The Swedish Deep-Sea Expedition. The geological material and its treatment with special regard to the eastern Pacific. Geol. Fdr Stoekh. F6rh, 72, 185-91.
BAILEY, E.B. 1936. Sedimentation in relation to tectonics. Bull. geol. Soe. Am., 47, 1713-26. BALLANCE,P.F. 1964. Streaked out mud ripples below Miocene turbidites, Puriri Formation, New Zealand. J. sed. Petrol., 34, 91-101. BERGER, W.H. 1974. Deep-sea sedimentation. In: Burk, C.A. & Drake, C.L. (eds), The Geology of Continental Margins. Springer-Verlag, New York. 213-41. BLATT,H., MIDDLETON,G. & MURRAY,R. 1980. Origin
Deep-water fine grained sediments," history, methodology and terminology of Sedimentary Rocks. 2nd edn. Prentice-Hall, New Jersey. 782 pp. B6GGILD, O.B. 1916. Meeresgrundproben der Sigboga expedition. Sigboda-Expeditie Monographie, 65, 1-50.
BOUMA, A.H. 1962. Sedimentology of some Flysch Deposits. Elsevier, Amsterdam. 168 pp. BRAMLETTE, M.N. & BRADLEY, W.H. 1940. Lithology and geologic interpretations. Pt. 1. In: Geology & Biology of North Atlantic Deep-Sea Cores. USgeol. Surv. Prof. Pap., 196, 1-24. CRAIG, G.Y. & WALTON,K. 1962. Sedimentary structures and palaeocurrent directions from the Palaeozoic rocks of Kirkcudbrightshire. Trans. Edin. geol. Soc., 19, 100-19. DALV, R.A. 1936. Origin of submarine 'canyons'. Am. J. Sci., 31, 401-20. DAVIES, T.A., MUSICH, L.F. & WOODBURY, P.B. 1977. Automated classification of deep-sea sediments. J. sed. Petrol., 47, 650-6. DE RAAF, J.F.M. READING, H.G. & WALKER, R. G. 1965. Cyclic sedimentation in the lower Westphalian of north Devon, England. Sedimentology, 4, 1-52.
DEAN, W.E., STOW, D.A.V., BARROW, E. & SCHALLREUTER, R. 1984. A revised sediment classification for siliceous-biogenic calcareous-biogenic and nonbiogenic components. Init. Repts. DSDP, 75, US Govt. Print. Off., Washington, DC. DIETZ, R.S. 1961. Continent and ocean basin evolution by spreading of the sea-floor. Nature, 190, 854-7. DUNBAR, C.O. & ROGERS, J. 1957. Principles of Stratigraphy. John Wiley, New York. 356 pp. DUNHAM, R.J. 1962. Classification of carbonate rocks according to depositional texture. Am. Ass. Petrol. Geol. Mem., 1, 108-21. DZULYNSKI, S. 8t~ KINLE, S. 1957. Problematic hieroglyphs of possible organic origin from the Belovoza beds. Soc. geol. Pol. Ann., 26, 266-9. -KSIAZKIEWICZ,M. & KUENEN, P.H. 1959. Turbidites in flysch of the Polish Carpathian Mountains. Bull. geol. Soc. Am., 70, 1089-118. EINSTEIN, H.A. & KRONE, R.B. 1962. Experiments to determine modes of cohesive sediment transport in salt water. J. geophys. Res., 64, 1451-61. GEALY, E.L., WINTERER, E.L. & MOBERLY, R. 1971. Methods, conventions and general observation. Init. Repts. DSDP, 7, U.S. Govt. Print. Off., Washington,-DC 9-26. FOLK, R.L. 1954. The distinction between grain size and mineral composition in sedimentary rock nomenclature. J. Geol., 62, 344-59. - 1959. Practical petrographic classification of limestones. Bull. Am. Ass. Petrol. Geol., 43, 1-38. -1968. Petrology of Sedimentary Rocks. Hemphills' Pub. Co., Austin, Texas. 170 pp. - 1973. Carbonate Petrography in the Post-Sorbian Age. In: Ginsburg, R.N. (ed.), Evolving Concepts in Sedimentology. Johns Hopkins University Press, Baltimore. 118-58. FucrlS, T. 1877. Ober die Entstehung der Aptychenkalke. Sber. Akad. Wiss. l/Vien Math. Nat. Klasse, Abt. I, 76, 329-34.
13
HEDBERG, H.D., (ed) 1976. International Stratigraphic Guide. Wiley-Interscience, New York. 200 pp. HEEZEN, B.C. 1959. Dynamic processes of abyssal sedimentation: erosion, transportation and redeposition on the deep sea floor. Geophys. J. R. astr. Soc., 2, 142-63. - & JOHNSON, G.L. 1963. A moated knoll in the Canary Passage. Dt hydrogr. Z., 16, 269 pp. , HOLLISTER, C.D. & RUDDIMAN, W.F. 1966. Shaping of the continental rise by deep geostrophic contour currents. Science, 152, 502-8. - - - , THARP, M. & EWING, M. 1959. The floors of the oceans. 1. The North Atlantic. Geol. Soc. Am. Spec. Pap. 165, 122 p.p. HEIM, A. 1924. Uber submarine Denudation und chemische Sedimente. Geol. Rdsch., 15, 1-47. HESS, H.H. 1962. History of ocean basins. In: Engel, A.E.J. et al. (eds), Petrologic studies: A Volume in Honour ofA.F. Buddington. Mem. geol. Soc. Am., Boulder, Co., 569-620. HESSE, R. 1975. Turbiditic and non-turbiditic mudstone of Cretaceous flysch sections of the East Alps and other basins. Sedimentology, 22, 387~,16. HINDE, G.J. 1890. Notes on Radiolaria from the Lower Palaeozoic rocks (Llandeilo-Caradoc) of the south of Scotland. Ann. Mag. nat. Hist., ser. 5, 6, 40-59. HOLLAND, C.H. et al. 1978. A guide to Stratigraphical Procedure. Geol. Soc. Lond. Spec. Rep., 11, 18 pp. HUBERT, J.F. 1964. Textural evidence for the deposition of many western North Atlantic deep sea sands by ocean-bottom currents rather than turbidity currents. J. Geol., 72, 757-85. HUXLEY, T.H. 1858. On some organisms living at great depths in the North Atlantic Ocean. Q. Jl. microsc. Sci., 8, N.S., 203-12. INGRAM, R.L. 1953. Fissility of mudrocks. Bull. geol. Soc. Am., 64, 869-78. -1954. Terminology for the thickness of stratification and parting units in sedimentary rocks. Bull. geol. Soc. Am., 65, 937-8. JOHNSON, D. 1938. The origin of submarine canyons. J. Geomorph., 1, 230-43. KELLING, G. 1958. Ripple-mark in the Rhinns of Galloway. Trans. Edinb. geol. Soc., 17, 117-32. KLEIN, G. DE V. 1966. Dispersal and petrology of sandstones of Stanley-Jackfork boundary, Ouachita fold belt, Arkansas and Oklahoma. Bull. Am. Assoc. Petrol. Geol., 50, 308-26. KUENEN, PH, H. 1937. Experiments in connection with Daly's hypothesis on the formation of submarine canyons. Leid. geol. Meded., 8, 327-35. -1950. Marine Geology. John Wiley, New York. 568 pp. -• MIGLIORINI, C.I. 1950. Turbidity currents as a cause of graded bedding. J. Geol., 58, 91-127. KUHN, T.S. 1970. The Structure of Scientific Ret'olutions. Univ. Chicago Press. 210 pp. LESLEY, J.P. 1892. Summary description of the geology of Pennsylvania. Penns. Geol. Surf,. 2nd Rep., 1, 1-720. LEWAN, M.D. 1979. Laboratory classification of very fine-grained sedimentary rocks. Geology, 6, 645-8. MIDDLETON, G.V. 1965. Introduction, Primary Sedimentary Structures and their Hydrodynamic Inter-
I4
D.A.V. Stow and D.J.W. Piper
pretation. Soc. econ. Paleont. Min. Spec. Pub., 12, 1-4. MIGLIORINI, C.I. 1946. L'eta del macigno dell'Appennino sulla sinistra del Serchio e considerozione sul rimaneggiamento dei macroforaminiferi. Bull. Soc. geol. Ital., 63, (1944), 75-90. MOLENCRAAF, G.A.F. 1922. On manganese nodules in Mesozoic deep-sea deposits of Dutch Timor. Proc. Sect. Sci. K. ned. Akad. Wet., 18, 415-30. MtJRRAY, J. & RENARD, A.F. 1891. Report on deep-sea deposits based on specimens collected during the voyage of HMS Challenger in the years 1873-1876. In: Challenger Reports, HMSO, Edinburgh. 525 pp. NATLAND, M.L. 1933. Depth and temperature distribution of some Recent and fossil Foraminifera in the southern California region. Bull. Scripps Instn. Oceanogr., La Jolla, Tech. Ser. 3, 225-30. PETTIJOHN, F.J. 1943. Archaean sedimentation. Bull. geol. Soc. Am., 54, 925-72. -1975. Sedimentary Rocks. 3rd edn. Harper & Row, New York. 628 pp. PICARD, M.D. 1971. Classification of fine-grained sedimentary rocks. J. sed. Petrol., 41, 179-95. POTTER, P.E., MAYNARD, J.B. & PRYOR, W.A. 1980. Sedimentology of Shale. Springer-Verlag, New York. 303 p.p. RUEDEMANN, R. 1935. Ecology of black mud shales of eastern New York. J. Paleont., 9, 79-91. RUNCORN, S.K. (ed.) 1962. Continental Drift. Academic Press, New York. 338 pp. SHEPARD, F.P. 1932. Sediments of the continental shelves. Bull. geol. Soc. Am., 43, 1017-40. -1948. Submarine Geology. Harper & Row, New York. 348 pp. - 1954. Nomenclature based on sand-silt-clay ratios. J. sed. Petrol., 24, 151-8. SORBY, H.C. 1861. On the origin of the so-called 'Crystalloids' of the chalk. Ann. Mag. nat. Hist. ser., 3, 8, 193-200.
1908. On the application of quantitative methods to the study of the structure and history of rocks. Quart. geol. Soc. Lond., 64, 171-233. SPEARS,D.A. 1980. Towards a classification of shales. J. geol. Soc. Lond., 137, 125-9. STEINMANN, G. 1905. Geologische Beobachtungen in den Alpen, II. Ber. naturf. Ges. Freiburg, 16, 18-67. STOCKS, T. 1933. Die Echo profile Wissenschaftliche Ergebnisse der Deutschen Atlantischen Expedition auf den "Meteor' 1925-1927, vol. 2. STOW, D.A.V. & LOVELL, J.P.B. 1979. Contourites: their recognition in modern and ancient sediments. Earth-Sci. Reviews, 14, 251-91. - - i n press. Deep-sea clastics. In: Reading, H.G. (ed.), Sedimentary Environments and Facies. 2rid edn. Blackwell Scientific Publications, Oxford. SUESS, E. 1875. Entstehung der Alpen. W. BraumfiUes, Vienna. 168 pp. VAN STRAATEN,L.M.U.J. 1967. Turbidites ash layers and shell beds in the bathyal zone of the southeastern Adriatic Sea. Revue G(ogr. phys. Gdol. dyn., 9, 219-39. VINE, F.J. & MATTHEWS,D.H. 1963. Magnetic anomalies over oceanic ridges. Nature, 199, 947-9. WALKER,R.G. 1973. Mopping up the turbidite mess. In: Ginsburg, R.N. (ed.), Evolving concepts in Sedimentology. Johns Hopkins University Press, Baltimore. 1-37. 1976. Facies Models: I. General Introduction. Geoscience Canada, 3, 21-4. WALTHER,J. 1893. Einleitung in die Geologie als historische Wissenschaft. 3 vols. Fischer Verlag. Vena. WENTWORTH, C.K. 1922. A scale of grade and class terms for clastic sediments. J. Geol., 30, 377-92. WESER, O.E. 1974. Sediment classification based on Deep Sea Drilling Project drilling. Geol. Soc. Am., Program Abstr., 6, 1073-4. -
-
D.A.V. STOW, Grant Institute of Geology, University of Edinburgh, West Mains Road, Edinburgh EH9 3JW, Scotland. D.J.W. PIPER, Atlantic Geoscience Center, Geological Survey of Canada, Bedford Institute of Oceanography, P O Box 1006, Dartmouth, Nova Scotia, Canada, B27 4A2.
A review of fine-grained sediment origins, characteristics, transport and deposition D.S. Gorsline SUMMARY: Fine-grained sediments and sedimentary rocks make up as much as 75% of the present and past sedimentary records. River discharge is the largest single source of fine-grained material, followed by biological, volcanic and aeolian sources. The composition, texture and bulk properties that characterize different sediment facies are controlled primarily by climate, tectonics, sediment supply, oceanic dispersal systems and biological activity. Fine sediments are transported into deep water by low-concentration nepheloid plumes, turbidity currents of a range of concentrations, flows and mass-movements. Mass-movement may be the major process delivering fine-grained sediment to the deep-sea floor over long time periods. Abyssal plains, the end of the passive margin transport system, are more affected by turbidity currents. Much bottom current reworking of sediments occurs at all depths in the ocean with results that may be very subtle. In shallower waters, balances between supply and dispersal may allow for the accumulation of mud belts on shelves and much fine sediments passes through such systems. Organic matter concentrations in fine sediment may be more a function of accumulation rates of organic versus terrigenous matter than local biological productivity. Fine-grained sediments require the application of methodology and results of a broad range of scientific disciplines. Effective understanding of the fine record will require interaction between several sciences.
The theme of this symposium is fine-grained sedimentation in deep water. Viewed from global sedimentation perspectives, the oceans are the ultimate sink for world sedimentation. Although continental fine sedimentary environments are not discussed here, some papers on these topics have been included in a complementary symposium edited by Reinhard Hesse (Sedimentary Geology, in press). Effective understanding of the oceanic fine sedimentation system requires cooperative studies by geologists, biologists, chemists and oceanographers. Therefore, the introductory discussion will, in part, be aimed at non-geologists. On the other hand, since many geologists are not familiar with oceanography, I will include some brief discussion of ocean circulation features that affect the distribution and deposition of fine sediments. This paper complements the paper in this volume by Stow & Piper which discusses deepwater fine-grained sedimentary facies. I will briefly touch upon facies and some special sedimentary features, but primarily from an oceanographer's view of the environmental conditions and characteristics rather than specific discussion of sedimentary structures and textures. Readers interested in the oceanic eddies and rings noted later should see the symposium volume by Robinson (1983).
Fine sediment mass, source and characteristics Mass and Source
The stratigraphic importance of fine-grained sediments is evident from conservative estimates of the proportion of silts and clays and their lithified equivalents in the geologic record. Recent authors (Potter et al. 1980; Blatt 1982) estimate that 70-75% of sediments and sedimentary rocks are fine particulate aggregates, including fine biogenic sediments and rocks. Years ago, Clarke (1924) noted that average shales contain one-third clay minerals, one-third quartz and one-third organics, carbonates and other minor minerals. The primary source of fine deposits is river discharge; and, secondarily, volcanic explosive eruptions, aeolian transport and biological production. There are other interesting sources such as chemical precipitates and meteoritic debris which contribute very minor amounts of material to the overall budget. Recent estimates for the bulk contribution of the major sources in present times give river discharge as 1.5-2 • 101~tons yr - l (Garrels & Mackenzie 1971; Milliman & Meade 1983); aeolian transport, exclusive of volcanic ejecta, as approximately 107 tons yr-1 (based on data from Prospero & Bonatti 1969; Windom 1969; Pewe 1981; Prospero 1981); the contribution of explosive volcaniclastics as approximately
17
I8
D.S. Gorsline
107 tons yr -1 (estimated from eruption sizes, Huang et al. 1979; Smith 1979); and fine-grained biogenic sediment sources of the order of 109 tons yr-1, based on stratigraphic estimates and ocean carbonate budgets (Broecker 1974; Blatt 1982). River discharge
The bulk of the annual river discharge comes from the world's 20 largest rivers, with the three largest Asian rivers contributing almost onequarter (Milliman & Meade 1983). Because of the limited number of large rivers, areas of major accumulation in the oceans are relatively few in number. For example, the few giant submarine fans are mostly tributary to the largest Asian rivers (Curray & Moore 1971). River load is primarily a function of drainage area for given climatic regions and river drainage areas are, to a first order, a function of tectonics. Inman & Nordstrom (1971) have shown that the major rivers drain into trailing edge or marginal seas due to the displacement of continental divides towards the collision edge of continental plates. As Strakhov (1970) has discussed, the amounts of sediment discharged can be related on a global scale to latitude which, in turn, controls the factors of temperature, rainfall, vegetation and evapo-transpiration. Undoubtedly some of these relationships were different for pre-Middle Palaeozoic time, but since most of the sedimentary record is younger than this (Garrels & Mackenzie 1971) we need not consider a nonvegetated world. Mid-latitude streams, or streams draining high plateaus in the case of some of the large Asian rivers, deliver the largest amounts of sediment per discharge area (Milliman & Meade 1983). The major rivers deliver very large loads of relatively fine grain-size. As noted earlier, these tend to debouch from the trailing edge of the cratons. Gibbs (1967) has shown the importance of the highlands in the headwaters of big rivers in determining the characteristics of the river particulate load. Undoubtedly the variation in land-sea surface area with time has produced matching variations in river load delivered. It is also likely that there is a minimum land area necessary to provide significant sediment discharge. Volcaniclastic contribution
Volcanic explosive eruptions tend to be associated with the andesitic and dacitic volcanoes of active margins and plate convergences and with large silicic eruptions derived from major batho-
liths (Smith 1979). Rate of plate subduction may also be a factor. Variable proportions of the erupted material are ejected as local low altitude air-fall and/or as high altitude injection into the stratosphere or high troposphere (Sheridan 1979). The deposits resulting from these events are strongly influenced by the local winds and upper atmosphere circulation, respectively. Although initially from point sources, the high atmospheric injections are mixed and dispersed so that the resulting air-fall deposits are globally distributed. Thus the general distribution of this fraction is essentially the same as the global distribution of dust from other sources (Windom 1969). The ash-falls of this large scale and their depositional record form an intriguing source of data regarding the periodicity of major volcanic events. Many authors have considered the periodicity of volcanism in the oceans and globally (e.g. Stille 1924; Umbgrove 1947; Huang et al. 1975; Ninkovitch & Donn 1975; Vogt 1979; Kennett 1981 ). In deep ocean distal pelagic sediments, the lithic contribution is probably almost entirely from wind-borne sources and oceanic volcanic input. This was probably also true in the past (e.g. Lindstrom 1974, for Ordovician pelagites). Aeolian contribution
Non-volcanic aeolian transport to the oceans tends to be from arid lands or from large outwash systems. The geographic arrangement of these types of areas and the major wind-belts determines whether material is transported to the deep ocean. Windom (1969) examined the dust falls deposited in ice of the polar caps and large glaciers and compared these values with estimates of the wind-borne contribution in lower latitude deep-sea sediments as determined by applying the composition data derived from the ice cap deposits to the components observed in the oceanic accumulations. He noted that the influx is affected by latitude, distance from source and climatic factors. (See also Rex & Goldberg 1958; Rex et al. 1969.) Loess is a major continental fine deposit in periglacial areas and downwind from arid regions. These deposits are eroded by the major stream systems draining the continental interiors. In the present mass budget, a large portion of the silt fraction may come from erosion of these glacial deposits. Analysis of the influence of arid regions upwind of ocean areas has been summarized by Pewe ( 1981), Prospero (1981) and earlier workers (see Rapp 1974). Windom (1969) states that
A review o f f i n e - g r a i n e d sediment aeolian input contributes between 10 and 75~o of the nonbiogenic fraction of deep ocean sediments depending on distance from source, latitude, dilution and effectiveness of the tracers used. Heath et al. (1973) have shown the shift in the pattern of wind-borne quartz in deposits in subtropical latitudes in the eastern Pacific over late Pleistocene time. On a shorter time-scale, the dust plumes from the Sahara into the Atlantic shift in latitude with the seasonal shift in windbelts. Prospero (1981) has noted the generally low windborne concentrations in the distant ocean areas outside the arid belts.
Biogenic contributions Biogenic sediments are a world unto themselves. Many pelagic tests, capsules, plates and spines are silt size and smaller, and so primary biological productivity is an important fine particulate source. Bioerosion is an important source of fine lime particulates (see Futterer 1974; Hein & Risk 1975). Accumulation ofbiogenic sediments represents an intricate balance between production and solution (Berger 1974). Variations in biogenic content and in the stratigraphic appearance of biogenic-influenced sedimentation are a function of dilution of biogenic production by terrigenous influx. Ginsburg & James (1974) have discussed the importance of biogenic contribution to ocean margins other than the tropics, and note that although diluted by terrigenous influx, the middle and higher latitude contribution represents a major addition to the tropical biogenic source that is most generally used as the basis for biogenic budgets. Biological production of fine particulates is, of course, related to areas of high productivity. The older view was that these were essentially in areas of major upwelling along eastern ocean boundaries, the eastern equatorial system and major ocean water mass convergences as in the Circumpolar Antarctic Convergence. More recent work has shown that western boundary currents can generate zones of quite high productivity (Dunstan & Atkinson 1976; Janowitz & Pietrafesa 1980; Blanton et al. 1981;) and that the central gyres of the surface ocean circulation are also much more productive than was thought a few years ago (Shulenberg & Reid 1981; Jenkins 1982; Martinez et al. 1983). Parrish (1983) has introduced an interesting but preliminary multicomponent model in which the spectrum of organic-rich sediments and sedimentary rocks, cherts, chalks, phosphate rocks and glauconite are related to a range of oceanographic factors and mass sediment budgets from
19
land. Certainly, as noted earlier, the older simple view of geographically restricted ocean upwelling cannot be simply applied to palaeogeographic reconstructions. Organic content alone is not simply a function of original organic production but must include consideration of lithic dilution.
Composition and bulk properties Clays Clay is defined texturally as all material finer than 4 /~m (Udden 1914; Wentworth 1922). In this discussion, clay will be used in its mineralogical sense with a grain-size usually less than 10 #m and typically less than 2 #m (referring to discrete grains and not aggregates). Mineralogically, clay is composed of a family of silicate minerals with sheet structures similar to the phyllosilicate micas. The clays represent the stable product of the weathering of feldspars and micas at earth surface conditions. Clays possess the property of ion exchange to a greater or lesser extent depending on the clay species and this property reflects the unsatisfied bonds on and within the sheet structures. This property is an important one in that it provides a mechanism for transfer of organics and metal ions and also is the primary factor in the characteristic aggregation of clay grains in ocean environments (Gibbs 1981; Gibbs & Heitzel 1982). Due to the charges on their surfaces, clays have cohesion, which is a basic factor influencing the strength of fine sediments of high clay content. Clay mineralogy reflects climate and relief, and secondarily lithology. Several authors (e.g. Loughnan 1969) have shown that a specific feldspathic rock can deliver a wide variety of clay species depending on such factors as temperature, rate of soil drainage, and chemical environment. For large areas with heterogeneous lithologies, climate is the main control. Storage time increases with increasing drainage area and this also influences composition and texture. Changes of global climate therefore exert a strong influence on fine sediment characteristics; the proportion of silt to clay and of silt +clay to coarser grain sizes. As has been shown by clay mineralogists (Loughnan 1969) and by large scale oceanic studies (Biscaye 1965; Griffin & Goldberg 1968) there is a primary latitudinal variation in dominant clay species that reflects climate. In general, tropical clays tend to be laterites or kaolinites; mid-latitudes produce montmorillonites and illites; and polar latitudes produce illites and chlorites.
20
D.S. Gorsline
Silts
Bulk properties
Texturally, silts range between 4 and 63 /~m in grain-size, although both silt and sand are part of a continuous spectrum of particle sizes. Silt is a particularly diagnostic grain-size class for studies of deposition and reworking by bottom currents and of general directions of fine sediment transport (e.g. McCave 1982). Silts are typically heterogeneous mixtures of detrital primary minerals of which quartz is most common, and feldspars and ferromagnesian minerals are common accessories. Silt origin has been attributed to glacial grinding (see Smalley 1966; Kuenen 1969), volcaniclastic eruptions (Smith 1979) and weathering processes in soils in tropical and mid-latitude regions (Pye & Sperling 1983). Many pelagic biologic tests, capsules, spines and plates are silt size, and so primary biological production produces much silt as well as clay size particulates (Nahon & Trompette 1982). Silts, because of the small particle size, have large grain surface areas and so weathering will probably rapidly remove metastable or unstable minerals leaving quartz as the dominant mineral. If, as has been suggested by Pye & Sperling (1983), much silt is formed by chemical fracturing of strained quartz in the soil profile, then silt initially begins with a high quartz content. Hydrodynamically, silts can be defined as particles that are moved primarily in suspension, while sands are moved in traction or saltation. The boundary is a broad one but probably lies somewhere between 30 and 100/~m (Inman 1949). In most rivers, fine sand is moved as suspension load at times of flood (Brownlie & Taylor 1981). In the ocean, the levels of turbulence and shear stress are lower and particles larger than about 30 ~m generally move as bedload. Under strong wave surge in coastal and inner shelf regions, sands are rarely suspended more than a few tens of centimetres above the bottom (Cook & Gorsline 1972). Wildharber (1966) found that the typical suspension load a few kilometres off southern California was already limited to particles with a coarsest diameter of 30/~m, and most were below 20 Ftm. Since fine sediments of silt size move in suspension, they will tend to preserve their original grain shape and this may be a means of pinpointing source in a given sedimentary system. Ehrlich and his associates (e.g. Ehrlich & Weinberg 1970; Ehrlich et al. 1980) have pioneered use of Fourier statistical methods to analyse ~grain shapes as a means of making source determinations. Silt grain shape analysis is a promising area for research.
There are three important bulk properties of fine-grained sediments: (a) the sediment strength, which is a function of the cohesive properties of clay particles, internal grain friction and grain packing; (b) the bulk density, which is a function of water content, sediment composition and void ratio; and (c) sediment fabric, or the stacking arrangement of clay particles and aggregates. The behaviour of fine-grained sediments on slopes depends on these three bulk properties, in addition to the slope angle and sediment load. The water content (or bulk density) appears particularly important (Field 1981b; Almagor 1982; Bennett & Nelson 1983). Thus cohesive sediments with high water contents (low bulk densities) may fail even on low gradients (Lewis 1971; Coleman 1976; Coleman & Garrison 1977); and cohesive sediments with low water content (high bulk density) may be stable on relatively high gradient slopes. To a first approximation, for a given textural type, the water content is directly related to rate of accumulation, thus it is evident that rapidly deposited sediments on slopes may be metastable or unstable and subject to mass failure. On a smaller scale, several workers have examined the microstructure of clay fabrics, and related these both to the depositional mode and to bulk properties (Bennett et al. 1981; Hein, in press). Bennett and his co-workers have examined the arrangement of clay flakes, domains (stacks of clay flakes) and other aggregate forms (pellets) in sediments of various bulk densities. As a rule, the lower density sediments have more open packing with larger voids; flakes and domains are arranged in loose chains. In denser sediments, the structures are flatter and more collapsed and voids are flat and lenticular. Hein and her associates are presently examining the three dimensional packing of fine grains in sediments from a variety of environments in the California Borderland. Preliminary work has shown that the bulk properties of the various facies and depositional environments are different. It will be interesting to see if the flake fabrics show the influence of creep or mass failure of more advanced stages of movement. O'Brien et al. (1980) have examined microfabric in older sediments and sedimentary rocks and see systematic arrangements even in quite compact and lithified sedimentary deposits.
Cycles in fine sedimentation Hemipelagic deposits may preserve almost complete records of depositional history because they
A review of fine-grained sediment usually (not exclusively) accumulate in areas of low energy. They can therefore record the results of cyclic forces driving deposition. These cyclic periods range from seconds to millions of years (e.g. Fischer & Arthur 1977; Hesse & Chough 1980). Common cycles include seasonal variations, larger term climatic variations ranging from decades to tens of thousands of years, tectonic cycles which can overlap the longer period climatic cycles and pass on to hundreds of millions of years. On a regional basis, local uplifts or depressions, stream capture and channel avulsion are examples of other quasicyclic driving forces. Cycles with periods of 103 yrs and more are usually preserved in bioturbated sediments, but may require careful analysis for their recognition. Resolution of cyclic sedimentation is limited by accumulation rates and bioturbation; records of periods of up to a thousand years or so (depending on accumulation rates) are usually destroyed by bioturbation except where anoxic bottom waters exclude benthos (e.g. Emery & Hulsemann 1962; Soutar & Crill 1977; Malouta et al. 1982; Savrda et al. 1983). Bed thickness is another factor since bioturbation is often most intense only in the top 5-10 cm (Nittrouer & Sternberg 1981) of the accumulating fine sediment, and thick rapidly deposited units may never be entirely bioturbated even in well aerated benthic environments (Savrda et al. 1983). Variations in oxygen content of bottom waters can be rapid events geologically and where accumulation rates are slow, the anoxic accumulations may be bioturbated during the following period of oxygenation. Benthic burrowing activity can be surprisingly intense even at oxygen levels of less than 0.5 ml/L (Savrda et al. 1983). Dean & Arthur (in press) and Arthur et al. (this volume) have described interactions between cycles in the Cretaceous black shales of the deep Atlantic that influence the timing and magnitude of reduction of the sediments and the degree of bioturbation. The change from dysaerobic to anaerobic conditions is apparently triggered by the shorter frequency climatic cycles at times when the longer climatic or tectonic cycles bring the environments close to anoxia. These features represent cyclic periods of 10 4 yrs and more. Human influence on sediment discharges One major problem in the use of recent deposits and mass budgets of contemporary rivers to develop analogs for ancient deposits is the pervasive influence of man's activities on erosion (Meade 1969, 1982). Most workers agree that this effect has been to increase sediment discharges
21
from rivers as a result of overgrazing and other agricultural practices, and the much more recent discharge of wastes and dredge spoil on a large scale (National Research Council 1976). Conversely, as in southern California, many rivers have been truncated in drainage area by flood control dams and spreading basins which trap sediments (Brownlie & Taylor 1981). Estimates of effects of human activities vary by an order of magnitude, but work on the sedimentary accumulations in contemporary offshore southern California basins suggests a two- to three-fold increase in accumulation rates since late Pleistocene time (Nardin 1981; Schwalbach 1982) which may reflect the initiation of human influences superimposed upon the natural climatic and sea-level changes which have also occurred during that time span. Meade (1982) has noted that after such events as intensive agricultural alteration of a drainage area, the resulting sediment is stored within the drainage system and released over time periods of as much as centuries. Bourrouilh (pers. comm. 1983) has noted that the construction of the Aswan Dam on the Nile has generated major problems over the Nile delta. These effects include the influence on natural fertilization of the agricultural lands by cutting off the annual deposition of silts, and changes in habitats of economically important fish species in the areas off the mouth of the Nile.
Controls on fine sediment deposition Tectonics The tectonic setting of the source area or depositional site exerts a first order control on finegrained sedimentation. It affects the rates of uplift and denudation drainage patterns and volumes of river discharge, coastal plain and shelf widths, margin gradients, gross sediment budgets, the morphology of receiving basins and local sealevel changes. The relative activity, or style and frequency of seismicity and faulting, is also of primary importance to sedimentation. It may exceed and mask the effects of the other controls or, in areas of low tectonic activity, play a secondary role. Supply As is true for all sedimentary systems, the interaction of supply and dispersal energy can produce a variety of fine sedimentary deposits, facies and structures for given tectonic and sea-level condi-
22
D.S.
Gorsline
tions (Sloss 1962). As we have seen, the principle fine sediment supply is from river contributions. Initially this is deposited almost entirely in the continental margin as a result of capture, filtering and pelleting processes that have been reviewed. Where this supply is constrained by climatic, oceanographic or bathymetric barriers, the secondary sources become dominant as for biogenie contributions of either carbonate or siliceous composition (Ginsburg & James 1974).
Oceanic dispersal systems Dispersal is achieved by different dominant processes or agents as we progress from shelf to deep ocean floor. Shelves lie within the mixed surface layer of the oceans where strong wave, tide and wind stress currents are active. Although part of the ocean wave spectrum, internal waves should be specifically mentioned here. These have their largest development at density discontinuities in the ocean water column. Cacchione & Southard (1974) have discussed these wave forms and Swift (1973) has discussed their influence on shelves. Slopes receive energy from gravity, large-scale eddy interaction with the substrate and slope currents. Deep ocean floors are affected by tidal currents and large-scale deep slow circulations and possibly by transient stronger flows perhaps associated with large-scale rings and eddies. Where topography reduces the flow cross sections as over gaps in mid-ocean ridges or over and around seamounts, the tidally driven and large water mass circulations are accelerated and can produce subtle to strong substrate effects as noted earlier. In all environments transient effects can be important as in the passage of a turbidity current, or a mass failure or storm effects in shallower waters.
Sea-level As has been noted by Rona (1973), Pitman (1978) and Watts (1982), sea-level position relative to the shelf edge is a major factor in the delivery of continental sediments to the deep ocean. At high sea-levels, decreased stream relief, broad shelves and possibly slower atmospheric and ocean circulation rates (in the case of climatically driven sea-level changes) tend to produce conditions of trapping of sediments on the submerged craton edge. When sea-level falls to or below the shelf edge, delivery is almost entirely to the slopes and the deep-sea floor. Sea-level can be both tectonically and climatically driven (Morner 1974; Sclater et al. 1977) and the oscillations over time will show significant spikes in power spectral analysis that correspond to the two major driving
factors. The tectonic signal will be of the order of 106 yr and the climatic signal will be of the order of 104-105 yr (the Milankovitch cycles). The onset of late Tertiary continental glaciation and the resulting rapid sea-level changes, has been a major factor in world sediment budgets. Laine (1980) has shown that much of the sediment in the North Atlantic abyssal plains probably was delivered during the glacial epochs. This would have required very high sediment discharges from periglacial streams. Emery & Uchupi (1972) have noted the concentration of submarine canyons along the edge of the northeast Atlantic margin of the US and Canada matches the area and latitude of the Pleistocene ice extent. This indicates that at low glacial sea-levels, large quantities of glacial outwash were dumped at the shelf edge and redeposited downslope by turbidity currents and other massmovements.
Biological productivity Biological production occurs over large areas of ocean surface waters where nutrient recharging is active (Eppley & Peterson 1979), therefore the sedimentation patterns will reflect these areal inputs rather than point or linear sources. This differentiation or classification is of interest because it will affect the concentration of particles in the water column and thereby control the accumulation rates on the underlying substrates. Plumes and jets associated with areas of coastal upwellings approximate point sources, and regional coastal upwellings are typically linear sources (Gorsline 1978). Large-scale eddies and rings usually move over trajectories of considerable distance and thus their sedimentation effects are integrated over large areas (Swallow 1976). An exception may occur in the California Current System where the seasonally developed large eddies appear to remain essentially fixed in position (Koblinsky et al. 1983; Simpson et al. 1983). If this pattern has been characteristic of the past few thousand years of roughly stable sealevel position and ocean circulation, then there should be some record of their existence preserved in the sediments. Such eddies can be areas of relatively high productivity (Chelton et al. 1982; Haury 1983).
Biological pelletization Planktonic organisms use several means to filter particulates from the water column, aggregate them and then produce large pellets that sink rapidly to the sea floor (e.g. Osterberg et al. 1964; McCave 1975). The processes include filtering,
A review of fine-grained sediment digestion and excretion (e.g. Osterberg et al. 1964; Frankenberg & Smith 1967); capture on films or extended nets (Gilmer 1972; Bruland & Silver 1981; Silver & Bruland 1981); generation of particulate organic matter from dissolution of bubbles in sea-water (Johnson & Cooke 1980); and large-scale pellets from fish feeding (Robison & Bailey 1981). Some small fraction of fine, dispersed grains do get through to the deep open ocean, but it is probably less than 5% of the total sediment injected from the continents. This is supported by the slow accumulation rate of deep pelagic nonbiogenic sediments. These contain mainly oceanic volcanic material and aeolian material, so the transfer of fine particulates through the coastal and margin systems must be very small. Porter & Robbins (1978) have recognized preserved pellets in older shales and mudstones and point out their close similarity to contemporary copepod faecal pellets. Bioturbation In the deep oceans of the present world, bioturbation is almost ubiquitous except in very restricted basins where biological oxidation demands are high or where sills intercept low oxygen water from the Intermediate Water masses of the world oceans (e.g. California Borderland, Cariaco Trough). The degree of benthic reworking of deep floor substrates is related to the oxygen content of the bottom waters. Bioturbation is limited only by the lowest oxygen content (less than 0.5 ml/L) and some organisms (e.g. nematodes) can continue to live in near-anoxic conditions. These do not completely bioturbate the sediment but leave characteristic fine networks of tiny burrows in the otherwise unreworked primary sedimentary laminations. In the ancient oceans, times of deep ocean anoxic conditions are not uncommon and have been noted above and in other papers in this volume. Dilution In the southern California borderland basins, the trapping effect of the inner basins screens out much of the tcrrigenous supply and so the outer basins are zones of relatively slow sediment accumulation and biogcnic fractions approach half the volume of the fine hemipelagic sediments. Cores in the central and outer basins reveal that the carbonate content decreases during times of glacially lowered sea-level due to augmentcd influx of detritals from exposed banks or the direct delivery of river load to the exposed shelf edge on the adjacent continent. Use of dating
23
methods shows that in actuality, both biogenic accumulation rates and terrigenous rates increased at times of lower sea-level corresponding to times of increased ocean circulation rate. The biogenics increased by perhaps twofold while the terrigenous output sometimes increased almost an order of magnitude (Gorsline & Prensky 1975; Gorsline 1981). Similar patterns have been observed in the deep ocean (e.g. Broecker et al. 1958).
Processes of fine sediment transfer to deep water The discussion will consider the large-scale primary processes of transfer of particulates resulting from water motion and the secondary transfer processes that are related primarily to massmovement processes. Stow & Piper (this volume) discuss sedimentary structures characteristic of various transport processes which complements and extends this discussion. The transport processes include low concentration (less than 10 /agm L-1) nepheloid plumes, high concentration plumes at or near the mouths of large rivers (more than 10 mg L-1), resuspension of fine sediments by strong bottom currents, turbidity currents, flow as matrix of debris flows (high concentration flows; Lowe 1976, 1979, 1982), and by massmovement of various types (Nardin et al. 1979a). These processes can be grouped into those operating throughout the water column, which are generally continuous although with seasonal and longer term variations, and those that occur along the bottom, which are usually discontinuous and discrete events including turbidity currents and other mass-movements (Gorsline & Emery 1959). The continuous processes usually involve large volumes but contain low concentrations per unit volume and thus produce relatively slow accumulation rates at any single locality. The discontinuous processes involve smaller water volumes, but because of their high concentrations and high input rates, produce much larger accumulation rates at a given point. Viscous boundary layer effects are likely to be preserved as sedimentary structures in the latter (Hesse & Chough 1980; Stow & Bowen 1980); the influence of boundary layer processes on the low concentration continuous depositional processes may appear as regional textural gradients in the surficial sediments (McCave & Swift 1976). Plumes and nepheloid-layers The discharges from the world's large rivers typically exit from the seaward mouths of deltaic
24
D.S.
distributaries as turbid plumes (Meade et al. 1975; Coleman 1976; Gibbs 1976; Coleman et al. 1981). Seasonal variations may be very important and long term climatic cycles that affect run-off will exert a major effect. The turbid plume interacts with the shelf circulation. In some instances, the season of flooding may be contemporaneous with shelf current patterns that are different from those of other seasons. This is the case off the Oregon-Washington coast where the floods typically occur at times when the shelf circulation moves coastal waters to the north in contrast to the mean annual flow to the south. Discharge via deltaic and estuarine systems is typically as flocculates resulting from ion exchange between the clays and saline ocean waters. High particle concentrations and biological factors facilitate the flocculating process. Combinations of tidal exchange and salinity gradient processes tend to hold the suspended fine particulates in the head of the estuaries, producing turbidity maxima in or near the zone of maximum salinity gradient (Postma 1967). Tidal effects interacting with river flow can alter this position to some degree (Allen et al. 1980). When fine particulates enter the ocean via rivers, coastal erosion, or airborne infall, they slowly fall through the water column. Since aggregation occurs by various processes, the initial particle size distribution is rapidly altered (see Kranck 1975; Paffenhofer et al. 1979). Depending on the bulk density of the aggregates, they may fall much faster than the discrete fine particles. It should be remembered that biologically aggregated particles are attractive sites for bacteriological colonization. Bacterial degradation of biogenic pellets can be well advanced after a few days. As these particles fall, therefore, they are increasingly broken down and become less dense in bulk with concommitant decrease in settling velocity (Pomeroy & Diebel 1980; Pomeroy, in press). Secondary aggregation near-bottom can be effective when near-bottom pelagic populations of pelagic filter feeders are present as in the example of benthopelagic holothuroids in deep waters of California Borderland Basins (Carney, pers. comm. 1983). The fall velocity depends on water viscosity also (viscosity varies inversely with temperature). In deeper colder waters, the velocity will decrease. In addition, the ocean is a density stratified system with many density discontinuities. Each of these tends to slow particle fall because of turbulence at the boundary. As a result, particle concentration in the ocean is non-uniform with depth as was noted by Kalle (1939) and as further discussed by Jerlov (1953~ 1968). Their work,
Gorsline
based primarily on optical methods, has been verified by later workers using both optical and direct sampling methods (see Eittrem et al. 1969; Drake 1972). Rodolfo (1964, 1970) used water sampling and filtration in continental margin waters to define these zonations. He noted a surface turbid layer most strongly developed near the coast and river sources (Manheim et al. 1972; Meade et al. 1975), mid-water layers, and a bottom layer. These zones are typically layers with particle concentrations of/~gm/L and maximum concentrations of a few mg/L near the months of rivers, and are commonly known as nepheloid-layers (Eittrem et al. 1969; Ewing & Connary 1970). They are now recognized to be an important pathway for much of the fine open ocean particulate transport. The bottom nepheloid-layer may be up to 2000 m in thickness although the highest concentration is typically within a few tens of metres of the bottom. Larger aggregates probably pass rapidly through these nepheloid-layers to the bottom, although bacterial degradation of pellets falling through the water column can increase the layer concentration (Honjo et al. 1982; Pomeroy, in press). In the Pacific (Ewing & Connary 1970), the bottom nepheloid-layers appear to be moved in large gyres within the major deep basins of that ocean and draw much of their lithic fraction from the western continental margins of North America. Variations in thickness may in part result from bottom erosion. Recent work in the eastern tropical Pacific (Lonsdale 1976; Heath et al. 1976) suggests that winnowing occurs in narrow passages between basins, with the subsequent transfer of suspended particulate material to the bottom nepheloid-layers of the deeper basins. Thus, some of the fine material deposited in one depression may have its source from an adjacent basin or sill. In any event, a conservative estimate of the total suspended load in ocean water at any given moment is of the order of 10 9 tons (Lal 1977; calculations by Gorsline). McCave (1982) has noted that the fine particulates must pass into the benthic boundary layer at the base of flows. As the bottom shear increases these small scale processes become increasingly important at given concentrations. When this layer is developed the processes within the layer are both erosional and depositional on a micro scale. Thus effects of even weak bottom flow may be subtlely recorded in the sediments deposited.
Resuspension by bottom currents Bottom currents can resuspend and sort fine deposits and particles at all depths in the ocean (Heezen et al. 1966). High velocity flows are able
A review of.fine-grained sediment
to shape large-scale bedforms and act as an agent of major fine sediment winnowing (Lonsdale & Malfait 1974). More subtle textural and grain orientation properties of the sediments can reveal data about flow direction and winnowing in the silt and clay particle size ranges by lower velocity currents (Johnson et al. 1977; Ledbetter & Ellwood 1980). Drake & Gorsline (1973) have reported observations in California submarine canyons that suggest periodic reworking of canyon sediments. Their studies were made at the same time as near-bottom current-metre measurements by Shepard and his associates (Shepard et al. 1979), which show that near-bottom tidally generated water motions can resuspend fine sediment and move it progressively up or down canyon. Karl (1976, 1981) has given numerous examples of the reworking of shelf substrates by bottom currents in shallow shelf waters off southern California. His work also supports the laboratory studies by Cacchione & Southard (1974) in which the strong water motions are associated with shoaling internal waves. These typically concentrate at the base of the thermocline. Work by J. D. Smith and his colleagues at the University of Washington has modelled the influence of bottom current shear on fine sediments on shelves and predicted the depth of reworking and the resulting textural changes (e.g. Jumars et al. 1981; Nowell et al. 1981;). Nittrouer & Sternberg (1981) have also noted the reworking of fine mid-shelf mud belts associated with the Columbia River discharge and the progressive winnowing of fines over a period of decades (also see Baker 1976). Larsen (1982) has discussed the influence of seaward directed dispersive flow generated by wave groups and longer waves working on the coast as a mechanism for reworking and seaward transfer of fine sediments. McCave (1982) and McCave et al. (1982) have reviewed the results of the high energy benthic boundary layer experiment (HEBBLE) studies on the deep continental rise off the Atlantic margin of the US and showed the influence of transient strong bottom currents on the substrate. Yingst & Aller (1982) have described the effects of bioturbation on the same substrates. Thus both biologic and physical processes may overprint to produce a final sedimentary depositional structure assemblage. The large scale eddies and rings associated with Gulf Stream turbulence can reach to the deep-ocean floor, at least at inception of the eddy or ring, and may be the source of the deep water transient flow events noted in the HEBBLE area.
25
Turbidity currents
Turbidity currents are turbulent, relatively high concentration flows (Nardin et al. 1979b; Middleton & Bouma 1973) that require a high mud content to carry the sands and gravels characteristic of the early phases of this mechanism. As the coarser fractions settle out leaving distinctive structures and facies as a record of flow conditions (Bouma 1962), the remaining fine suspensates are laid down as distal turbidites on the lower slope or fan and abyssal floor, or in overbank deposits away from the conducting channels (Walker 1967; Haner 1971). These fine turbidites are typically finely laminated, graded, homogeneous or combinations of these features (Piper 1978; Stow 1979; Stow & Bowen 1980; Hesse & Chough 1980; Thornton 1981a). As contemporary turbidite basin deposits have been more closely studied, it is evident that thicker layers (up to several metres thick) are also common. Stanley (1981) and Stanley & Maldonado (1981) have called these massive layers unifites. They apparently represent rapid deposition from large muddy turbidity flows generated by slumping of unstable slope sediments. Thornton (1981 a) has described smaller examples from Santa Barbara Basin in the California Borderland, that contain transported microfauna from shallower slope depths. Since the recognition of the subtle grading and occasional faint lamination requires X-radiography and detailed textural analysis, these have probably been missed in the stratigraphic record. It is likely that turbidity current-deposited fine sediments are much more common in shales than previously supposed. Bioturbation may erase the thin-bedded forms, but the thicker units (more than 10 cm) will probably survive biologic stirring. Debris-flows and other high concentration flows
In a theoretical sequence, canyons store sediment from shelf and coastal sources and periodically release masses of the stored sediment in response to a variety of triggering forces (see Gorsline 1980). The sequence of processes that results is slides and slumps, progressing to high concentration flows and debris-flows and then turbidity currents. Changes in slope, concentration, channelization and textural composition produce a variety of facies and sedimentary structures (e.g. Mutti & Ricci Lucchi 1978; Nardin et al. 1979; Hein 1982, in press). Deposition of sands and gravels from high concentration flows and of pebbly muds from debris-flows is probably rapid. As the mass velocity slows to some critical level, the coarse mass 'freezes' as a slug.
26
D.S.
Gorsline
mud belts. The shift in position will be the response to the interaction of the two main Slope sediments are strongly affected by mass- variables (supply and dispersal) at given sea-level movements (Dott 1963; Emery & Uchupi 1972; positions. Inner shelf mud belts are exemplified by the Jacobi 1976; Embley 1976; Haner & Gorsline 1978; Nardin et al. 1979; Damuth 1980). This is Amazon muds that deposit along the Surinam recorded in the geologic record and in acoustic coast (Wells & Coleman 1981). These form seismic profiles in the contemporary oceans by aggrading belts that cause a general progradation contorted and sheared bedding, and by smaller of the coast. High mud supply is the basic factor scale evidence of displacement based on displaced (also see Augustinas 1978). The muds are of low faunal remains and textural anomalies (Douglas bulk density initially and actually damp out 1981). The magnitude of the material lost from a incoming wave energy. The cusps of mud accregiven slope or palaeoslope surface in a sedimen- tionary deposits have wave lengths of about 100 tary section can be determined by study of the km. Mid-shelf mud belts (of the order of 100 km bulk properties of the section. If a zone is overcompacted as compared to immediately long) have been studied offthe Washington coast, overlying beds, or if surface sediments are over- the New England shelf and off the Eel-Klamath compacted, estimates can be made of the unload- Rivers of northern California (Emery & Uchupi ing that has occurred (Booth 1979). Almagor 1972; Baker 1976; Field 1981a, respectively). All (1976) has shown that a regional slope profile can three areas represent dynamic accumulations be drastically modified by large-scale mass-move- where seasonal flood discharge of fine particuments (also see Emery & Uchupi 1972; Seibold & lates is greater than the capacity of shelf dispersal Hinz 1973). Mass-movement is a major factor processes. Offthe Columbia and the Eel-Klamath volumetrically in continental slope transport and Rivers, the flood stages come at times when the shelf current set is to the north and thus both mud is particularly characteristic of fine sediments. A variety of triggering forces can generate belts trend north from the source river mouths. mass-movement including earthquake shocks, The fines are constantly recycling through the sudden or rapid loading, undermining or under- deposits with periods of a few decades (Nittrouer cutting, and gas charging. Several writers have et al. 1979; Nittrouer & Sternberg 1981). The considered the influence of benthic sediment Washington deposit is graded with depth as a reworking on sediment strength. Richardson & result of the reworking of the accumulating silts Young (1980) have found that bioturbation can and clays over time and the basal, somewhat either increase or decrease strength, whereas coarser particles probably represent an equilibDrake (1976) suggested that bioturbation de- rium residuum of the process. The east coast belt creases cohesion and strength. Modelling studies probably has its source as the fine sediments of of the erodibility of substrates on shelves and the the Bay of Fundy and Gulf of Maine rather than a effects of burrowing organisms by Nowell et al. discrete river source. This deposit appears to be in (1981) and Jumars et al. (1981) showed the dynamic equilibrium. On many shelves, small mud patches are found theoretical effects on strength of a range of biological influences, and predict the resulting of scales an order of magnitude or more smaller substrate sensitivity, erosion and redeposition in than the mid-shelf mud belts discussed above. On response to strong near-bottom fluid stress as the west coast of the US the shelf systems are probably sufficiently close to dynamic equilibrium during storms. so that major changes in shelf sediment character can occur when the systems are perturbed. An example is the effect of the 1969-70 floods in the Depositional sites for fine southern California area which delivered the sediments in the ocean largest sediment discharges since 1938 (Curtis et al. 1973). These discharges overwhelmed the Shelves dispersal processes and the result was a mud Fine sediments can form appreciable deposits on deposit on the shelf that was over 30 cm thick shelves and shallow marine platforms when the inshore (Kolpack 1971; Drake et al. 1972) and supply exceeds the dispersal rate (Sloss 1962; required almost two years of normal wave reworkMcCave t972). Stanley et al. (1983) have dis- ing to remove. This shelf is normally sandy with an cussed the mudline as a textural marker that outer silty sand zone and occasional compact mud demarcates the positions of mud belts on shelves. patches a few hundred square metres in area. The Several scenarios have been examined: coastal reworking of the shelf deposits had a secondary mud belts, mid-shelf mud belts and shelf edge effect on the adjacent basin sedimentation which
Mass-movement: slumps and slides
A review o f f i n e - g r a i n e d s e d i m e n t has been discussed by Fleischer (1970b), Kolpack (1971) and Drake (1972). Outer shelf mud belts are also common. Off southern California, in the San Pedro area, Gorsline & Grant (1972) have described an outer shelf belt of very fine silty sand that is bioturbated and appears to be accumulating. Recent work by Drake et al. (in press) gives dynamic support to this since they show that the seasonal shift in the threshold depth for wave stirring of fine sediments essentially bounds the outer shelf fine silty belt. That some reworking by low velocity currents does occur is shown in profiles of water turbidity collected over a multi-year period by Karl (1976) in which increased turbidity at and above the bottom is seen at the depth of intersection of the local seasonal thermocline and the substrate. This may be the result of breaking internal waves striking the shelf at those depths. The dispersion of wave energy seaward and return flow generated by build-up against the coast by onshore wind stresses can be the sources of offshore net motion of fine particulates once they have been stirred into the water column by storm wave surge of tidal currents (Larsen 1982; Drake et al., in press). Slopes
Fine sediments moving off the shelf are typically deposited on the adjacent slopes. Both along slope currents (Drake, pers. comm.) and biological pelletization tend to favour their accumulation on the slope. Haner & Gorsline (1979) have reported that the slopes with most mass-movement activity off southern California are those that are in the path of turbid plumes passing seaward off promontories and from gyres centred over broad shelf segments (also see Davis 1980; Thornton 1981b). Those that are not so affected are typically the areas of highest stability, least mass-movement and slowest sediment accumulation rate. Similar convergences of nepheloid plumes and slopes along other coasts must determine the sites of highest accumulation rates and least stability. The extreme end member of the possible series of slope sedimentation types is the slope off the mouth of a major delta distributary. In these areas (e.g. Mississippi Delta, Coleman 1976) the slope sediments are deposited on gentle gradients at very high rates and creep and fluid failures are common. As discussed earlier, mass-movement and gravity influence dominates this depositional environment. Where sediment accumulations are large, the scale of the mass-movements also increases as witnessed by the large-scale failures on the Atlan-
27
tic margin of the US described by Emery & Uchupi (1972) in their comprehensive review of that margin. Such mass-movements must contribute a large amount of sediment to continental rises on passive margins and to trench floors on active margins. In the California Borderland, a transform margin, the basins of the northern inner borderland in the zone of highest fine sedimentation (Gorsline 1981; Schwalbach 1982) typically have slopes that are dominated by mass-movement features (Gorsline 1978; Nardin et al. 1979a). Deep sea and basin floors
The floors of marginal basins, trenches and abyssal plains are the end of the slope gradient from continental platform to deep ocean floor and, as such, represent the ultimate sediment sink in the oceans. Reworking and sediment transfer can still take place, particularly in gaps and sills in seamounts and oceanic ridges (Lonsdale & Malfait 1974; Chamley 1975; Diester-Haas 1975). In small marginal basins, sea-level position is an important control on sedimentation rate in general (Rona 1973; Pitman 1978; Watts 1982). This was also true of the Atlantic in Late Neogene times (Laine 1980) when glacial influx was a major source for the construction of the western abyssal plains and the North American slope and rise areas. In Pacific type oceans with active margins, the deep basins are starved and receive mainly hemipelagic sediments plus locally derived turbidity current and mass-movement deposits. In marginal basins, trenches and Atlantic type abyssal plains, turbidity current deposition is a major factor together with continuous nepheloid-layer transport. Bennetts & Pilkey (1976), Bornhold & Pilkey (1971) and Malouta et al. (1981) have described the area and volumes of a number of recent turbidites in Atlantic abyssal plains and marginal basin regimes. Elmore et al. (1979) described a turbidite from Hatteras Abyssal Plain that extended over a distance of at least 500 km. All of these turbidites involve volumes of sediment that range from 108 to 101~m 3, and that represent very large initial sediment accumulations in the source areas, although the resulting deposits are thin turbidites. Thus the important characteristic of this environment is an area which is large compared to bed thicknesses that are relatively small (centimetre range). As Ewing & Connary (1970) have stated, the nepheloid plumes in deep ocean basin plains are controlled by the deep-water circulation. As a result of earth rotation and tidal influences these deep circulations are gyres and thus suspension
28
D.S. Gorsline
transport is not directly to the basin centre but rather slope-parallel with much longer travel paths than a simple downslope flow over the gentle basin slopes. Such circulations could lead to radial transfer of suspension load to the low velocity centre of the basin-centred gyre with slightly higher sedimentation rates in mid-basin. Some limited evidence of this has been accumulated for small margin basins (Malouta et al. 1981). Isopach maps of Holocene basin sedimentation in the basins of the California Borderland show similar thicker accumulations in mid-basin rather than in the areas of associated fans (Schwalbach 1982). This reflects the low level of depositional activity in fans in the present time of high sea-level and also the probable influence of deep-basin water circulation on nepheloid transport. Data reported by Drake (1972) for Santa Barbara Basin in the borderland also suggests the presence of gyral flow in the deep-basin water as well as some possible long wave effects perhaps resulting from tidal resonance. Large-scale rings and eddies associated with large ocean boundary current turbulence and instability may affect the deep-sea floor. As shown in data from Koblinsky et al. (1983) the rings have diameters of 100-200 km and extend to depths of at least 1500 m (the maximum sampling depth) in the California Current System. The Gulf Stream eddies and rings may extend to the sea floor at times of formation and since many of these progress into the western north Atlantic, they may have an important effect on deep-sea floor substrates. The effect may be subtle and probably requires careful textural analysis together with grain orientation studies and represents an interesting problem for study. Fine-grained biogenic sediments are an important facies in pelagic sediments of the contemporary deep-sea floors. They occur in areas of high productivity, low terrigenous influx and relatively slow accumulation rate compared to ancient biogenous sediments (less than 5 cm/1000 yr versus 15 cm/1000 yr; Hakansson et al. 1974). The chalks and diatomites which are widely distributed in some parts of the geologic record appear to have been shelf or platform deposits (Hakansson et al. 1974; Surlyk & Birkelund 1979; Bottjer, pers. comm.). For example, the extensive chalks of the Cretaceous were deposited in water depths of less than 200-300 m. Diatomites represent times of high nutrient supply whereas the chalks formed at times of relatively reduced nutrient supply. Both are formed in starved environments where other factors have screened out the continental detrital contribution. Analysis of chalk stratigraphy has produced evidence for cycles with periods of the order of 104 yr that may match
the climatic cycles resulting from inequalities of solar insolation at high latitudes related to variation in the Earth's rotation, axial wobble and precession (Milankovitch cycles).
Conclusions It is obvious when looking at the broad spectrum of fine-grained sediment research, that it is an area of sedimentary petrology that requires interactions with several other disciplines. At least since late Palaeozoic time, most fine terrigenous particles have probably been worked upon by organisms either in transit or after deposition. Since fine sediments move in suspension for at least part of their transport history, oceanography is a necessary consideration in any palaeogeographic reconstruction of ancient depositional environments. Since fine sediments are associated with organic matter, partly due to similar transport pathways and also due to absorption of organics on clay flake surfaces, fine sediments are hosts to biotas that feed on this food source. Biological activity, including bacterial processes, alters the chemical environment after deposition and is a major factor in diagenetic change. Thus, microbiology and marine biology are important data sources for the stratigrapher and sedimentologist. Bioturbation can erase the primary physical structures that diagnose transport processes and so require very detailed analysis of mudstone and shale sections to see the evidence of textural or compositional differences that signal different processes such as pelagic infall versus fine turbidite deposition. It is evident that associations of organic matter and fine sediment are not restricted to a given ocean condition; e.g. upwelling in the classic sense. Rather, the organic content may be more a function of dilution by terrigenous input rather than high productivity (Gorsline 1981). The work to date and much of the work reported in this volume shows that the fine sedimentary record must be extensively restudied. Much of the shale record may actually be composed of small scale mass-movement (creep) and fine turbidites. Reworking may be much more extensive and it will require careful and very detailed analysis to define small hiatuses. Micropalaeontologists (e.g. Douglas 1981) are discovering that microfaunal remains are much more mobile than might be thought in the absence of evidence of large-scale turbidity current flow or large mass-movements. Thus many supposed hemipelagic records may contain mixed fossil assemblages in what may appear to be pelagic
A review of fine-grained sediment infall a c c u m u l a t i o n with no evidence of lateral transport. W h e n fine sediment depositional records are examined in detail, cyclicity is evident at a variety o f scales. This is an i m p o r t a n t research area in which m u c h w o r k needs to be done. W h e n fine sediments are deposited in anoxic environments, the fine detail of cycles as small as seasons can often be deciphered from the preserved primary structures. The bulk of the geologic record is fine-grained sediment. Therefore most of geologic history is in this part of the section. We have only begun to read this record.
29
ACKNOWLEDGEMENTS; M u c h of the research u p o n which this discussion is based has been sponsored by various grants over the past 10 years from the N a t i o n a l Science F o u n d a t i o n . Their continued support is most gratefully acknowledged. Drs. R. Bourrouilh, R. D u t t o n , R. Douglas, D. D r a k e and S. T h o r n t o n all read one or more versions of this paper and their c o m m e n t s were most helpful. Drs. D. Stanley, J. Syvitski and D. Stow all read earlier versions and also m a d e useful comments. The degree of incompleteness or u n d e r s t a n d i n g is solely the responsibility of the author.
References ALLEN, G.P., SALOMON, J.C., BASSCULLET, P., DU PENHOAT,Y. and DE GRANDPRE,C. 1980. Effects of the tide on mixing and suspended sediment transport in macrotidal estuaries. Sed. Geol., 25, 51-68. ALMAGOR, G. 1976. Physical properties, consolidation processes and slumping in recent marine sediments in the Mediterranean continental margin slope of southern Israel. Geol. Survey Israel, Rep., MG/76/4, 1 3 2 pp. - 1982. Marine geotechnical studies at continental margins: a review--Part II. Appl. Ocean Res., 4, 130-50. AUGUSTINAS,P.G.E.F. 1978. The Changing Shoreline of Surinam (South America). Unpub. Ph.D. Dissertation, Univ. of Utrecht, Netherlands. 232 pp. BAKER,E.T. 1976. Distribution, composition and transport of suspended particulate matter in the vicinity of Willipa Submarine Canyon, Washington. Bull. geol. Soc. Am. Bull., 87, 625-32. BENNETT, R.H., BRYANT,W.R. & KELLER,G.H. 1981. Clay fabric of selected submarine sediments: fundamental properties and models. J. seal. Petrol., 51, 217-32. - & NELSON,T.A. 1983. Sea floor characteristics and dynamics affecting geotechnical properties at shelf breaks. In: Stanley, D.J. & Moore, G.T. (eds) The Shelf Break: Critical Interface on Continental Margins. Soc econ paleo. Min Spec. Pub., 3 3 . BENNETTS, K.R.W. & PILKEY, O.H. 1976. Characteristics of three turbidites, Hispaniola-Caicos Basin. Bull. geol. Soc. Am., 87, 1291-300. BERGER,W.H. 1974. Deep-sea sedimentation. In." Burk, C.A. & Drake, C.L. (eds), The Geology of the Continental Margins. Springer-Verlag, Berlin. 213-41. BISCAYE, P.E. 1965. Mineralogy and sedimentation of recent deep-sea clay in the Atlantic Ocean and adjacent seas. Bull. geol. Soc. Am., 76, 803-32. BLANTON, J.O., ATKINSON, L.P., PIETRAFESA, L.J. & LEE, T.N. 1981. The intrusion of Gulf Stream Water across the continental shelf due to topographicallyinduced upwelling. Deep Sea Res., 28, 393-405. BLATT, H. 1982. Sedimentary Petrology'. W.H. Freeman & Co., San Francisco. 564 pp.
BOOTH,J.S. 1979. Recent history of mass wasting on the upper continental slopes, northern Gulf of Mexico as interpreted from the consolidation states of the sediment. In: Doyle, L.J. & Pilkey, O.H. (eds), Geology of the Continental Slope. Soc. econ. Paleo. Min. Spec. Pub. 27, 153-64. BORNHOLD, B.D. & PILKEY, O.H. 1971. Bioclastic turbidite sedimentation in Columbus Basin, Bahamas. Bull. geol. Soc. Am., 82, 1341-54. BOUMA, A.H. 1962. Sedimentology of some Flysche Deposits. Elsevier, Amsterdam. 168 pp. BROECKER, W.S. 1974. Chemical Oceanography. Harcourt Brace Jovanovich Inc., New York. 214 pp. BROECKER, W.S., TUREKIAN, K.K. & HEEZEN, B.C. 1958. The relation of deep-sea sedimentation rates to variations in climate. Am. J. Sci., 256, 503-1~/. BROWNLIE, W. R. t~ TAYLOR, B.D. 1981. Sediment management for southern California Mountains, coastal plains and shorelines; Part C--Coastal sediment delivery by major rivers in southern California. Calif. Inst. of Technology, Report EQL 17C, Pasadena. 314 pp. BRULANO, K.W. • SILVER,M.W. 1981. Sinking rates of fecal pellets from gelatinous zooplankton (salps. pteropods, doliolids). Marine Biol., 63, 295-300. CACCHIONE, D.A. & SOUTHARD,J.B. 1974. Incipient sediment movement by shoaling internal gravity waves. J. geophys. Res., 70, 2237-42. CHAMLEY,H. 1975. Influence des courants profonds au large du Bresil sur la sedimentation argileuse recente. IXth International Sedimentological Congress, Nice, Resumes des Publications, VIII, 1. CHELTON,D.B., BERNAL,P.A. & McGOWAN, J.A. 1982. Large-scale interannual physical and biological interaction in the California Current. J. mar. Res., 40, 1095-125. COLEMAN, J.M. 1976. Deltas - Processes of Deposition and Models for Exploration. Continuing Education Publications, Inc., P.O. Box 2590 Sta. A, Champagne, Illinois. 102 pp. -8~ GARRISON, L.E. 1977. Geological aspects of marine slope stability, northwestern Gulf of Mexico. Marine Geotech., 2, 9-44. --, ROBERTS, H.H. MURRAY, S.P. & SALAMA, M.
3o
D.S. Gorsline
characterization of grain shape. J. sed. Petrol., 40, 205-12. Err'rREM, S.L., EWING, M. & THORND1KE, E.M. 1969. Suspended matter along the continental margin of the North American basin. Deep Sea Res., 16, 613-24. ELMORE, R.D., PILKEY,O.H., CLEARY,W.J. & CURRAN, H.A. 1979. Black Shell Turbidite, Hatteras Abyssal Plain, western North Atlantic Ocean. Bullgeol. Soc. A m . , 90, 1165-76. EMBLEY, R.W. 1976. New evidence for the occurrence of debris flow deposits in the deep sea. Geology, 4, 371-74. EMERY, K.O. & HULSEMANN,J. 1962. The relationships of sediments, life and water in a marine basin. Deep Sea Res., 8, 165-88. 38, 51-76. & UCHUVl, E. 1972. Western North Atlantic DAVIS, C.C. 1980. LANDSA T Image Analysis of CircuOcean: topography, rocks, structure, water, life and sediments. Am. Ass. Petrol. Geol. Mere., 17, 532 pp. lation and Suspended Sediment Transport; California Continental Borderland. Unpub. M.S. Thesis, Univ. EPPLEY, R.W. & PETERSON, B.J. 1979. Particulate So. Calif., Los Angeles. 216 pp. organic matter flux and planktonic new production DEAN, W.E. & AgTI-IUR, M.A., in press. Origin and in the deep ocean. Nature, 202, 677-80. diagenesis of Cretaceous deep-sea black shales and EWlNG, M. & CONNARY, S.D. 1970. Nepheloid layer in multicoloured claystones in the Atlantic Ocean. US the North Pacific. Geol. Soc. Am. Mem., 126, 41-82. geol. Surv. Prof. Pap. FIELD, M.E. 1981a. Deposition on Pacific shelf DEMAlSON, G.J. & MOORE, G.T. 1980. Anoxic environe d g e - zone of constrasts. Bull. Am. Ass. Petrol. ments and oil source bed genesis. Bull. Am. Ass. Geol. Bull., 65, 924. Petrol. Geol., 64, 1179-209. FIELD, M.E. 1981b. Sediment mass-transport in basins: DIESTER-HAAS, L. 1975. Influence of deep oceanic controls and patterns. In: Depositional Systems of currents on calcareous sands off Brazil. IXth InterActive Continental Margin Basins. Pacific Section national Sedimentological Congress, Nice, Resumes Soc. econ. Paleo. Min Short Course, 61-84. des Publications, VIII, 2. FISCHER, A.G. & ARTHUR, M.A. 1977. Secular variaDox'r, R.H. 1963. Dynamics of subaqueous gravity tions in the pelagic realm. Deep-water carbonate depositional processes. Bull Am. Ass. Petrol. Geol., environments, Sue. econ. Paleo. Min. Spec. Pub., 25, 47, 104-29. 19-50. DOUGLAS, R.G. 1981. Palaeoecology of Continental FLEISCHER, P. 1970b. Mineralogy and sedimentation Margin Basins: a modem case history from the history, Santa Barbara Basin, California. J. sed. Borderland of Southern California. In: Depositional Petrol., 42, 49-58. Systems of A ctive Continental Margin Basins. Pacific FRANKENBERG,D. & SNXTH,K.L. 1967. Coprophagy in marine animals. Limnol. Oceanogr., 12, 443-50. Section, Soc. econ. Paleo. Min. Short Course, 121-56. FUrrERER, D.K. 1974. Significance of the boring sponge DRAKE, D.E. 1972. Distribution and Transport of SusCliona for the origin of fine-grained materials of carbonate sediments. J. sed. Petrol., 44, 79-84. pended Matter, Santa Barbara Channel, California. Unpubl. Ph.D. Dissertation, Univ. So. Calif. Los GARRELS, R.M. & MACKENZIE, F.T. 1971. Evolution of Sedimentary Rocks. W.W. Norton & Co., Inc., New Angeles. 358 pp. York. 397 pp. - 1976. Suspended sediment transport and mud deposition on continental shelves. In: Stanley, D.J. GIBBS, R.J. 1967. The geochemistry of the Amazon & Swift, D.J.P. (eds), Marine Sediment Transport River system- Part I, the factors that control the salinity and the composition and concentration of and Environmental Management. John Wiley, New York. 127-58. the suspended solids. Bull. geol. Sue. Am., 7 8 , 1203-32. - & GORSLINE, D.S. 1973. Distribution and transport of suspended particulate matter in Hueneme, - 1976. Distribution and transport of suspended particulate material of the Amazon River in the Redondo, Newport and La Jolla Submarine ocean. In: Wiley, M. (ed.), Estuarine Processes. Canyons, California. Bull. geol. Soc. Am., 84, Academic Press, New York. 35-47. 3949-68. DUNSTAN, W.M. & ATKINSON, L.P. 1976. Sources of - 1981. Sites of river-derived sedimentation in the ocean. Geology, 9, 77-80. new nitrogen for the South Atlantic Bight. In." & HEL'rZEL, S. 1982. Coagulation and the deposiWiley, M. (ed.) Estuarine Processes. Academic - Press, New York. 69-78. tion of mud. Abstracts Vol., XIth Intl. Sed. Congress, EHRLICH, R., BROWN, P.J., YARUS, J.M. & PRZYGOCKI, Hamilton, 4. GILMER, R.W. 1972. Free-floating mucus webs: a novel R.S. 1980. The origin of shape frequency distribufeeding adaptation for the open ocean. Science, 176, tion and the relationship between size and shape. J. sed. Petrol., 50, 475-84. 1239-40. - & WEINBERG, B. 1970. An exact method for GINSBURG, R.N. & JAMES, N.P. 1974. Holocene car1981. Morphology and dynamic sedimentology of the eastern Nile Delta shelf. In: Nittrouer, C.A. (ed.), Sedimentary Dynamics of Continental Shelves. Developments in Sedimentology. Elsevier. 301-26. COOK, D.O. & GORSLINE,D.S. 1972. Field observations of sand transport by shoaling waves. Marine Geol., 13, 31-55. CURRAY, J.R. & MOORE, D.G. 1971. Growth of the Bengal Deep-sea Fan and denudation in the Himalayas. Bull. geol. Soc. Am., 82, 563-72. CURTIS,W.F., CULBERTSON,J.K. & CHASE,E.B. 1973. Fluvial-sediment discharge to the conterminous United States. US geol. Surv. Circular, 670, 17 pp. DAMUTH, J.E. 1980. Use of high-frequency (3.5-12 kHz) echograms in the study of near-bottom sedimentation processes in the deep-sea. Marine Geol.,
A review of fine-grained sediment bonate sediments of continental shelves. In: Burk, C.A. & Drake, C.L. (eds), The Geology of Continental Margins, Springer Verlag, Berlin. 137-55. GORSUNE, D.S. 1978. Anatomy of margin basins. J. sed. Petrol., 48, 1055-68. - 1980. Deep-water sedimentologic conditions and models. Marine Geol., 38, 1-21. - 1981. Fine sediment transport and deposition in active margin basins. In: Depositional Systems of Active Margin Basins, Pacific Section Soc econ Paleo Min. Short Course, 39-60. - & EMERY,K.O. 1959. Turbidity current deposits in San Pedro and Santa Monica Basins off southern California. Bull. geol. Soc. Am., 70, 279-88. & GRANT, D.J. 1972. Sediment textural patterns on San Pedro Shelf, California (1951-1971): reworking and transport by waves and currents. In: Swift, D.J.P., Duane, D.B. and Pilkey, O.H. (eds), Shelf Sediment Transport. Dowden, Hutchinson & Ross, Stroudsburg, Pa., 575-600. - - & PRENSKY, S.E. 1975. Palaeoclimatic inferences for Late Pleistocene and Holocene from California Borderland basin sediments. In: Suggate, R.P. Cresswell, M.M. (eds), Quaternary Studies. Royal Soc. New Zealand. 147-53. GRIFFIN, J.J. & GOLDBERG, E.D. 1968. Clay mineral distribution in the Pacific Ocean. The Sea, 3, 728-4 1. HAKANSSON, E., BROMLEY, R . & PERcH-NIELSEN, K. 1974. Maastrichtian chalk of northwest E u r o p e - a pelagic shelf sediment. Spec. Pubs. lnternl. Assoc. Sed., 1,211-33. HA~R, B.E. 1971. Morphology and sediments of Redondo Submarine Fan, southern California. Bull. geol. Soc. Am., 82, 2413-32. -& GORSLINE, D.S. 1979. Processes and morphology of continental slope between Santa Monica and Dume Submarine Canyons, Southern California. Marine Geol., 28, 77-87. HAURV, L.R. 1983. An offshore eddy in the California Current System. Part IV: plankton distributions. Progress in Oceanography, in press. HEATH, G.R., DAUPHIN, J.P., OPDYKE, N.P. & MOORE, T.C. 1973. Distribution of quartz, opal, calcium carbonate and organic carbon in Holocene, 600,000 yr and Brunhes/Matuyama age sediments of the North Pacific. Geol. Soc. Am. Abstracts with Programs, 5, 662. --, MOOSE, T.C. & DAUPHIN, J.P. 1976. Late Quaternary accumulation rates of opal, quartz, organic carbon and calcium carbonate in the Cascadia Basin, northeast Pacific. Investigations of Late Quaternary oceanography and Paleoclimatology, Geol. Soc. Am. Mem., 145, 393-410. HEEZEN, B.C., HOLLISTER,C. & RUDDIMAN,W.F. 1966. Shaping of the continental rise by deep geostrophic contour currents. Science, 152, 502-8. HE1N, F.J. 1982. Depositional mechanisms of deep-sea coarse clastic sediments, Cap Enrage Formation, Quebec. Can. J. Earth Sci., 19, 267-87. --, in press. Conglomeratic deposits of braided channel systems: deep-sea and fluvial settings. In." Gravels and Conglomerates, Internl. Assoc. Sed. Mem. --, in press. Fine-grained deep-sea mass flow depo-
31
sits, sedimentary facies, mass physical properties and depositional mechanisms, offshore California Continental Borderland. J. sed. Petrol. -& RISK, M.J. 1975. Bioerosion of coral heads: inner patch reefs, Florida reef tract. Bull. Marine Sci., 25, 133-8. HESSE, R. & CHOUGH, S.K. 1980. The northwest Atlantic mid-ocean channel of the Labrador Sea. II. Deposition of parallel laminated levee muds from the viscous sublayer of the low density turbidity currents. Sedimentology, 27, 697-711. HONJO, S., MANGANIN1, S.J. & COLE, J.J. 1982. Sedimentation of biogenic matter in deep ocean. Deep Sea Res., 29, 609-25. HUANG, T.C., WATKINS, N.D. & SHAW, D.M. 1975. Atmospherically transported volcanic glass in deepsea sediments: volcanism in subAntarctic latitudes of the south Pacific during the late Pliocene and Pleistocene time. Bull. geol. Soc. Am., 86, 1305-15. --, WATKINS, N.D. & WILSON, L. 1979. Deep-sea tephra from the Azores during the past 300,000 yrs: eruptive cloud height and ash volume estimates. Summary, Part I. Bull. geol. Soc. Am., 90, 131-3. INMAN, D.L. 1949. Sorting of sediments in the light of fluid mechanics. J. sed. Petrol., 19, 51-70. - & Nordstrom, C.E. 1971. On the tectonic and morphologic classification of coasts. J. Geol., 79, 1-21. JACOBI, R.D. 1976. Sediment slides on the northwestern continental margin of Africa. Marine Geol., 22, 157-73. JANOWITZ, G.S. & PIETRAF~A, L.J. 1980. A model and observations of time-dependent upwelling over the mid-shelf and slope. J. geophys. Res., 10, 1574-83. JENKINS, W.J. 1982. Oxygen utilization rates in North Atlantic subtropical gyre and primary production in oligotrophic systems. Nature, 300, 246-48. JERLOV, N.G. 1953. Particle distribution in the ocean. Reports of the Swedish Deep Sea Expedition 1947-48, Ill, 71-98. 1968. Optical Oceanography. Elsevier, Amsterdam. 194 pp. JOHNSON, B.D. & COOKE, R.C. 1980. Organic particles and aggregate formation resulting from the dissolution of bubbles in sea water. Limnol. Oceanogr., 25, 653-61. JOHNSON, D.A., LEDBETTER,M. & BURCKLE, I.H. 1977. Vema Channel palaeooceanography-Pleistocene dissolution cycles and episodic bottom water flow. Marine Geol., 23, 1-33. JUMARS,P.A., NOWELL,A.R.M. & SELF, R.F.L. 1981. A simple model of flow-sediment-organism interaction. Marine Geol., 42, 155-72. KALLE, K. 1939. Bericht 2. Teilfahrt der deutsche nordatlantische Expedition "Meteor". V. Die chemischen Arbeiten auf der "Meteor"-fahrt, Januar-Mai, 1938. Ann. Hydrograph. Maritimen Meteorol., Beih. Z. 1939, 23-30. KARL, H.A. 1976. Processes Influencing Transportation and Deposition of Sediment on the Continental Shelf, Southern California. Unpubl. Ph.D. Dissertation, Univ. So. Calif., Los Angeles. 331 pp. -1981. Response of the suspended sediment trans-
32
D.S. Gorsline
port system to continental shelf dynamics. GeoMarine Letters, 1, 243-8. KENNETT, J.P. 1981. Marine Tephrochronology. In: Emilian, C. (ed.), The Sea. John Wiley, New York. 1373-437. -& THUNELL, R.C. 1975. Global increase in Quaternary explosive volcanism. Science, 187, 497-503. KOBLINSKY, C.J., SIMPSON, J.J. & DICKEY, T.D. 1983. An offshore eddy in the California Current System. Part II: Surface manifestations. Progress in Oceanography, in press. KOLPACK, R.L. 1971. Biological and oceanographical survey of the Santa Barbara Channel oil spill 1969-1970. II. Physical, chemical and geological studies. Allan Hancock Foundation, Univ. So. Calif. Los Angeles. 477 pp. KRANCK, K. 1975. Sediment deposition from flocculated suspensions. Sedimentology, 22, 111-23. KUENEN, PH.H. 1969. Origin of quartz silt. J. sed. Petrol., 39, 1631-33. LA1NE, E.P. 1980. New evidence from beneath the western North Atlantic for the depth of glacial erosion on North America and Greenland. Quat. Res., 144, 188-98. LAL, D. 1977. The oceanic microcosm of particles. Science, 198, 997-1009. LARSEN, L.H. 1982. A new mechanism for seaward dispersion of midshelf sediments. Sedimentology, 29, 279-83. LEDBETTER, M.T. & ELLWOOD, B.B. 1980. Spatial and temporal changes in bottom water velocity and direction from analysis of particle size and alignment in deep-sea sediment. Interpretative modelling of deep-ocean sediments and their physical properties. Marine Geol., 38, 245-62. LEWIS, K.B. 1971. Slumping on a continental slope inclined at 1-4 ~ Sedimentology, 16, 97-110. LINDSTROM,M. 1974. Volcanic contribution to Ordovician pelagic sediments. J. Ned. Petrol., 44, 287-91. LONSDALE,P. 1976. Abyssal circulation of the southeastern Pacific and some geological implications. J. geophys. Res., 81, 1163-76. & MALFAIT, B. 1974. Abyssal dunes of Foraminiferal sands on Carnegie Ridge. Bull. geol. Soc. Am., 85, 1697-1712. LOUGHNAN, F.C. 1969. Chemical Weathering of the Silicate Minerals, Elsevier, Amsterdam. 154 pp. LOWE, D.R. 1976. Grain flow and grain flow deposits. J. Ned. Petrol., 46, 188-99. 1979. Sediment gravity flows: their classification and some problems of its application to natural flows and deposits. In: Doyle, L.J. & Pilkey, O.H. (eds) Geology of the Continental Slopes, Soc. econ. PaleD. Min. Spec. Publ. 27, 75-84. 1982. Sediment gravity flows: II. Depositional models with special reference to deposits of highdensity turbidity currents. J. Ned. Petrol., 52, 279-97. MALOUTA, D.N., GORSLINE, D.S. & THORNTON, S.E. 1981. Processes and rates of basin filling in an active transform margin: Santa Monica Basin, California Continental Borderland. J. Ned. Petrol., 51, 1077-95. MANHEIM, F.T., HATHAWAY,J.C. & UCHUPI, E. 1972. Suspended matter in surface waters of the Northern Gulf of Mexico. Limmol. Oceanog., 17, 17-27.
-
-
-
-
MARTINEZ, L.-A., SILVER, M.W., KING, J.M. & ALLDREDGE, A.L. 1983. Nitrogen fixation by floating diatom mats: a source of new nitrogen to oligotrophic ocean waters. Science, 221, 152-4. MCCAVE, I.N. 1972. Transport and escape of finegrained sediment from shelf areas. In: Swift, D.J.P., Duane, D.B. & Pilkey, O.H. (eds), Shelf Sediment Transport. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 225-48 pp. -1975. Vertical flux of particles in the ocean. Deep Sea Res., 22, 491-502. 1982. Processes of transport and deposition of fine-grained suspended sediment in the sea. -
-
Abstracts of Papers, Xlth International Congress of Sedimentology, Hamilton, Ontario, Canada. 3 pp. HOLLISTER,C.D., LAINE, E.P., LONSDALE,P.F. & RICHARDSON,M.J. 1982. Erosion and deposition on the eastern margin of the Bermuda Rise in the late Quaternary. Deep Sea Res., 29, 535-61. & SWIFT, S.A. 1976. A physical model for the rate of deposition of fine-grained sediments in the deep sea. Bull. geol. Soc. Am., 87, 541-6. MEADE, R.H. 1969. Errors in using modern stream load data to estimate natural rates of denudation. Bull. geol. Soc. Am., 80, 1265-74. -1982. Sources, sinks and storage of river sediment in the Atlantic drainages of the US. J. Geol., 90, 235-52. - - , SACHS,P.L., MANHEIM,F.T., HATHAWAY,J.C. & SPENCER,O.W. 1975. Sources of suspended matter in waters of the Middle Atlantic Bight. J. sed. Petrol., 45, 171-88. MIDDLETON, G.V. & BOUMA, A.H. (eds) 1973. Turbidites and Deep-water Sedimentation. Pacific Section Soc. econ. PaleD. Min. Short Course, 157 pp. MILLIMAN, J.D. & MEADE, R.H. 1983. World-wide delivery of river sediments to the oceans. J. Geol., 91, 34-42. MORNER, N.-A. 1974. Sea level variations and climatic fluctuations. Colloques Internationaux du CNRS, 219, 135-41. MUTTI, E. & RICcI-LUCCHI, F. 1978. Turbidites of the northern Apennines: Introduction to facies analysis. Internl. Geol. Rev, 20, 125-66 (Translation). NAHON, D. & TROMPETTE,R. 1982. Origin of siltstones glacial grinding versus weathering. Sedimentology, 29, 25-36. NARDIN, T.R. 1981. Seismic stratigraphy of Santa --,
-
-
Monica and San Pedro Basins, California Borderland. Late Neogene history of sedimentation and tectonics. Unpubl. Ph.D. Dissertation, Univ So. Calif., Los Angeles. 295 pp. , EDWARDS, B.D. & GORSLINE, D.S. 1979a. Santa Cruz Basin, California Borderland: dominance of slope processes in basin sedimentation. In: Doyle, L.J. & Pilkey, O.H. (eds), Geology of Continental Slopes. Soc. econ. PaleD. Min. Spec. Pub., 27, 209-22. --, HEIN, F.J., GORSLINE, D.S. & EDWARDS, B.D. 1979b. A review of mass movement processes, sediment and acoustic characteristics and contrasts in slope and base-of-slope systems versus canyonfan-basin floor systems. In: Doyle, L.J. & Pilkey,
A review of fine-grained sediment
33
Effect on Man. Geol. Soc. Am. Spec. Pap., 186, C.H., (eds.), Geology of Continental Slopes, Soc. 71-86. econ. PaleD. Min. Spec. Pub. 27, 61-73. & BONATTI, E. 1969. Continental dust in the NATIONAL RESEARCHCOUNCIL 1976. Disposal in the . - atmosphere of the eastern equatorial Pacific. J. Marine Environment. National Academy of geophys. Res., 74, 3362-71. Sciences, NBC Report, Washington, DC. 76 pp. PYE, K. & SPERLING,C.H.B. 1983. Experimental invesN1NKOVITCH, D. & DONN, W.L. 1975. Explosive Cenotigation of silt formation by static breakage prozoic volcanism and climatic interpretations. Science, cesses: the effect of temperature, moisture and salt 194, 899-906. on quartz dune sand and granitic regolith. SedimenNITTROUER,C.A. & STERNBERG,R.W. 1981. The formatology, 30, 49-62. tion of sedimentary strata in an allochthonous shelf environment: the Washington Continental Shelf. RAPP, A. 1974. A review of desertization in Africa -Water, vegetation and man. Secr. Int. Ecology, Marine Geol., 42, 201-32. Rep., 1, Stockholm, 77 pp. , STERNBERG, R.W., CARPENTER, R. & BENNETT, J.T. 1979. The use of Pb-210 geochronology as a REX, R.W. & GOLDBERG,E.D. 1958. Quartz contents of pelagic sediments in the Pacific Ocean. Tellus, 10, sedimentological tool: application to the Washing153-9. ton Continental Shelf. Marine Geol., 31,297-316. SYERS, J.K., JACKSON, M.L. & CLAYTON, R.N. NOWELL, A.R.M., JUMARS, P.A. & EKMAN, J.E. 1981. - - , 1969. Aeolian origin of quartz in soils of Hawaiian Effects of biological activity on the entrainement of Islands and in Pacific pelagic sediments. Science, marine sediments. Marine Geol., 42, 133-54. 163, 277-9. O'BRIEN, N.R., NAKAZAWA,K. & TOLUHASHI,S. 1980. RICHARDSON,M.D. & YOUNG, D.K. 1980. Geoacoustic Use of clay fabric to distinguish turbiditic and models and bioturbation. Marine Geol., 38, 205-18. hemipelagic siltstones and silts. Sedimentology, 27, ROBINSON,A. 1983. Eddies in Marine Science. Springer47-62. Verlag, New York. 609 pp. OSTERBERG,C., eARLY, A.G. & CURL, H. 1964. AccelerROBISON, B.H. & BAILEY,T.G. 1981. Sinking rates and ation of sinking rates of radionuclides in the ocean. dissolution of midwater fish fecal matter. Marine Nature, 200, 1276-77. Biol., 65, 135-42. P A F F E N H O F E R , G.-A. & KNOWLES,S.C. 1979. Ecological RODOLFO, K.S. 1964. Suspended Sediment in Southern implications of fecal pellet size, production and California Waters. Unpubl. M.S. Thesis, Univ. So. consumption by copepods. J. mar. Res., 37, 35-49. Calif., Los Angeles. 91 pp. PARRISH, J.T. 1983. Upwelling deposits: nature of association of organic-rich rock, chert, chalk, phos- - - 1970. Annual suspended sediment supplied to the California Borderland by the southern California phorite and glauconite. Bull. Am. Ass. Petrol. Geol., watershed. J. Ned. Petrol., 40, 666-71. 67, 529. PEWE, T.L. 1981. Desert dust: an overview. In: Pewe, RONA, P.A. 1973. Worldwide unconformities in marine sediments related to eustatic changes in sea level. T.L. (ed.), Desert Dust: Origins, Characteristics, and Nature, 244, 25-6. Effect on Man. Geol. Soc. Am. Spec. Pap., 186, 1-10. PIPER, D.J.W. 1978. Turbidite muds and silts on SAVRDA, C.E., BOTTJER, D.J. & GORSLINE, D.S. 1983. Euxinic biofacies in anoxic basins: San Pedro and deep-sea fans and abyssal plains. In: Stanley, D.J. & Santa Barbara Basin, California Continental BorKelling, G. (eds), Sedimentation in Submarine derland. Bull. Am. Ass. Petrol. Geol., 67, 544. Canyons, Fans and Trenches, Dowden, Hutchinson SCHWALBACH, J.R. 1982. A Sediment Budget for the & Ross, Stroudsburg, Pa. 163-76. Northern Region of the California Continental BorPITMAN, W.C. III 1978. Relationships between eustacy derland. Unpubl. M.S. thesis, Univ. So. Calif., Los and stratigraphic sequences of passive margins. Angeles. 212 pp. Bull. geol. Soc. Am., 89, 1389-1403. POMEROY,L.R., in press. Microbial processes in the sea: SCLATER, J.G., HELLINGER, S. & TAPSCOTT, C. 1977. The palaeobathymetry of the Atlantic Ocean from diversity in nature and science. In: Hobbie, J.E. & the Jurassic to the present. J. Geol., 85, 509-52. Williams, P.J. Leb. (eds.), Heterotrophic Activity in SEIBOLD, E. • HINZ, K. 1973. Continental slope the Sea. Plenum Press, New York. construction and destruction. West Africa. In: Burk, - & DEIBEL,D. 1980. Aggregation of organic matter C.A. & Drake, C.L. (eds), The Geology of the by pelagic tunicates. Limnol. Oceanogr., 25, Continental Margins. Springer-Verlag, Berlin. 643-52. 179-96. PORTER, K.G. & ROaalnS, E.I. 1978. Zooplankton fecal pellets link fossil fuel and phosphate deposits. SHEPARD, F.P., MARSHALL,N.F., McLOUGHLIN, P.A. & SULLIVAN, G.G. 1979. Currents in Submarine Science, 212, 931-3. Canyons and other Sea Valleys. Am. Ass. Petrol. POSTMA, H. 1967. Sediment transport and sedimenGeol. Studies in Geology., 8, 173 pp. tation in the estuarine environment. In. Lauff, H.G. (ed.) Estuaries. Am. Assoc. Advancement Sci. Publ., SHERIDAN, M.F. 1979. Emplacement of pyroclastic flows: a review. In: Chapin, C.E. & Elston, W.E. 83, 158-79. (eds), Ash-flow Tuff`v, Geol. Soc. Am. Spec. Pap., POTTER, P.E., MAYNARD, J.B. & PRYOR, W.A. 1980. 180, 125-36. Sedimentology of Shales. Springer-Verlag, Berlin. SHULENBERG,E. & REID, J.L. 1981. The Pacific shallow 306 pp. oxygen maximum, deep chlorophyll maximum, and PROSPERO,J.M. 1981. Arid regions as sources of mineral primary productivity reconsidered. Deep Sea Res., aerosols in the marine atmosphere. In: Pewe, T.L. (ed.), Desert Dust: Origins, Characteristics, and 28, 901-20.
34
D.S. Gorsline
SILVER, M.W. & BRULAND, K.W. 1981. Differential stratigraphical study of fossil assemblages from the feeding and fecal pellet composition of salps and Maastrichtian white chalk of northwestern Europe. pteropods and the possible origin of the deep-water In." Kauffman, E.G. & Hazel, J.E. (eds), Concepts flora and olive green 'cells'. Marine Biol., 62, and Methods of Biostratigraphy. Dowden, Hutchin263-73. son & Ross, Stroudsburg, Pa. 257-76. SIMPSON, J.J., DICKEY,T.D. & KOBLINSKY,C.J., 1984. SWALLOW,J.C. 1976. Variable currents in mid-ocean. Mesoscale eddies in the California Current and their Oceanus, 19, 18-25. effects on climate. Am. Meteorol. Soc. International SWIFT, D.J.P. 1973. Continental shelf sedimentation. Conf. on Air-Sea Interactions of the Coastal Zone, In: Burk, C.A. & Drake, C.L. (eds), The Geology of The Hague. the Continental Margins. Springer-Verlag, Berlin. SLOSS, L.L. 1962. Stratigraphic models in exploration. 1 I%36. J. sed. Petrol., 32, 415-22. THORNTON, S.E. 1981a. Holocene Stratigraphy and SMALL, L.F., FOWLER, S.W. • UNLU, M.Y. 1979. Sedimentary Processes in Santa Barbara Basin - InSinking rates of natural copepod fecal pellets. fluence of Tectonics, Oceanography, Climate and Marine Biol., 51,233-41. Mass Movement. Unpubl. Ph.D. Dissertation, SMALLEr, I.J. 1966. The properties of glacial loess and Univ. So. Calif., Los Angeles. 351 pp. the formation of loess deposits. J. sed. Petrol., 36, 1981b. Suspended sediment transport in the 669-76. surface waters of the California Current off southSMITH, R.L. 1979. Ash flow magmatism. In: Chapin, ern California. Geo-Marine Letters, 1, 24-8. C.E. & Elston, W.E. (eds), Ash-flow Tufts. Geol. UDDEY, J.A. 1914. Mechanical composition of clastic Soc. Am. Spec. Pap., 180, 5-27. sediments. Bull. geol. Soc. Am., 25, 655-744. SOUTAR, A. & CRILL, P.A. 1977. Sedimentation and UMBGROVE, J.H.F. 1947. The Pulse of the Earth. climatic patterns in the Santa Barbara Basin during Martinus Nijhoff, The Hague. 358 pp. the 19th and 20th Centuries. Bull. geol. Soc. Am., 88, VOGT, P.R. 1979. Global magmatic episodes: new 1161-72. evidence and implications for steady-state midSTANLEY, D.J. 1981. Unifites- structureless muds of oceanic ridges. Geology, 7, 93-8. gravity-flow origin in Mediterranean basins. Geo- WALKER, R.G. 1967. Turbidite sedimentary structures Marine Letters, 1, 77-83. and their relationship to proximal and distal ..- , ADDY, S.K. & BEHRENS,E.W. 1983. The mudline: depositional environments. J. sed. Petrol., 37, variability of its position relative to shelf break. In: 25-43. Stanley, D.J. & Moore, G.T. (eds), The Shelf Break: WATTS, A.B. 1982. Tectonic subsidence, flexure and Critical Interface on Continental Margins. Soc. econ. global changes of sea level. Nature, 297, 469-74. Paleo. Min. Spec. Pub., 33, 279-98. WELLS, J.T. & COLEMAN,J.W. 1981. Physical processes - 81. MALDONADO,A. 1981. Depositional modes for and fine-grained sediment dynamics, coast of Surfine-grained sediment in the western Hellenic inam, South America. J. sed. Petrol., 51, 1053-68. Trench, Eastern Mediterranean. Sedimentology, 28, WENTWORTH, C.K. 1922. A scale of grade and class 273-290. terms for clastic sediments. J. Geol., 30, 377-92. STILLE, H. 1924. Grundfragen der t'ergleichenden Tek- - 1933. Fundamental limits to the sizes of clastic tonic. Verlag Gebruder Borntraeger, Berlin. 443 pp. grains. Science, 77, 633-34. STow, D.A.V. 1979. Distinguishing between fine WILDHARBER, J.L. 1966. Suspended Sediment over the grained turbidites and contourites on the Nova Continental Shelf oft Southern California. Unpubl. Scotian deep water margin. Sedimentology, 26, M.S. Thesis, Univ. So. Calif., Los Angeles. 98 pp. 371-88. WISDOM, H.L. 1969. Atmospheric dust in permanent --& BOWEN, A.J. 1980. A physical model for snowfields; implications to marine sedimentation. transport and sorting of fine-grained sediments by Bull. geol. Soc. Am., 80, 761-82. turbidity currents. Sedimentology, 27, 31-46. YINGST, J.Y. & ALLER, R.C. 1982. Biological activity STRAKHOV,N.M. 1970. Principles of Lithogenesis. 1. and associated sedimentary structures in HEBBLE Consultants Bureau. Plenum Press, New York. 535 area deposits, western North Atlantic. Marine Geol., PP. 448, M7-M 15. SURLYK, F. & BIRKELUND, T. 1979. An integrated D.S. GORSLINE,Department of Geological Sciences, University of Southern California, Los Angeles, California 90089-0741, USA.
Erosion, transport and deposition of fine-grained marine sediments I.N. McCave S U M M A R Y : Fine-grained marine sediments are cohesive but their degree of cohesion is not simply determined by grain size. Cohesion controls erodibility, and water content, mineralogy, cation exchange capacity, salinity of interstitial and eroding fluid, organic mucus content and Bingham yield strength are all parameters relating to cohesion that have been proposed. No unique relation to erodibility has emerged and for erosion of slowly deposited sediment, modified by biota, it seems that measures of surface properties such as aggregate strength should be more relevant than bulk properties such as yield strength. The latter may be more appropriate for rapidly deposited estuarine muds. Once eroded, suspended sediment is subject to flocculation and biological aggregation, which alters size and settling velocity distributions, thereby controlling distribution in the flow and rate of deposition. Disaggregation may also occur through turbulent straining of particles, but in most marine situations this will permit stable particles over 1 mm in diameter, though under higher shear close to the bed the upper limit may be 100-200/~m. For deposition where ~o < 0.1 Pa the maximum stable aggregate size is probably /> 100/~m. Deposition of fine sediment on smooth beds probably occurs by entrapment in, and settling through, the viscous sublayer. On rough beds particles are trapped in the interstices between roughness elements. We have only a rough idea of critical deposition conditions but can give a fairly good estimate of deposition rate in given conditions. The longer term net deposition rate as measured by radiometric methods over weeks to months probably owes as much to frequency of erosion as to frequency and rate of deposition and is not yet well predicted from fluid and sediment parameters.
Our ability to predict the behaviour of finegrained sediment is extremely poor because the particles stick together and the deposits are a wonderful medium for sustaining life. Most problems stem from one or both of these factors. Particles may adhere to one another because the van der Waals forces or electrostatic attractions are large relative to the weight of the particles. Alternatively they may be bonded by mucus secretions of bacteria, algae, diatoms or other members of the benthic in- and epiflora and fauna. Neither the physicochemical nor the biochemical bonding forces can be simply predicted or measured and thus the fluid stress required to remove particles from the bed cannot be predicted either. Moreover the same assemblage of particles may vary in water content or degree of compaction and biological components, thus the same type of mud may have several critical erosion stresses. Transport and deposition of fine-grained sediment depends to a large extent on settling velocity of the suspended particles and that depends on their state of aggregation which is part biological, part physico-chemical in origin. There are no good ways of predicting the state of aggregation of polydisperse, multimineralic suspensions without biota, thus in natural situations empiricism reigns, but based on very few measurements. This review considers mainly the behaviour of mud beds under Newtonian fluid flows and
deposition from dilute suspensions ( < 300 g m -3 in concentration) where particles do not interfere with flow structure. At high concentration ( > 5 - - 1 0 kg m -3) hindered settling begins, the suspension becomes fluid mud and behaves more like a consolidating soil than a turbulent suspension. The flow structure found in low-concentration cases over a smooth bed is the classic turbulent boundary layer comprising a viscous sublayer adjacent to the bed succeeded by a transition (buffer layer) and tubulent core whose velocity profile is fitted by the logarithmic von Karman-Prandtl relation (Rouse 1937). Both erosion and deposition involve interactions in the viscous sublayer whose thickness is ~ ' ~ 10 v/u. where v is kinematic viscosity (1 to 1.4 • 10 -6 m 2 s-1 in shallow to deep water) and u. is the shear velocity ( = ~ ) where ~o is the skin friction and p is the fluid density. Low values of u. in the deep ocean are ~ 1 mm s -1 while on the shelf they generally range up to 100 mm s-1. So ~' ranges from 14 mm down to 0.1 mm. Under depositional conditions the range is 6"= 1-14 mm, i.e. larger than almost all particle aggregates. Deposition then, and much erosion, involves transfer of particles across a viscous sublayer which covers a large part of most mud beds. Viewed at the sublayer scale the erosion and depositional processes should be similar to estuary, shelf, deep-sea and laboratory situations. The driving forces and concentrations will vary in
35
36
I.N. McCave
magnitude and frequency between these locations but at least for slowly varied flow, comparisons should be valid. Clearly we cannot be restricted to deep-sea information as there would be very little to report. This paper is thus general in character, avoiding only the very high concentrations typical of some estuaries, and high frequency oscillatory flow found under waves on the shelf. Although this paper concerns the behaviour of muds, the complexities which afflict them are not completely absent from sands. In particular, sands may suffer adhesion due to mucus and epipsammic flora. The latter are treated by Webb (1969) and Meadows & Anderson (1968), and their effect was strikingly demonstrated by de Boer (1981) who killed off the flora on a sandy tidal-flat (using CuSO4) and found the bedforms completely changed due to an altered erosion threshold. Nowell's (1982) experiments show that the erosion threshold of sands varies according to their muco-polysaccaride content and may be double the value for clean sand with no bacterial mucus. Grant et al. (1982) also present field and
laboratory data showing thresholds raised by biological activity on tidal-fiat sands. The curves of Miller et al. (1977) should be viewed as reference standards only rather than providing appropriate values for field situations.
Erosion Two facets of erosion are of particular interest: the critical conditions for initiation of erosion and the subsequent rate of erosion as a function of excess shear stress.
Critical conditions--physical parameters Erosion and the boundary shear stress at its onset can be measured with adequate precision in laboratory flumes. However, the property of the fine sediment to which this critical stress should be related is not generally agreed, and there have been many suggestions. In one of the most influential but misleading diagrams ever pub-
FIG. 1. Erosion, transport and deposition diagram according to Hjulstrom (1935, 1939) with mean velocity as the ordinate. Curve A is for critical erosion conditions while curve B is for critical deposition with the dashed part extrapolated.
Erosion, transport and deposition of.fine-grained marine sediments lished relating to fine sediment erosion (and deposition), Hjulstrom (1935, 1939) plotted critical erosion velocity against sediment size d (Fig. 1). This is quite acceptable for non-cohesive sediment but cohesion is by no means simply a function of size. Thus the left-hand limb of Hjulstroms curve, rising with decreasing size, expresses a truth that the sediments may be harder to erode than sand, (and in the case of boulder clay much harder); but smaller size is not the simple cause. We cannot predict critical erosion u,~ from d unless the sediment is noncohesive. The examination of fine non-cohesive sediments, pure glacial and crushed quartz silt, was made by Rees (1966), White (1970), Mantz (1977) and Uns61d (1982). This showed that critical values of u, continued to decrease with decreasing d down to sizes of 10/~m quartz in water. The range of u,c is from 8.8 mm s- 1 at 50 /~m to 6.1 mm s- l at 10/~m. For sizes smaller than 10 #m, quartz in water becomes cohesive (Uns61d 1982). The water content of the sediment has been related to U,c by several authors. Migniot (1968) plotted u,c versus sediment concentration C (in kg m -3) but found that different muds plotted on different curves and concluded the plot was not sufficiently general (Fig. 2). He also found that, for the same concentration, U,c was 1 to 2 times greater under salt water than under fresh water. The range of u,~ values, from 4 mm s- 1 to 60 mm
37
s - l for C = 80 to 700 kg m -3, is wide. The two regions shown in Fig. 2 have, for C > 300 kg m -3, critical shear stress r c ~ C 4 and for C < 3 0 0 kg m -3, rczcC 2 to C TM. Thorn & Parsons (1980) and Thorn (1981) summarizing work on four muds (Hydraulics Research Station 1977, 1979a, b) show a range ofz~ between 0.1 and 1 Pa (u,~= 10 to 31 mm s -l) for concentrations of 75 to 200 kg m -3 that are well fitted by a single line (Fig. 3). However, the lowest value ofz~ for each mud after two days consolidation was 0.05, 0.01 0.13 and 0.19 Pa (U,c=6, 10, 11.3, 13.8 mm s-l). Several American workers in the late 1950s and early 1960s sought to express critical erosion conditions for freshwater muds and agricultural soils in terms of the mechanical parameters plastic limit (Pt) and liquid limit (Lt) and plasticity index (Iw=Lt-Pi), (Dunn 1959; Smerdon & Beasley 1959; Gibbs 1962; Carlson & Enger 1963). Many samples of cohesive sediment, when plotted on a graph of Iw versus Lt, fall close to the 'A line', Iw=0.73 ( L / - 2 0 ) (Terzaghi & Peck 1968). These authors show that ~ increases with Iw, the data of Smerdon & Beasley giving -co= 1.31 Iw0"84 Pa for Iw between 10 and 50%. In general z~ increases along the A line, though at higher values of Iw Gibbs (1962) suggests some soils may be expandable and show lower thresholds. This apparent simplicity was confounded by Lyle & Smerdon (1965) who showed that the degree of compaction expressed as voids ratio e also exerts
60 50 40
30
'= 2 0 E E
100
| 1 200 300 C kg m -3
I
! 500
700
Fig. 2. Curves of critical erosion shear velocity versus bed concentration for four different muds from Migniot (1968).
38
I.N. McCave I
I
I
I
I
[
I
1.0
I
I
Y "
oo
I-*
/
0.1
FIG. 3. Curve of critical erosion shear stress versus bed concentration for three muds from Thorn & Parsons (1981).
0.01
I
]
I
i
it
BED
a strong influence (Fig. 4). No work of this type appears to have been published for marine sediments. There have been only two investigations using deep-sea sediments, calcareous ooze (Southard et al. 1971) and red clay (Lonsdale & Southard 1974). The range of concentrations examined is 150 to 800 kg m -3 (47-86~ water of wet weight) and the cohesion displayed by the ooze is slight (u,c = 4 to 8 mm s-1) while that of the red clay is very great (u,c = 8 to 36 mm s - 1(Fig. 5). Failure of the red clay when compacted to < 68% water content was by detachment of lumps, presumably as the bulk shear strength of the material was exceeded. This mode of erosion may result from form-drag on roughness elements created by an initial phase of particle-by-particle erosion. The lack of unique relationship between critical erosion u, and bed concentration for several muds caused Migniot (1968) to use the yield strength r.,. obtained from a rotating viscometer as FIG. 4. Critical erosion stress versus plasticity index with values of voids ratio as an additional parameter for Texas soils (Lyle & Smerdon 1965).
Ill
i
I
I
100
10
kg m -3
CONCENTRATION
2-0
i
T
!
z
I
1.5
1-0
/
.,.q,,
0"5
0
s
i
l
2's
plasticity index
,[w
35
Erosion, transport and deposition ojfine-grained marine sediments 40
!
\
l
!
(o) - " - - - - - M ass erosion (.)
30
Y ~
!
RED CLAY
20 ~4--Particle
ii
39
erosion
CALCAREOUS OOZE " ~ . , , ~ . ~ .
9
9
9
FIG. 5. Critical erosion shear velocity versus water content (percent water/total weight) for deep-sea sediments from McCave's (1978) recalculation of original data in Southard et al. (1971) and Lonsdale & Southard (1974).
--e
I
I
I
I
i
50
60
70
80
90
Water content {% by we/yht]
1962; 9 1975) (Fig. 6). These show for C = 2 0 - 3 0 0 k g m -3, ryocC 25, a dependence which Krone justifies theoretically. The Hydraulics Research Station (1979a) study of Brisbane mud also included some critical erosion shear stresses r,. and yield strengths r,, as a function of C, which were found to plot on the same line (Fig. 7). Nevertheless there is not one line here or on Fig. 2 so neither C nor water content relate uniquely to r,.. Comparisons between determinations of rB from viscometry and vane shear are made difficult by the variability associated with differing
the x-axis. This has the effect of compressing the data for 10 muds onto a single two-part curve with values ofu,c from 3 to 65 mm s -l (Fig. 6). On this curve at low ry, rc=0.32 ry89and for ~>.> 1.5 Pa, "cc=0.26 Vy. The yield strength measured in this way cannot simply be related to that obtained from shear vanes, and no measurements have been reported of in situ yield strength measured by viscometer. However, there have been measurements relating the Bingham yield strength, "rb extrapolated from viscometer measurements to the bed concentration C (Krone IO0
,
, w,,,
I
,
,
,
,
,,,l
I
i
!
i
,
,lit|
!
!
u
,
,,l,l
,
9
!
i
llllj
,
9
~ 9
. "~
/
9
e9
~ o~/" 9 ol~# ,~ilP
9
~ ''~
9
I0
9
9
r / , , , .
0 OiD 9
~
9
,
~:-~"
Ii 9
o.~
..~.~'~,,
i
i ,,,,I
9. 9
"~~
"
i
i
i
i
|llli
o.ol
i
|
i
i
i,l,I
i
I
I
oJ
i
iii
II
JD
i
I
1
i
zJltJ
j
J9
i
i
i
i1!
Joo
Y/e/d Strength Pa FIG. 6. Migniot's (1968) data for many muds collapsed onto a single curve using yield strength determined by vise 9
I.N. M c C a v e
4o
i
i
i
!
!
i
i
i
i
I
i/
I
/
1
// / /
0.8
/ / / ~.~/
0.6
ec
t.,i,J I-w
0.4
/u/
/
/
U
~7 ~)'/
0.2
I-Z 0.1
"" 0.08
,.-I,
0.06
~0.04
0.02
FIG. 7. Yield strength as a function of bed concentration showing many different lines with the same trend (slope ocC5/2 from Krone (1962) and Owen (1975). Also plotted is a line relating critical erosion stress to bed concentration on which 3 determinations of yield strength by rheogoniometer also plot, from Hydraulics Research Station (1979a).
0.at 0.00s o.oo6
0.004 10
equipment and operating conditions. The other experiments listed above are also hard to relate to natural situations because equality of bed concentration does not necessarily imply equality of all the factors contributing to sediment cohesion. In the late 1950s and early 1960s Krone (1959, 1962, 1963) conducted a classic series of experiments on erosion, transport and deposition of cohesive material. A more accessible summary of some of this is in Krone (1976, 1978). Repeated measurements of aggregate shear strength in a
Ill
2o
i
i
,o
I
r162
100
~o
~o
'
500
BED CONCENTRATION(Ckgm-3)
concentric cylinder viscometer yielded a few discrete values for shear strength and density for a given mud, rather than a continuum. Krone (1963) interpreted this in terms of distinct orders of aggregation--an assemblage of primary particles yields a zero-order aggregate (ps = 1269 kg m -3, ~y=2.2 Pa), several zero-order aggregates yield a first order aggregate (ps = 1179 kg m-3, Zy= 0.39 Pa), these combine to yield second order aggregates (ps = 1137 kg m -3, Zy= 0.14 Pa) and so on (values are for San Francisco Bay mud). A
Erosion, transport and deposition of fine-grained marine sediments series of erosion experiments conducted on the beds of Bay mud formed under varying deposition conditions showed that values of suspension concentration versus bed shear stress could be connected by a series of lines whose intercepts on the x-axis gave critical erosion stresses corresponding closely to the set of aggregate strengths deduced from the viscometer measurements (Krone 1962) (Fig. 8). These are interpreted by supposing that the bed has a discrete number of structural arrangements corresponding to the orders of aggregates 9Thus a zero-order aggregate at the surface of a bed of other zero-order aggregates will require a force equivalent to the strength of a first order aggregate to pull it off (Krone 1976). Krone's work leads us to think of erosion as the breaking of bonds between aggregates and the bed, requiring determination of bond strength, an idea that extends beyond the notion of pure electro-chemical flocs attached to a pure silt-clay bed. Compaction and erosion
With increasing overburden the aggregates appear to collapse and Krone (1963) shows a change in structure, probably to zero order at about 25 mm depth. The mass of sediment above the plane of collapse of first- to zero-order aggregates was 5.5 kg m -2. The shear strength of zero-order aggregates is in excess of 2.0 Pa, and in some cases up to 4.6 Pa for sediment with 75% < 2 /~m in diameter. This means that once the aggregates have been compacted to zero-order by overburden they become very difficult to erode, requiring values of u, > 50 mm s - 1. By contrast,
9
,,,-c~
A
4I
even first-order aggregates are much weaker, the greatest measured by Krone (1963) being 0.94 Pa. Aggregates of greater than zero-order are unlikely to be able to resist forces due to waves and tides combined, but could probably withstand those due to tides alone. The production of a bed of zero-order seems to be a prerequisite for the survival of mud deposits in high-stress zones, i.e. at least 5 kg m -2 should have been accumulated for the lowest layers to be ultimately preserved. A related feature of beds in high-stress areas is that they require higher stresses to erode them than materials of similar size and mineralogy in low-stress areas. A bed sample taken in an improperly closed box corer from the Nova Scotian Rise was found to have been eroded by water flowing through the corer leaving a rather irregular but firm mud surface. Samples of the same mud, taken undisturbed had a very soft surface and shear-vane tests showed the mud layer to have uniform and low shear strength down to 80 mm depth, below which it was hard and of different composition. It seems that the bed had 'hardened' under stress, the weak surface aggregates presumably being replaced by stronger ones of zero-order. The process is probably analogous to the stiffening of soils under sustained loading described by Mitchell (1976, p. 292). The sand content of mud increases its critical erosion shear stress and the presence of sand layers assists the drainage of mud layers resulting in their more rapid compaction (Terwindt & Breusers 1972). These authors found that a 20 mm thick mud layer with 37% sand, consolidated for 2 hours, could withstand stresses of u, up to
9
8
B
ev- c..~
~'- 0.15 Pa. Most estuarine and coastal muds have a significant admixture of several tens of percent of sand thereby giving the right conditions for preservation.
Erosion rate--physico-chemical parameters Several workers consider that the strength of the binding force between clay particles can be simply parameterized by the cation exchange capacity (m equiv/100 g) of the material. Sargunam et al. (1973) showed that the same soil but with different ionic concentration of interstitial fluid (0.004N-0.14N NaCI and 0.14N CaCI2) had different critical erosion stresses (z,= 1.1 to 8.5 Pa) under distilled water. However, when the eroding and interstitial fluids have the same composition critical erosion stress became very high. They concluded that migration of distilled
1
water into the soil and swelling due to the osmotic pressure gradient played an important part in the erosion process. This may be of some importance in estuaries where tidal and seasonal changes in the limit of saline intrusion occur. However, cation exchange capacity (CEC) alone does not parameterize erodability, as clay with adsorbed Ca + + is harder to erode than with adsorbed Na+: the sodium adsorption ratio S A R = N a + / (Ca + + + Mg + +) is important too. Arulanandan (1975) showed that Zc increased with increasing CEC for low SAR but decreased for high SAR. No single parameter indicative of critical shear stress emerges from this approach. Plots of erosion rate b (kg m -2 s- 1) versus shear stress show that for a given bed and fluid at a fixed temperature a straight line is obtained described by = M (vo - re)
where, if0 is in kg m -2 s -1 and -c is in Pa (Nm -2) M has units s m -1. Examples are given by Sargunam et al. (1973), Owen (1975), Ariathurai & Arulanandan (1978) and Thorn & Parsons (1980). Owen (1975) gives values of M of 1.07 to 2.04x 10 -3 s m -l for beds of density C = 199 to 246 kg m -3 at zero salinity, and 2.04 to 0.31 x 10 -3 s m -l for beds of 246 k g m -3 density at salinities from 0 to 32%. In his experiments,
1
1
13. Q. ,m
E
E O)
/~ //,//~
00
INDIVIDUALMUDS: o Brisbane 9 Grangemouth A Belawan 9Scheldt
I
10
(1)
I
20 /Xt (minutes)
3'0
FIG. 9. Rate of erosion per unit shear stress excess as a function of the measurement time-step drawn from Thorn (1980), Thorn & Parsons (1981).
E r o s i o n , t r a n s p o r t a n d deposition critical conditions were not affected by salinity but erosion rate was. Thorn & Parsons (1980) show that when a step change is made in shear stress the suspended sediment concentration initially rises rapidly but then levels off giving erosion rate constants M of 2.63 • 10 -3 s m -1 averaged over 10 min., 1.78 • 10 -3 S m -I over 20 min. and 1.39 • 10 -3 S m -1 over 30 min. for beds of C ,-~250 kg m-3 (Fig. 9). The experiments at the Hydraulics Research Station (Thorn & Parsons 1980; Thorn 1981) were conducted by depositing a mud bed from flowing water and allowing it to compact for two days. Such a bed shows a density profile increasing downwards. The step changes made in shear stress allow both the erosion rate and the critical erosion stress to be obtained because the concentration becomes steady after a while at which time the amount of sediment removed from the bed and thus, using the density profile, the density of the exposed surface, can be determined. The values of density C and zc of Fig. 3 were determined in this way. There may be some ambiguity here in that behaviour under clear water at the given stress would differ by lower "co (Allersma et al. 1967). Thus for clear water flows the curve of Fig. 3 would not be so steep. Ariathurai & Arulanandan (1978) show the erosion rate constant M to vary as a function of CEC, SAR and temperature. With increasing temperature the erosion rate rises more steeply and the critical erosion stress decreases (Fig. 10). These authors express their results using = MT(~0/1W--1)
(2)
thus their Mr = M~c. Values of M, are 0.5 to 5 • 10 - 3 kg m -2 s -1 corresponding, with ~c = 1 to 2.5 Pa, to M of 0.2 to 5 • -3 s m -~,
i
4
I
I
I
/
of fine-grained marine sediments
43
comparable with the results of Owen and Thorn & Parsons. Although in the deep sea, pore and eroding fluid composition, SAR and temperature may be regarded as constant, CEC is not. In estuaries the system may be very variable via several parameters. Clearly both the erosion rate and critical stress are functions of several physicochemical variables. Finally Ariathurai & Arulanandan (1978) note that presence of organic matter is also a factor in critical shear stress, and Young & Southard (1978) using natural marine muds, found that critical stress increased with organic carbon content, varying linearly between U,c = 8.3 mm s -1 at 0.8~ organic C and 12.9 mm s-1 at 2.0~o C. The observation of the temperature dependence of erosion rate has suggested to several authors, most recently Gularte et al. (1980), that rate-process theory should apply to erosion 0=X-~-exp-
~
exp
(3)
where k, h and R are the Boltzmann, Plank and general gas constants, T is absolute temperature, AF activation energy, Vz volume of 'flow units' (atoms or molecules in chemical reaction rates but probably clay particles here) and X a constant related to the frequency of activation. The process is thus considered in the framework of chemical theory developed for reaction rates but with application in soil mechanics (Mitchell 1964). If an experimental activation energy E is substituted for AF--V/rN/2 where N is Avogadro's number (Mitchell et al. 1968) then
lakT
= X-:- exp (--E)_~_
(4)
I
% x
~C
~,,,3 E
~2 nZ
o m
/,S
1
o r LU
0
"
i
0
1
2
3 SHEAR
I
1
I
4
5
6
STRESS
Pa
FIG. 10. T e m p e r a t u r e d e p e n d a n c e o f erosion rate and critical erosion stress 7
of compacted muds in a Moore & Masch apparatus from Ariathurai & Arulanandan (1978).
44
I.N. McCave
These two equations give a basis for determination of E and Vf from (4) ?In(b/T)
E
~(1/7)
R
(5)
and from (3) Oln b - -
~z
VU 2kT
-
(6)
From plotting log ~/T versus 1/T(Fig. 11) E may be determined, and from plotting log ,~ versus z, (Fig. 12) V; is found. The data of Gularte et al. (1980) determined from flow over an artificial illite-silt mixture in a water tunnel yield activation energies of 73 to 107 kJ mol-1 and flow volumes of 1.5 to 6.1 x 10 -19 m 3 (equivalent spherical diameter 0.66 and 1.05 pm, possibly illite particles) which decrease with increasing salinity. The activation energies are little less than those reported for soil creep but much greater than those for viscous flow of water. Thus Gularte et al. (1980) argues that breaking of strong interparticle forces is indicated. A flow volume decrease with increasing salinity (i.e. smaller 81n ~ / ~ at higher salinity) is much harder to explain convincingly in terms of flow (particle) volume but is
10 "3
E o~ 10 "4 v, I-
9
.,,.,o
10-5 3.2
i
.
I
I
I
I
3.3
3.4
3.5
3.6
lit
~
x 10 3
FIG. 11. Log-linear Arrhenius plot of erosion rate--temperature relationship shown by Gularte et al. (1980).
consistent with compression of the double layer resulting in stronger cohesion of aggregates. The approach gives some insights into the control of erodibility for pure materials but the actual values may not be realistic for the natural environment with biota, and in the constancy of the deep sea there is little relevance in notions of control of erosion by temperature. Experimental variation of temperature in situ would undoubtedly increase erosion rate (if only through the death-throes of the biota) but inference of rateprocess parameters would not be facilitated. Substantial data from coastal environments shows that suspended sediment concentration, probably due to sea-bed and tidal-flat erosion, becomes greater as the temperature falls (Buchan et al. 1967; Newton & Gray 1972). It seems unlikely that this is controlled by temperature, sodium absorption ratio, cation exchange capacity, water content, grain-size or any other of the mechanical properties investigated so far. Biological
influences
Few experiments have provided hard data on the subject but many field observations indicate that organisms may both stabilize and destabilize sediment. Stabilization of sediments by benthic diatoms and other micro-algae was investigated using an in situ bottom flume in the Bahamas by Scoff=in (1968, 1970) and Newmann et al. (1970). Mats of Enteromorpha and Lyngbya are extremely tough, the former withstanding flows up to 1 m s-J in a 0.1 • 0.1 • 3 m duct (Scoff=in 1970) probably equivalent to u,=50-100 mm s -l depending on roughness. These mats become so tough and thick (10 ram) that one cannot speak of critical erosion conditions for the sediment (which is not in contact with the flow), but rather 'mat breakup conditions' which is what happens in practice. The stabilizing influence of diatom-produced mucilage on sediment was shown by simple paddle-wheel stirring in culture flasks by Holland et al. (1974). Critical erosion stresses were not measured, but a significant difference in erodibility was demonstrated. Diatom populations on tidal-fiats produce mucus threads shown by Gouleau (1976) and Robert & Gouleau (1977) and presumed to be responsible for stabilizing the surface and promoting rapid mud deposition. Similar arguments were made by Coles (1979) and Frostick & McCave (1979) concerning mobile epipelic algae that leave a trail of mucus in sediment. Both authors found seasonal (summer) build up of mud on tidal-flats and winter erosion correlated with live/dead or absent algae. This suggested to Frostick & McCave (1979) an
Erosion, t r a n s p o r t a n d deposition o f f i n e - g r a i n e d m a r i n e s e d i m e n t s 10-4 -
45
/
/ /
r
/ /
J~
.
/ /
/
10--~
/
/
/Io
9 / /.
//. -
-
~
2', E
" "
~ /
.
/
9
/o
/ /.
J
9
/
/0
Io-eL /::)~ / i/. I- I /, I
10-)
I
0
I
I
0.1
0.2
I
0.3 l-, Pa
I
I
I
0"4
0"5
0"6
explanation for the winter increase in coastal suspended sediment concentrations mentioned above. Stabilization and destabilization may occur at different times of the year in subtidal sediments also. Rhoads et al. (1978) took undisturbed samples of the muddy bottom of Long Island Sound and inserted them into the bottom of a laboratory flume. Critical erosion experiments revealed a seasonal trend of low threshold in summer and high in winter during 1974-1975, but this trend was not continued into 1976 and the changes may be random. Nevertheless the values are significantly different, going from a minimum of Oc = 0.07 up to 0.11 m s- l in a flow 0.1 m deep, where fT 5-10 kg m -3, Krone 1962; Diephus 1966; Owen 1975) the mud may display hindered settling, i.e. a reduction in settling velocity, and start to develop cohesive strength (Krone 1963). This property is essential to the formation of 'fluid mud', high concentration suspensions (often 50-200 kg m-3) with a yield strength (i.e. non-Newtonian), form-
The first term on the right accounts for formation of particles of volume t~ by collision of particles of volumes Vi and (r i-vi) and the second term gives the loss of rJ sized particles by collision with all others. The size distributions are given by n(v) = dN/dv where dN is the number concentration of particles in volume v to v + dr. The terms K are the 'coagulation kernels' and are specific for different mechanisms (Friedlander 1977; Pruppacher & Klett 1978). Thus for Brownian motion with particles i and j, diameter d, and d o. = (di + dj) 2kT~j KBij "~" 3]2 di4
(8)
where T is absolute temperature and k is the Boltzmann constant. For turbulent shear the kernel is Ksij = m.d3ij (~/v) 89
(9)
Erosion, transport and deposition of fine-grained marine sediments where A is a constant equal to 0.163 based on Saffman & Turner (1956), and e is the turbulent dissipation rate. The kernel for differential settling is
time to halve the number of particles in suspension, can be calculated. For Brownian motion this is t,.B --
(10)
K~o = rc~J Aw~ij ( E c o + EDo)
47
4 where Aw~.o is the settling velocity difference between the particles, E c is the collision efficiency of the particles and ED is the collection efficiency by Brownian diffusion of particles onto a larger particle settling through smaller ones. Plotting these kernels for deep ocean conditions of temperature (viscosity) and turbulence shows (Fig. 13) that Brownian motion dominates for sizes below 1 pm, succeeded by capture through differential sedimentation and turbulent shear with increasing particle size. Using a somewhat different set of assumed values (high shear, constant particle density with size), Hunt (1982) suggests a region of shear dominance from 1 pm to 30/~m. However, oceanic suspensions usually contain a few large particles and these should be efficient scavengers even at low shear rates, because of the term d~ij in the turbulent shear kernel (McCave, in press). For more pure mud suspensions in regions of high shear (e.g. tidal estuaries) there may still be large particles present (e.g. Biddle & Miles (1972) show 800 pm flocs) whose bulk density is considerably less than that of smaller aggregates (Tambo & Watanabe 1979). If suspensions are initially monodisperse (all particles the same size) a 'coagulation time', the
3~t 4k TNo
(11)
and for turbulent shear t,.s -
1.04
.~ , Nod (e/v)~
(12)
where N,, is the initial particle concentration. For high-concentration deep-sea nepheloid-layers in the range d = 0.5-1 pm, No--~4.105 cm -3 and tcB "-, 889days. However, with a more realistic coalescence efficiency of 10%, t,8 becomes about 3 months. Coastal water concentrations are at least two orders of magnitude higher and tcB drops to 2 to 20 hours there. For shear with No = 6.103 cm -3 in the range d = 4 - 8 pm and deep-sea conditions, (e/v)l ,,, 0.084 s- I, &s = 4.4 months or, at 10% efficiency, 3.6 years. In coastal water with both a higher shear rate (-,- 1 s- 1) and concentration (N,, = 6.105 cm-3), t,s is in the range 2.7 to 27 hours. In both cases the figures suggest that significant coagulation can occur on the timescale of the tidal cycle in shallow water but that months to years are more appropriate to oceanic nepheloid-layers. In midwater ocean depths McCave (in press), shows that calculated coagulation rates are extremely slow with t,. in the region of tens to hundreds of years. Whether particles will be brought close enough together to stick depends on the surface chemistry
-6-7
-
di = 5 / z m .
.
.
.
-8
-9
kG kB
-10
k6
-11 -12
R = B~OWNIAN 5" = S'HEA~
ks
# = DIFFU~ENTIAL E~LDIMENTATION -13
--t
0.1
I
I
I
1
10
100
I
103
I
104
FIG. 13. Coagulation kernels plotted for collision of a 5 p m particle with particles of 1 to 104 pm diameter using particle and fluid parameters detailed by McCave (in press). Diagrams of this type were introduced by Friedlander (1965) and have been presented for aqueous suspensions by Lerman (1979) and Hunt (1980).
48
I.N. McCave
of the particles and the solution. In simple outline the balance between van der Waals attractive forces and electrical double-layer repulsion determines whether mineral particles in a pure solution will stick. If the solution has high ionic strength the particles' surface charge is balanced by a thin layer of counter-ions and the repulsive force is smaller a few nanometres away from the surface than it is in low ionic strength conditions. This permits particles to come close enough for van der Waals forces to dominate. The theory behind these models due to Derjagin & Landau and Verwey & Overbeek (DLVO) is given in Stumm & Morgan (1981) and van Olphen (1977). However, this is not the only mode of particle binding. Bacteria are observed to aggregate because they exude biopolymers. The exudates of both bacteria and algae give rise to 'bioflocculation' which may also involve mineral matter as the polymer can form a bridge between particles (Stumm & Morgan 1981; Busch & Stumm i968). This flocculation by polymers is an important process in waste water treatment dealt with by O'Melia (1972, 1978), Gregory (1978) and Lyklema (1978). Given the substantial bacterial populations of bottom sediments, binding by polymer bridging may be important in accounting for some of their cohesive properties. Binding of suspended clay and silt particles by mucus threads associated with diatoms is shown by Ernissee & Abbott (1975) (Thallasosira), and by Lewin et al. (1980) (Chaetoceras). Feeding by several groups of zooplankton (e.g. salps, doliolids, some pteropods, larvaceans) involves mucus nets used as particle traps which are then ingested or from which food is removed (Jorgensen 1966; Madin 1974; Gilmer 1974; Alldredge 1977; Muliin 1980; Bruland & Silver 1981; Alldredge & Madin 1982). These mucus structures are sometimes discarded and settle, acquiring more particles on the way down (Alldredge 1976). Thus
there are several ways in which organisms can promote aggregation in suspensions, apart from ingestion and pellet formation (Haven & Morales-Alamo 1972; Honjo & Roman 1979), and the phenomenon should certainly not be considered exclusively to be inorganically electrochemical. A further feature of natural suspended particles is that they tend to show the same (slightly negative) electrophoretic mobility in sea water, suggesting that some organic coating may determine important surface electrical properties (Niehoff& Loeb 1972, 1974; Hunter & Liss 1982). If all the particles are similar it may have the effect of reducing the proportion of collisions that result in coalescence. There are scarcely any determinations of the actual influence of all these effects on coalescence efficiency of natural particles. Swift & Friedlander (1964) obtained an efficiency of 37~ for polystyrene latex in artificial salt water for Brownian motion and laminar shear. Edzwald et at. (1974) obtain a value of 10% for natural estuarine sediment suspended in salt water, which is the best value available, although real world values are probably very variable.
Aggregate strengths and sizes Krone (1978) gives a pr6cis of his original (1963) elegant hypothesis of orders of aggregation. In essence he found that the same suspension could show different shear strengths and of volume concentrations of aggregates according to the shear rate. Furthermore the variation was discrete rather than continuous. Thus he supposed that at high shear only primary aggregates (zeroorder) could survive but at lower shear aggregates of these aggregates (first-order) would be stable and so on at even lower shears. He also calculated the densities of these aggregates as shown in Table 1. In most cases four orders are recognized
TABLE 1. Properties of sediment aggregates (from Krone 1976). Aggregate order 0 1 2 3 4 5 6
San Francisco Bay p kgm -3 r Pa 1269 2.2 1179 0.39 1137 0.14 1113 0.14 1098 0.082 1087 0.036 1079 0.020
Delaware Estuary p r 1250 2.1 1132 0.94 1093 0.26 1074 0.12
Brunswick Harbour, GA p r 1164 3.4 1090 0.41 1067 0.12 1056 0.062
Gulfport Channel MS p r 1205 4.6 1106 0.69 1078 0.47 1065 0.18
Bulk densities p are with interstitial fluid pw= 1025 kg m -3 Aggregate shear strength ~ was determined by viscometer
White River (saline) p 1212 4.9 1109 0.68 1079 0.47 1065 0.19
Erosion, transport and deposition o f f i n e - g r a i n e d marine sediments
49
FIG. 14. Particle volume size spectra determined by Coulter counter from Eisma & Kalf (1979), southern North Sea. Note higher concentration peaked and low concentration fiat winter distributions with no biological interference, and peaked late summer spectra dominated by biota. though seven are found for San Francisco Bay samples. Krone (1976) relates the strengths of these orders of aggregation to critical erosion stresses and the relation between aggregate strengths and stresses underlies the suggestion of 'stress-hardening' made in the previous section here. Size distributions from shallow and deep water have been principally determined using the Coulter Coulter (Brun-Cottan 1971; Sheldon et al. 1972; Eisma & Kalf 1979; Pak et al. 1980; McCave 1983). Recently some measurements
with the HIAC counter (Dickson & McCave 1982) and the laser forward-scattering particle counter (Bale et ai., in press) have appeared. Some authors have also used microscopic enlargement and particle counting (Schubel 1968, 1969, 1971; Wellershaus et al. 1973; Harris 1977; Baker et al. 1979; Lambert et al. 1981) to produce number spectra. Estuarine volume-size distributions at moderate concentrations ( ~ 100 g m -3) show a single well-developed peak (Kranck 1981) as do samples from the shelf when free of biological influences
5~
I.N.
McCave
100
00
5
m, E
T, ~ 1 7 6
10 ~ 9
~,g
"
o
'uWO
9 o l9l
...
.Ab
9
.o 0% o8 9
/
Iii
Cm -t" - - - - - - / " ~
9o
~.".~- b. "" "t
/ Ct
9"+" I' .% eee,.e-j
~
/\
\ \
- -
~ o
FIG. 15. Relationship between peakedness of distribution and concentration, from Kranck ( 1981).
I
(Kranck 1973, 1975; Eisma & Kalf 1979). The modal peak occurs mainly between 5 and 20/~m, but where plankton are important several modes may be encountered (Eisma & Gieskes 1977; Eisma & Kalf 1979) (Fig. 14). In the southern North Sea the peaked distributions are found nearshore while offshore at low concentrations ( < 10 g m -3) the size spectra are flat (Fig. 14, Eisma & Kalf 1979). This peakedness is clearly related to concentration in Kranck's (1981) estuarine data (Fig. 15). She maintains that this peak is produced by flocculation and that the kinetics of the process only result in a well developed peak when concentrations are high enough for it to develop rapidly. Kranck (1973, 1975) and McCave (1981) show aggregated and disaggregated distributions to demonstrate this essential feature of (all) suspended sediment size distributions (Fig. 16), a feature which in one data set from shelf and estuaries yields a regular relation between the flocculated (d/) and deflocculated (da) modal sizes, df=2.79 dd0'772 (Fig. 17). Some of this flocculation may involve organic binding of particles as samples may be readily deflocculated with hydrogen peroxide (Fig. 16). Data from Chesapeake Bay have figured in arguments that flocculation is not important in estuaries (Schubel 1968; Meade 1972) and that it is (Zabawa 1978). Although the microscopically counted size distributions from there appear not to relate closely to volume size distributions
I
10
I 100
C~/Ct (McCave 1979a, Fig. 6.11), nevertheless the data of Schubel (1968, 1969, 1971) do demonstrate the fact of fine background and intermittently resuspended populations of particles. The larger intermittently resuspended particles are shown by Zabawa (1978) to contain aggregates of both faecal pellet and electrochemical coagulation origins. Krone (1972, 1978) also argues persuasively that flocculation is a significant process in modifying size distributions and enhancing sedimentation through the consequent increase in settling velocity. Few determinations of oceanic particle size distributions have been made. Bader (1970) observed that particle-number size distributions in water tended to follow a power law relationship, and several workers (Brun-Cottan 1971, 1976; Sheldon et al. 1972: McCave 1975; Lerman et al. 1977; Baker et al. 1979; Pak et al. 1980; Richardson 1980; McCave 1983) showed that the exponent in the distribution N = a d -m was about m = 3 (i.e. a Junge distribution), where dis particle diameter and N is the number of particles bigger than d, for the region between the intense bottom and surface nepheloid-layers (Fig. 18). The presence of distributions comprising two segments of slopes m < 3 for sizes finer and m > 3 for sizes coarser than 5 pm is shown for shallow water by Brun-Cottan (1971), Bader (1970) and McCave (1975) and for nepheloid-layers both on the shelf (Pak et al. 1980) and in deeper water (Richardson
Erosion, transport and deposition o f fine-grained marine sediments 0.3
1.2
d. 0.2
0.8
g~
0.4
I-
,
~
i
1.4
t~ 0.6
1.2
0.4 O
0.8 0.4
o
9
0
'
1
~'~'
10
0
100
!
~
1
10
100
DIAMETER (pm)
Gently w e t
4 ~ (63~Jm)
sieved
==s
phi r ",2
4
8 /Jm
16
32
It]
Treated
with
H202
5
phi
Sample of suspended sediment taken at bottom +2m in 18m w a t e r depth off Walcott,Norfolk
FIG. 16. Raw and disaggregated size distributions from Nova Scotian coastal waters (Kranck 1973) (above) and southern North Sea coastal waters (McCave 1981) (below). 1980; McCave 1983) (Fig. 18). The m = 3 distribution is equivalent to a flat volume distribution, i.e. equal volumes of particles in logarithmic size grades (Sheldon et al. 1972) which appears to be the case in clear water, while the two-segment distribution is a peaked volume distribution with
5I
the peak in the region 4 to 8 pm found in more turbid water in direct comparison with the shelf and estuarine data given above, though at much lower concentrations. Lambert et al. (1981) measured and counted particles retained on 0.4/~m pore size Nuclepore filters. They find that the individual components (e.g. aluminosilicate grains, aggregates, o p a l + quartz, goethite particles) fit log-normal particle number distributions. In particular they find decreasing numbers of particles per logarithmic size interval for sizes < 0.8 pm, in contrast to the findings of Harris (1977). Similar results from the same method are reported by Baker et al. (1979) where decrease in dN/dd is demonstrated for d < 2.94 pm. Above that size the distributions are close to a power law with m > 3, thus there is a peak in the volume size distribution at about d-- 3 pro. Harris' (1977) data show a region where m = 1.65, close to the Brownian coagulation prediction of 1.5. The discrepancy between his data and those of other authors in the submicron range may either be undercount in the others as suggested by Wellershaus et al. (1973) or a true lack of such particles because they have already passed on up the size spectrum by aggregation. Aggregation in these deep-sea suspended sediment samples is also shown by the ultrasonic disaggregation experiments of McCave (1983) (Fig. 19). In these there were two modes of behaviour. High concentration, peaked distributions showed a substantial drop in the fine peak at 4 pm and an overall loss of material. This was thought to be due to disaggregation into submicron particles not now counted, being below the threshold. On the other hand, several of the very low concentration flat distributions actually gained material (Fig. 19). It was thought that in the process of becoming flat, much material in these distributions had been transferred into larger aggregates than the counter was registering (> 32 pm). Disaggregation of these particles was suggested to provide more material to the distribution than was lost by breakup of particles in the 1.26-32 #m counted window. Thus the flat distributions have gone beyond the stage of dominance by a single flocculated peak and have material more widely spread across the size spectrum (McCave, in press).
Breakup Analogous to the coagulation kernels of equation 7 are breakup kernels which account for the breakup ofvj to yield (L~-v;) and vi sized particles such that the rate of breakup of~!j per unit volume of suspension is
dN/dt
=
J(t~-; vi,
Vj--Vi) n (vj) dr~ dvi
(13)
I.N. McCave
52
64
16 E 9 0 /
m
9
8
0
0-.a U,.
4
de= 2"79d0772
FIG. 17. Relationship between modal diameters of flocculated and disaggregated suspensions from Nova Scotian estuaries (Kranck 1973).
I 2
according to Jeffrey (1982). A similar formulation giving the production rate of vi sizes from breakup ofvj sized particles was given by Valentas et al. (1966) and Valentas & Amundson (1966) (summarized by Spielman (1978)) as
dN/dt = J(vi, vj) n (vj) dvi dvj
(14)
but this presumes breakup into two equal halves. Jeffrey's expression yields these terms in the population balance equation for production and loss of vSs by breakup
~n(v) Ot
i
i
| J(vi+vfi vj, vi) n (?.)j--~ui) d v i Q/
o vj
f J(t); vi, v j - vi) n (vj) dvi o
(15)
However, it is by no means clear what is the functional form of J. This depends to some extent on the mode of particle breakup. Parker et al. (1972) proposed that particles either are fractured or are eroded particle-by-particle through surface shear. Their analysis showed that for particles smaller than the Kolmogorov microscale 2k = (I,'3//~) ~ the maximum size stable to surface erosion should be dm oc -~ and for d > 2k it should be dmocF -2, while in both regimes the
i
32
I
64
breakup size should be dm oc F- 89 (Note, as shear rate F oc e89where e is turbulent dissipation rate, these relations are equivalent to d,,,oce- 89dmoceand dmoc e-~ respectively). Parker et al. (1972) cite some support for these dependancies. Examination of floc size distributions shows many flocs larger than 2k which Parker et al. (1972) conclude must be undergoing surface erosion but are also maintained by high rates of aggregation. The rate of production of particles by surface erosion below the microscale was shown to be
dN/dt = KeF 2,
oo t'*
I
4 8 16 DEFLOCCULATED MODE ( d 4 p m )
(dN/dt oc e)
(16)
where Ke is a coefficient. Somewhat different relationships for floc fracture are proposed by Tambo & Hozumi (1979). They suggest that dmoC,g-3/2(3+kp) in the viscous subrange and dmocg-l/(l+kp ) in the inertial subrange, kp is the exponent in the relationship between floc size and density having a range kp = 1-1.5, so dm OCg -(0"38-0"33) and dmoCg -(0"5-0"4). If we say for the viscous subrange din oc e- 89then dmocU, -l or Zo- 89rather than z -1 suggested by Krone (1962). Their relationships for the viscous range are well supported by data from a stirred flocculator. They found dm~ 1.5 mm at e =2.10 -2 cm 2 s -3 and 200/~m at e = 10 cm 2 s -3 (F = 1.4 and 32 s -~) for clay-aluminium flocs. This suggests that in the generally low shear environment of the sea, flocs of a millimetre and more will be stable
Erosion, transport and deposition of fine-grained marine sediments
53
10 4
o~ 103"
|
E o
SAMPLE I-.z uJ O n"
15
KN83112110,
1000
|
mab
n..
("
)
2
u.= 10
,|
z
~ 101
~ lO.
N=7206d 3 2 3 2
..i
~10 ~
_o 0
'2
1
4.
8
1E;
1#
32
2 I
105.
4
8
16
32
@ (.) (.) (.)
104 .
(.) (.)
SAMPLE
2O
KN83/13/01
, 3 mab
(.)
~E 103,
9
(.) f,ILl
Iz
~
~ 15. LU Q.
Z LLI
"-- 10-
...I
w
rO
i1:
1 0 2,
_>
.~ 5"
101.
10o (.)
0
1(~ 1
1
2
4
8
16
32
dt~'n
,
~)
4.
8
16
',")
32
dlJm
FIG. 18. Examples of high and low concentration oceanic size distributions determined by Coulter counter from McCave (1983). (a) and (c) are particle-volume frequency curves while (b) and (d) are the corresponding cumulative particle-number distributions. while in higher shear zones near the bed 100-200 pm may be the limit. Under depositional conditions zo ~ "clat concentrations 1 kg m -3 (Mehta & Partheniades 1973).
Erosion, transport a n d deposition o f fine-grained marine sediments ('Co--Zt)/zt and against t/tso where t is time and ts0 is the time at which Ceq/Co= 50%. Thus there is a fraction of a given mud that will deposit at shear stresses well above the critical deposition stress (as defined by Einstein & Krone (1962)). This is of importance in estuarine deposition and the origins of fluid mud, but it also casts further doubt on critical erosion experiments where there is a large suspended load. There are no measurements of the properties of the material selectively deposited in this way. It is likely that only the strongest aggregates are deposited at the higher values of stress and that those particles capable of forming only weak bonds are continually disrupted near the bed and thus do not deposit. The process of deposition at the high values of stress (e.g. ~o= 0.6 Pa which could move 1 mm sand) must involve adhesion of sediment to the bed by electrostatic forces. Thus it seems likely that a bed deposited under such high shear conditions would have a much greater strength and critical erosion shear stress than one deposited under a slower flow. Field measurements
A large amount of the sediment in suspension in shelf and deep ocean waters is resuspended. Thus the short term deposition rate calculated from equation (18) rarely agrees with that obtained from radionuclides. In shallow marine environments where wave stresses are intermittently high, prediction of the net rate of deposition would require coupled models for fluid shear stress and flux and for rates of erosion and deposition. These exist for some estuarine areas, for example the Thames (Odd & Owen 1972), but not yet for shelves or the deep-sea, though progress is being made (Nowell et al. 1982). Thus we are forced back to measurement of accumulation rate. The radionuclides in sediments used for dating them are emplaced in two ways. Part of the signal at a given depth below the sediment surface arrives by being mixed down by biological activity and part arrives by the sediment surface moving up, by deposition. In conditions of no burrowing, for example in Santa Barbara Basin, the signal is entirely advective. But in most cases it is partly advective, partly bio-diffusive. In steady state this is expressed by
f v OpsU~ ~3psU -~xkr,~ff~-x j - ~ - ~ x + P s P - 2 p s N = 0
(21)
where KB is the biological diffusivity, S the sedimentation rate (LT-~), ps the sediment bulk density, N atoms g- 1 of nuclide and the third and fourth terms give nuclide production and decay
6I
rates (Goldberg & Koide 1962). The 21~ method has achieved great popularity recently because of its 22 year half-life. However, work with several tracers of differing half-lives shows that the biological diffusivity is not constant but decreases with depth. Sedimentation rates in Long Island Sound originally thought to be 1.1 mm y-~ on the basis of 21~ were halved to 0.5 mm y-~ when the diffusivity in the surface layer was found from 234Th (t~= 24d) profiles (Benninger et al. 1979). While 2t~ profiles may show a surface mixed layer underlain by a region of 21~ decrease, that decrease may not be due only to 2~~ decay. It may also contain some mixing effects on a rather longer time-scale. Thus the multi-layer model of Olsen et al. (1981) needs results from two or three nuclides to establish both depth-dependent diffusivities and deposition rates. Radionuclides with short half-lives are best in areas of rapid sediment accumulation. For example Kuehl et al. (1982) have determined rates from 6.5 to over 20 mm y-~ off the Amazon and Nittrouer et al. (1979) give 1-4 mm y-i (2-14 kg m-2 y-l) in the mid-shelf mud belt of the Washington Shelf using 21~ However, Bothner et al. (1981) found biological mixing too severe for that method and gave 0.2-0.5 mm y-1 using 14C for the mud belt off southern New England. Several measurements have been made on Long Island Sound sediments, giving values from 0.5 mm y-l to 3 mm y-l using zl~ 234Th and 7Be (Benninger et al. 1979; Aller et al. 1980; Krishnaswami et al. 1980). The advantage of using these short-lived isotopes is that short term rates of deposition can be determined. 2~~ (t 89 138d) and the bomb-produced ~37Csare also valuable in this respect. A comparison between them, the longer term rates given by ~4C and calculations based on equation (18) would give some insight into the amount of resuspension involved in the formation of shelf mud deposits. The same may also be possible in the deep-sea though few areas are likely to give fast enough rates of deposition. A notable exception is on the Nova Scotian rise where the creation of bedforms is believed to occur on a time-scale of a few weeks.
Conclusions Modern marine muds are complex mixtures of organic and inorganic components. Most experimental work has sought to characterize them in physical or chemical terms--Bingham yield strength, concentration, cation exchange capacit y - a n d relate these parameters to critical erosion stress and erosion rate constants. This approach
62
I.N. McCave median settling velocities generally fall in the range 0.05-0.5 mm s-~ an increase of up to two orders of magnitude over that for the disaggregated population. Shelf suspensions have velocities that fall at the lower end of this range, but there are few determinations and none at all for the deep sea. Aggregates are both produced and broken up by fluid shear. The breakup is rather poorly known at present but it may restrict aggregate sizes during deposition to be less than about 50/~m unless they have strong biological binding. The process will perform some sorting function so that only particles bound by strong bonds will be deposited in high shear zones. On this basis some clay mineral sorting may also be achieved by resuspension. Deposition is dominated by the concentration of the suspension and the settling velocity of the particles. The parameters all interact as settling velocity and critical deposition stress are functions of concentration. Nevertheless given these properties, it is reasonably predictable and can be incorporated in numerical models. Rates of shelf deposition would fall around 0.5 to 0.05 mg m -2 s-~ while deep-sea rates would be predicted to be
has had some success with some investigations in estuaries but has been less successful at shelf and oceanic depths because it is difficult to obtain in situ measurements comparable with those taken in the laboratory. Sediments in situ may have an important element of organic binding and their slow deposition rates render them most unlike the rapidly deposited slurries used in the laboratory. Thus we have a good idea of the general level of erosion stress (0.05-2 Pa) and erosion rate constant (10-4--2.10 -3 kg m -2 s -l Pa -l) but no good way of predicting it. It appears likely that deposits from fast flowing water will have a high erosion stress and that fast flowing water will 'harden' a previously deposited bed. Experiments in the field and laboratory currently being conducted will show the extent and importance of organic binding on clastic sediments. Such experiments have already shown the significance of sub-tidal algal mats on carbonates. Many changes occur to muds during transport but the most important is aggregation following particle collision by physical and biological means which increases particle settling velocity. For fully flocculated estuarine suspensions,
1000
I
% TRANSPORT SUSPENDED
lOO
AS LOAD
> W
~5
?0 ,0 k.
/ 10 F
I
tT I~
O
NO
MOVEMENT
I~n
t !
1.0 lffrn
tO0/~m
"lO/.~m
Imm
lOrnrn
GRA/N S I Z E d L
10-3
I
I
I
10 -2
10 -I
1.0
I
10
I
50
I
100
I
J
I
I
200 300 400 500
J
rnrns -7
SETTLING VELOCITY % F[G. 26. Proposed transport and deposition diagram for fine suspended sediment with erosion and transport fields ['or coarser materials.
Erosion, transport and deposition of fine-grained marine sediments one or two orders of magnitude less. Departure from this reflects variability in the environment; both of the depositing suspension's concentration and of the temporal distribution of resuspension. Earlier authors have sought to incorporate both erosion and deposition on one diagram (Hjulstrom 1935; Sundborg 1956; Postma 1967). However, as observed above, while deposition may be a function of settling velocity and thus size, erosion is not and so one cannot plot erosion and deposition on the same diagram, except for non-cohesive sediment. One can plot transport and deposition, however, and this is done in Fig. 26. The suspension transport curves are based on the exponent in the suspended load equation ( = w/tcu, describing the gradient of suspension concentration (where ~,-=0.4 is the von Karman constant), following the suggestion made first by Inman (1949) and subsequently taken up by others (Sundborg 1956). The critical deposition is the same as the critical erosion curve from White (1970) and Mantz (1977) following the arguments
63
of McCave & Swift (1976). The deposition field is contoured with lines of equal deposition rate divided by concentration giving a ~deposition velocity'. For erosion a separate diagram is required; however, it will appear from Figs 1-8 that no unique basis for such a diagram has yet been devised. ACKNOWLEDGEMENTS: In a synthesis of this type one draws heavily on the work of colleagues and other workers in the field. My sincere thanks go to them and my apologies in cases where credit is inadequately given. The manuscript benefited from reading by Drs. A.R.M. Nowell, R. Parker and J.P.M. Syvitski who have my thanks for the onerous chore. Thanks also to Barbara Slade who persuaded our word processor to accept an increasingly bulky manuscript. Finally I am grateful to the University of East Anglia and the US Office of Naval Research (Contract N00014-82-C-0019 NR083-004) for their support.
References ALLDREDGE, A.L. 1976. Field behaviour and adaptive strategies of appendicularians. Marine Biol., 38, 29 39. 1977. House morphology and mechanisms of feeding in the Oikopleuridae (Tunicata, Appendicularia). J. Zool., 181, 175-88. & MADIN, L.P. 1982. Pelagic tunicates: unique herbivores in the marine plankton. Bioscienee. 32, 655-63. ALLEN, J.R.L., in press. Fluid dynamics and sediment structures. In: Brenchley, P. & Williams, B.J. (eds) Sedimentology--Current Concepts and Applied Aspects. Geol. Soc. Lond. Spec. Pub. ALLER, R.C., BENNINGER,L.K. & COCHRAN,J.K. 1980. Tracking particle-associated processes in nearshore environments by use of 234Th/238U disequilibrium. Earth planet. Sci. Lett., 47, 161-75. ALLERSMA, E., HOEKStRA, A.J. & BIJKER, E.W. 1967. Transport patterns in the Chao Phya estuary. Delft Hydraulics Lab. Publ. no. 47, 24pp. AmAtHURAI, R. & ARULANANDAN,K. 1978. Erosion rates of cohesive soils. J. Hydraulics Die., Proc. Am. Soc. cir. Engrs., 104, 279-83. ARml, L. 1978. Some evidence for boundary mixing in the deep ocean. J. geophys. Res., 83, 1971-9. ARULANANDAN,K. 1975. Fundamental aspects of erosion of cohesive soils. J. Hydraulics Div.. Proc. Am. Soc. cir. Engrs., 101 (HY5), 635-9. BADER,H. 1970. The hyperbolic distribution of particle sizes. J. geophys. Res., 75, 2822-30. BAKER,E.T. 1976. Distribution, composition and transport of suspended particulate matter in the vicinity of Willapa submarine canyon, Washington. Bull. geol. Soe. Am., 87, 625-32. - - , FELLY, R.A. & TAKAHASHI,K. 1979. Chemical composition, size distribution and particle morpho-
-
-
logy of suspended particulate matter at DOMES sites A, B and C: relationships with local sediment composition. In: Bischoff, J.L. & Piper, D.Z. (eds), Marine Geology and Oceanography of the Pacific Manganese Nodule Province, Plenum Press, New York. 163-201. BALE, A.J., MORRIS, A.W. & HOWLAND, R.J.M., in press. Measuring the size characteristics of suspended particles in an estuary by laser Fraunhofer diffraction. In: Parker, W.R. & Kinsman, D.J.J. (eds), Transfer Processes in Cohesive Sediment Systems. Plenum Press, New York. BENNINGER, L.K., ALLER, R.C., COCHRAN, J.K. & TUREKIAN,K.K. 1979. Effects of biological sediment mixing on the 21~ chronology and trace metal distribution in a Long Island Sound sediment core. Earth planet. Sci. Lett., 43, 241-59. BIDDLE, P. 8r MILES, J.H. 1972. The nature of contemporary silts in British estuaries. Sed. Geol. 7, 23-33. BISCAYE, P.E. & EITTREIM, S.L. 1974. Variations in benthic boundary layer phenomena: nepheloid layer in the North American Basin. In: Gibbs, R.J. (ed.), Suspended Solids in Water. Plenum Press, New York. 227-60. -& -1977. Suspended particulate loads and transports in the nepheloid layer of the abyssal Atlantic, Marine Geol., 23, 155-72. - - , GARDNER,W.D., ZANEVELD,J.R.V., PAK, H. & TUCHOLKE, B.E. 1980. Nephels! have we got nephels! EOS, Trans. Am. Geophys. Un., 61, 1014 (abstract). DE BOER. P.L. 1981. Mechanical effects of microorganisms on intertidal bedform migration. Sedimentolog:v, 28, 129-32. BOTHNER,M.H., SPIKER,E.C., JOHNSON,P.P., REYDIGS, R.R. & ARUSCAVAGE,P.J. 1981. Geochemical evi-
64
I.N. McCave
dence of modern sediment accumulation on the continental shelf off southern New England. J. sed. Petrol., 51, 281-92. BRULAND,K.W. AND SILVER,M.W. 1981. Sinking rates of faecal pellets from gelatinous zooplankton (salps, pteropods, doliolids). Marine Biol., 63, 295-300. BRUN-COTTAN, J.C. 1971. Etude de la granulometrie des particules marines, measures effectuees avec un compteur Coulter. Cah. Oceanogr. 23, 193-205. - 1976. Stokes settling and dissolution-rate model for marine particles as a function of size distribution. J. geophys. Res., 81, 1601-6. BUCHAN, S., FLOODGATE, G.D. & CRISP, D.J. 1967. Studies on the seasonal variation of the suspended matter in the Menai Straits. I. The inorganic fraction. Limnol. Oceangr., 12, 419-31. BUSCH, P.L. & W. STUMM, 1968. Chemical interactions in the aggregation of bacteria: bioflocculation in waste treatment. Environ. Sci. Technol., 2, 49-53. CARLSON, E.J. & ENGER, P.F. 1963. Studies of tractive forces of cohesive soils in earth canals. U.S. Bureau of Reclamation, Denver, Hydraulic Branch Report, 504.
COLES, S.M. 1979. Benthic microalgal populations on inter-tidal sediments and their role as precursors to salt marsh development. In: Jeffries, R.L. & Davy, A.J. (eds), Ecological Processes in Coastal Environments. Blackwell, Oxford. 25~12pp. DEUSER, W.G., Ross, E.H. & ANDERSON, R.F. 1981. Seasonality in the supply of sediment to the deep Sargasso Sea and implication for the rapid transfer of matter to the deep ocean. Deep Sea Res., 28, 495-505. , BREWER,P.G., JICKELLS,T.D. & COMMEAU,R.F. 1983. Biological control of the removal of abiogenic particles from the surface ocean. Science, 219, 388-91. D1CKSON, R.R. & I.N. MCCAVE, 1982. Properties of nepheloid layers on the upper slope west of Porcupine Bank. Int. Council Explor. Sea, CM 1982/C:2, IIydrography Committee, 9pp. D1EPHUIS, J.G.H.R. 1966. The Guiana coast. T(jdschr. Kon. Ned. Ardrijksk. Gen., 83, 145-52. DRAKE, D.E. 1976. Suspended sediment transport and mud deposition on continetal shelves. In: Stanley, D.J. & Swift, D.J.P. (eds), Marine Sediment Transport and Environmental Management. Wiley, New York, 127-58. DRAKE, D.E., KOLPACK, R.L. & FISCHER, P.J. 1972. Sediment transport on Santa Barbara-Oxnard Shelf, Santa Barbara Channel, California. In: Swift, D.J.P., Duane, D.B. & Pilkey, O.H. (eds), Shelf Sediment transport: Process and pattern. Dowden, Hutchinson & Ross, Stroudsburg, PA. 307-32. DUNBARR.B. & BERGER,W.H. 1981. Fecal pellet flux to modern bottom sediment of Santa Barbara Basin (California) based on sediment trapping. Bull. geol. Soc. Am., 92, 212-18. DUNN, I.S. 1959. Tractive resistance of cohesive channels. J. Soil Mech. Found. DIE., Proc. Am. Soc. cir. Engrs., 85 (SM3). EDZWALD, J.K., UPCHURCH, J.B. & O'MELIA, C.R. 1974. Coagulation in estuaries. Environ. Sci. Technol., 8, 58-63.
EINSTEIN, H.A. & KRONE, R.B. 1962. Experiments to determine modes of cohesive sediment transport in salt water. J. geophys. Res., 67, 1451-61. EISMA, D. & G1ESKES,W.W.C. 1977. Particle size spectra of non-living suspended matter in the southern North Sea. Netherlands Institute .for Sea Research, Internal Report 1977-7, 22pp. & KALE, J. 1979. Distribution and particle size of suspended matter in the Southern Bight of the North Sea and the eastern Channel. Neth. J. Sea. Res., 13, 298-324. EITTREIM, S.L. & M. EWING, 1972. Suspended particulate matter in the deep waters of the North American Basin. In: Gordon, A.L. (ed.), Studies in Physical Oceanography. Gordon & Breach, New York. 123-67. - - - , THORNDIKE, E.M. & SULLIVAN, L. 1976. Turbidity distribution in the Atlantic Ocean. Deep Sea Res., 23, 1115-27. EKMAN, J.E., NOWELL, A.R.M. & JUMARS, P.A. 1981. Sediment destabilisation by animal tubes. J. mar. Res., 39, 361-74. EMERY, K.O. 1960. The Sea off Southern California: A Modern Habitat of Petroleum. John Wiley, New York. 366pp. ERNISSEE,J.J. & ABBOTT,W.H. 1975. Binding of mineral grains by a species of Thalassiosira. In: Simonsen, R. (ed.), Third symposium on recent and fossil marine diatoms, Kiel 1974, Nova Hedwigia, Beiheft, 53, 241-8. EWING, M., EITTREIM, S., EWING J. & LE PICHON, X. 1971. Sediment transport and distribution in the Argentine Basin: part 3, Nepheloid layer and processes of sedimentation. In: Ahrens, L.H. (ed.), Physics and Chemistry of the Earth, 8, Pergamon Press. 49-77. FALCO, R.E. 1977. Coherent motions in the outer region of turbulent boundary layers. Phys. Fluids, Suppl., 20, 124-30. FOWLER, S.W. & SMALL, L.F. 1972. Sinking rates of euphausiid faecal pellets. Limnol. Oceanogr., 17, 293-96. FRANCIS, J.R.D. 1973. Experiments on the motion of solitary grains along the bed of a water-stream. Proc. Roy. Soc. Lond., A332, 443-71. FRIEDLANDER, S.M. 1965. The similarity theory of the particle size distribution of the atmospheric aerosol. In: Spurny, K. (ed.), Aerosols, Physical Chemistry and Applications, Czech. Acad. Sci., Prague. 115-30. 1977. Smoke, Dust and Haze: Fundamentals of Aerosol Behaviour. John Wiley, New York. 317pp. FROSTICK, L.E. & MCCAVE, I.N. 1979. Seasonal shifts of sediment within an estuary mediated by algal growth. Estuar. coast, mar. Sci., 9, 569-76. GIBBS, H.J. 1962. A study of erosion and tractive force characteristics in relation to soil mechanics properties. U.S. Bureau of Reclamation, Report Em-643. GILMER, R.W. 1972. Free-floating mucus webs: a novel feeding adaptation for the open ocean. Science, 176, 1239-40. - 1974. Some aspects of feeding in Thecosomatous Pteropods. J. Exp. Mar. Biol. Ecol., 15, 127-44. GOLDBERG, E.D. & KOIDE, K. 1962. Geochronological
Erosion, transport and deposition of fine-grained marine sediments studies of deep-sea sediments by the ionium-thorium method. Geochim. eosmochim. Acta, 26, 417-450. GOULEAU, D. 1976. Le role des diatomees benthiques dans l'engraissement rapide des vasieres Atlantiques decouvrantes. C. R. hebd. Seanc. Acad. Sci., Paris, D283, 21-3. GRANT, W.D., BOYER,L.F. & SANFORD, L.P. 1982. The effects of bioturbation on the initiation of motion of intertidal sands. J. mar. Res., 40, 659-77. GREGORY, J. 1978. Effects of polymers on colloid stability. In: Ives, K.J. (ed.), The Scientific Basis of Flocculation. Sijthoff & Noordhoff, Alphen a.d. Rijn. 101-39. GULARTE, R.C., KELLY, W.E. & NACC1, V.A. 1980. Erosion of cohesive sediments as a rate process. Ocean Engrg., 7, 539-51. GUST, G. 1976. Observations on turbulent-drag reduction in a dilute suspension of clay in sea water. J. Fluid Mech., 75, 29-47. HARTLETT, J.C. & KULM, L.D. 1973. Suspended sediment transport on the northern Oregon continental shelf. Bull. geol. Soc. Am., 84, 3815-26. HARRIS, J.E. 1977. Characterisation of suspended matter in the Gulf of Mexico--II. Particle size analysis of suspended matter from deep water. Deep Sea Res., 24, 1055-61. HAVEN, D.S. & MORALES-ALAMO,R. 1972. Biodeposition as a factor in sedimentation of fine suspended solids in estuaries. In: Nelson, B.W. (ed.) Environmental Framework of Coastal Plain Estuaries., Geol. Soc. Am. Mem., 133, 121-30. HEAD, M.R. & BANDYOPADHYAY,P. 1981. New aspects of turbulent boundary-layer structure. J. Fluid Mech., 107, 297-338. HEEZEN, B.C. & HOLLISTER, C.D. 1964. Deep-sea current evidence from abyssal sediments. Marine Geol., 1, 141-74. HESSE, R. & CHOUGH, S.K. 1980. The Northwest Atlantic Mid-Ocean Channel of the Labrador Sea: II. Deposition of parallel laminated leveemuds from the viscous sublayer of low density turbidity currents. Sedimentology, 27, 697-711. HJULSTROM, F. 1935. The morphological activity of rivers as illustrated by the river Fyris. Bull. geol. Inst. Upsala, 25, 221-527. 1939. Transportation of detritus by moving water. In: Trask, P.D. (ed.), Recent Marine Sediments. Am. Ass. Petrol. Geols, Tulsa, 5-31. HOGG, N., BISCAYE,P., GARDNER, W. & SCHM1TZ,W.J. 1982. On the transport and modification of Antarctic bottom water in the Vema Channel. J. mar. Res., 40 Supplement, 231-63. HOLLAND, A.F., ZINGMARK, R.B. & DEAN, J.M. 1974. Quantitative evidence concerning the stabilisation of sediment by benthic diatoms. Marine Biol., 27, 191-96. HONJO, S. 1976. Coccoliths: production, transportation and sedimentation. Mar. Micropaleont., l, 65-79. 1980. Material fluxes and modes of sedimentation in the mesopelagic and bathypelagic zones. J. mar. Res., 38, 53-97. 1982. Seasonality and interaction of biogenic and -
-
-
-
-
-
65
lithogenic particulate flux at the Panama Basin. Science, 218, 883-4. & ROMAN, M.R. 1978. Marine copepod fecal pellets; production, preservation and sedimentation. J. mar. Res., 36, 45-57. - - - , MANGANIN1, S.J. & POPPE, L.J. 1982a. Sedimentation of lithogenic particles in the deep ocean. Marine Geol., 50, 199-220. - - , SPENCER,D.W. & FARR1NGTON,J.W. 1982b. Deep advective transport of lithogenic particles in Panama Basin. Science, 216, 516-8. HUNT, J.R. 1980. Prediction of oceanic particle size distributions from coagulation and sedimentation mechanisms. In: Kavanaugh, M.C. & Leckie, J.O. (eds), Particulates in Water. American Chemical Society, Advances in chemistry series No. 189, 243-57. -1982. Self similar particle-size distributions during coagulation: theory and experimental verification. J. Fluid Mech. 122, 169-86. HUNTER, K.A. & L~SS, P.S. 1982. Organic matter and the surface charge of suspended particles in estuarine water. Limnol. Oceanogr., 27, 322-35. HYDRAULICS RESEARCH STATION, 1977. Properties of Grangemouth mud. HRS, Wallingford, Report EX781, 13pp. 1979a. Port of Brisbane siltation study. Fifth report: properties of Brisbane Mud. HRS, Wallingford, Report EX860, 18pp. -1979b. Properties of Belawan mud. HRS, Wallingford, Report EX880, 15pp. INGLIS, C.C. & ALLEN, F.H. 1975. The regimen of the Thames estuary as affected by currents, salinities and river flow. Proc. Inst. civil. Eng., 7, 827-78. INMAN, D.L. 1949. Sorting of sediments in the light of fluid mechanics. J. sed. Petrol., 19, 51-70. ISEKI, K. 1981. Particulate organic matter transport to the deep sea by salp faecal pellets. Mar. Ecol.-Progr. Ser., 5, 55-60. IvEs, K.J. & BI4OLE,A.G. 1973. Theory of flocculation for continuous flow system. J. Environ. Eng. Div., Proc. Am. Soc. cir. Engrs., 99, (EEl), 17pp. JANKE, N.C. 1966. Effects of shape upon the settling velocity of regular convex geometric particles. J. sed. Petrol., 36, 370-6. JEFFREY, D.J. 1982. Aggregation and breakup of clay floes in turbulent flow. Adv. Colloid. Interface Sei., 17, 213-8. JORGENSEN, C.B. 1966. Biology of Suspension Feeding. Pergamon Press, Oxford. KAJIHARA, M. 1971. Settling velocity and porosity of large suspended particles. J. Oceangr. Soc. Japan, 27, 158-62. KIRBY, R. & PARKER, W.R. 1983. Distribution and behaviour of fine sediment in the Severn Estuary and inner Bristol Channel, U.K. Can. J. Fish. Aquat. Sci., 40 (suppl. 1), 83-95. KNAUER, G.A., MARTIN, J.H. & BRULAND,K.W. 1979. Fluxes of particulate carbon, nitrogen and phosphorous in the upper water column of the northeast Pacific. Deep Sea Res., 28, 921-36. KOMAR, P.D., MORSE, A.P., SMALL, L.F. & FOWLER, S.W. 1981. An analysis of sinking rates of natural
66
I.N. McCave
copepod and euphausiid faecal pellets. Linmol. Oceanogr., 26, 712-80. KRANCK, K. 1973. Flocculation of suspended sediment in the sea. Nature, Lond., 246, 348-50. -1975. Sediment deposition from flocculated suspensions. Sedimentology, 22, 111-23. 1981. Particulate matter grain-size characteristics and flocculation in a partially mixed estuary, Sedimentology, 28, 107-14. KRISHNASWAMI, S., BENNINGER, L.K., ALLER, R.C. & VON DAMM, K.L. 1980. Atmospherically-derived radionuclides as tracers of sediment mixing and accumulation in near-shore marine and lake sediments: evidence from 7Be, 21~ and 239,24~ Earth planet. Sci. Lett., 47, 307-18. KRONE, R.V. 1959. Second annual report on silt transport studies utilizing radioisotopes. Hydraulic Engineering and Sanitary Engineering Research Laboratory, University of California, Berkeley. 123pp. -1962. Fhmw studies of the transport q[sediment in estuarial shoaling processes. Hydraulic Engineering and Sanitary Engineering Research Laboratory, University of California, Berkeley. 110pp. 1963. A study of rheologic properties of estuarial sediments. Sanitary Engineering Research Laboratory Report 63-8, University of California. Berkeley, 91 pp. 1972. A .field study of flocculation as a fiwtor in estuarial shoaling processes. Tech. Bull. 19, U.S. Army Corps of Engineers Committee on Tidal Hydraulics, Waterways Experiment Station, Vicksburg, Miss. 62pp. 1976. Engineering interest in the benthic boundary layer. In: I.N. McCave, (ed.), The Benthic Boundary Layer. Plenum Press, New York. 143-56. 1978. Aggregation of suspended particles in estuaries. In: Kjerfve, B. (ed.), Estuarine Transport Processes. University of South Carolina Press, Columbia, S.C. 177-90. KUEHL, S., NITTROUER,C.A. & DE MASTER, D.J. 1982. Modern sediment accumulation and strata formation on the Amazon continental shelf. Marine Geol., 49, 279-300. LAMBERT, C.E., JEHANNO, C., SILVERBERG,N., BRUNCOTTAN, J,C. 81, CHESSELET, R. 1981. Log-normal distributions of suspended particles in the open ocean. J. mar. Res., 39, 77-98. LERMAN, A. 1979. Geochemical Processes." Water and Sediment Environments. John Wiley, New York. 481pp. , CARDER, K.L. & BETZER, P.R. 1977. Elimination of fine suspensoids in the oceanic water column. Earth planet. Sci. Lett., 37, 61-70. LEWIN, J., COLVIN, J.R. & MCDONALI), K.L. 1980. Blooms of surf-zone diatoms along the coast of the Olympic Peninsula, Washington. XII. The clay coat of Chaetoceros armature T. West. Botanica Marina, 23, 333-41. LONSI)ALE, P. & SOUTHARD, J.B. 1974. Experimental erosion of North Pacific red clay. Marine Geol., 17, M25-M34. LORENZEN, C.J., SHUMAN, F.R. & BENNETT,J.T. 1981. In situ calibration of a sediment trap. Linmol. Oceanogr., 26, 580-4. -
-
-
-
-
-
-
-
LYKLEMA, J. 1978. Surface chemistry of colloids in connection with stability. In: Ives, K.J. (ed.), The Scientific Basis of Flocculation, Sijthoff & Noordhoff, Alphen a.d. Rijn. 3-36. LYLE, W.M. & SMERDON, E.T. 1965. Relation of compaction and other soil properties to erosion and resistance of soils. Trans. Am. Soc. Agric. Engrs., 8. MAD1N, L.P. 1974. Field observations on the feeding behaviour of Salps (Tunicata: Thaliacea). MarhTe Biol., 25, 143-7. -1982. Production, composition and sedimentation of salp fecal pellets in oceanic waters. Marine Biol., 67, 39-45. MANTZ, P.A. 1977. Incipient transport of fine grains and flakes by fluids--extended Shields diagram. J. Hydraulics Dir., Proc. Am. Soc. cir. Engrs., 103 (HY6), 601-15. -1978. Bedforms produced by fine cohesionless, granular and flakey sediments under subcritical water flows. Sedimentology, 25, 83-103. MCCAvE, I.N. 1970. Deposition of fine-grained suspended sediment from tidal currents. J. geophys. Res., 75, 4151-9. 1975. Vertical flux of particles in the ocean. Deep Sea Res., 22, 491-502. 1978. Sediments in the abyssal boundary layer. Oceanus, 21 (1), 27-33. 1979a. Suspended sediment, Chapter 6. In: Dyer, K.R. (ed.), Estuarine Hydrography and Sedimentation. Cambridge University Press. 13 l-185. 1979b. Suspended material over the central Oregon continental shelf in May 1974. I: Concentration of organic and inorganic components. J. sed. Petrol., 49, 1181-94. 1981. Location of coastal accumulation of fine sediments around the Southern North Sea. In: Postma, H. (ed.), Proceedings of the Workshop on Sediments and Pollutant Exchange. Rapp. ProcesVerbaux Reun. Cons. Int. Explor. Mer., 181, 15-27. 1983. Particulate size spectra, behaviour and origin of nepheloid layers over the Nova Scotian Continental Rise. J. Geophys. Res., 88, 7647-66. --, in press. Size spectra and aggregation of suspended particles in the deep ocean. Deep-Sea Res., 31. --, LONSDALE, P.F., HOLLISTER, C.D. & GARDNER, W.D. 1980. Sediment transport over the Hatton and Gardar contourite drifts. J. sed. Petrol., 50, -
-
-
-
-
-
1
0
4
9
-
6
2
.
& SWlFT, S.A. 1976. A physical model for the rate of deposition of fine-grained sediments in the deep sea. Bull. geol. Soc. Am., 87, 541-6. MEADE, R.H. 1972. Transport and deposition of sediments in estuaries. In: Nelson, B.W. (ed.), Environmental Framework of Coastal Plain Estuaries. Geol. Soc. Am. Mem., 133, 91-120. - - , SACHS, P.L., MANHEIM,F.T., HATHAWAY,J.C. t~ SPENCER, D.W. 1975. Sources of suspended matter in waters of the Middle Atlantic Bight. J. sed. Petrol., 45, 171-88. MEADOWS, P.S. & ANDERSON, J.G. 1968. Microorganisms attached to marine sand grains. J. Mar. Biol. Ass. UK, 48, 161-175. MEHTA, A.J. & PARTHENIADES, E. 1973. Depositional -
-
Erosion, transport and deposition o/ifine-grained marine sediments Behaviour of Cohesive Sediments. Coastal and oceanographic engineering laboratory, University of Florida, Gainesville, Tech. Rept., 16, 274pp. MIGNIOT, C. 1968. Etude des proprietes physiques de differents sediments tres fins et de leur comportement sous des actions hydrodynamiques. La Houille Blanche, 7, 591-620. MILLER, M.C., MCCAVE, I.N. & KOMAR, P.D. 1977. Threshold of sediment motion under unidirectional currents. Sedimentology, 24, 507-27. MITCHELL, J.K. 1964. Shearing resistance of soils as a rate process. J. Soil Mech. Found. Div., Proc. Am. Soc. cir. Engrs., 90 (SM1), 29-61. -1976. Fundamentals of Soil Behaviour. John Wiley, New York. 422 pp. , CAMPANELLA,R.G. & SINGH, A. 1968. Soil creep as a rate process. J Soil Mech. Found. Div., Proc. Am. Soc. cir. Engrs., 94, (SM1), 231-53. MULLIN, M.M. 1980. Interactions between marine zooplankton and suspended particles, a brief review. In: Kavanaugh, M.C. & Leckie, J.O. (eds), Particulates in Water. American Chemical Society, Advances in Chemistry Series, No. 189, 233-41. NEUMAN, A.C., GEBELE1N,C.D. & SCOFFIN,T.P. 1970. The composition structure and erodability of subtidal mats, Abaco, Bahamas. J. sed. Petrol., 40, 274-97. NEWTON, A.J. & GRAY, J.S. 1972. Seasonal variation of the suspended solid matter off the coast of North Yorkshire. J. Mar. Biol. Ass. UK., 52, 33-47. NIEHOFF, R.H. & LOEB, G.I. 1972. The surface charge of particulate matter in seawater. Limnol. Oceanogr., 17, 7-16. & LOEB,G.I. 1974. Dissolved organic matter in sea water and the electric charge of immersed surfaces. J. mar. Res., 32, 5-12. NITTROUER,C.A., STERNBERG,R.W., CARPENTER,R. & BENNETT, J.T. 1979. The use of Pb--210 geochronology as a sedimentological tool: application to the Washington continental shelf. Marine Geol., 31, 297-316. NOWELL, A.R.M. 1982. Measuring and modelling organism effects on sediment transport. Geol. Soc. Lond. Newsletter, II, (2), 38-9 (abstract). • CHURCH, M. 1979. Turbulent flow in a depthlimited boundary layer. J. geophys. Res., 84, 4816-24. , JUMARS, P.A. & ECKMAN, J.E. 1981. Effects of biological activity on the entrainment of marine sediment. Marine Geol., 42, 133-53. ODD, N.V.M. & OWEN, M.W. 1972. A two-layer model of mud transport in the Thames Estuary. Proc. Instn. cir. Engrs., Suppl, 9, 175-205. VAN OLPHEN, H. 1977. An Introduction to Clay Colloid Chemistry. 2nd edn. John Wiley, New York. OLSEN, C.R., SIMPSON,H.J., PENG, T.H., BoPP, R.F. & TRIER, R.M. 1981. Sediment mixing and accumulation rate effects on radionuclide depth profiles in Hudson Estuary sediments. J. geophys. Res., 86, 1020-8. O'MELIA, C.R. 1972. Coagulation and flocculation. In: Weber, W.J. (ed.) Physiochemical Processes for -
-
-
-
-
67
Water Quality Control. Wiley-Interscience, New York. 61-109. 1978. Coagulation in wastewater treatment. In: Ives, K.J. (ed.), The Scientific Basis of Flocculation, Sijthoff& Noordhoff, Alphen a.d. Rijn, 219-68. OWEN, M.W. 1970. A Detailed Study of the Settling Velocities of an Estuary Mud. Hydraul. Res. Station, Wallingford, Rep. INT 78, 25pp. 1971. The effect of turbulence on the settling velocities of silt flocs. Proc. 14th Congress int. Ass. Hydraulics Res., Paris, Paper D-4, 27-32. -1975. Erosion of Avonmouth mud. Hydraul. Res. Station, Wallingford, Rep. INT 150, 17pp. PAK, H. & ZANEVELO,J.R.V. 1977. Bottom nepheloid layers and bottom mixed layers observed on the continental shelf off Oregon. J. geophys. Res., 82, 3921-31. --, ZANEVELD,J.R.V. & KITCHEN, J. 1980. Intermediate nepheloid layers observed off Oregon and Washington. J. geophys. Res., 85, 6697-708. PARKER, D.S., KAUFMAN,W.J. &JENKINS, D. 1972. Floc breakup in turbulent flocculation processes. J. Sanit. Eng. Div., Proe. Am. Soc. cir. Engrs., 98 (SAI), 79 99. PARKER, W.R. & KIRBY, R. 1981. The Behaviour of Cohesive Sediment in the Inner Bristol Channel and Secern Estuary in Relation to Construction 0[" the Sez'ern Barrage. Inst. Oceanogr. Sci. (Wormley U.K.) Rept. 117, 54pp. Unpublished M.S. PARTHENIADES, E., CROSS, R.H. & AYORA, A. 1968. Further results on the deposition of cohesive sediments. Proc. l lth Conf Coastal Engineering, London. Am. Soe. civil Engrs., 723-42. PIERCE, J.W. & SIEGEL, F.F. 1979. Suspended particulate matter on the southern Argentine shelf. Marine Geol., 29, 73-91. POSTMA, H. 1961. Transport and accumulation of suspended material in the Dutch Wadden Sea. Neth. J. Sea Res., 1, 148-90. 1967. Sediment transport and sedimentation in the estuarine environment. In: Lauff, G.H. (ed.), Estuaries, Am. Ass. Advmt. Sci. Pub., 83, Washington, 158-79pp. PRUPPACHER, H.R. & KLETT, J.D. 1978. The Microphysics of Clouds and Precipitation. Riedel, Dordrecht. 714pp. REES, A.I. 1966. Some flume experiments with a fine silt. Sedimentology, 6, 209-40. RHOADS, D.C., YINGST, J.Y. & ULLMAN, W. 1978. Seafloor stability in central Long Island Sound: Pt. I. Temporal changes in erodibility of fine-grained sediment. In: Wiley, M.L. (ed.), Estuarine Interaction. Academic Press. 221-44. & BOYER, L.F., in press. The effects of marine benthos on physical properties of sediments; a successional perspective. In: Tevesz, M.J.S. & McCall, P.L. (eds.), Animal-Sediment Relations." The Biogenic Alteration of Sediments. Plenum Press, New York. RICHARDSON, M.J. 1980. Composition and Characteristics of Particles in the Ocean." Evidence for Present Day Resuspension. Ph.D. Thesis, Mass. Inst. Technol., Woods Hole Oceanogr. Inst., WHOI-80-52. ROBERT, J.-M. & GOULEAU, D. 1977. Confirmation -
-
-
-
-
68
I.N. McCave
exp6rimentale du r61e d'une diatom+e benthique, Navicula ramosissima (Agardh) Cleve, dans la production de formations mucilagineuses stabilisatrices des vasieres marines littorales. C.R. hebd. Seanc. A cad. Sci., Paris, D284, 1915-18. ROUSE, H. 1937. Modern conception of the mechanics of fluid turbulence. Trans. Am. Soc. cir. Engrs., 102, 463-543. ROWE, G.T. & GARDNER, W.D. 1979. Sedimentation rates in the slope water of the northeast Atlantic Ocean measured directly with sediment traps. J. mar. Res., 37, 581-600. SAFFMAN,P.G. & TURNER,J.S. 1956. On the collision of drops in turbulent clouds. J. Fluid Mech., 1, 16-30. SARGUNAM, A., RILEY, P., ARULANANDAN, K. & KRONE, R.B. 1973. Physico-chemical factors in erosion of cohesive soils. J. Hydraulics Div., Proc. Am. Soc. cir. Engrs., 99, (HY3), 555-8. SCHAFFERNAK, F. 1922. Neue Grundlagen fur Berechhung der Geschiebefuhrung in Flusslaufen. Franz Deutike, Leipzig & Vienna. SCHRADER, H.J. 1971. Faecal pellets: role in sedimentation of pelagic diatoms. Science, 174, 55-7. SCHUBEL, J.R. 1968. Suspended sediment of the northern Chesapeake Bay. Chesapeake Bay Institute, Johns Hopkins Univ. Technical Report, 35, 262pp. 1969. Size distributions of the suspended particles of the Chesapeake Bay turbidity maximum. Neth. J. Sea Res., 4, 283-309. 1971. Tidal variation of the size distribution of suspended sediment at a station in the Chesapeake Bay turbidity maximum. Neth J. Sea Res., 5, 252-66. SCOFFI~, T.P. 1968. An underwater fume. J. sed. Petrol., 38, 244-7. 1970. The trapping and binding of subtidal carbonate sediments by marine vegetation in Bimini Lagoon, Bahamas. J. sed. Petrol., 40, 249-73. SHANKS, A.L. & TRENT, J.B. 1980. Marine snow: sinking rates and potential role in vertical flux. Deep Sea Res., 27, 137-43. SHELDON,R.W., PRAKASH,A. & SUTCLIFFE,W.H. 1972. The size distribution of particles in the ocean. Limnol. Oceanogr., 17, 327-40. SILVER, M.W. & ALLDREDGE,A.L. 1981. Bathypelagic marine snow: deep-sea algal and detrital community. J. mar. Res., 39, 501-30. SMAYDA, T.J. 1969. Some measurements of the sinking rate of faecal pellets. Limnol. Oceanogr., 74, 621-5. -1971. Normal and accelerated sinking of phytoplankton in the sea. Marine Geol., 11, 105-22. SMERDON, E.T. & BEASLEY, R.P. 1959. Tractive force theory applied to stability of open channels in cohesive soils. Agricultural Expt. Sta., Unit'. Missouri, Columbia, Research Bull. 715. SMOLUCHOWSK1, M. 1917. Versuch einer mathematischen Theorie der Koagulation kinetik kolloider L6sungen. Ann. Physik. Chem., 92, 129-68. SOUTHARD, J.B., YOUNG, R.A. & HOLLtSTER, C.D. 1971. Experimental erosion of calcareous ooze. J. geophys. Res., 76, 5903-9. SPIELMAN,L.A. 1978. Hydrodynamic aspects of flocculation. In: Ives, K.J. (ed.). The Scientific Basis of -
-
-
-
Flocculation. Sijthoff and Noordhoff, Alphen a.d. Rijn. 63-88. STow, D.A.V. & BOWEN, A.J. 1980. A physical model for the transport and sorting of fine-grained sediment by turbidity currents. Sedimentology, 27, 31-46. STUMM, W. & MORGAN, J.J. 1981. Aquatic Chemistry (chapt. 10, the solid--solution interface). Wiley, New York, 2nd Edn. 780pp. SUND~ORG, A. 1956. The River Klaralven, a study of fluvial processes. Geogr. Annaler, 38, 125-316. SUZUKI, N. & KATO, K. 1953. Studies on suspended materials. Marine snow in the sea. I. Sources of marine snow. Bull. Fac. Fisheries, Hokkaido Unit,., 4, 132-5. SWIFT, D.L. & FRIEDLANDER,S.K. 1964. The coagulation of hydrosols by Brownian motion and laminar shear flow. J. Colloid Sei., 19, 621-47. TAMBO, N. & HOZUMI, H. 1979. Physical characteristics of flocs--II. Strength offloc. Water Res., 13, 421-7. -t~r WATANABE,Y. 1979. Physical characteristics of flocs--I. The ftoc density function and atuminium floc. Water Res. 13, 409-19. TERWINDT, J.H.J. & BREUSERS,H.N.C. 1972. Experiments on the origin of flaser, lenticular and sandclay alternating bedding. Sedimentology, 19, 85-98. TERZAGHI, K. & PECK, R.B. 1968. Soil Mechanics in Engineering Practice. John Wiley, New York. THORN, M.F.C. 1975. Monitoring silt movement in suspension in a tidal estuary. Int. Assoc. Hydraulic Res. 16th Cong., SaD Paulo, Pap C71. 596-603pp. 1981. Physical processes of siltation in tidal channels. In: Hydraulic Modelling Applied to Maritime Engineering Problems. Inst. Civil Engrs., London. 47-55pp. & PARSONS, J.G. 1980. Erosion of cohesive sediments in estuaries: an engineering guide. In: Third International Symposium on Dredging Technology, British Hydraulics Research Assocation, Cranfield. 349-58. UNSOLD,G. 1982. Der Transportbeginn rolligen Sohlmaterials in gleichformigen turbulenten Stromungen: eine kritische Uberprufung der Shields-Funktion und ihre experimentelle Erweiterung auf feinstkornige, nicht-bindige Sedimente. Doctoral Dissertation, University of Kiel, 145pp. VALENTAS, K.J. & AMUNDSON, N.R. 1966. Breakage and coalescence in dispersed-phase systems. Ind. Eng. Chem. Fund., 5, 533-52. - - , BILOUS,O. & AMUNDSON,N.R. 1966. Analysis of breakage in dispersed phase systems. Ind. Eng. Chem. Fund., 5, 271-9. WEATHERLY, G. & KELLEY, E.A. 1982. 'Too cold' bottom layers in the HEBBLE area. J. mar. Res., 40, 985-1012. WEBB, J.E. 1969. Biologically significant properties of submerged marine sands. Proc. Roy. Soc. Lond., B174, 355-402. WELIKANOFF, M. 1932. Eine Untersuchung uber erodierende Stromgeschwindigkeiten. Wasserkraft u. Wasserwirtschaft, 27 (17), 196-9. WELLERSHAUS, S., GOKE, L. & FRANK, P. 1973. Size distribution of suspended particles in sea water.
Erosion, transport and deposition of fine-grained marine sediments Meteor Forschungs-Ergbenisse, Series D, no. 16, 1-16. WELLS, J.T. & COLEMAN,J.M. 1981. Physical processes and fine-grained sediment dynamics, coast of Surinam, South America. J. sed. Petrol., 51, 1053-68. WIalTE, S.J. 1970. Plane bed thresholds of fine grained sediments. Nature, Lond., 228, 152-53. WIEBE, P.H., BOYD, S.H. & WINGET, C. 1976. Particulate matter sinking to the deep-sea floor at 2000 m in the Tongue of the Ocean, Bahamas, with a description of a new sedimentation trap. J. mar. Res., 34, 341-54.
69
Y1NGST,J.Y. & RHOADS,D.C. 1978. Seafloor stability in central Long Island Sound. Part II. Biological interactions and their potential importance for seafloor erodibility. In: Wiley, M.N. (ed.), Estuarine Interactions. Academic Press, New York. 245-60. YOUNG, R.A. & SOUTHARD, J.B. 1978. Erosion of fine-grained marine sediments: Sea floor and laboratory experiments. Bull. geol. Soc. Am., 89, 663-72. ZABAWA, C. 1978. Microstructure of agglomerated suspended sediments in Northern Chesapeake Bay Estuary. Science, N.Y., 202, 49-51.
I.N. MCCAVE, School of Environmental Sciences, University of East Anglia, Norwich, NR4 7T J, UK.
Methods and observations in the study of deep-sea suspended particulate matter S.L. Eittreim S U M M A R Y : Forward light scattering and light transmission are the two best techniques, coupled with direct sampling, to measure the spatial and temporal variations in suspended sediments of the ocean. It has been difficult to calibrate these methods with satisfactory precision in terms of suspended particulate matter (SPM) concentration; the best precision is approximately _+ 10 ~gL -I. The poor resolution is due to the many unknown factors in the relationship between optical effects and concentration of SPM, in addition to errors in direct sampling of SPM concentration. Because typical deep-sea concentrations are in the range of 5 to 50 FtgL-1, extreme care must be taken to avoid contamination by airborne dust and salt filter residues. As a short-lived tracer of water motions, SPM is useful because its settling rates range from about 10 I to 102 m day -I, rates giving half-life time-scales roughly comparable to time-scales of deep-water eddy motions and boundary layer events. More needs to be learned, however, to define these settling rates with some confidence. Much progress has recently been made in studies with sediment traps which directly measure the downward vertical flux of particles. The work with sediment traps is prompted by the realization that large fast-settling aggregates and faecal pellets are important in the delivery of material to the bottom, and due to their short residence time and hence low concentrations in the water column, they are not sampled adequately by water bottles. Introduction
The interest in fine-grained suspended particulate matter (SPM) studies in the deep sea stems from a n u m b e r of sources. First, SPM studies give information about the transport paths of deep-sea sediments. These transport paths span great distance and time-scales, and little is k n o w n of them aside from gross characterization of turbidity distributions and inferences from fine-grained sea floor sediments. Second, SPM can be of use in studies of deep-sea circulation as a tracer o f water motions. Because the particles settle out, SPM acts as a non-conservative property of the water whose 'half-life', if it is known, can be used to derive mixing rates. This is proving especially useful in benthic boundary-layer studies where brief boundary-layer mixing events can be detected and examined with this short-lived tracer. Third, it is now widely appreciated that element transports in settling particles are important links in the geochemical and biological cycles of the ocean. Therefore the ocean's dynamics of particle aggregation, dissolution, and settling, as well as particle chemistry and nutrient content, are important frontiers of study to understand these cycles. In this brief review paper I focus on deep-water methods, including optical methods used and the principles behind each. For a complete and rigorous treatment of optical methods in oceanography, see Jerlov (1976). Gibbs (1974) presents studies of applied optical methods in oceanography ranging from studies of ocean SPM to studies aimed specifically at the optical properties of sea
water. In addition, Sackett (1978) gives a comprehensive review of the state of the art of SPM work in sea water. Simpson (1982) presents a recent review of SPM work and some novel approaches toward sampling. The deep-sea environment is in general relatively time-invarient and the SPM concentrations are low, ranging from 1 to 103 /~gL-~. The coastal, shelf, and estuarine environments, on the other hand, are d o m i n a t e d by tidal and short-term storm effects, and in general higher concentrations prevail. Examples of data from different parts o f the world ocean will also be presented to give an idea of the range of values encountered in the deep sea. A strong impetus was provided towards the study of deep-sea SPM by the Swedish Deep Sea Expedition in the late 1940s, which systematically sampled the world ocean at all depths and found significant geographic variations as well as pronounced vertical stratification of SPM as determined by light-scattering properties of samples analysed o n b o a r d ship (Jerlov 1953). These studies contributed to the realization (e.g. Heezen & Hollister 1964) that the deep sea was not necessarily a 'stagnant' environment devoid of significant currents and that boundary-layer frictional effects are likely to be important in suspending sediments. Inspired by the Swedish Expedition data, two different approaches have evolved for studying the variations in ocean water turbidity ~. The two different approaches have 1 It should be noted here that 'turbidity', like cloudiness, is a term to describe the net optical effect of particles in water and is a non-quantitative term.
71
72
S.L.
Eittreim
been aimed at quite different objectives, and it is important to separate these different objectives in looking at instruments used today. One objective is to determine ocean water optical properties. These optical properties are strongly determined by, among other things, the SPM present. The other allied objective is to determine the amount of SPM in ocean water and its spatial and temporal variations. The traditional method of instrument deployment, that of vertical profiling from ships, augmented by discrete water samples, has provided the most geologically and oceanographically useful data. The recent evolution towards deployment of moored arrays reflects the fact that temporal information is needed on SPM concentrations as well as details of spatial variations near bottom in order to better understand boundary-layer processes which may be 'eventdominated'. Fine-scale spacing of sensors and long-term measurements can best be achieved by avoiding attachment to a surface vessel.
Optical methods Since direct sampling of SPM is for practical reasons spatially limited, optical methods offer the best means of obtaining information on SPM variations with high resolution. Another method for continuous data acquisition of SPM concentrations is acoustic scattering, but acoustic methods have proved useful only for relatively high concentrations of coarse material due to limitations imposed by sound wavelength and attenuation (Orr & Hess 1978). Optical methods can of course be applied in vitro, but their main advantage is in situ where continuous sampling in time and space can be achieved as opposed to discrete sampling, the latter of which is limited in interpretive value to an oceanographer. The two optical methods presently in widest use for determining SPM concentrations in the ocean are forward scattering and transmission. Forward scattering offers the highest sensitivity due to the fact that particles scatter light preferentially in a forward direction. However, absolute calibration is difficult, and most measurements based on scattering to date have been recorded in relative units. Instruments which measure relative scattering in a forward direction coupled with water samples taken for SPM concentration have been employed by Lamont-Doherty Geological Observatory (Lamont) and Woods Hole Oceanographic Institution (Woods Hole) through the 'GEOSECS' program, a reconnaissance geochemical survey of the world ocean. Reconnaissance surveys were carried out by Lamont in the
late 1960s and 1970s, occupying several thousand stations worldwide at which vertical nephelometer profiles were made (e.g. Sullivan et al. 1973, 1975; Jacobs et al. 1974, 1980; Eittreim et al. 1976). At a limited number of these stations, water samples were also collected for SPM extraction. GEOSECS included vertical nephelometer profiles and water samples for SPM extraction at most of their several hundred stations (Bainbridge et al. 1976, 1977). Transmission, or its reciprocal, attenuation, offers the attractiveness of simplicity of measurement and ease of optical calibration. The drawback of transmission measurements for deepocean work is their lower sensitivity to suspended particles and the relatively stronger effect of dissolved substances on the measurement. Thus, in very low concentration environments ( < 5 0 pgL-l), their utility may be less than forward scattering measurements. However, R. Zaneveld (pers. comm. 1981) of Oregon State University recently developed a red-light-emitting-diode transmissometer for which he claims _+0.1% absolute accuracy. This instrument may be able to resolve low concentration variations in the deep sea. To quantitatively define the optical properties of ocean water, well-calibrated instruments are required. Careful control on angles, irradiated volumes, wavelengths, source strength, and window transmission and reflectance must be maintained in order to define the two optical factors in an absolute sense: light-scattering as a function of angle and light absorption (Jerlov 1976). To determine scattering, the measurement must be made with high precision as a function of angle because of the steep increase in scattering with decreasing small angles. Quantitative in situ scattering measurements have been made to great depths in limited areas (e.g. Beardsley et al. 1970; Matlack 1972), but no broad reconnaissance surveys have yet been made. Advancement towards standardized measurements of factors that cause ocean turbidity variations, using a rugged package for survey work, would be a timely development (Austin 1974). Much of the complexity of the measurement is eliminated if one is simply interested in SPM concentration. Devices which measure a relative optical property that is strongly SPM-dependent and which are coupled with water samples to calibrate the measurement in terms of SPM concentration can be made simple and rugged, and such instruments have proved to be productive in the past. Scattering measurements About half of the scattered light produced by
The study of deep-sea suspended particulate matter particles is in a forward cone of 9 ~. Forward scattering observed against the black background of the deep sea below the photic zone is thus a highly sensitive method of measuring suspended particles. This pronounced forward lobe means that quantitative measurements of scattering must be made with respect to angle, 0./~(0) is the conventional symbol for scattering per unit volume as a function of angle. Most relative scattering instruments measure an integral of/~(0) over some range of angles. However, if the volume from which the scattering emanates is not limited to a small size and defined with precision, calibration in absolute units is not possible. By limiting the sample volume to a small size, the sampling statistics degrade (because of the relative rarity of the larger particles) so that the objective of obtaining useful SPM data which is representative of a spatial average is somewhat incompatible with obtaining quantitative scattering measurements- hence the advent of scattering instruments ('nephelometers' or 'turbidimeters') which measure in relative units over a range of forward angles (Thorndike 1975; Meade et al. 1975). The Lamont instrument uses light scattered from approximately 8 ~to 2 4 ~from an incandescent light and records photographically. The film for each station is calibrated prior to lowering to determine its optical density-exposure relationship (Thorndike 1975). The Woods Hole nephelometer uses a laser light source with similar small and adjustable angles of scatter, is self-contained, and has an acoustic telemetered link to a surface vessel (Meade et al. 1975). An instrument designed by R. Koehler of the Ocean Engineering Group at Woods Hole and used in abyssal benthic boundary layer studies by Armi & D'Asaro (1980) also uses forward light-scattering (20~'-40 '~) from a detector and light source of an industrial turbidimeter and also measures in relative units. Measurement of total scattering at all angles, an integral of/~(0) from 0 ~=to 180, is conventionally denoted by b in units of m - 1. Instruments that measure b, excluding for practical reasons the far forward and far aft angles, have been built and used at sea by Jerlov (1953), Beardsley et al. (1970), and Sternberg et al. (1974). These instruments also have generally recorded in relative units.
= 1/r In I / T , or = In l I T for a 1-m path-length instrument.
Workers have used both units, %T, or attenuation in m - l . Attenuation is a somewhat more convenient unit because its value varies directly with SPM concentration rather than inversely. Drake (1971, 1974), Pak et al. (1980), and McCave (1983) have reported on oceanic sampiing programs in which transmission was measured in conjunction with water sampling for SPM concentrations.
Water sampling objectives and methods Concentrations of SPM, particle size spectra, and particle composition are the three principal kinds of information sought through direct water sampiing. Concentrations of SPM in the deep sea are commonly measured gravimetrically and expressed in/~gL -1 (or #gkg-1), although parts per million (ppm) by volume is often used when the measurement is of particulate volume by Coulter Counter* or other electronic particle counting device. Particle number concentration NL -~, is also used when data are obtained with particle-counting instruments. Because of the steep increase in N with decreasing particle size, these units are highly sensitive to the limits of the size window being measured. Figure 1 gives the p.gL -1
measurements
Because light attenuates as a function of e ~", where ~ is the attenuation coefficient and r is a distance, the relation between light transmission, T, and ~ is
8
--
IM
~
-b.
A
I O
O
..4
-
\
r
~O
Attenuation
73
O
ppb by volume (assuming p= 2gcm -3) FIG. 1. Volume or mass concentration versus number concentration for various diameter spherical particles. * Any use of trade names is for descriptive purposes only and does not imply endorsement by the USGS.
74
S.L.
Eittreim
relationship between volume or weight concentration and number concentration for different assumed spherical particle sizes. Note that for a given constant mass or volume concentration, the number concentration will change by four orders of magnitude for a shift from 20 to 1 ~m in particle size. An adequate description of oceanic SPM usually requires information on its size spectra. The settling velocity and hence residence time of particles is determined largely by their size; thus to use SPM as a quantitative tracer of water motions requires knowledge of the size characteristics of the SPM. Studies of the size spectra naturally concern the state of aggregation of particles. Collection of particles into first, second, and higher order aggregates is common in the ocean, and unless the aggregation state is known, predictions regarding settling velocities as well as size definition of SPM with respect to bottom sediments may be incorrect (Krone 1962; Kranck 1973; McCave, 1983). The composition of SPM in ocean water can be examined by microscopic, chemical, X-ray or electron microprobe techniques (Sackett 1978). Automated counting methods coupled with a scanning electron microscope now allow extremely small samples to be dealt with quantitatively (Bishop & Biscaye 1982). Although centrifugation was a popular means of extracting particles for study in the past (Jacobs & Ewing 1969; Lisitzin 1972), since the advent of membrane filters with uniform and small pore size, most workers now use filtration as a means of collection. In situ filtration methods have had some success where large water volumes were necessary due to very low concentrations or when the desired analysis required a large sample (Bishop et al. 1977; Peterson 1977). The GEOSECS program produced a systematic sampling and analysis method for SPM which might be regarded as a model for future work (Brewer et al. 1976). Systematic procedures were designed to avoid contamination. Duplicate sampling and analyses were done by separate institutions (Lamont and Woods Hole) to assess the accuracy and completeness of the extraction techniques. The fact that these duplicate analyses showed significant differences indicates their sensitivity to sampling technique and the importance of procedure to good results (Brewer et al. 1976): Most sampling for ocean SPM is now done with open-ended polyvinylchloride bottles, which offer the least resistance to water passing through the bottles as the sampling apparatus is lowered. Where concentrations are less than 50 itgL -~, 20-30-1itre bottles are normally used to obtain
enough sample for handy analysis. 'Rosette' samplers, actuated by command from the surface, are the most convenient for accurately locating samples. Messenger-tripped bottles are commonly used and give good results as long as sufficient concern is given for possible contamination by the wire. The awareness of possible contamination, or the opposite, incomplete extraction, cannot be overstressed. It should be assumed that each sampling bottle has its own "signature' of contamination, and this should be measured by rotating bottles and taking replicate samples to determine what its 'blank' contribution is before accepting data from it. Incomplete extraction, due to particles settling out below the bottle spigot or sticking to bottle walls, has been documented (Gardner 1977) and can be minimized by agitation of the sample bottle and by filtering as soon as possible after bringing the sample aboard. When working with SPM concentrations in the tens of pgL-N which is the most common value for open-ocean SPM below the euphotic zone, serious errors can be introduced by laboratory contamination. Wherever filters are exposed, a clean air environment produced by an air filtration hood system or 'clean room' is desirable. Residual salts on filters are an equally serious contaminant which can increase filter weight concentrations by a factor of 10 or more. The GEOSECS program procedures call for washing filters with 7 to 10 aliquots of distilled, filtered water. Oceanic SPM studies prior to 1970 did not generally follow such rigorous procedures to avoid contamination and to completely rinse out salts, and generally reported higher concentration values (Harris 1972; Sackett 1978). Polycarbonate micro-filters, which are made by etching out neutron bombardment holes made in plastic sheet, became available about 10 years ago. These filters, as discussed by Biscaye & Eittreim (1974), make particularly good extraction devices for low concentrations of fine SPM because of (1) low or negligible weight loss of filter material during filtration and (2) a lower tare weight and less change with humidity than other filter materials. Brewer et al. (1976) estimated a sample weight precision of ___5 pg, based on repeated weighings, which translates to _+0.5 /tgk -~ for a 10-L sample. Standard deviation of SPM concentrations from replicate samples was 5.2 pgL -~ (Brewer et al. 1976) which gives a maximum error of the overall sampling and weighing techniques; a maximum because of the possibility that real variations or patchiness in SPM concentrations in the water contributed to the variations. Microscopic particle counting and identifica-
The stud), of deep-sea suspended particulate matter tion offers another means of analysis of filtered SPM. Before the advent of automated particle counting and scanning electron microscope methods of particle analysis (Lambert et al. 1981: Bishop & Biscaye 1982), such work was extremely slow and tedious but is now a useful technique, although in its infancy. Using an energy-dispersive X-ray spectrometer coupled to a scanning electron microscope, Lambert et al. (1981) and Bishop & Biscaye (1982) obtained information on A1, Si, P, S, K, Ca, Cr, and Fe which can be roughly translated to relative percentages of clay mineral species and biogenic versus mineralogenic percentages. Harris (1972) obtained size spectra data down to particle sizes of 0.02 /xm using scanning electron micrographs and manual measurements of particles. Number versus particle size can most efficiently be obtained with Coulter Counters or other electronic sizing devices, which measure volumes of particles as they pass through an orifice that is part of a monitored electrical path. These volumes are converted to diameters of equivalent spheres. In order to cover the size spectrum adequately, several different apertures must be used and the data overlapped, and in order to obtain data on particles smaller than 1 /ma, methods other than the Coulter Counter must be used. Bader (1970), Carder et al. (1971), and Sheldon et al. (1972) and others have successfully used Coulter Counters at sea. Measures must be taken to minimize mechanical and electrical noise which the instrument is quite sensitive to. The volume concentrations measured, when combined with concentrations by weight from filtration data, can be used to calculate apparent SPM bulk density (McCave 1983). As a general rule, oceanic as well as atmospheric particle sizes are distributed according to a power-law distribution, N = A D .... , where N is the number of particles greater than diameter D and A is a constant (Junge 1963; Bader 1970). A slope, m, of 3 implies equal volumes of particles in all size classes, a circumstance which Sheldon et al. (1972) found to hold generally for biogenic ocean particles ranging in size from bacteria to whales. For inorganic as well as organic SPM such a power law has been found to hold true in general, and attention is usually focused on cases of where and why the slope m deviates from a value of 3. Thus the emphasis in recent years has been on the description of SPM size spectra in terms of m and its variations up and down the water column and up and down the size spectrum. Figure 2 shows the correlations obtained between various kinds of optical measurements and SPM concentrations determined either by
75
filtered weight or by Coulter Counter volume. For low concentrations (e.g. < 50/tgL -~) it has been difficult to obtain a good empirical calibration with a low scatter of data points. There are many reasons for this. Primary among them is the fact that scattering as well as attenuation is sensitive to total particulate surface area rather than mass or volume. Thus, changes in the particle size spectra will change the optical effect per unit volume of particulates. Colour and index of refraction differences in the particle population will also have an effect, especially on attenuation, although this effect is less than that caused by size differences (Jerlov 1976). Finally, scatter introduced by sampling errors is probably roughly + 5 /~gL -t even for the most careful work. In any case, demonstrations of empirical calibrations such as those in Fig. 2 are necessary for deriving SPM concentrations from optical measures, and the scatter in these relationships should be kept in mind in the subsequent data analysis.
Sediment trap measurements The residence times of particles in the water must be considered when relating SPM trapped in sampling bottles to sedimentation on the bottom (McCave 1975). Bottles have a bias towards collection of particles with long residence time, since these particles occur in greater proportion in water than in the sediment. Sediment traps, which capture the downward flux of particles, offer another means of collecting oceanic SPM. Sediment traps collect particles which may, although contributing significantly to sediment, have a very short residence time in the water column. Hence, bottle samples are biased toward the small, slowly settling population of particles, whereas traps are biased toward the larger, high-flux population. Dymond et al. (1981) compared data from four different types of sediment traps deployed in Santa Barbara Basin. The maximum difference in flux measured among the four traps was a factor of 2, and the average flux measured over the 48-day period was in fair agreement with the longer term sedimentation rate in the basin determined by 21~ profiles in the sediment. Bruland et al. (1981) determined a ratio of SPM flux to bottom sediment flux of 0.93 for sediment traps set in Santa Barbara basin. It should be kept in mind that the fine SPM in the water column is scavenged to some unknown extent by the faster-falling large particles, the faecal pellets and particles which Honjo et al. (1982) called 'amorphous organic aggregates'. Honjo et al. (1982) demonstrated a consistent increase in the flux of lithogenic particles with
S.L. Eittreim
76 1000
(a) 9
600
O)
(c)
100
:k
60
400
d
: - .,,=|:..
t,-
~
80
(b)
-':~
9
40
".:i?i': " "
lO
200
Q.
20
1
I 10
0
I 100
0
I 600
Scattering ratio
I 400
I 200
I 0
-
I 2
0 0
Scattering pulse delay time
I 4
I 6
I 8
Relative scattering
(ms)
(d) 1200
L.j 8oo
(e)
9 " " : ...
1000 -
v
---.
"
9
6 c 0
. ; 9 ~149
(9
o
400
-
9. : :
9
: ""
9
9149149
..~.!~.~
""
o 9 ~~
9 ~ '-'.~ -
lOO
-.
9
9 o ,It,
0 0
Jil'l
I
0.5
1.0
I
I
1.5 2.0 Attenuation coeff., ( x ( m - 1 )
100
I
I
I
I
I
80
60
40
20
0
% Transmission
FIG. 2. In situ calibrations of various SPM-sensing instruments: (a) Lamont--Thorndike nephelometer (Biscaye & Eittreim 1974) and (b) Woods Hole---GEOSECS nephelometer (Meade et al. 1975) are both foward scattering meters. (c) University of Washington b-meter (Sternberg et al. 1974). (d) and (e) transmissometer calibrations of McCave (in press) and Drake (1974), respectively. Note the differences in scales of SPM concentrations. For (d), conversion of SPM concentration from ppb to/~g L- ] is based on a particle density of 2.0 g cm -3. depth that can best be explained by progressive scavenging during descent of large organic aggregates. The 'faecal pellet express' may be the primary means by which the fine SPM reaches the sea floor (Honjo & Roman 1978; Honjo 1982), and sediment traps offer the principal means of studying this process. Sediment trap studies are opening up a new field which directly addresses the question of residence time of particles in the water column. The VERTEX (Vertical transport and exchange of materials) program, for example (Martin & Knauer 1982), is an interdisciplinary study which attempts to define the downward flux
of particles and their chemical and biological changes during descent by sampling the upper and mid-water column with large sediment-trap arrays. Many questions still have to be answered regarding trapping efficiencies and the meaning of what is caught in the traps (Gardner 1980a,b). No trap deployments have yet attempted to catch non-vertical flux, for example. In a near-bottom nepheloid layer, where eddy diffusion in all directions may be important, the three-dimensional flux maybe important to understand. Although a trap may establish a downward net vertical flux, the true gross flux is unknown, since
The study of deep-sea suspended particulate matter it is possible, for example, that more (unsampled) particles moved up during the sampling period then moved down. To get at these kinds of questions, more work needs to be done in the future and the reader is referred to the above references as starting points.
Composition and particle size of ocean S P M Sackett (1978) reviewed the state of knowledge on the geochemistry, composition and concentrations of oceanic SPM, radionuclide estimates of residence times, and other aspects of SPM work which I will not repeat here. A few recent developments are worth mentioning, however. Automated particle counting and analysis techniques (Lambert et al. 1981, Bishop & Biscaye 1982) will soon produce more knowledge of SPM compositions. The two sources of ocean SPM are terrigenous detritus, transported via the atmosphere (Windora 1975), surface currents, mid-water turbid plumes emanating from the shelf break (Drake 1971) or bottom turbid flows, and biogenic detritus from the photic zone. In the mid-water regions, greater proportions of biogenic opal and Ca have been found (Bishop & Biscaye 1982), whereas bottom nepheloid-layers are known to be enriched in clay-mineral elements such as Fe (Betzer & Prison 1971) and Al (Feely et al. 1971) relative to the rest of the water column. The limited information that exists regarding particle size spectra of SPM suggests that aggregation state is important and that a bimodal distribution is often caused by the presence of a large aggregate mode in addition to the smaller primary particle mode. Second, third, and fourthorder aggregates are found, sometimes biologically bound, sometimes simply electrostatically bound (Kranck 1973; McCave, 1983). In general, the number vs. diameter distributions are loglinear, often with a break in slope at about 4/am, indicating a peaked volume spectrum in the vicinity of 4 ~tm (Harris 1972; McCave 1975).
Regional and temporal variations in the oceans For a compilation of worldwide SPM concentrations, the reader is referred to the GEOSECS volumes (Bainbridge et al. 1976, 1977) and to Brewer et al. (1976) for discussion of the Atlantic GEOSECS data. Water of lowest SPM concentration usually occurs in the middle of the water column (Fig. 3). The surface mixed layer and
77
euphotic zone supplies the upper water column with airborne and water-borne clay and silt and biogenic debris, all of which settle and diminish their concentration downward by aggregation (which increases their settling velocity), dissolution, and grazing by organisms (Lal & Lehrman 1973). The near-bottom part of the water column is supplied with resuspended fine sediment from the sea floor. The mid-water clearest zone is the region of overlap between the downward decrease from the surface and the upward decrease from the bottom and is usually a kilometre or more from these two sources. In areas where bottom resuspension is minimal the mid-water clear zone is very deep, or even extends to the bottom, such as in the central gyres of the north and south Pacific (Ewing & Connary 1970). The lowest concentrations in the mid-water zones occur in the low productivity areas of the temperate latitudes away from the continental margin sources of sediment or productive zones associated with divergences (Eittreim et al. 1976). The exact concentrations in these mid-water clear zones are difficult to establish because values are near the level of resolution allowed by sampling methods, about 5 #gL -1 (Biscaye & Eittreim 1974; Brewer et al. 1976). Figure 3 shows a range of turbidity profile types from low-energy, Iow-SPM concentration areas to high-energy, high-concentration areas. Note the broad clear zone in the mid-water region and the occurrence of bottom boundary layers capped by high gradients in the high-energy areas. Mid-water stratification is commonly associated either with advecting water masses of turbidity higher or lower than the surrounding water, such as the Mediterranean outflow, or due to a concentration of SPM at high density gradients caused by the reduction in mean settling speed of particles at the high-gradient regions (Jerlov 1959; Drake 1971; Pak et al. 1980). The most common high-turbidity feature of the water column is that associated with the seasonal thermocline. As Jerlov (1959) and Pak et al. (1980) note, and as is seen in many of the Lamont nephelometer profiles which were taken at night and thus avoid daylight interference (Eittreim 1970), the thermocline is commonly associated with a strong turbidity maximum which is presumed to be largely biogenic. Particle maxima can also be produced in theory by the effect of dissolution on settling speed of particles in the presence of slow upwelling (Lal & Lehrman 1973; Brun-Cottan 1976), but this effect has not been documented by measurements. The increase in turbidity commonly observed adjacent to bottom, termed a 'nepheloid-layer' by Ewing & Thorndike (1965), is observed more
78
S.L.
PACIFIC
~
J
Eittreim
7
o. S
ATLANTIC
J
U,o~ 4
I
I
I
o c .E s
15
112 38
/'xg L-1
? ....
183
~ 293
1
Log 6 -
( s c a t t e r i n g ratio)
}
138
FIG. 3. Selected nephelometer profiles (Eittreim & Ewing 1972; Sullivan et al. 1973, 1975) showing regional variations in profile shapes from a low-energy area (north and south-eastern Pacific) on the left to high-energy area (western Atlantic basins) on the right. SPM concentrations in pg L - l are indicated for bottom 100 m of water column, mostly from data of nearby GEOSECS stations (Bainbridge et al. 1976, 1977). Concentration values for the two profiles on the far right from Biscaye & Eittreim (1974) and McCave (in press), respectively. Numbers at the top of the profiles refer to station numbers of cruises Conrad 15 and 16 and Vema 23. Note that the units are on a log scale which suppresses the high- and enhances the low-amplitude features.
often than not in the ocean. In areas of extremely weak bottom water flow (e.g. < 5 cm s -1) this layer is sometimes not observed, the clear intermediate water extending to the bottom. But in most areas, some effects of bottom water interaction with sediment results in at least a minor increase in turbidity even if only a few p g L - 1 an amount difficult to document by sampling (Fig. 2). The exponential increases in SPM usually observed towards bottom often describe a series of straight-line segments on log-linear plots (Eittreim 1970; Sullivan et al. 1973, 1975), suggesting a hierarchy of layers on top of one another, with different exponents governing the diffusive particle transport within each layer. One explanation proposed by Armi & D'Asaro (1980) involves detachment from the bottom of turbulent and turbid benthic layers. Lateral migration of these detached layers along isopycnal surfaces, perhaps from relative topographic highs, into the interior of the water column may provide a mechanism for diffusing the SPM away from the bottom without invoking the unrealistically high coeffi-
cients of vertical eddy diffusion computed by Eittreim & Ewing (1972). Biscaye & Eittreim (1977) examined advected SPM fluxes for the bottom waters of the western Atlantic basins. The western Atlantic has a generally continuous bottom nepheloid-layer associated with the western boundary currents of polar bottom waters and a large data base of Lamont nephelometer stations exists in this area. Fluxes ranging from about 1 to 8 x 106 tons yr -1 were found at six section locations where bottom water transports have been measured. These fluxes are the same order of magnitude as the non-carbonate sediment deposition in the basins "downstream' of these sections (based on longterm sedimentation rates), suggesting that such fluxes are a significant factor in deposition of fines in these Western Atlantic basins. These fluxes are small, however, in comparison to loads carried by major rivers near their mouths such as the Amazon (900x 10 6 tons yr - l ) (Milliman & Meade 1983). However, most of the river particulate load, as is pointed out by Milliman & Meade, is deposited on subareal deltas, in estuaries, or on
The study of deep-sea suspended particulate matter the inner shelf. Data that will give average terrigenous sediment output beyond the shelf break to the deep sea is sorely needed. Until estimates, even good to an order of magnitude, can be made of offshelf terrigenous transport, attempts at making deep-sea sediment budgets in such areas as the western Atlantic basins, where turbid bottom water transport is significant, will be precluded. Another major unknown factor in constructing deep-sea sediment budgets is the residence times of nepheloid-layer particles. Estimates are needed of the number of deposition/resuspension cycles the average particle experiences before final deposition. If these estimates are not made, the link cannot be made between what is observed in the water column today and what is deposited on the sea floor. This question is presently being addressed by sediment trap work mentioned above. Extreme temporal variations such as those shown in Fig. 4 and those found in investigations in the HEBBLE area (40~ 62~ Shor et al., this volume) are probably confined to high energy regions ( > 10 cms -1 flow). Although seasonal fluctuations in biogenic production should be
""';
(a)
~5 far)
""
Po"~". \"...
:
m
I
_
. ....
I'
-(b) I _o q \
M
," 1
...
\:
.:
\
"
,
\
I oNo.,s, .on"]
400
A
200 ~
..
~oo~ ~ --9
(c)
/,-,
B60[:, ..~...o." ....~
~'~401," , .6
~
,'
\1
/
~ 20 0
5
I0
15
20
Days
FIG. 4. Examples of temporal variations in SPM observed in an active or high-energy area at two stations 110 km apart on the flank of the Blake-Bahama Outer Ridge (modified after Biscaye & Eittreim 1974). (a) and (b) are taken from bottle-sampled and filtered SPM. m.A.B, means metres above bottom. (c) Data from Lamont-Thorndike nephelometer bottom-most data point, approximately 10 metres above bottom.
79
expected as well as seasonal changes in river and airborne clay and silt debris, no significant seasonal changes have yet been detected in midocean areas. Perhaps this is because these changes are either too small to detect, or biogenic particles settle slowly enough so that the changes are smoothed out vertically in the water column. On the Lamont nephelometer stations repeatedly occupied, only the near-bottom part of the water column in high-velocity regions showed significant variations (Eittreim et al. 1976). Nephelometer profiles taken repeatedly over a 19-day period on the Blake-Bahama Outer Ridge flank showed a striking agreement in the region of the water column shallower than 3500 m, above the benthic boundary layer (Biscaye & Eittreim 1974). Studies in the HEBBLE area have shown that high concentrations ( > 1000 #g L -], Amos 1979; Biscaye et al. 1980) and short-term particle resuspension events ( < 1 day, Pak et al. 1983) do occur in the deep sea in restricted locations, such as along the narrow 'filament' of Western Boundary Undercurrent on the continental rise of the north-west Atlantic. To understand sedimentation of fines in the ocean, these local event-dominated processes must be more clearly understood. There is evidence that frequent high turbidity events can be caused by both downslope turbidity currents (Amos & Gerard 1979) and resuspension by contour-following bottom currents (Biscaye et al. 1980). The broad regional distributions shown in the GEOSECS and Lamont data reflect water mass interactions and mixing processes on a larger and slower scale than the processes observed in the HEBBLE area. An integration of these two different scale phenomena must ultimately be made.
Summary A few directions for future studies of fine SPM stand out as particularly fruitful at this time. Further work on aggregation and settling dynamics is important. Such studies may reveal new methods for characterizing resuspended versus 'fresh' SPM, i.e. that which has not yet experienced bottom contact. If the sea floor acts as an efficient aggregator as one might expect, then resuspended sediments might be predominantly in the form of aggregates, although the preliminary work of McCave (1983) does not suggest this. More studies of elemental characterization of SPM at all levels in the water column and comparison with adjacent bottom sediments may yield new insights into sources and pathways. Detailed high-resolution profiles in the bottom
8o
S.L. Eittreim
b o u n d a r y - l a y e r m a y yield i m p o r t a n t i n f o r m a t i o n on the nature of b o u n d a r y - l a y e r turbulence and h o w sediment is carried in suspension. Studies of particle flux m e a s u r e m e n t s with sediment traps will help our u n d e r s t a n d i n g of particle residence times and mechanisms of d o w n w a r d transport. In order to construct sediment budgets for deep-sea areas where resuspension of terrigenous sediment is c o m m o n , better data on offshelf terrigenous transport is needed to improve the u n d e r s t a n d i n g of the possible contribution from steady or aperiodic turbidity flows. Studies focused on particular pathways of S P M will be important. For example, what roles do canyons play, if any, in the offshelf transport of fine sediment? New types of deep-sea bedforms created by
u n k n o w n mechanisms related to steady and perhaps aperiodic b o t t o m flows have recently been found (Hollister et al. 1974; D a m u t h 1980; F l o o d 1981). These features are likely related to suspended loads carried in benthic b o u n d a r y layers, and their study by methods of SPM m e a s u r e m e n t with improved resolution and precision will be necessary. Needless to say, rigorous adherence to careful procedures in S P M w o r k to avoid c o n t a m i n a t i o n and to completely extract particles will help advance the state of the art. ACKNOWLEDGMENTS: D a v i d E. D r a k e and D a v i d Z. Piper of the US Geological Survey reviewed the manuscript and offered m a n y helpful suggestions.
References AMOS, A.F. & GERARD,R.D. 1979. Anomalous bottom water south of the Grand Banks suggests turbidity current activity. Science, 203, 894-97. ARMI, L. & D'ASARO, E. 1980. Flow structures of the benthic ocean. J. geophys. Res., 85, 469-84. AUSTIN, R. 1974. Instrumentation used in turbidity measurement. Proc., NOIC Turbidity Workshop. May 6-8. 1974, National Oceanographic Instrumentation Center, Washington, DC, 45-74. BADER, H. 1970. The hyperbolic distribution of particle sizes. J. geophys. Res., 75, 2822-30. BAINBRIDGE, A.E., BISCAYE, P.E., BROECKER, W.S., HOROWITZ,R.M., MATHIEU,G., SARMIENTO,J.L. & SPENCER, D. 1976. GEOSECS Atlantic bottom hydrography, Radon and suspended particulate atlas. GEOSECS Operations Group, Scripps hlstitution of Oceanography, Internal Report. , , , , , - -1977. GEOSECS Pacific bottom hydrography, Radon and suspended particulate atlas. GEOSECS Operations Group, Scripps Institution of Oceanography, Internal Report. BEARDSLEY, G.F., JR., PAR, H., CARDER, K. & LUNDGREN, B. 1970. Light-scattering and suspended particles in the eastern equatorial Pacific Ocean. J. geophys. Res., 75, 2837-48. BETZER, P.R. & PILSON, M.E.Q. 1971. Particulate iron and the nepheloid layer in the western North Atlantic, Caribbean and Gulf of Mexico. Deep Sea Res., 18, 753-61. BISCAYE, P.E. & EITTREIM, S. 1974. Variations in benthic boundary layer phenomena: nepheloid layer in the North American Basin. In: Gibbs, R. (ed.), Suspended Solids in Water. Plenum Press, New York. 227-60. -& -1977. Suspended particulate loads and transports in the nepheloid layer of the abyssal Atlantic Ocean. Marine. Geol., 23, 155-72. -& ZANEVELD, J.R.V., PAK, H. & TUCHOLKE, B. 1980. Nephels! Have we got Nephels! (abstract) LOS Trans. Am. geophys. Un., 61, 1014.
BISHOP,J.K.B. & BISCAYE,P.E. 1982. Chemical characterization of individual particles from the nepheloid layer in the Atlantic Ocean. Earth planet. Sci. Lett., 58, 265-75. --, EDMOND,J.M., KETTEN, D.R., BACON, M.P. & SILKER, W.B. 1977. The chemistry, geology and vertical flux of particulate matter from the upper 400 m of the equatorial Atlantic Ocean. Deep Sea Res., 24, 511-48. BREWER, P.G., SPENCER,D.W., BISCAYE,P.E., HANLEY, A., SACHS,P.L., SMITH, L.L., KADAR,S. & FREDERICKS, J. 1976. The distribution of particulate matter in the Atlantic Ocean. Earth planet. Sci. Lett., 32, 393-402. BRULAND, K.W., FRANKS, R.P., LANDING, W.M. 8~ SOUTAR, A. 1981. Southern California inner basin sediment trap calibration. Earth planet. Sci. Lett., 53, 400-8. BRUN-COTTAN, J. 1976. Stokes settling and dissolution rate model for marine particles as a function of size distribution. J. geophys. Res., 81, 1601-06. CARDER, K.L., BEARDSLEY,G.F., JR. & PAK, H. 1971. Particle size distribution in the eastern equatorial Pacific. J. geophys. Res., 76, 5070-7. DAMUTH, J.E. 1980. Use of high-frequency (3.5-12 kHz) echograms in the study of near-bottom sedimentation processes in the deep-sea: a review. Marine Geol., 38, 51 75. DRAKE, D.E. 1971. Suspended sediment and thermal stratification in Santa Barbara Channel, California. Deep Sea Res., 18, 763-9. 1974. Distribution and transport of suspended particulate matter in submarine canyons off southern California. In: Gibbs, R. (ed.), Suspended Solids in Water. Plenum Press, New York. 133-53. DYMOND, J., FISCHER, K., CLAUSEN, M., COBLER, R., GARDBER, W., RICHARDSON, M.J., BERGER, W., SOUTAR, A. & DUNBAR, R. 1981. Sediment trap intercornparison study in the Santa Barbara Basin. Earth planet. Sci. Lett., 53, 409-18. EITTREIM, S. 1970. Suspended Particulate Matter in the
The study of deep-sea suspended particulate matter Deep Water of the North American Basin. Ph.D. Thesis, Columbia Univ., New York (available in microfilm). 166 pp. -& EWING, M. 1972. Suspended particulate matter in deep waters of the North American Basin. In: Gordon, A. (ed.), Studies in Physical Oceanography, A Tribute to George Wust on his 80th Birthday, 2. Gordon & Breach, New York. 123-67. , THORNDIKE, E.M. & SULLIVAN, L. 1976. Turbidity distribution in the Atlantic Ocean. Deep Sea Res., 23, 1115-27. EWING, M. & CONNARY,S. 1970. Nepheloid layer in the North Pacific. In: Hays, J. (ed.)., Geol. Soc. Am. Mere., 126, 41-82. & THORNDIKE, E.M. 1965. Suspended matter in deep ocean water. Science, 147, 1291-94. FELLY, R.A., SACKETT, W.M. & HARRIS, J.E. 1971. Distribution of particulate aluminium in the Gulf of Mexico. J. geophys. Res., 76, 5893-902. FLOOD, R.D. 1981. Longitudinal triangular ripples in the Blake-Bahama Basin. Marine Geol., 39, M 13-M20. GARDNER, W.D. 1977. Incomplete extraction of rapidly settling particles from water samplers. Limnol. Oceanogr., 22, 764. 1980a. Sediment trap dynamics and calibration: a laboratory evaluation. J. mar. Res., 38, 17-39. 1980b. Field assessment of sediment traps. J. mar. Res., 38, 41-52. GIBBS, R.J., (ed.) 1974. Suspended Solids in Water. Plenum Press, New York. 320 pp. HARRIS, J.E. 1972. Characterization of suspended matter in the Gulf of Mexico - 1. Spatial distribution of suspended matter. Deep Sea Res., 19, 719-26. HEEZEN, B.C. & HOLL1STER,C. 1964. Deep-sea current evidence from abyssal sediments. Marine Geol., l, 141-74. HOLLISTER, C.D., FLOOD, R.D., JOHNSON~D.A., LONSDALE, P. & SOUTHARD, J.B. 1974. Abyssal furrows and hyperbolic echo traces on the Bahama Outer Ridge. Geology, 2, 395-400. HONJO, S. 1982. Seasonality and interaction of biogenic and lithogenic particulate flux at the Panama Basin. Science, 218, 883-4. -& ROMAN, M.R. 1978. Marine copepod faecal pellets; production, preservation and sedimentation. J. mar. Res., 36, 45-57. --, MANGANINI, S.J. & POPPE, L.J. 1982. Sedimentation of lithogenic particles in the deep ocean. Marine Geol., 50, 199-219. JACOBS, M.B. & EWING, M. 1969. Suspended particulate matter: concentrations in the major oceans. Science, 163, 380-3. JACOBS, S.S., BAUER, E.B., BRUCHHAUSEN,P.M., GORDON, A.L., ROOT, T.F. & ROSSELOT, F.L. 1974. Eltanin Reports; cruises 47-50, 1971; 52-55, 1972: Hydrographic stations, bottom photographs, current measurements, nephelometer profiles. Tech. Rept. Lamont-Doherty Geological ObservatoJ3, CU-2-74. , GEORGI, D.T. & PATLA, S.M. 1980. Conrad 17; Hydrographic stations, sea floor photographs, nephelometer profiles in the southwest Indian-
-
-
-
-
-
81
Antarctic Ocean. Tech. Rept. Lamont Doherty Geological Observatory CU-I-80- TR1. JERLOV, N.G. 1953. Particle distribution in the ocean. Rep. Swedish Deep-Sea Expedition, 3, 73-97. 1959. Maxima in the vertical distribution of particles in the sea. Deep Sea Res., 5, 178-84. -1976. Marine Optics. Elsevier, Amsterdam. 231 PP. JUNGE, C.E. 1963. Air Chemistry and Radioactivity. Academic Press, New York. 382 pp. KRANCK, K. 1973. Flocculation of suspended sediment in the sea. Nature, 246, 348-50. KRONE, R.B. 1962. Flume studies of the transport of sediment in esturarial shoaling processes. Hydraulic Engineering Laboratory and Sanitary Engineering Research Laboratory, Univ. of California, Berkeley. 110 pp. LAL, D. & LERMAN,A. 1973. Dissolution and behavior of particulate biogenic matter in the ocean; some theoretical considerations. J. geophys. Res., 78, 7100-11. LAMBERT, C.E., JEHANNO, C., SILVERBERG, N., BRUNCOTTON, J.C. & CHESSELET, R. 1981. Log-normal distribution of suspended particles in the open ocean. J. mar. Res., 39, 77-98. LISITZIN,A.P. 1972. Sedimentation in the World Ocean. Soc. Econ. PaleD. Min. Spec. Publ.., 17, 225 pp. MARTIN, J. • KNAUER, G.A. 1982. A comparison of particulate reactivities of Ag, Cd, Co, Cu, Fe, Mn, Mo, Ni, V and Sn observed during Vertex II. (abstract) LOS Trans. Am. geophys. Un., 63, 960. MATLACK,D.E. 1972. Deep ocean optical measurement (DOOM) report; Bahama Channels and northwestern Atlantic Ocean. Naval Ordnance Lab., Maryland, NOLTR-72-284, 1-33. MCCAVE, I.N. 1975. Vertical flux of particles in the ocean. Deep Sea Res., 22, 491-502. 1983. Particle size spectra, behavior and origin of nepheloid layers over the Nova Scotian Continental Rise. J. geophys. Res. 7647-66. MEADE, R.H., SACHS, P.L., MANHEIM, F.T., HATH-
-
-
8 8 ,
AWAY, J.C. & SPENCER, D.W. 1975. Sources of suspended matter in waters of the Middle Atlantic Bight. J. sed. Petrol., 45, (Appendix) 186-8. MILLIMAN, J.D. & MEADE, R.H. 1983. Worldwide delivery of river sediments to the oceans. J. Geol., 91, 1-21. ORR, M.H. & HESS, F.R. 1978. Remote acoustic monitoring of natural suspensate distributions, active suspensate resuspension, and slope shelf water intrusions. J. geophys. Res., 83, 4062-8. PAK, H., ZANEVELD, R.V. & KITCHEN, J. 1980. Intermediate nepheloid layers observed off Oregon and Washington. Jour. geophys. Res., 85, 6697-708. -& ZANEVELD,R.V. 1983. Temporal variations of beam coefficient on the continental rise off Nova Scotia. J. geophys. Res., 88, 4427-32. PETERSON, R.E. 1977. A Study of Suspended Particulate
Matter." Arctic Ocean and Northern Oregon Continental Shelf Ph.D. Thesis, Oregon State Univ., Corvallis. SACKETT, W.M. 1978. Suspended matter in sea water.
82
S.L. Eittreim
In: Riley, J.P. & Chester, R. (eds.), Chemical Oceanography. Academic Press, London. 127-72. SHELDON, R.W., PRAKASH, A. & SUTCL1FFE,W.H. JR. 1972. The size distribution of particles in the ocean. Limnol. Oceanogr., 17, 327-40. SIMPSON, W.R. 1982. Particulate matter in the oceans - sampling methods concentration, size distribution and particle dynamics. Oceanogr. Mar. Biol. Ann. Ree., 20, 119-72. SMITH, C.L., KADAR, S. • FREDERICKS, J. 1976. The distribution of particulate matter in the Atlantic Ocean. Earth planet. Sci. Lett., 32, 393-402. STERNBERG, R.W., BAKER, E.T., MCMANUS, D.A., SMITH, S. & MORRISON, D.R. 1974. An integrating nephelometer for measuring particle concentrations in the deep sea. Deep Sea Res., 21,887-92.
SULLIVAN, L., THORNDIKE, E., EWING, M. & EITTREIM, S. 1973. Nephelometer measurements, Hach turbidimeter measurements and bottom photographs from Conrad cruise 15. Tech. Rept. Lamont-Doherty Geological Observatory 8-CU-8- 73. --, THORNDIKE, E. & EITTREIM, S. 1975. Nephelometer measurements and bottom photographs from Conrad cruise 16. Tech. Rept. Lamont-Doherty Geological Ohsert'atory CU- 1 I- 75. THORNDIKE, E.M. 1975. A deep-sea photographic nephelometer. Ocean Engrg., 3, 1-15. WINDOM, H.L., 1975. Eolian contribution to marine sediment. J. sed. Petrol., 45, 2-8.
S.L. EITTREIM, US Geological Survey, Menlo Park, California 94025 USA.
Grain-size characteristics of turbidites Kate Kranck S U M M A R Y: Detailed sampling using a very small sample size and grain-size analysis with a Coulter Counter of three fine-grained turbidites enabled a distinction to be made between the well-sorted single-grain Stokes'-deposited and unsorted "whole suspension' floc-deposited grain-size populations. The results indicate that each turbidite is a continuous sequence deposited from the same source suspension. Particles settle to the bottom as flocculated masses. Initially flocs are broken up by near-bottom shear forces and only the coarsest silt and sand remains on the bed. The remaining mud forms a temporary mud suspension near the bottom which reflocculates and intermittently deposits at some critical concentration producing mud interlayers between silt laminae. Decrease in current velocity eventually allows simultaneous deposition of single grains and flocs with the latter becoming progressively more abundant resulting in formation of graded beds. Eventually, all deposition occurs in the form of mud flocs and a massive sediment results. The study essentially confirms the basic mechanism of the fine-grained turbidite deposition model proposed by Stow & Bowen (1978). The abundant presence of turbidites in the geological column is a result of the fact that transport of continental material to the ocean depressions is not a continuous one-step process. Terrestrial erosional products from river and aeolian transport initially sediment out relatively close to their source. Subsequent and intermittent resedimentation of this material downslope provides a large amount of clastic material to the deep sea. As a result approximately 50-70~o of oceanic sediment consists of continental clastics. Many of these are turbidites whose sole marks, graded bedding, mud-silt lamination and other characteristic features identify them as deposits from high turbidity subaqueous flows which travel distances on the order of tens to hundreds of kilometres. The 90% turbidite component of abyssal sediments attests to the importance of this process and justifies the attention given to turbidite sedimentation in the geological literature. Starting with the early work of Kuenen (1951, 1966a,b), studies of ancient, modern and experimental turbidite beds have provided a basic knowledge of the origin and physical characteristics of these deposits. The work of Bouma (1962) has proved especially useful by providing an idealized sequence which has formed a basis for most subsequent regional descriptions of turbidite deposits as well as for discussions of the hydrodynamic process creating turbidites (for reviews see, Middleton 1970; Allen 1982). It is noteworthy that almost all of this previous work has been based on descriptions of gross structural or stratigraphic features which can be identified directly in outcrops or sediment cores. These features include variability in apparent grain-size such as graded bedding and silt-mud laminae, but they can all be identified without the
aid of grain-size analysis. For example, the work of Harms & Fahnestock (1965), on the basis of similarities in the layering and surficial bedforms, compares Bouma's divisions with laboratory and river sediment deposited under different decreasing flow conditions. Some size distributions are presented in this and other work but more to give a general idea of grain-size range than as a diagnostic tool. Exceptions are Scheidegger & Potter (1965) and Potter & Scheidegger's (1966) studies of grain-size and vertical variability, and Middleton's (1967) distinction of distribution grading and coarse-tail grading. Middleton found that coarse-tail grading characterizes turbidites deposited from high concentration surges and distribution grading deposits from low concentration suspensions, interpretations which could not have been made from visual inspection alone. The value of grain-size analysis is excellently demonstrated by Stow & Bowen's recent use of detailed size analysis to postulate a model for the origin and grain-size distribution of the silt-mud laminae in turbidites from the Scotian margin (Stow & Bowen 1978, 1980). They postulate a mechanism of shear sorting whereby flocculated sediment arriving at the bottom-water interface is broken up by shear forces. Initially only the largest silt size of particles settle through the boundary layer and are deposited. The fine mud mud portion forms a layer of suspended mud of increasing concentration which eventually produces aggregates strong enough to withstand shear break up and rapidly deposits as a mud 'blanket'. The progressive fining of the silt laminae and the correspondence of the size at which particle percentages fall off at the coarse end of the mud grain-size distribution with the
83
84
K. Kranck
increase at the fine end of the silt peaks were seen as evidence that a series of mud silt laminae were deposited from one suspension in a waning flow. The purpose of this paper is to analyse further the grain-size characteristics of fine-grained turbidites and to examine the mechanism for laminae formation proposed by Stow & Bowen (1978, 1980).
overall stock distribution gives all grains the same relative settling rate and results in a bottom sediment which also has this same size distribution. The absolute settling rate will depend on the rate at which the flocs form, i.e. for any given suspension on the concentration. Stokes' settling, one size at a time, tends to produce well sorted sediment whereas the absence of sorting in floc settling produces a broad flat grain-size distribution similar to that of the parent suspension. Sediment formed through a combination of both types of settling will have a well sorted modal peak and a tail offloc deposited sediment (Fig. 1). Although these experimental results are derived from settling in still water there is a close parallel to turbidity current deposition. In both cases sediment settles from an initially high concentration suspension to form a deposit characterized generally by decreasing grain-size with time. In still water some Stokes' settling always occurred so that the modal and maximum sizes always decrease. In a turbulent flow, however, sediment may be kept in suspension by turbulence and floc settling will decrease the overall concentration, but the grain-size range does not change as in case B in Fig. 1. The spectral form of the Stokes'-settled portion of a sediment may be predicted from the size distribution of the source suspension. From a well mixed suspension the flux of given size particles to the bottom is given by
Settling model The grain-size analyses of sediments used in this study were examined in the light of a grain-size settling model (Fig. 1) built on earlier studies of depositional behaviour of sediments and described below. A series of settling experiments (Kranck 1980) performed with different concentrations of sediment suspended in both salt and fresh water showed that whereas in normal Stokes' settling, one size class at a time disappears from a given depth in the suspension, during settling of flocculated material all constituent grain-sizes settle at the same rate. In the former case, grain-size is the controlling factor of settling rate but in the latter case, the concentration of sediment in the suspension controls settling rate. This difference is due to the formation of flocs or settling entities each of which is made up of grains which have the same proportional size distribution as the whole suspension. This duplication within each floc of the STOKES'SETTLING
F=Cw
(1)
BOTH
FLOC S E T T L I N G
(.,9
Z
~
LOG d
LOG d
~
LOG d
---~ LOG d
LOG d
~
LOG d
FIG. 1. Sketch illustrating settling model used to interpret grain-size analysis. Numbers refer to portions of sediment populations removed from suspension and deposited progressively with time or distance away from source. Stokes' settling from an unsorted sediment in suspension leads to deposition of one grain-size at a time and produces a well sorted bottom sediment. Floc settling leads to deposition of flocs which contain a proportional amount of all grain-sizes and deposits a sediment with the same relative grain-size distribution as the parent source suspension. Sediment formed by a combination of both settling processes will be a mixture of both settling types depending on the relative importance of each for the particular locality and grain-size range.
Grain-size characteristics
where C = concentration and w = settling rate of a given grain-size in the source suspension. According to Stokes' Law the settling rate is proportional to the square of the diameter. If C were constant for all grain-sizes (i.e. independent of d) the size distribution of the resulting bottom sediment would be quadratic i.e. it would exhibit a slope of two on a logarithmic plot. The size distribution of the source suspension may be determined from the finest grained portion of the bottom sediment formed exclusively from grains which all deposit at the same rate while parts of flocs, and thus duplicate the size composition of the source suspension. If it is assumed that the size distribution of a source suspension is also described by a power law relationship, a bottom sediment may be seen as a two-component mixture of two sub-populations defined by equations (2) and (3) below (Fig. 2).
V f = aD'"
(2)
Vs = bD ''+z
(3)
where V(and V, are the volumes in logarithmic
of turbidites
85
intervals of particle diameter as percent of total volume, a and b are the intercepts of the floc and Stokes' settled sub-populations respectively, D is grain diameter and m, the slope of the source suspension (Fig. 2). The function of the whole population distribution becomes V = aD m + bD m+ 2
(4)
Equation (4) describes only the curve for particles smaller than the mode, i.e. positive slope segment of the distribution curve. The coarser negative slope to the right of the mode should be very steep reflecting truncation at the maximum grain-size carried in the suspension. This results in a strongly negatively skewed distribution.
Description of material Turbidites were sampled from three cores; two from the Laurentian Fan (Stow 1981) and one from the Sohm Abyssal Plain (Vilks et al., unpublished).
I0
$= f
0:2.4
3
=, o .I b :.062 rl
.01
I
1oo DIAMETER,(ffm)
FIG. 2. Example of grain-size distribution spectra illustrating method of deriving constants in equations 2, 3 and 4 (see text).
86
K. Kranck
VVvx I
I
0%
2
0%
6OO
UNIFORM MUD E3
2 2% 610 8%
3
20%
GRADED MUD E2
--620
40%
-4.
58%
-5
0%
63O
SILT-MUD LAMINAE El
-6
66%
I 640
,
#
,
0%
FIG. 3. Muddy turbidite subdivided according to sedimentary structures (Piper 1978) with grain-size spectra for selected samples. Units of axis same as in Fig. 2 with 10% the highest gradation shown on Y-axis of each graph. Percent values give floc settled proportion of sample. Three digit numbers give core depth in centimetres.
.....- _78 SILT-MUD D - - ' - ' - ' - - - - -9 ..... -IOll
65O
97%
/VVvv', i
I I I
#
i
Grain-size characteristics o[ turbidites Core 80-016-015 from the Sohm Abyssal Plain consisted of a 11.8 m sequence of distal silt and mud turbidites 0.5 to 0.2 m thick, bottoming out in sandy beds also of turbidite origin (Vilks et al. unpublished). In the turbidite chosen for sampling (Fig. 3) a thin (4 cm) sequence of silt and fine mud interlayered as 2-5 mm laminae forms the lowest layer (tentatively identified as a Bouma D division). The mud and silt are well separated and uncontaminated samples could be obtained of each. This is overlain by a 15 cm sequence of very finely laminated silt and mud. The laminations are diffuse and faint and no attempt was made to sample particular features of the layers. This appears to correspond to Piper's (1978) Et subdivision. Above this is another approximately 15 cm thick unit in which the laminations disappear and an even gradation of increasingly finer material is indicated by change in colour (Subdivision E2). A fairly distinct boundary separates this from a 15 cm uniform mud with no visable structures (Subdivision E3). In Core 73-011-9 from the Laurentian Fan one complete 25 cm long turbidite was sampled (Fig. 4). Over half the sequence consisted of sandy material. The lowermost clean-looking fairly uniform sand with only minor faint lamination is identified as Bouma B division. Next is a Bouma C division sand with several intervals of crosslamination and finally a sequence with distinct sand-mud laminae each several mm thick and easily sampled (Bouma D). Overlying this D division is a graded mud similar in appearance to the E2 sequence of core 80-016-015. These two turbidite beds from the Sohm Abyssal Plain and Laurentian Fan were chosen as representative of the cores as well as for their conformity to the ideal Bouma turbidite sequence. Between them they contain all the classical Bouma (1962) divisions except the A layer. In order to investigate further the origin of fine-grained bedding structures, a short less classical sequence was sampled in detail from Core 79-021-37, also from the Laurentian Fan. Much of this core was characterized by irregular intervals of short graded silt-mud sequences alternating with silt-mud laminations on different scales. A mud and a silt sample were collected from each of three silt-mud couplets and five samples from a graded bed near the interlayered sequence (Fig. 5).
Analytical methods All the cores were subsampled with a I mm wide spatula which was inserted parallel to the bedding to obtain approximately 1 g of sample. A minis-
87
cule amount of this was subsampled for size analysis. The grain-size distributions of the samples were analysed using a Model T A l l Coulter Counter interfaced with a HP-85 computer system. Calibration methods and sample processing followed methods as described in Kranck & Milligan (1979). The samples were disaggregated in a small amount of 10% (NaPOD6 solution, suspended in 3% NaCI solution and analysed using a 30 gin, and a 200 #m orifice tube for all samples as well as a 400/ml tube for the coarser sands. The samples were deflocculated by ultrasonification and the size analyses represent the single mineral grain distributions. The volume in each size channel (designated by size of channel midpoint) was calculated as a percentage of the total in all channels. Particle volume doubles in each consecutive channel (equivalent to l/3q~), resulting in 24 to 36 data points per analysis including some overlap between tubes. The results were plotted as log-log frequency distribution-spectra (Figs 3-5). This display method was chosen rather than the more commonly used arithmetic-log axis plots because portions of different samples with similar relative size composition plot as similar shaped curves irrespectively of the nature of the rest of the size distribution. For each turbidite the slope of the source suspension (equation (1)) was determined by measuring the slope angle of the size spectra between 1 /Lm and 2 #m. The slope values were averaged and a distribution for the floc settled portion of the sample was calculated from equation (1) using the 1 pm volume percent as the intercept. The calculated floc settled distribution was subtracted from the total size distribution to obtain the size distribution of the material that did not settle as flocs (i.e. the Stokes' settled material). In Figs. 3 to 5, each sub-population is marked on the spectra plot of the total distribution and the percent Stokes' settled material in each sample is listed beside the graph.
Results In all three turbidite sequences analysed, there is a general upward decrease in modal grain-size although in the D and El divisions the intermittent mud layers temporarily interrupt this trend. The size distributions are negatively skewed. The mud samples have a gentle positive slope which varies by less than five degrees from sample to sample, and a steeper negative slope. With increase in modal size, a modal peak becomes more prominent and in the coarser samples only the finest grain-sizes have the same slope as the fine
88
K. Kranck ~VVVX/v
210
••,
!
~
, 0%
O%
-2
GRADED MUD E2
~
-3
0%
I
SAND-MUD . . . . . . LAMINAED~.. / ~/"
I
I
-4 14%
-
/ t
CROSS- V . ' / i ' BEDDING
.Z 1-8
C I"/" I.'1." l
........... ~
/,.~
' 9
83%
!
230crnt
!
!
I
I
I
I
I
I
SAND B U
tt~ll L~
96%
/ ,
,
~
94% I I
I
I
I
, ~ s /I I
FIG.4. Sandyturbidite subdivided according to Bouma(1962)with grain-sizespectra for selectedsamples. Units on axis sameas in Fig. 2 with 10% The highestgradationshownon Y-axisof each graph. Percentvalues give flocsettledproportion of each sample. Three digit number givecore depth in centimetres.
I
92% .
I
95%
I0 /I
'
Grain-size characteristics of turbidites
89
FIG. 5. X-radiographs of turbidite portion sampled for graded layer (1-5) and sand mud cuplets (6-11) with grain-size spectra. Units on Y-axis same as in Fig. 2 with 10~ the highest values marked on Y-axis of each graph. Percent values give floc settled proportion of each sample. Three digit number give core depth in centimetres.
90
K. Kranck
muds. A few of the sand samples show a rise in the slope in the very finest sizes although this may be an experimental error related to electrical noise. Only in the E3 division of Core 80-016-015 was there no perceivable difference in grain-size between samples confirming that this material was indeed uniform. Subtraction of the postulated floc sub-population from the total population produced a nearly straight line for the Stokes' deposited sub-population. It is noteworthy that the slope of this exponential distribution is close to two in the non-laminated beds but becomes steeper and close to four in the lower coarser inter-layered portions of the turbidites. The percentage of Stokes' deposited material varies fi'om 97?,o to 0'~,o and reflects the general variation in modal size trend and increasing peakedness of the spectra.
Discussion The grain-size data from the three cores sampled appear to confirm the two-component model of settling (Kranck 1980): i.e. unsorted floc-deposited material deposited alternately or intermittently with single grain or Stokes'-deposited well-sorted coarser material. The presence of both of these types in one sample produces the negatively skewed modal peak and a fine, nearly straight tail characteristic of silty muds (Kranck 1975, 1980). These distributions plot as two nearly straight segments on cumulative probability plots and have been frequently described for various sediments (e.g. Inman & Chamberlain 1955) including turbidites (Piper 1973; Jipa 1974; Kepferle 1978; Stow 1981). The dissection of grain-size distribution into sub-population components is heavily dependent on high accuracy and high resolution analysis. Although the fine end portion of the size distribution could be satisfactorily defined mathematically, the function for the negative slope is not clear. It is difficult to determine empirically because in present methods of analysis the number of grains analysed in this region are very few, because an increase in sample concentration tends to cause coincidence from excessive numbers of small particles. Also it is not known if the rise in relative particle volumes at the fine end of the many sand samples (Fig. 4) is due to experimental errors associated with electrical noise or to some other effect such as trapping of very fine particles in pore spaces. The relationship of these results to the mechanics of turbidite deposition may be assessed by considering each division of the Bouma sequence in turn.
A Dit'ision: None of the core portions analysed contained examples of division A so its detailed granulometry could not be assessed. Kranck (1980) has likened the unsorted material which initially settles out of a flocculating experimental suspension to the basal portion, i.e. A division of natural deposits. Essentially this only adds flocculation to the earlier views (summarized by Middleton, 1970) of the basal or A division as a product of rapid deposition from a high concentration suspension where size sorting is inhibited. This division should have preserved within it the original size configuration of all but the coarsest size particles and a detailed study of its grain-size composition should prove profitable. B Dirision: This division, characterized by some plane parallel laminae, may be seen as a transition between division C, where traction transport of sand along the bottom is well established as indicated by ripple bedforms and the massive totally unsorted A layer, where transport presumably ceased when grains reached the bed. If there is sufficient floc disruption by bed shear to create a clean sand, this when superimposed on muddier A material may form the lower reverse graded portion of some turbidites. C Dirision: There is no perceptible difference between the size spectra of samples from B and C divisions which therefore is a classification based solely on structural evidence. Both B and C divisions have modal peaks almost entirely consisting of Stokes'-deposited material although a distinct tail of mud is also present. The slope close to 4 indicates that this material has been deposited and subsequently resuspended presumably to form a near-bottom traction-saltation load before again settling. Such near-bottom transport is in accord with the structures present and the need to expel the mud fraction. It is in agreement with reports of strong near-bottom turbulence reported from experimental turbidite flows. D Division: This is the division in which mud deposition first becomes a significant factor and the texture to which Stow & Bowen's (1978, 1980) lamination mechanism is most relevant. During its formation, turbulence has apparently decreased to a point where mud can deposit intermittently. A detailed examination of Stow & Bowen's model is outside the scope of this paper, but these results in general confirm their contention that the mud and silt laminae are deposited out of the same suspension. The rapid deposition of the mud suspension built up near the bottom is probably promoted by two factors: turbulence dampening and floc maturation. Heathershaw (1979) suggests that a concentration of suspended sediment (0.01 mm in diameter) as low as 15 rag/!
Grain-size characteristics of turbidites is able to modify turbulence and 130 rag/1 is the amount required to dampen it completely. The exact critical concentration threshold will be a function of many factors including grain-size and composition. Floc maturity is a term which may be used to describe the shear strength of the floc. Kranck & Milligan (1980) demonstrated experimentally that given sufficient time a suspension of fine-grained sediment will form flocs able to withstand very high shear rates. An important factor in this is the adhesive action of organic detritus and surface bacteria. The combination of turbulence dampening and floc maturity may cause deposition of mud laminae from even very dilute suspensions. E~ Division: The major difference between this subdivision and the D division may be one of scale. In the latter, mud uncontaminated by silt could be sampled and analysed; in the former the 1 mm thick randomly extracted samples all contained a portion of Stokes'-settled grains. Greater sampling resolution may or may not succeed in isolating the true sedimentation unit: the question has only limited relevance in explaining the depositional mechanism, which is assumed to be similar for both divisions. E: Division: As was the case in differentiating Divisions B and C, the El and E2 divisions can only be distinguished visually. The samples in both El and E2 have Stokes sub-populations with positive slopes of 2 indicating direct settling without resuspension. E.~ Division: The complete lack of Stokes'-settied fraction and of any changes in the grain-size indicates this sequence has settled entirely by flocculation. The suspension from which it has been deposited must be in equilibrium with the turbulence in the water so that only by forming flocs can the particles overcome upward advection. The only conditions under which a uniform sediment will form from a non-constant source is when the suspension is completely flocculated so that each particle which arrives at the bottom is a floc containing equal proportions of all grains from the suspension. In turbidite sedimentation
91
this may occur at the beginning of a flow when concentrations are very high, so that floc settling rates greatly exceed Stokes' settling rates, or alternatively at the end when there are no grains left which are able to settle singly from in the suspension. F Dirision: In these cores there were no signs of bioturbation or other features characteristic of normal pelagic sedimentation designated as F division by Piper (1978). If such material was laid down between the turbidite episodes it must have been eroded or cannibalized by the succeeding flow.
Conclusions The examination of the fine detail of the textural changes in individual turbidites shows the importance of grain-size properties in understanding depositional mechanisms. In this paper only deep-sea turbidites have been examined but similar studies are a prerequisite to the development of a truly genetic classification of fine-grained sediment. Modern electronic particle analysers and computer processing of results allow a large number of samples to be processed rapidly. This is especially valuable in the case of fine-grained sediments where individual grains cannot be seen directly. In the deep-sea sediments where depositional conditions are hard to observe directly, fine-grain granulometry should be a standard supplement to conventional studies. Similar investigations in the future of contourites, red and black pelagic muds and other facies would help settle many of the problems regarding their origin and stratigraphic significance. ACKNOWLEDGEMENTS: I am indebted to D. Piper and G. Vilks who provided the turbidite material used in this study. The size analyses were performed by T. Milligan and D. Nelson. Many colleagues have been helpful in the preparation of the final manuscript.
References ALLEN, J.R.L. 1982. Sedimentary Structures. Their Character and Physical Basis. Vol. II. Elsevier, Amsterdam, BOUMA, A.H. 1962. Sedimentology of some Flysch Deposits. Elsevier, Amsterdam, 162 pp. HARMS,J.C. & FAHNESTOCK,R.K. 1965. Stratification, bed forms and flow phenomena (with an example from the Rio Grande). Soc. econ. Palaeo. Min. Spec. Pub., 12, 84-115.
HEATHERSHAW,A.D. 1979. The turbulent structure of the bottom layer in a tidal current. Geophys. J. R. astr. Soc., 58: 395-430. INMAN, D.L. & CHAMBERLAIN,T.K. 1955. Particle-size distribution in near-shore sediments. Soc. econ. Palaeo. Min. Spec. Publ., 3, 106-29. JIPA, D.C. 1974. Graded bedding in recent Black Sea turbidites: a textural approach. In: Degens, E.T. & Ross D.A. (eds), The Black Sea-Geology, Chemistry
92
K. Kranck
and Biology. Am Ass. Petrol. Geol. Mem. 20. KRANCK,K. 1975) Sediment deposited from flocculated suspensions. Sedimentology, 22, 111-23. 1980. Experiments on the significance of flocculation in the settling of fine-grained sediment in still water. Can. J. Earth Sci., 17, 1517-26. -& MILLIGAN, T. 1979. The use of the Coulter counter in studies of particle size distributions in aquatic environments. Bedford Institute of Oceanography, Report Series BI-R-79-7. 1980. Macroflocs: Production of Marine Snow in the Laboratory. Marine Ecol, Progress Series 3: 19-24. KEPFERLE,R.C. 1978. Prodelta Turbidite Fan Apron in Borden Formation (Mississippi), Kentucky and Indiana. In: Stanley, D.J. & Kelling, G. (eds), Sedimentation in Submarine Canyons, Fans and Trenches. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 224-38. KUENEN,PH.H. 1951. Properties of turbidity currents of high density. Soc. econ. Paleo. Min. Spec. Pub., 2, 14-33. 1966a. Matrix of turbidites: experiment approach. Sedimentology, 7, 267-97. 1966b. Experimental turbidite lamination in a circular flume. J. Geol., 74, 523-45. MIDDLETON, G.V. 1967. Experiments on density and turbidity currents III. Deposition of sediment. Can. J. Earth Sci., 4, 475-505. 1970. Experimental studies related to problems of -
-
-
-
-
-
-
-
-
flysch sedimentation. In: Lajoie, J. (ed.), Flysch Sedimentology in North America. Geol. Ass. Can. Spec. Pap., 7. 253-72. PIPER, D.J.W. 1973. The sedimentology of silt turbidites from the Gulf of Alaska. In: Kulm, L.D., von Huene, R. et al., Initl. Repts. DSDP, 18, 847-67. 1978. Turbidite Mud and Silt on Deep-sea Fans and Abyssal Plains. In: Stanley, D.J. and Kelling, G. (eds), Sedimentation in Submarine Canyons, Fans and Trenches. Dowden, Hutchinson & Ross, Stroudsburg, Pa. POTTER, P.E. & SCHEIDEGGER,A.E. 1966. Bed thickness and grain size: graded beds. Sedimentology, 7, 233-40. SCHEIDEGGER, A.E. & POTTER, P.E. 1964. Textural studies of graded bedding, observations and theory. Sedimentology, 5, 289-304. STOW, D.A.V. 1979. Distinguishing between finegrained turbidites and contourites on the Nova Scotian deep water margin. Sedimentology, 26, 371-87. & BOWEN, A.J. 1978. Origin of lamination in deep-sea, fine grained sediments. Nature, 274, 324-28. 1980. A physical model for the transport and sorting of fine-grained sediment by turbidity currents. Sedimentology, 27, 31-46. 1981. Laurentian Fan: morphology, sediments, processes and growth pattern. Am. Ass. Petrol. Geol. Bull., 65, 375-93. -
-
-
KATE KRANCK, Atlantic Oceanographic Laboratory, Bedford Institute of Oceanography, Dartmouth, Nova Scotia, B2Y 4A2 Canada.
Fine-grained sediments of the Zaire deep-sea fan, southern Atlantic Ocean T.C.E. van Weering and J. van Iperen SUMMARY: The Zaire deep-sea fan is a large mud-dominated system fed mainly by the Zaire (Congo) River. A submarine canyon deeply incised into the outer estuary funnels part of the river's high suspended load directly out to the deep sea. Sediment is also contributed to the fan via slumping and debris-flows initiated on the upper slope. A number of 11-17 m long piston cores have been collected from this part of the SE Atlantic Ocean, and three of these are described in detail in terms of structure, texture and composition. Four sediment facies are recognized: (1) turbidite muds with mainly TEl, TE2 and TE3 structural divisions (after Piper 1978): (2) pelagic and hemipelagic biogenic muds: (3) homogeneous uniform muds that probably result from slow settling of dilute suspensions: and (4) debris-flow deposits. There are marked differences in facies distribution, turbidite thickness and organic carbon content between the upper and outer regions of the fan. The marine geology group of the Netherlands Institute for Sea Research has carried out a research programme in the Zaire (formerly Congo) River, its estuary, the adjoining part of the continental shelf and slope and in the Angola Basin during cruises held in 1976, 1978, and 1980 (Fig. 1). The aim of the investigations was the study of the physical and chemical processes involved in the outflow of the Zaire River to the Southern Atlantic Ocean and the study of the recent and sub-recent sediment dispersal pattern in the adjacent part of the continental slope and the Angola Basin. Results on the chemical and hydrographic processes in the Zaire outer estuary and plume have been reported by Eisma et al. (1978). This area was selected as the Zaire River has the world's second largest drainage basin (3.7x 106 km 2) and mean annual discharge (45 000 m3s -~) and is the main contributor of terrigenous sediments to the eastern South Atlantic Ocean, together with the River Niger. Moreover, although it is known that a part of the sediment load of the river is funnelled directly to the deep sea via a canyon which has incised itself already down to 100 m in the lower estuary, the Zaire deep-sea fan and its sediments have been studied only to a limited extent (Heezen et al. 1964; Bornhold 1973; Shepard & Emery 1973). In this article some selected cores will be described which are considered as representative of the depositional areas and processes on the Zaire fan.
Geological and oceanographical setting Geologicalsetting The Zaire fan lies in the Angola Basin, the deepest depression of the eastern South Atlantic. This
basin is a restricted basin bounded by the Walvis Ridge to the south, the Mid Atlantic Ridge to the west and the Guinea rise to the north (Fig. 1). The continental shelfofZaire and Angola is about 100 km wide and has its break at a depth of 100-120 m. Sediment studies of the Angolan shelf are not known to us; the shelf of Zaire, Cabinda and Congo has been extensively studied by Giresse and his co-workers (Giresse & Odin 1973; Giresse & Moguedet 1974: Giresse & Cornen 1976; Giresse et al. 1979) and is partly summarized in a recent article by Giresse et al. (1982). The typical shelf sediments are mainly finegrained, forming a blanket of muddy sands and sandy muds over the entire shelf. Clays are derived from the south and consist of 65% kaolinite, 251~0 illite-montmorillonite and some 10~ illite. Faecal pellets are found all over the shelf and are glauconitic at depths of less than 25 m and more than 110 m. Between 100 and 120 m relict shelly sands form a submarine terrace. The shelf is dissected by the Zaire canyon, which has eroded down to 100 m depth at 30 km from the river mouth in the outer estuary and down to 425 m at the river mouth. The canyon and fan valley have been studied by Heezen et al. (1964) and in greater detail by Shepard & Emery (1973), showing that the canyon and fan valley continue as far as 800 km to the west. The canyon has eroded deeply into the continental slope; in a zone landward of a belt of diapirs, the incision below sea bottom is from 400-1400 m and has a characteristic steep-sided V shape. Within the diapir belt, presumably salt diapirs of Aptian age (Baumgartner & van Andel 1971~ Emery 1972; Lehner & de Ruiter 1975), the canyon decreases in depth to about 300 m below sea bottom; its slopes are irregularly-shaped probably because of slumping and/or local uplift 95
96
T.C.E. van Weering and J. van Iperen
FIG. 1. SE Atlantic Ocean and study area. Depth contours are in metres. Main topographical features are the Mid-Atlantic Ridge, Walvis Ridge and the Guinea Rise.
through doming of the diapirs. Locally the canyon becomes less deep and more flat-floored seaward of the diapir belt; here it is flanked by levees up to 70 m high. The diapir zone has produced a sea bottom relief of several hundreds of metres through local updoming; it ends abruptly to the west with a steep escarpment, the Angola escarpment. The lower continental rise has some relict hills at depths between 4000 and 4500 m (Bornhold 1973). In the western part of the Angola Basin various seamounts interrupt the flat-lying sediments. Heezen et al. (1964) and Bornhold (1973) studied the sediments to some extent.
Hydrography The surface water circulation in the Angola Basin is primarily influenced by two main current systems, the Benguela Current and the South Equatorial Counter Current (SECC). The Benguela Current, dating from about 10 million years BP (Siesser 1980), transports relatively cold waters with velocities of 15-20 cms -1 northwards in a broad zone along the African coast up to about 15S., where the current curves to the west and meets the SECC (Fig. 2). The interaction between the currents leads to a complex pattern of gyres and local upwelling phenomena; the distribution
Fine-grained sediments of the Zaire deep-seafan
97
FIG. 2. Surface current pattern in the SE Atlantic Ocean. of silicoflagellates in a number of selected cores of the Angola Basin also points to local upwelling conditions (de Ruiter, pers. com.). A small branch of the Benguela Current continues towards the north and induces a northerlydirected longshore drift along the coast of Angola, probably resulting in the formation of beach barriers at the mouth of the Zaire River. There is evidence for a subsurface countercurrent beneath the Benguela Current, along the edge of the continental shelf with its core at 300-400 m water depth and with a southerly direction (Hart & Currie 1960). The SECC flows to the south-east at about 7-8~ and forms a gyre
where it meets the Benguela Current. A southerlydirected branch continues off the coast of Angola as the Angola Current, while another branch turns northwards off the coast of Zaire and Congo and probably influences the direction of the Zaire River plume. Recent research (van Bennekom & Berger 1983) shows that the freshwater outflow of the Zaire River is noticeable as far as 800 km offshore, the plume meanders and forms lobes in response to variations in windstress, river discharge and current strength. The direction of the plume is south-West or south-south-west in February and March, west or south-west from
98
T.C.E. van Weering and J. van Iperen
April to August and to the north-north-west in October and November and part of the southern summer. The deeper waters in the Angola Basin consist of North Atlantic deep water and Antarctic bottom-water which enter the basin via sills in the Romanche and Chain fracture zones (Wrist 1935). Current velocities at the bottom are very low (van Bennekom & Berger, 1983). Suspended matter Data on suspended load (Eisma & Kalf 1983) show that the Zaire transports annually an average of 43 x 106 tons of suspension load. About half of this load settles in the head of the canyon, and only 5% reaches the deep sea, the remainder being deposited on the shelf and slope or remaining near the bottom. The total content of suspended matter decreases from 10 m g L - 1 to 2 m g L - 1, going from the river mouth to the edge of the shelf. According to Eisma & Kalf (1983) flocculation affects only a few percent of the total suspension load. In the open ocean the concentrations of suspended matter are well below 1 mgL -l. The amount of organic matter in the suspension load varies from about 30% in the estuary to over 60% on the shelf. In the open ocean the suspended sediment comprises 70-90% biogenic and organic material (Emery & Honjo 1979), mainly organic aggregates (90~o), skeletal debris (8%) and mineral grains (2%).
Methods Piston cores of up to 18 m length were obtained by means of a modified Alpine piston corer. For the deep-sea coring a special Kevlar cable was used with the same breaking strength and stretch as a steel cable but considerably lighter. On board, the cores were cut in 120 or 150 cm long sections and stored at 4 C . The sections were split as soon as possible after retrieval, described and photographed. Colours were noted using the standard Rock Colour Charts of the Geological Society of America. On board, X-radiographs were made of the split sections in order to delineate internal sedimentary structures. The X-ray apparatus used was a Hewlett Packard Faxitron N-407 system, using Agfa Gevaert D-7 film. In the laboratory a number of fine-grained turbidites were selected for further detailed study and description, as well as for grain-size analysis and mineralogical determination. Grain-size analysis was performed with standard sieve and pipette methods, applying whole phi-intervals (Folk 1968). The carbonate content
of the samples was calculated through Ca determination of the samples by fluorometric titration with EGTA. Organic carbon was measured by means of a Perkin-Elmer CHN Analyser type 240-B. The sedimentary structures were described from the X-radiographs, following the modified turbidite terminology of Piper (1978). He subdivided Bouma's (1962) E division into turbidite mud (TEl laminated, TE2 graded, TE3 ungraded), and pelagic mud (F).
Results During the cruises a number of piston cores were taken in order to characterize sediments and depositional patterns of the shelf, the upper and lower slope, the deep-sea fan and the abyssal plain (Fig. 3). Only part of the cores revealed the presence of turbidite structures, in contrast to the rather structureless, homogeneous sediments forming the greater part of the shelf and slope (Fig. 4). The cores presented in some detail below were selected for further study as they show systematic changes in fine sediment dispersal across the fan. The main aim of the coring programme was to study the lateral extent and variation of turbidite sedimentation. (1) Core 38 was taken at a water depth of 5490 m in the outer deep-sea fan abyssal plain area, which is interrupted by some sea-mounts in the direct neighbourhood. Within the core, three turbidite intervals can be recognized, separated by rather homogeneous structureless, locally bioturbated and mottled pelagic sediments (Figs 5 and 6). (2) Core 42, obtained from a water depth of 4460 m, is considered representative of the outer part of a middle-fan area; probably this is an interchannel deposit cut off from the main turbidite supply. The lower 530 cm of the core is entirely turbiditic, whereas above this level, homogeneous partly bioturbated sediments are found (Figs 7 and 8). (3) Core 32 at a depth of 3600 m, was taken on a levee near the main channel of the upper fan. This core consists entirely of turbidites (Figs 9 and
10). Colour In core 38 the sediments range from dark greenish-grey (10 GY 4/4) via dark green (10 G 4/1) pelagic intervals, into olive-grey (5 GY 6/1), olive-brown (2.5 Y 4/3) and dark brown homogeneous sediments (10 YR 3/3). The turbidite intervals at the top are, in general, darker olive-
Fine-grained sediments of the Zaire deep-sea fan
99
"0
0 "0
O
0
"0
0 "0
"0
R r
.=. o~ fz~
R.6
0
.~ "0
9
0
I oo
T.C.E. van Weering a n d J. van Iperen
Fro. 4. Correlation of climatic zonation and occurrence of turbidite intervals (horizontal rule) in cores from the SE Atlantic Ocean.
grey and greenish-grey. The turbidites in core 42 have a grey (10 Y 4/1) to dark olive-grey colour (5 GY 3/1) and are only slightly darker than the greenish-grey (7.5 GY 4/1) to olive-grey (2.5 GY 5/1) non-turbidite sediments. The sediments of core 32 are generally dark and range in colour from olive-black (5 GY 2/1) in the lower part of the turbidites to slightly lighter, though still olive-black (7.5 Y 3/2) sediments, forming the top of a single turbidite. Within the lighter sediments, mottling occurs on a minor scale Structure
The turbidite intervals in core 38 show alternations of To (occasionally), TEl, TE2 and TE3
divisions. In the lower part of the turbidite interval TE1/TE2 alternations are most frequent, the TEl division having a sharp, clearly recognizable lower boundary. The TEl division is mostly thickly-laminated (3-6 mm thick laminae) or very thin-bedded (6-15 mm). The TE2 division is occasionally reversely-graded. Towards the top of the turbidite intervals, TEl divisions become scarce, TE2 and TE3 have a greater thickness and are slightly burrowed (Fig. 5). The TE3 interval grades gradually into the pelagic F division; upper boundaries of the turbidite intervals are therefore difficult to recognize. Syn-sedimentary deformation is seldom observed (Fig. 6). Towards the top of the core the turbidites are covered by rather homogeneous bioturbated
Fine-grained sediments o f the Zaire deep-sea f a n
I 01
FIG. 5. Columns showing the climatic zonation, the carbonate and organic carbon content and the turbidite intervals (T) of core 38. Climatic zonation and age after Jansen et al. 1983. Length of the core indicated in centimetres. mottled sediments, forming the greater part of the sediment column. The turbidite sections of core 42 consist almost entirely of alternations of TEz/TE3 divisions and of TE3/F divisions, in a setting which changes from turbidite-dominated into pelagic. TEl divisions occur only rarely, and are thickly-laminated (3-6 mm). The TE2 and TE3 divisions are thicker in the lower part (with a mean of 35 ram), changing gradually into thinner-bedded turbidites towards the top of the turbiditic interval (Fig. 8). Almost all the turbidites are covered by F divisions, separated by a clear boundary. The F divisions increase in thickness towards the top of the turbidite section; in the lower part they vary between 5 and 13 mm, with a mean of 7.5 mm, whereas upwards the F division is at maximum 32 mm and has a mean thickness of 16 mm. Over the complete turbidite interval, the ratio of pelagic sediments to turbidite sediments is 1: 3. Core 32 shows laminated TE~ divisions, TE: graded muds and TE3 muds. F intervals are scarce, but increase towards the top (Fig. 10). The
TE~ divisions are mainly thickly-laminated (3-6 mm), although many are thinner ( < 3 ram) and more rarely, thicker. Most Tvz divisions occur together with TE2 divisions, sometimes followed by TE3. Sequences without TEl divisions are also present. The mean thickness of a complete turbidite sequence is at maximum in the lower part of the core and decreases towards the top. Although the F intervals increase towards the top of the core, their thickness seldom exceeds 10 mm, and is mainly around 5 mm. The ratio of pelagic to terrigenous sediments is 1 : 20. Various levels with clay pebbles are found. The clasts are oval to circular, more rarely angular, and have a longest axis of 0.3-2.5 cm. Burrowing activity in the sediments is scarce in the lower part and somewhat more evident in the upper part. Most burrows are syn-sedimentary.
Grain-size and composition Grain-size determinations of some selected turbidite intervals were carried out. In core 38, the TD
T.C.E. van Weering and J. van Iperen
[o2
CORE
38
30-150
cm
IIo 3
7(
4
%
9
:'~ '~w~,~" ....
....-
"S
?
"S
i
"S
a. 0F-
T 50"
:
i i I
:
....
....
,S
']
~
70-
,,v-
,S .S
FIG. 6. Radiograph of sedimentary structures in lower turbidite interval of core 38. Scale in centimetres. S denotes samples used for grain-size determinations shown in Fig. 11. Note horizontal layering disturbed by bioturbation in lower part of core. Turbidites are TD/TEI and TEI/TE2 alternations changing into TE3 interval. Note also occurrence of burrows in TE3 intervals at 95 and 123 cm.
of the Zaire deep-sea fan
Fine-grained sediments
IO3
CORE 42 sediment column
•
g
o N
carbonate 0
2 ,
I
4 j
I
6 ,
I
8 ,
I
10 ~
I
12 ,
I
3m
,0-
00"
I
0 A
organic 1
J
carbon 2
I
5 % I
B, 29.7
18- 00_ 0
'3-
14 % ,
~9 - . ~ 2 0 9
'0)5-
O0" bi, 2 0 . 8
;20 O-
goo. I o~ to
_:--T
o. g. -9- ' ~
57.5
o
F~G. 7. Columns show climatic zonation, carbonate and organic carbon content, and turbidite interval (T) of core 42. Climatic zonation and age after Jansen et al. (1983). Length of the core indicated in centimetres. interval samples at 83 89 cm has a sand content (> 63/~m) of 20%, and a coarse silt (32-63 ~m) mode of about 55~Jo (Fig. 11). Several TEl intervals samples show lower sand content, still have coarse silts as the modal fi'action but contain a somewhat higher percentage of very fine silt and clay. There is apparently a change into a bimodal grain-size distribution from TD via TEt into TE2. The sand and coarse silt fraction contains abundant quartz, some mica, many diatoms, and sponge spicules, together with some radiolarians, and rare sponge spicules embedded in faecal pellets. Pyritization is relatively common. The finer fractions are mainly clays and siliceous ooze (fragments of diatoms, radiolarians and silicoflagellates). In core 42, the TEl division consists primarily of coarse and very coarse silts (63-16 pm) in addition to a considerable amount of fines ( < 4 #m), although the 4-8 ~m fraction is very small, and the distribution is therefore slightly bimodal. The TE2 and TE3 divisions are sometimes difficult to separate visually, and this is also reflected in their grain-size distributions, which are very similar,
both with around 80,~ of very fine silt and clay-sized particles (Fig. 12). In the TEl division, the sand fraction contains predominantly quartz and mica (rounded flakes), and considerable amounts of pyrite, diatoms (some partly pyritized) and sponge spicules. Some of the quartz grains are coated with haematite, as are some of the diatoms. Some of the diatoms in the turbidite intervals are clearly freshwater species. The silt and clay fractions consist of micas, clay minerals, coccoliths, fragments of foraminifera and phytoliths, and rare radiolarians. Plant remains occur throughout. The TEl divisions of core 32 characteristically have a coarse or very coarse silt modal grain-size, with a small amount of finer material. The TE2 divisions contain far more very fine silt and clay ( < 4 l~rn), and in the TE3 divisions this fraction accounts for over 60'~o of the sediment with only a minor coarser fraction (Fig. 13). The sand fraction consists of quartz at the base of some of the TEl divisions, with more mica in the upper part of the laminated units. The micas are often abraded and have rounded edges. Sponge spicules, dia-
IO4
T.C.E. van Weering and J. ~'an Iperen
FIG. 8. (a)-(b) Radiographs of core sections from core 42 showing increasing bioturbation in TE3 intervals towards the top of the core together with a decrease in thickness of TEl and TE2 intervals. Some of the light spots are pyritized burrows. Note local disturbance (at 25 cm) of TE2 interval.
Fine-grained sediments of the Zaire deep-seafan
Io5
Io6
T.C.E. t'an Weering and J. t'an Iperen CORE 5 2 sediment column
!I
0
carbonate 2 4%
organic carbon 5 4
5% I
~n
[,.,.
-=T~_
8-
g_ oO. FI~. 9. Columns of core 32 indicating climatic zonation, carbonate and organic carbon content. The entire core comprises turbidites. Core length in centimetres. Climatic zonation after Jansen et al. 1983.
S_0
toms and some foraminiferal fragments are also present. Plant fragments occur in all TEl divisions, orientated with their long axis parallel to the bedding plane. The silt and clay fraction consists of diatoms, coccoliths, mica, minor amounts of quartz, clay minerals, organic remains and phytoliths.
crease with increasing distance from the source (Fig. 14), being greatest in core 32 and least in core 38. The turbidites contain more organic carbon than non-turbidites in core 38, whereas there is almost no difference between the two in core 42.
Carbonate and organic carbon
Discussion
The carbonate content shows great variations due to a combination of dissolution and dilution with terrigenous material. In core 38 (Fig. 5), the peak value of 20.8% carbonate at the top of the core reflects recent depositional conditions, and indicates a slow dissolution of carbonates at the sea bottom. Other peaks in core 38 are the result of the occurrence of some carbonate rich clay pebbles of allocthonous origin. The carbonate in core 42 (Fig. 7) shows a number of consecutive peaks in the non-turbidite section, with peak values of 29.7~ at the top of the core. The turbidite section has a lower carbonate content with rather less clear peaks as a result of dilution by the increased terrigenous supply. This effect is still clearer in core 32 (Fig. 9) where the amount of carbonate is low compared to other cores, with values between 0.5 and ~o/ The content of J / o" organic carbon shows a clear tendency to de-
Piston cores from the continental slope, the upper part of the Zaire submarine canyon, and of the Zaire fan were studied by Heezen e t al. (1964). They recognized four sediment facies on the basis of sedimentary structures. Facies I is a homogeneous silty iutite, facies II a crumbly silty lutite, facies III contains graded silts and sands and facies IV is formed by laminated, alternately lutite-rich and lutite-free silt. These sediments are covered in the northern part by carbonate rich, grey and black lutites, and in the western and southern part of the study area by a reddishbrown lutite. The occurrence of facies IV and III was related to turbidity currents, triggered probably by high discharges from the Zaire River. A study of cable breaks between 1886 and 1937 showed that these invariably occurred in periods of high river discharge. Heezen et al. inferred that the Zaire
Fine-grained sediments of the Zaire deep-sea fan
IO7
FIG. 10(a)-(b). Radiographs of core sections from core 32. In the lower part of the core, thin To and TEl intervals alternate with TE2 graded muds. Distinction between TE2 and TE3 is difficult. F intervals increase in frequency and thickness towards the top. Light spots at 360 cm are angular and oval clay pebbles. Note slightly inclined lamination above 363 cm. Burrowing is scarce.
I08
T.C.E. van Weering and J. van Iperen
Fine-grained sediments of the Zaire deep-seafan
lO9
core58
% 60
83 ~,~- 8 4 cm
40 ~
8 6 - 8 6 ~2
Cm
93-93
~2
cm
~
200
63
4
63
4
,1 ~2-,2:m
0
63
4
6.3
4
63
4
_ _ 1131/2-,4 :r~
63
4
FIG. 11. Grain-size distribution of samples from core 38 showing decrease of very coarse silt component and increase in the < 4 #m fraction from TD--*TE1--*TE2.
submarine canyon today experiences 50 turbidity currents per century. This is a strikingly higher number than, for example, that suggested for basins off California, where a frequency of 1 per 400 years is estimated (Gorsline & Emery 1959), or for the western Mediterranean where a frequency was estimated of at least 3 per 2000 years (Rupke & Stanley 1974). In recent years the study of fine-grained deposits covering the deep-sea floor has increased considerably. A further subdivision of Bouma's (1962) turbidite division was made by Piper (1978). He subdivided fine turbidite muds into a laminated Ej division, a graded E2 division and an
ungraded E3 division; the overlying hemipelagic sediment he referred to as the F division, following Van der Lingen (1969) and Hesse (1975). Piper also stressed that, volumetrically, turbidite muds are two to ten times more important than turbidite sands. Previously, Rupke & Stanley (1974) subdivided fine-grained deposits from the Mediterranean into two genetically different mud types. Type A mud typically showed the same structures as Piper's El E3 and F divisions, and this type was recognized as turbidity current derived. Type B muds formed through pelagic settling. Stanley (1981) and Blanpied & Stanley (1981) described
core 42 % 80-
60 84
,~
,2-42~2cm
40
[~
,,_,,~0m
TE2
TE3
20,
o-
_
65
4
63
111-111 ~2cm
63
4
113~2-114cm
TE2
63
4
TE3
4
63
4
FIG. 12. Grain-size distribution of samples from core 42. Increase of the < 4 #m fraction from TEl ~TE2--*TE3 divisions reflects upward-fining trends.
T.C.E. ~,an Weering and J. t,an Iperen
I I0
core 3 2
i
6 0 84 2 1 - 211/2cm
231/2- 2 4 cm
27-271/2cm
40-
o
__.4-
63
4
63
4
63
4
63
4
63
4
63
4
o
60 :348 - 3 4 8 ~'2 cm
351 - 351 ~'2cm
40
TE 3
20
63
4
63
4
FIG. 13. Grain-size distribution of samples from core 32. Note difference in grain-size between TD and TEl intervals.
% orgonic ~ carbon ~
_
i
J
r r,-
~
I--
5
t--
i
613 b.-
I--
i
i
oo I,.-
co t.-
t--
I--
9 = meon volue
4
1
0
o
!
,oo
2~o
3oo'
4~o
-..
5oo'
6~o km
from
40 scource
FIG. 14. Relation between content of organic carbon and distance from source for a number of cores. T78-46 and T78-45 are indicated on Fig. 3 as 46 and 45 respectively.
Fine-grained sediments of the Zaire deep-sea fan "unifites' as 'lutite layers, often thick, comprising both structureless muds and faintly laminated muds revealing a fining-upward trend', from restricted basins in the Mediterranean. They ascribed the origin of unifites to a sediment gravity-flow continuum related to mud-charged turbidity currents and less dense turbid layer flows. They inferred extremely rapid sedimentation rates for the structureless muds of > 200 cm/1000 yrs. However, the unifite muds are not truly homogeneous and are partly comparable to Piper's laminated to graded and ungraded turbidite muds. An 'ideal' sequence for fine-grained turbidities comprising nine subdivisions (T0-T~) has been proposed by Stow & Shanmugan (1980), based on a study of both recent and ancient sediments. The complete sequence is rarely observed, and could not be established in the cores described in this paper either. Our own observations of the Zaire fan deposits show that four types of fine-grained deposits can be recognized in the area. Type 1 are turbidite muds, type 2 are hemipelagic sediments forming the F division on top of turbidites, type 3 are homogeneous, uniform sediments, and type 4 are the muddy debris-flow deposits found intercalated in core 32.
core 49
80-
II
I
Within the cores studied, only TD (occasionally), TEl, TE2 and TE3 divisions are present. TE~ divisions contain small amounts of sand and have their main mode in the coarse and very coarse silts. The TE~ division characteristically contains a small fraction of the 4-8 ~m fraction. The thickness of the TEj division varies but is < 20 mm in core 38 and < 10 mm in cores 42 and 32. The TE2 and TE3 divisions have a considerably greater thickness in cores 42 and 32. Most logically, the fine-grained turbidites from the Zaire fan form the distal end members of channelized, gravity-induced mass-flow in the canyon and upper fan. On the upper fan the channel levees are up to 70 m high so that only the finest sediments settle on the flanks, resulting in the deposition of thin TEl and mainly of TEe and TE3 turbidites. Any sand-sized material found outside the channels will occur as very thin sheet sands, interbedded with silts and silty clays. In the outer fan area, where the levees are less pronounced, the dilute turbid cloud or turbidity current tail will more easily overtop the levees, resulting in somewhat thicker TD and TEl intervals. However, the greater part of the turbid cloud will probably not reach that far into the deep sea, and so thinner TE2 and TE3 turbidites will be deposited. A contour current origin for the
53 cm
5 4 6 cm
6 8 5 cm
_
60-
I
I 40
20
I
65
80
63
4
core 44
6B
4
i8 cm
4
8 0 cm
4 2 5 cm
60 1
40
20
65
4
6
4
6B
4
FIG. 15. Characteristic grain-size distribution of samples from the continental slope, uninterrupted by turbidite sedimentation. Compare with Figs 11, 12 and 13.
I 12
T.C.E. van Weering and J. van Iperen
finely-laminated and graded silts can be excluded because the bottom current velocities in the Angola Basin are very low (van Bennekom, pers. comm.). For the deposition of the homogeneous finegrained deposits encountered in many cores from the Angola Basin and Zaire fan (Fig. 15) another transport mechanism is suggested. These deposits most likely include the crumbly silty lutites and the homogeneous lutites of Heezen et al. (1964); although the crumbly homogeneous muds that we cored can be ascribed to the formation of gas through rapid decomposition of organic-rich sediments. Typically, these crumbly sediments are found in the upper part of the fan, where the content of organic carbon is twice as high as in the outer fan area (Fig. 14). This type of homogeneous sediment could form through the settling out of low-density turbid-layer flows, perhaps caused by slumping on the continental slope. Slumping followed by debris-flows have resulted in the debris-flow deposits with oval, circular and angular mud clasts found in core 32. The slumping is probably caused by two principal mechanisms. Large amounts of the fine suspension load from the Zaire River are deposited on the continental shelf and slope on an uneven sub-bottom (Eisma & Kalf 1983). This uneven and irregular sea bottom has been induced partly by the intermittent updoming of salt diapirs. Sea floor instability is enhanced by both the high sedimentation rates and the irregularity of the sea floor, causing slumping followed by debris-flows across the slope and progressively more dilute suspension clouds further into the basin. Normal hemipelagic settling would occur at the same time as settling from these dilute suspensions, producing the homogeneous fine-grained sediments encountered between the turbidite intervals in core 38 and overlying the turbidites in core 42. Low-density suspension flows in the bottom boundary layer might also result in the deposition of the homogeneous muds. However, optical
measurements during one of our cruises (Zaneveld et al. 1979) gave no indication of nepheloidlayer sediment transport close to the bottom at that time.
Conclusions Four sediment types, reflecting different depositional mechanisms, can be recognized in cores from the Zaire deep-sea fan. Type 1: fine-grained turbidites, comprising thinly-laminated silts forming the TEl division, graded silts forming the TE2 and ungraded silts and clays forming the TE3 division (after Piper 1978). Type 2: pelagic sediments between turbidite beds particularly on the outer fan. Type 3: homogeneous, uniform sediments, which are thought to have accumulated slowly through settling out of dilute suspensions. These suspensions may have been caused by either slumping or by low density flows in the bottom boundary layer. Evidence of slumping is present in the debris-flow deposits found in a core at the base of the slope; bypassing of fines may have resulted in the formation of low-density turbid clouds responsible for the deposition of type 3 sediments. Type 4: debris-flow deposits found at the base of slope and upper fan area as mud clasts in a silty-clayey matrix, deposited from debris-flows derived from slumping on the slope. Difference in thickness between the various turbidite dl Asions over the fan is most likely the result of the interaction of sea floor morphology and the volume and height of the turbid cloud passing over the sea bottom. The ratio of pelagic to turbidite sediments is 1 : 3 on the outer fan and 1:20 on the upper fan. The content of organic carbon in turbidite sediments of the Zaire fan decreases considerably with increasing distance from the source; on the upper fan, organic carbon contents are slightly more than 4%.
References BAUMGARTNER, T.R. & VAN ANDEL, TJ. H. 1971. Diapirs of the continental margin of Angola, Africa. Bull. geol. Soc. Am., 82, 793-802. BENNEKOM,A.J. VAN& BERGER,G.W. 1983. Hydrography of the Angola Basin and modifications caused by the Zaire river. Neth. J. Sea Res., 17, (in press). BLANPIED, C. & STANLEY, D.J. 1981. Uniform mud (unifite) deposition in the Hellenic Trench, eastern Mediterranean. Smithson. Contrib. Marine Sci., 13, 40pp. BORNHOLD,B.D. 1973. Late Quaternary sedimentation
in the eastern Angola Basin. Techn. Rep. Woods Hole Oceanogr. Instit, 73--8. BOUMA, A.H. 1962. Sedimentology of Some Flysch Deposits. Elsevier, Amsterdam. 168 pp. EISMA, D. & KALr, J. 1983. Dispersal of Zaire river suspended matter in the estuary and the Angola Basin. Neth. J. Sea Res., 17, (in press). - - , DE BLOK, J.W., DORRESTEIN,R., POSTMA,H., NIENHUIS, P.H. & WEBER, R.E. (eds.) 1978. Geochemical Investigations in the Zaire River, Estuary and Plume. Neth. J. Sea Res., 12(3/4), 255-420.
Fine-grained sediments of the Zaire deep-sea fan
-
EMERY, K.O. 1972. Eastern Atlantic Continental Margin: Some results of the 1972 cruise of the R.V. Atlantis II. Science, 178, 298-301. -& HONJO, S. 1979. Surface suspended matter off Western Africa: relation of organic matter, skeletal debris and detrital minerals. Sedimentology, 26, 26-775. FOLK, R.L. 1968. Petrology of Sedimentary Rocks. Univ. of Texas. 170 pp. GIRESSE, P. & CORNEN, G. 1976. Distribution, nature et origine des phosphates miocen+s et +oc6nes sousmarins des plate-formes du Congo et du Gabon. Bull. BRGM. (2me-Serie), Section IV-no 1, 5-15. - - & MOGUEDET,G. 1974. La mati6re organique dans les s6diments du plateau congolais, facteurs de distribution, cons6quences sur les authigen6ses min6rales. Annls Univ. Brazzaville., T 10 (C), 15-29. & ODIN, G.S. 1973. Nature min6ralogique et origine des glauconies du plateau continental du Gabon et du Congo. Sedimentology, 20, 457-188. , KOUYOUMONTZAKIS,G. & MOGUEDET, G. 1979. Le quaternaire superieur du plateau continental congolais-exemple d'evolution paleooceanographique d'une plate-forme depuis environ 50 000 ans. In: van Zinderen Bakker, E.M. & Goetzee, J.A. (eds), Palaeoecology of Africa. Balkema Publishers, Rotterdam. 193-217. , JANSEN, F., KOUYOMONTZAK1S,G. & MOGUEDET, G. 1982. Les fonds du plateau continental congolais et le delta sous-marin du fleuve Congo. Traveaux et Documents de I'ORSTOM Nr., 138, 13-45. GORSLINE, D.S. & EMERY, K.O. 1959. Turbidity current deposits in San Pedro and Santa Monica Basins off Southern California. Bull geol. Soe. Am., 70, 279-90. HART, T.J. & CURRIE, R.I. 1960. The Benguela Current. Discovery reports., 31, 123-298. HEEZEN, B.C., MENZIES, R.J., SCHNEIDER,E.D., EW1NG, W.M. & GRANELL1,N.C.L. 1964. Congo Submarine Canyon: Bull. Am. Assoc. Petrol. Geol., 48, no. 7, 1126-49. HESSE, R. 1975. Turbidites and non-turbiditic mudstones and Cretaceous flysch sections of the Eastern Alps and other basins. Sedimentology, 22, 387-416. -
I 13
JANSEN, J.H.F., VAN WEERING, TJ.C.E., GIELES, R. & VAN IPEREN, J. 1983. Middle and Late Quaternary paleoceanography and climatology of the Zaire (Congo) fan and the adjacent eastern Angola Basin. Neth. J. Sea Res., 17, (in press). LEHNER, P. & DE RUITER, P.A.C. 1975. Structural history of Atlantic margin of Africa. Bull. Am. Ass. Petrol. Geol., 61(7), 961-81. VAN DER LINGEN, G.J. 1969. The turbidite problem. N.Z. Jl. Geol. Geophys., 12, 7-50. PIPER, D.J.W. 1978. Turbidite muds and silts on deep-sea fans and abyssal plains. In: Stanley, D.J. Kelling, G. (eds), Sedimentation in Submarine Canyons, Fans and Trenches. Dowden Hutchinson & Ross, Stroudsburg, Pa. RUPKE, N.A. 1975. Depostion of fine-grained sediments in the abyssal environment of the Alg6ro-Balearic Basin, Western Mediterranean Sea. Sedimentology, 22, 95-109. -t~ STANLEY, D.J. 1974. Distinctive properties of turbiditic and hemipelagic mud layers in the Alg6roBalearic Basin, western Mediterranean Sea. Smithson. Contr. Earth Sci., Washington, 13, 40pp. SHEPARD, F.P. & EMERY, K.O. 1973. Congo Submarine Canyon and Fan Valley. Bull. Am. Ass. Petrol. Geol., 57 1679-91. SIESSER, W.G. 1980. Late Miocene Origin of the Benguela Upwelling System off Northern Namibia. Science, 208, 283-85. STANLEY, D.G. 1981. Unifites: structureless muds of gravity flow origin in Mediterranean Basins. GeoMarine Letters, (1), 77-83. STOW, D.A.V. & SHANMUGAM, G. 1980. Sequence of structures in fine-grained turbidites: comparison of recent deep-sea and ancient flysch sediments. Sed. Geol., 25, 23-42. WOST, G. 1935. Schichtung und Zirkulation des Atlantischen Ozeans II. Die Stratosph~ire. Wiss. Ergebn. D.A.E. Meteor VI, (1) 288 pp. ZANEVELD, J.R.V., SPINRAD, R.W. & MENZIES, D.W. 1979. Optical and hydrographical observations in the Congo river and the Angola basin during May 1978. Oregon School of Oceanography Data Report, Ref. 79-3. 202 pp.
TJ.C.E. VANWEERING• J. VANIPEREN,Netherlands Institute for Sea Research, PO Box 59, Texel, The Netherlands.
Sedimentary sequences on the north-west Mediterranean margin during the Late Quaternary: a dynamic interpretation A. Monaco and Y. Mear SUMMARY: The methods we have used in this study of pelitic sediments are based on the concept of fine-grained particle populations, and include: the counting of grains in the > 40 ~m fraction (lithoclastic, biogenic, authigenic and aggregates); grain-size analysis of the pelitic fraction expressed as a bi-logarithmic frequency curve giving the 'evolution index' of three granulometric classes--silt, clayey-silt and clay; mineralogical indices based on the ratios between particles with different hydrodynamic behaviour--quartz, mica-illite and smectite; and geochemistry of Mn, Cu and Fe to characterize the biogenic nature of the sediment. A succession of turbiditic (heterogeneous population), hemipelagic and pelagic (homogeneous population) sediments were examined from the NW Mediterranean margin. We identify three main sequences in the Pleistocene to Recent succession, each of which may be interrupted by episodic events: (1) a palaeoclimatic sequence ( ,-- 5 m thick) of Pleistocene to Holocene age, marking the end of deep channelized lobe construction on the Rhone deep-sea fan; (2) a dynamic sequence (repeated intervals of 20-50 cm thick) of Wfirmian and Holocene age respectively, in the lower fan and in the central basin: and (3) a uniform sequence (without granulometric, mineralogical or geochemical grading) corresponding to a ponded unit filled with reworked sediments (contourites) and slumps. Using several kinds of analysis, many authors have attempted to recognize the various divisions of sedimentary sequences and to establish a nomenclature for them (Rupke & Stanley 1974; Piper 1978; Stow & Shanmugan 1980; Stanley & Maldonado 1981). Their recognition in varied environments has often led to the association of these sequences with their fossil counterparts (e.g. Bouma 1972; Mutti & Ricci Lucchi 1978). The time-scale of such sedimentary phenomena remains poorly understood. In the present study, analysis of the coarsefraction composition, sediment grain-size, mineralogy of the < 2 #m fraction and the geochemistry (Fe, Mn, Cu) of the < 40 #m fraction of pelitic sediments has allowed the definition of various pelagic, hemipelagic and turbiditic facies. This method is applied to deposits from different physiographical domains (slope, deep-sea fan and basin) of the NW margin of the Mediterranean Sea. Based on these results and on morphological, structural, palaeographic and ecological considerations, we reconstruct the sedimentation mechanisms that have operated during the last Pleistocene-Holocene glacial-interglacial cycle. A standard sequence is recognized and its palaeoclimatic and dynamic significance are discussed.
Methods The methods used are specially adapted to the
fine-grained nature of the pelitic deposits studied. The principle of these methods, both granulometric and mineralogical, is based on the idea of populations of particles and their respective behaviour. Composition of the > 40/~m fraction
The overall proportion of this fraction in the sediments studied is generally low, but its composition is often revealing. This includes heavy minerals, quartz, feldspar, mica, calcite, microfauna and macrofauna (Fig. 1). Particular attention is given to two kinds of component that are indicative of different facies: clastic or biogenic aggregates (faecal pellets) and the products of diagenesis. They have been studied using a scanning electron microscope to reveal their fine structure and a microprobe for their chemical composition. Granulometry of the < 40 pm fraction
The grain-size distribution of a sediment plotted as a semilogarithmic graph or a 'canonic graph' (Rivi~re 1952) allows an estimation of the degree of 'evolution' or sorting of the sediment. The method using the slope of a bi-logarithmic frequency graph (Rivi~re 1960) has the advantage of permitting the calculation an 'evolution index' for a given granulometric class (Desprairies 1974). A facies can be defined from: the relative behaviour of three size fractions, silt (10-40/~m), clayey-silt II 5
I 16
A. Monaco
and
Y. Mear
FIG. 1. Vertical profiles of the pelitic facies for different grain-size fractions. Left, > 40/~m, relative proportions of elastic, biogenic and authigenic grains; Centre, < 2/~rn, variation in smectite/illite (rl) and illite/quartz (r2) ratios; Right, > 40 #m, variation in Mn, Cu and Fe.
(1-10 #m) and 'clay' ( < 1 pm); and the degree of sorting or 'evolution index' of each fraction in ascending order, unsorted facies (n=0), parabolic facies ( - l < n < 0 ) , logarithmic facies (n = - 1), and hyperbolic facies (n < - 1). In Fig. 2 we show an example of three typical facies. The fine-grained turbidite is a heterogeneous deposit composed of three differentlyevolved fractions. The silt fraction has a hyperbolic facies (n=2.8), the clayey-silt has a parabolic facies (n = - 0 . 5 ) characteristic of sediments deposited by excess loading, and the clay fraction is apparently unsorted, or has undergone subsequent disaggregation. The hemipelagite and pelagite both show hyperbolic facies for the silt fraction (n = - 2 . 4 or a little less), whereas the clayey-silt is logarithmic (n = - 1) for the hemipelagites, and hyperbolic (n = - 1.2 to - 1.4) for the pelagites. The clay fraction is parabolic ( n = - 0 . 7 ) and hyperbolic ( n = - 1.4) respectively, showing improved sorting for the hemipelagites, often coinciding with an enrichment in the finest clay (smectite). In fact, the graphs of bi-logarithmic frequency reflect the population dynamics of grains with different properties (shape, form, density) and the greater or lesser homogenization of terrigeneous
material. The degree of evolution or sorting of particles is a function of the energy and duration of transport. Horizontal or vertical variations in these indices allow us to follow the evolution of a sediment type of process in terms of its continuity or discontinuity.
Mineralogical criteria The method of mineralogical indices uses the ratios between characteristic peaks obtained by X-ray diffraction on oriented aggregates from the < 2/~m fraction. We have used minerals which have different hydrodynamic behaviours due to their different shape or size: smectite (Sin), illite (I), and quartz (Q) (Monaco 1981; Monaco et al. 1982). We can thus define rl = S m / I =
smectite 17 A (001) illite 10 A (001)
illite 5 A (002) r2 = I / Q = quartz 4.2 A (100) When these ratios change in parallel through a sedimentary section they indicate a homogeneity of the clay fraction, the absolute value being greater when the selection of clayey particles by
Sedimentary sequences on the north-west Mediterranean margin -I-0
160
FIG. 13. Distribution of the azimuth of maximum magnetic susceptibility axis for successive stratigraphic intervals. All cores combined.
CANARIES
0
s
I STADE 2
CV1 011~ ~ J,,~U N E HOLOCE[
4,'~ .3AP
/ BLANC
FIG. 14. Probable pathways for turbidity currents across the N W African margin to the study area (CVI) and adjacent areas (from Sarnthein & Diester-Hass 1977).
I65
Sedimentation in the southern Cape Verde Basin 2
Age 0
2
0
4
6
8
~ _ _ L . - - _ _ I L
I0
I
9
12 m.
I
I
I
(103 71
GO -[-
ol c~I[
Z
zl
n,"
zl
nn
500
- -
KS
7
....
KS
9
---
KS
tl
-------. . . .
KS KS KS
12 13 14
%..
9..
\ .9
\
"..\
.
N ',
"~.
"~ \
\
\
\ \
\ \
Cr~l
\
- - i
\
'\
\
\
\ \ \
\
\
i
\ -
1500
\
\ \
I--
\
\
I | I I I I
Fic. 15. Sedimentation rates of cores from area CV1.
reversed magnetic polarity interval. A hiatus of the same age has been reported by Pujol et al. (1976) in a core from the Great Meteor Seamount. This may well correspond to higher bottom current activity, perhaps related to intensification of the circum Antarctic Bottom Water current, which is known to affect bottom circulation in many parts of the oceans (Huang & Watkins 1977).
Conclusion This detailed study of a small area on the lower rise off N W Africa has shown the complexity and variability of sedimentation in an area of lowrelief open ocean. Despite the flatness of the sea floor, areas of deposition and erosion co-exist within a few kilometres of each other. Low-relief channels are filled by both fine-grained hemipela-
166
G.A. A uffret et al.
gic sediment and thick sand layers. Accurate reconnaisance of this complex interfingering of sediment facies necessitates a detailed deep-tow investigation. This complexity is related to the interplay of sedimentary processes, which include gravity-controlled hemipelagic deposition, high-
density and high-velocity turbidity currents emplaced during periods of lowered sea-level, and m u d d y contourites c o m p o s e d mostly of calcareous clayey m u d emplaced between 1 500 000 and 900 000 yrs BP and, to a lesser extent at a b o u t 500 000 yrs BP.
References ABBOTT, D.H., MENKE, W., HOBART, M., ANDERSON, R.N. & EMBLEYR.W., in press. Processes influencing slope stability in marine sediments. J. geophys. Res. ARTHUR, M.A. & YON RAD, O. 1979. Early Neogene base-of slope sediment at site 397, DSDP Leg 47A: sequential evolution of gravitative mass transport processes and redeposition along the northwest African passive margin. In: von Rad, U. & Ryan. W.B.F. et al., Init. Rept. DSDP, 47, part 1. US Govt. Print. Off., Washington, DC. 603-618. AUFFRET, G.A., SICHLER, B. & COLENO, B. 1981. Deep-sea sediments texture and magnetic fabric. indicators of bottom currents regime. Oceanologica Acta, 4, No. 4, 475-88. - - & SICHLER,B. 1982. Holocene sedimentary regime in two sites of the north-eastern Atlantic continental slope. Bull. Inst. Gdol. Bassin d'Aquitaine, Bordeaux, 31, 181-194. BE1N,A & FUTreRER, D. 1977. Texture and composition of continental shelf to rise sediments off the northwestern coast of Africa: an indication for downslope transportation. "Meteor' Forch. Erg., C. 27, 46-74. BELDERSON,R.H. & LAUGHTON,A.S. 1966. Correlation of some Atlantic turbidites. Sedimentology, 7, 103-16. CROWLEY. T.J. 1981. Temperature and circulation changes in the eastern North Atlantic during the last 150000 years: evidence from the planktonic foraminifer record. Mar. Micropalaeont., 6, 97-129. DIESTER-HAAS,L. 1979. DSDP Site 397: Climatological, sedimentological and oceanographic changes in the Neogene autochthonous sequence. In: Von Rad, U. & Ryan, W.B.F. et al., Init. Rept. DSDP, 47, part 1. US Govt. Print. Off., Washington, DC. 647-70. - - 1 9 8 0 . Upwelling and climate off northwest Africa during the Late Quaternary. Palaeoecol. Afr., 12, 229-38. DILLON, W.P. & SOUGY, J.M.A. 1974. Geology of west Africa and Canary and Cape Verde Islands. In: Nairn A.E.M. & Stehli, F.G. (eds), The Ocean Basins and Margins, 2, The North Atlantic. Plenum Press, New York, London. 315-90. EMBLEY,R.W. 1975. Studies of Deep-Sea Sedimentation Processes Using High Resolution Seismic Data. Ph.D. Thesis, Columbia University. New York. 334 PP. & JACOBI, R.D. 1977. Exotic Middle Miocene sediment from Cape Verde Rise and its relation to piercement structure. Am. Ass. Petrol. Geol. Geol. Notes, 2004-9. 1982 Anatomy of some Atlantic margin sediment slides and some comments on ages and mechanisms. -
-
-
-
In: Saxov, S. & Niewenhuis, J.K. (eds), Marine Slides and other Mass-Movements. Plenum. Press, New York and London. 189-213. HUANG, T.C., & WATKINS, N.C. 1977. Contrasts between the Bruhnes and Matuyama sedimentary records of bottom water activity in the South Pacific. Marine Geol., 23, 113-32. JACOBI, R.P. & HAVES,D.E. (1982). Bathymetry, Microphysiography, And Reflectivity Characteristics of the West African Continental Margin. Sierra Leone to Mauritania. In: yon Rad, U. et al. (eds), Geology of the Northwest African Continental Margin. Springer-Verlag, Berlin. 182-212. KOOPMAN, B. (in press). Sedimentation von Saharastaub in subtropischen Atlantik w/ihrend der letzten 25 000 J~ihre. 'Meteor'Forsch. Erg., C. 35. LABRACHERIE~ M. 1980. Les radiolaires tbmoins de l'6volution hydrologique depuis le dernier maximum glaciaire au large de Cap Blanc (Afrique du Nord-Ouest). Palaeogeo. Palaeoclimat. Palaeoecol., 32, 163-84. LA BRECQUE,D., KENT, V. & CANDE,S.C. 1977. Revised magnetic polarity time scale for late Cretaceous and Cenozoic time. Geology, 5, 330-5. LANCELOT, Y. & EMBLEV, R.W. 1977. Piercement structures in deep oceans. Bull. Am. Ass. Petrol. Geol., 61, 1191-2000. LONSDALE,P. 1978. Bedforms and the benthic boundary layer in the North Atlantic: A cruise report of indomed Leg 11. SIO Reference 78-30. LUTZE, G.F. 1980. Depth distribution of benthonic foraminifera on the continental margin off NW Africa. 'Meteor' Forsch. Erg., C. 33, 31-80. - - , SARNTHEIN,M., KOOPMAN, B., PFLAUMANN,V., ERLENKENSER,H. & THIEDE, J. 1979. Meteor cores 12309: Late Pleistocene reference section for interpretation of the Neogene of site 397. In: Von Rad, U., Ryan, W.B.F. et al., Init. Rept. DSDP., 47, part l, US Govt Print. Off., Washington, DC. 727-39. MARTINI, E. 1971. Standard Tertiary and Quaternary calcareous nannoplankton zonation. 2nd Plankt. Conf. Proc. Rome 1970., 2, 739-85. MOLLER, C. & ROTHE, P. 1975. Nannoplankton contents in regard to petrological properties of deep-sea sediments in the Canary and Cape Verde areas. Marine Geol., 19, 259-73. PUJOL, C. 1980. Les foraminif~res planctoniques de l'Atlantique Nord au Quaternaire. Ecologie, Stratigraphie, Environnement. Mem. Inst. Gbol. Bassin d'Aquitaine, Bordeaux, 10, 254 p. --, DUPRAT, J. & PUJOS-LAMY, A. 1976. R~sultats pr61iminaires de l'&ude effectuOe par l'Institut de G~ologie du Bassin d'Aquitaine concernant la mis-
Sedimentation in the southern Cape Verde Basin sion MIDLANTE A dans l'Atlantique oriental entre 51 ~ N. et 35 ~ N. Bull. Inst. Gdol. Bassin Aquitaine, Bordeaux, 19, 3-32. SARNTHE1N,M. & DIESTER-HAAS,L. 1977. Aeolian sand turbidites. J. sed. Petrol., 47, 2, 868-90. --, THIEDE, J., PFLAUMANN,U., ERLENKEUSER,H., FUTTERER, D., KOOPMAN,B., LANGE, H. & SEIBOLD, E. 1982. Atmospheric and oceanic circulation patterns offnorthwest Africa during the past 25 million years. In: von Rad, U. et al (eds) Geology of the Northwest African Continental Margin. SpringerVerlag, Berlin, Heidelberg. 545-604. SHACKLETON,N.J. 1977. The oxygen isotope stratigraphy record of the late Pleistocene. Phil. Trans. Roy. Soc. Lond., 13, 280, 169--82. SCHMINCKE, H.V. & VON RAD, U. 1979. Neogene
I67
evolution of Canary Island volcanism inferred from ash layers and volcaniclastic sandstones of DSDP site 397 (Leg 47 A)./n: Von Rad, U., Ryan, W.B.F. et al., Init. Rept. DSDP, 47, part 1: US Govt. Print. Off. Washington, DC. 703-25. TmEDE, J. 1977. Aspects of the variability of the glacial and interglacial North Atlantic eastern boundary current (last 150 000 years). 'Meteor' Forsch Erg. C. 28, 1-36. THIERSTEIN, H.R., GEITZENHAUER,K.R., MOLFINO, B. SHACKLETON, N.J. 1977. Global synchroneity of late Quaternary coccolith datum levels: validation by oxygen isotopes. Geology, 5, V.S. 400-404. VANGRIESHEIM, A. & MADELAIN, F. 1982. Rapport contrat C.E.A. TMC.'14575 fitsc. 2, Oc&mographk' physique. (unpublished).
G.A. AUFFRET, R. LE SUAVE,R. KERBRAT& B. SICHLER,Centre Oc6anologique de Bretagne, B.P. 337, 29273 Brest, Cedex, France. S.RoY, C.LAJ, CNRS-CEA Centre des Faibles Radioactivit6s, B.P. l, 91190, Gif-sur-Yvette, France. C.MULLER, Consultante, 1 rue Martignon, 92500, Rueil Ma/maison, France. Contribution 824 du Centre Oc6anologique de Bretagne.
Sedimentary processes on the west Hellenic Arc margin
H. Got S U M M ARY: High-resolution seismic surveys across the western Hellenic Arc show that the
Plio-Quaternary cover is highly discontinuous along most of the margin. The slope displays a series of basins separated by ridges generally devoid of sediment. Sedimentological analyses of the recent sediment permit the origin and processes of sedimentation to be defined. The sediments provided by a specific area on land are successively trapped, released by the slope basins and transferred via channels and canyons towards the deeper trench basins. Several main provinces can be distinguished. Each of them includes one or more slope basins and trenches without either lateral connection or the mixing of sediment inputs. The diversity and complexity of sedimentary structures, the granulometric characters, and the variability in rates of sedimentation argue for the prevalence of mass gravity processes. The transfer of sediment from the upper slope basins to the outer trench is effected by an alternation of short periods of giant mudflows and regular but slight reworking between slope basins. This cascade-feeding process, characteristic of the Hellenic subduction margin, does not appear to prevent terrigenous material from reaching the deep trenches. We compare this briefly with sedimentation on passive margins in the western Mediterranean Basin.
This paper is a synthesis of earlier work that describes the distribution and nature of unconsolidated sediments covering the Ionian margin of Peloponnesus, between the islands of Zakinthos and Crete (Fig. 1), and attempts to identify sedimentation mechanisms. This margin represents the outer part of an active margin, which formed as a result of the convergence between the European and African plates in the region of the Ionian Sea and the Eastern Mediterranean basins. The data discussed here have been collected during four cruises (1974, 1976, 1977 and 1978) and include 4000 km of high-resolution seismic profiles, using a 3000-J sparker (IUN-Naples) and a 40 in 3 air-gun (Station de Geodynamique sous-marine, VillefranGhe), and 30 piston cores from the Peloponnesus margin and adjacent trenches. Previous studies of this area have primarily focused both on the. structure of the arc, including structural studies of the margin and Ionian trench (Le Quellec et al. 1978; Le Quellec & Mascle 1979; Le Quellec et al. 1980), and on its patchy sedimentary cover. Wong & Zarudzki (1969) and Ryan et al. (1971) have studied the general deposition and thickness of recent sediments, Ryan (1972) and Hinz (1974), their age and Pastouret (1970), Blanc & Blanc-Vernet (1971), Chamley (1971), Emelyanov (1972), Ryan (1972), Stanley & Knight (1978), Stanley et al. (1978), Vittori et al. (1981), Got et al. (1981), Blanpied & Stanley (1981), their nature.
Morphology and structural configuration The margin of the western Hellenic Arc is morphologically very complex, comprising a series of basins and terraces on the slope and isolated segments of a discontinuous trench at the base of the slope (Fig. l, Carter et al. 1972). Two major structural trends are evident, one at 140~ and the other at 70~176 These two trends divide the margin into numerous tectonically-controlled sedimentary units. From north to south the main basins and terraces are: (l) the Cephalonia-Zakinthos plateau (Hinz 1974) wedged between the Peloponnesus slope and Strophades Islands eastern slopes; (2) the Messenia and Laconia basins, separated by the Cape Matapan foreland; (3) and numerous smaller basins and sedimentary terraces primarily affecting the Kithira-Antikithira margin where a 10~ to 2 0 ~ trend, related to alpine tectonics is locally present (e.g. Cape Matapan). From north-west to south-east the trench segments are: The Zakinthos Trench (Got et al. 1977) and the North and the South Matapan Trenches. These trench basins, in particular the South Matapan Trench, are subdivided into a series of ponds ranging from 4200 to 4600 m in depth, and separated by ridges related to the base of the slope (Le Quellec et al. 1978; Vittori 1978; Le Pichon et al. 1979). A recent detailed seismic survey of the South Matapan Trench slope (Got 1981) together with a precision (Sea-Beam) bathymetric survey (Le
I69
I7o
H. Got
0
50km
PELOPONNESUS
AEGEAN
%
.,2o
SMT CRETE
Fig. 1. Chart of study area showing the core locations and DSDP sites 126, 127, 128. ZT = Zakinthos Trench; NMT =North Matapan Trench; SMT = South Matapan Trench; KB= Kiparissia Bay; MB = Messenia Bay; LB = Laconia Bay. Bathymetry in metres. Pichon et al. 1979) enable the relationship between slope basins and trenches to be more accurately defined, and it appears that the basins and trenches are connected by a network of tectonically-controlled channels and canyons which are generally devoid of sediments.
Stratigraphy Most of the seismic air-gun and sparker profiles show two major acoustic units (Fig. 2): a lower acoustic unit and a stratified sedimentary cover. These two units display different acoustic characteristics in different areas.
Deep stratigraphic unit South-west of the trench, along the Mediterranean Ridge, the lower acoustic unit is a severalhundred-milliseconds thick stratified sequence
(Fig. 2). Previous seismic (Wong & Zarudzki 1969: Hinz 1974) and sedimentological studies (Biscaye et al. 1972) relate this unit to the top of the Messinian (M Reflector). This series was cored on the Mediterranean Ridge during DSDP legs 13 and 42A at sites 125 and 374, respectively; the sediments recovered at site 126 contain Messinian evaporitic series (Ryan et al. 1973) and middle Miocene sediments. Furthermore, a submersible study of the Hellenic Trench (Le Pichon et al. 1981) has revealed outcrops of probably evaporitic nature attributed to the Messinian. The penetration obtained with our seismic systems suggests such an evaporite sequence. Locally, particularly in the zone of cobblestone morphology, this lower unit loses its stratified character. The transition between these two characters is gradational and can be best explained by active deformations affecting the whole sedimentary sequence.
Sedimentary processes on the west Hellenic' Arc margin
~7I
0
0
~9
o o
O
8
g
8
0
.~
9
,.~
t",lr,~
I72
H. Got
The lower unit along the inner wall of the trench is generally considered as an acoustic basement composed of alpine consolidated rocks. In slope basins, between the acoustic basement and the unconsolidated sedimentary cover, there are a series of parallel high-amplitude reflections, called unit C. This unit can be tentatively attributed to Miocene sediments trapped in subsiding basins. Upper stratigraphic units
The overlying sedimentary section includes two acoustically distinct units with different geometrical distributions. The lower acoustic unit (termed B) is characterized by weak internal reflections with poor lateral continuity. For this reason, the unit is referred to as a 'transparent layer'. However, near coastal areas, and particularly near areas of sedimentary input, there are more internal low and medium amplitude reflectors, although in areas of thin sediment cover, the series conserves its weakly reflective character. This unit generally overlies the deep units (S, M or C); the unconformity between the basement and unit B is related to a Miocene-Pliocene erosional surface, although on the Mediterranean Ridge, the two units are concordant. Structures which are interpreted as normal faults, commonly linked to faulted underlying basement, are observed within this weaklyreflective unit. In contrask the upper sedimentary unit (termed A) is characterized by parallel, highamplitude reflectors. Typically, these lie horizontally, and offsets of the order of 100 ms occur on the Zakinthos terrace, across vertical faults. Based on the seismic reflection profiles of the Mediterranean Sea, the results of the DSDP drilling at sites 125, 126, 127 and 128 (Ryan et al. 1973) and the structural configuration of the unconsolidated cover on land (Sorel 1976; Angelier 1976; Kelletat & Schroeder 1976: Dufaure 1977; Mercier et al. 1978) the acoustic stratigraphy can be dated as follows: Unit
Seismic character
Lithology
Age
Unit A
internally reflective unit weakly reflective unit unit
mud, sand, ash mud, marls
Quaternary
Unit B
Pliocene
Sediments Distribution
Mapping of the sediment thickness reveals continuity of fill from basins to trenches. The margin comprises a series of slope basins connected by narrow channels generally devoid of sediment.
These connected basins determine sedimentary units extending from a specific source terrain on land to a trench basin. Three main units can be distinguished (Fig. 3): Kiparissia basin to North Matapan Trench; Messinia and Laconia basins to northern basin of the South Matapan Trench; Kithira Strait to southern basin of the South Matapan Trench. In each of these connected basin systems the sediments contain lenticular layers, slump scars, gravitational slides, tensional scars, perturbed morphology and compressionally contorted beds (Fig. 4), all of which indicate that a large proportion of sediments have been reworked by mass gravity processes. Based on sedimentological and 3.5 kHz data, a ~giant' mud flow emplacement mechanism has been proposed by Stanley & Knight (1978) in the Hellenic Trench system. The overall nature of this sedimentation is referred to as cascade feeding (Got et al. 1981; Vittori et al. 1981) in which the continental slope both traps and/or serves as a conduct for sediments. The tectonic influence is clearly marked, not only in the erosion of the sedimentary units but also in the development of the valley and canyon network that funnels the sediments through the slope basins to the trenches. In the basin-valley system across the Kithira margin (Fig. 5), the deep-stratigraphic unit or acoustic basement (S) is overlain by a thick and relatively transparent upper unit. This upper unit contains a thin stratified series sandwiched between the more acoustically-transparent layers. This series is present in the successive slope basins from 1000 to 3000 m depth and can be identified as a key horizon of probably sandy composition. However, we note a change in the acoustic character of the stratified series. In the shallower basin, it is well marked by strong reflectors, whereas downslope its character tends to attenuate and its thickness increases. This evolution can be interpreted as a dilution of the reflective unit with the progressive reworking of sediments from basin to basin. General lithology
The study area is mainly located in a pelagic domain, and biogenic muds are predominant. However, within this superficial cover, two more distinctive lithotypes have been recognized: sapropels and ash layers. The sapropels were formed during 3 major episodes, at 7900 yrs, 23-25000 yrs and 30-38 000 yrs m~ (Nesteroff 1973; Stanley et al. 1978). The carbon content increases from the most recent layer (1.2-1.8%) to the most ancient (2.3-3.8%). These layers are present in about 50%
S e d i m e n t a r y p r o c e s s e s on the w e s t H e l l e n i c A r c m a r g i n
I73
FIG. 3. Reconstruction of the provenance and processes of sedimentation on the south-western Peloponnesus margin. This map shows: (a) the major sedimentary basins on the slope as delineated by the 200 ms isopach (after Vittori et al. 1981); (b) the main mineralogical provinces based on heavy and clay minerals; (c) the origin of inputs and the main axes of sediment dispersal (arrows). UCZB = Upper Cephalonia-Zakinthos Basin; LLB = Lower Laconia Basin; LCZB = Lower Cephalonia-Zakinthos Basin; KB = Kithira Basins; MB = Messenia Basin; WCB = Western Cretan Basin; of the cores, mainly on the continental margin. Their thickness varies from several centimetres in core 3 to 3 m in core 6 (Fig. 6). In addition to their high organic-carbon content, these sapropelic layers are characterized by mainly pyritic mineralization. The problems of their origin is not yet fully solved (Sigl et al. 1978). Preliminary results, based on element analysis, carbon content, and infrared spectrometry suggest that the organic matter is essentially of marine origin (Vittori et al. 1978). Two main ash layers have been recognized in
the Eastern Mediterranean: the Lower Tephra and the Upper Tephra, respectively dated at 25 000 and 3400 yrs BP (Ninkovitch & Heezen 1965). In cores from the south Peloponnesus margin, only the Lower Tephra occurs. The geographic distribution of the Lower Tephra is highly variable (Fig. 7); it is generally present on the slope, where its thickness is usually about 1 cm, but it can also occur as a succession of several centimetric layers. In channels (core 14, Fig. 6), ash is dispersed over more than 30 cm. This lithotype is rarely found in the trenches.
FIG. 4. Slumps and slump scars affecting the upper sedimentary unit in the Messenia Basin.
Sedimentary processes on the west Hellenic Arc margin
I7 5
FIG. 5. Line drawing of a seismic reflection profile across the Kithira margin showing the widespread correlation of the acoustic refectors. The arrows indicate the approximate core locations. In addition to these distinctive horizons, the biogenic muds show several variations (Fig. 6). Chromatic changes from yellow to grey are common and well marked. However, except for the black layers that result from the organic content, colour changes do not seem to be correlated with mineralogic or textural variations. The ash layers and biogenic accumulations form the main coarse layers. However, some cores retrieved in channels (core 14) or at channelmouths (core 11 l) show accumulation of detrital elements sometimes with graded bedding. We have also noted cross-bedding (core 3a), slumping affecting pieces of sapropels covered by horizontal deposits (cores 17 and 18), sharp contacts between muddy layers (core 6), mud and sapropel clasts, and changes of bed inclination in the same core (core 6).
Texture and composition The cores have been analysed for grain-size ( < 40 /~m), carbonate content, organic carbon, clay
minerals, and light and heavy minerals. Grainsize and mineralogy provide the most significant data. The percentage of the >40 /tm fraction is generally low. There is a slight decrease in the whole coarse fraction (biogenic and terrigenous) from the coastal gulfs to the trench basins. However, if the terrigenous content is considered alone, there appears to be no significant change in the percentage with depth (Fig. 8). One of the trench cores (111) contains more than 60~ terrigenous material. Similarly, the median grain-size does not change consistently with depth but is distributed according to the physiographic domain; the highest values (6-8/am) are encountered in transport zones (channels and canyons) and the lowest (1-2 /~n) are found in trenches. However, if we consider the evolution of the median grain-size in a sedimentary system (Fig. 9) such as the Laconia system (cores 14-12-11) or the Kithyra margin (cores 18-19), the values decrease slightly with depth. But it is noteworthy that the main pheno-
~76
H. Got
FIG. 6. Sections of cores showing the main lithological and structural features. The length is expressed in centimetres; f= foraminifera concentrations; s = sapropel layer; arrows indicate features discussed in text. menon is improved sorting with depth rather than a reduction of size. The cumulative grain-size curves (Fig. 10) are presented in a semi-logarithmic diagram. This permits the mode of deposition to be defined on the basis of the form of the curve (Riviere 1952). No sediment of the Peloponnesus margin, including trenches, shows a hyperbolic grain-size distribution characteristic of pelagic settling. On the
contrary, the form of this curve is generally parabolic, indicative of dominant gravity transport mechanisms. Clay minerals are dominated by illite (40-50%) followed by smectite and chlorite-kaolinite (30%); this association is characteristic of the Ionian Basin (Chamley 1971). From the quantitative standpoint, two main areas can be distinguished: (1) from Zakinthos to Cape Matapan,
Sedimentary processes on the west Hellenic Arc margin
I77
o
o
~9
0
0
0 0
2~
0
~2
o'~
o
o 2
I78
H. Got 9 Gulf and t r a n s i t zones
1oo2
9 slope basins
E9
9 Trenches
11'
q4
2O 9
12
50
36
4
6
9
5 9 18 9
nm
32
3
9
32
"19 9 9 35 34 9 9 9
37
9
9
22
2
3
4
5 x 1000m. Depth FIG. 8. Diagram showing the random variation of the detrital terrigenous content of the sediments versus depth. where the values of illite are around 50%; and (2) the Kithira area which is characterized by the highest values of smectite (30~I~) of probable Aegean origin (Venkatarathnam & Ryan 197 l). The clay-mineral assemblages cannot be correlated with either facies or colour; only the sapropelic layers show a decrease of smectite but this correlation has a diagenetic origin (Chamley 1971). Likewise, there are no systematic differences in clay mineralogy as a function of depth. ~5-
14
v
11
~4z _J oe"
o3 r
1'0
2'0
3b
20
AGE (X 1000 years)
FIG. 9. Evolution of the mean size of the pelitic fraction of sediments versus depth in two sedimentary units. Cores 14-12-11: Laconia Basin--Northern Basin of South Matapan Trench; Cores 18-19: Kithira Ridge--Southern Basin of the South Matapan Trench. Full lines: trenches. Dashed lines: Slopes.
On the basis of the heavy-mineral frequency distribution in the studied samples, four assemblages have been identified. These contain the following mineral species, listed in decreasing order of abundance: (1) garnet, spinel, and epidote (cores 8, 3, 1); (2) chloritoid, garnet, epidote, and blue amphiboles (glaucophane and crossite) (cores 4, 6); (3) epidote, chloritoid, and blue amphiboles (cores 17, 12, 14, 112); (4) magnesite, epidote, chloritoid, green hornblende (cores 19, 110. These associations, together with the clay mineral assemblages outlined above, allow us to define the three mineralogical provinces shown in Fig. 3. (a) Province 1 comprises Kiparissia Bay, CephaIonia-Zakinthos plateau, and North Matapan Trench. It is characterized by garnet, spinel and illite derived from the Alphee basin (Kiparissia Bay). (b) Province 2 includes Messenia and Laconia basins and the South Matapan Trench. It is defined by an association of chloritoid, epidote, glaucophane, illite, smectite and chlorite and the main source area is the Eurotas Basin (Messenia and Laconia Bays). (c) Province 3, coinciding with the Kithira area, is marked by an association ofchloritoid, epidote, glaucophane, hornblende, magnesite and smectite; this assemblage occurs in the slope basins of the Kithira Margin and in the eastern segment of the South Matapan Trench. The authigenic magnesite is probably linked, as well as the smectite, to the exchange of water masses across the Kithira Ridge (Lacombe & Tchernia 1972; Stanley & Perissoratis 1977).
I79
Sedimentary processes on the west Hellenic Arc margin 100~
/ < . 51'o MgCO3) with only minor amounts of calcite (Milliman 1974; Bathurst 1975). In such shallow, sunlit seas, fine-grained, metastable carbonate sediments are organically produced, primarily by calcareous algae (Neumann & Land 1975). In contrast, fine-grained pelagic carbonate sediments of the deep-sea consist almost entirely of calcite with only minor amounts of aragonite (Garrison 1981). These carbonate sediments are also produced organically; calcite by planktonic foraminifera and coccolithophorids, and aragonite by pteropods. The only deep-sea environments where aragonite and/or magnesian calcite are suspected of being precipitated inorganically as primary grains are the warm, hypersaline environments of the Mediterranean and Red Seas (Milliman et al. 1969; Milliman & Mi.iller 1973). Thus, in modern, open oceans there are two mineralogically distinct types of fine-grained carbonate sediments. However, around the margins of shallow-water carbonate platforms these two mineralogically unique sources of carbonate sediment mix to create yet a third type of fine-grained carbonate sediment, referred to as 'periplatform' ooze (Schlager & James 1978). Although we have been aware of periplatform ooze sedimentation in deep-water Bahamian troughs for some time (Pilkey & Rucker 1966; Rucker 1968; Kier & Pilkey 1971; Lynts et al. 1973), there have been very few studies that have attempted to quantify the contribution of bankderived versus pelagic-derived fine-grained sediment. Boardman (1978) first approached this
problem from a geochemical perspective for Holocene sediments in North-west Providence Channel, Bahamas. He concluded that more than 8000 of the total mud fraction is bank-derived, and that pteropods, the only known pelagic source of aragonite, do not contribute significantly to the fine-grained fraction. Thus, the two major sources of fine-grained periplatform oozes can readily be distinguished, simply on the basis of carbonate mineralogy. The very high percentages (80 + %) of bank-derived fines in North-west Providence Channel surface sediments may at first seem surprising, unless one considers the tremendous volumes of sediment being produced by calcareous algae on tropical shallow-water platforms. Neumann & Land's (1975) budget study demonstrated that calcareous algae alone are not only capable of producing all fine-grained carbonate sediments in the Bight of Abaco, a typical Bahamian lagoon, but that there is an overproduction of carbonate sediment by a factor of 1.5 to 3 times. Boardman's (1978) study convincingly showed that much of this excess shallow-water produced sediment is deposited along adjacent deep-water slopes and basins. However, Boardman's (1978) study was conducted on samples collected from North-west Providence Channel, an interplatform trough which receives fine-grained shallow-water carbonate sediment from both Little and Great Bahama Banks. Thus, it is still not known how far laterally fine-grained shallow-water carbonate sediments can be transported along open-ocean settings before being sedimented, and what compositional changes occur with increasing distances from their source. I99
K.C. Heath and H.T. Mullins
200
To help answer such questions, we have conducted a detailed study of fine-grained carbonate sediments along the open-ocean northern margin of Little Bahama Bank (Fig. 1). Our study area extends parallel to the northern margin of central Little Bahama Bank for a distance of approximately 100 km and extends 'offshore' for a distance of approximately 50 km. Our results confirm Neumann & Land's (1975) and Boardman's (1978) suggestions that large volumes of bankderived fines are presently being deposited along deep-water Bahamian slopes and basins. Our results also suggest that lateral transport in excess of 100 km occurs, and that the composition of periplatform muds varies in a linear fashion with distance from the shallow-water source.
Methods Thirty-five surface sediment samples were col-
lected from the slope north of Little Bahama Bank in July-August 1979 during cruise E-3A-79 of Duke University's R/V Eastward (Fig. 2). Twenty-four of the samples were collected by a Shipek grab sampler, and 11 samples were collected from the top 5 cm of pilot cores at piston core localities. Samples collected from the axes of submarine canyons, or from pilot cores that indicated evidence of sediment gravity-flow deposits at the sea floor, were excluded from our present analysis to minimize any possible effects of resedimentation by gravity-flow processes. Thus, we assume that our samples are mostly representative of only very recent hemipelagic sedimentation. However, at some grab sites it is possible that we have sampled parts of finegrained non-channelized turbidity-current deposits. Subsamples from each station were first soaked in chlorox to remove organics and various size fractions separated by wet sieving, isolating the 76 ~
78 ~
80 ~
i i
BLAKE
~
28 ~
LATEAU
NORTH
T
o o oJ
0
50 km
28 ~
INNNrn
LITTLE~ -..-(..:?
BAHAMA r ~BANK
9
n
200m
GREAT
BAHAMA~ BAN K
80 ~
25 ~
78 ~
76 ~
FIG. ]. Index map and generalized bathymetry of the northern Bahamas illustrating location of study area.
Open-ocean, off:bank transport o/i/ine-grained carbonate sediment
201
r~
%
,ez o~
.r
E
E
E ~5
",o~
(1)
o~
r-
~)
E
0
to
0
7-
~..
'%~
0
0
6 C
o
FIG. 3. 3.5 kHz profile record across the Obock Trough near to the KS 01 position (after Clin & Pelissier-Hermitte 1981).
FIG. 4. Lithologv of core KS 01 (after Moves et al. 1981).
Late Quaternary calcareous clayey-silty muds in the Obock Trough O
g._L 0ll,
--
001,
--
o~ ^ -Iv.
~
o
--
08
--
0Z
--
09
--
0g
--
0t;,
--
0s
--
-
-
e-
"--
J
~ e~
A
--
0 Ol
o00
E
Q;
DJaJ!U!LUDJOJ. 0!q~,uaq
--
00~
L
'~ :>Z
......'~
O00g
"O
A L
--
O
~
:n
~F::, .2.~ > ._,.2 E9 E
--
0~
::n
.s
"Z: "a
.~O
0~
2E~
G) I
"E 06
213
,&
O
o
r
(..)
i~.~
z
000
--
--
0o~ o
--
Oi
z
//j tA/L-
,
Oi
n
A
--
o00g
000
spodXoel~d
--
00~ 0
n"l
spodo~a~,d
N I ~'-
- -
OOOg
E~ .~
t-
w
--
ooog
.,..
0
r
I
o!UOPlUOld oo~
- -
,
,
o
g
"~
,
o
=_
\
--
~0
]
I
I
I
T
T
1
c:O0 ~ 9
r
~1) g~
'-g
~ 0 r
C: 0 ---
I!
IIIIitl
Jil - !11
o "'"
'"-
~ ~
~
~ "^
,'," w o
"~
z
r.:
,~'E
',-?;
N
,
..
z w ~o
.
'-~
g
E~
rS~
214
J.-C. Faugbres et al.
sian minerals), which outcrop around the Gulf of Tadjoura (Faug6res & Gonthier 1981: Faug+res et al. 1983). The Obock Trough sediments appear therefore to be a typically pelagic or hemipelagic facies. However, pelagic settling from the surface cannot explain the high sedimentation rates observed nor their important lateral variations, for there is a scarcity of continental input within the desert climatic zone and relatively low productivity in an area free of any upwelling effects. How, then, were these sediments emplaced?
Turbidite characteristics Detailed examination of one of the cores has revealed three sets of characteristics that suggest turbidity currents were responsible for the deposition of much of the sequence.
Sedimentary structures and microstructures of dynamic origin Three types of structures and microstructures have been recognized on X-radiographs and in
thin sections (Fig. 5; Fig. 6(a), 1, 2, 3; Fig. 6(b) and Fig. 6(c)). (1) Sharp often erosional contacts separate different-coloured muds. Overlying these contacts the sediment mostly shows horizontal or oblique plane lamination. It is richer in sand- or silt-sized particles and is sometimes graded. (2) Plane lamination is common at several levels in the core and mainly results from an alternation of laminae rich in sand-sized paralleloriented bioclasts with laminae rich in finegrained matrix and with smaller bioclasts. These laminae may be horizontal or oblique. (3) Parallel orientation of bioclasts or of siltsized terrigenous particles along horizontal or oblique laminae. All these structures are indicative of bottom dynamic processes. Moreover, the recurrence of laminated levels through the core suggests a periodic dynamic process.
Cyclicity in the abundance and composition of the sand fraction Analysis of the sand-fraction content through the core shows weak cyclic variation in abundance (Fig. 5). The 'peaks' correspond to an increase in
FIG. 6(a) Above: main components of sand (left) and silt (right). Below: X-radiographs of thin slabs of core KS01 (core width 5 cm); (1) erosive contact and lamination, (2) fine-scale lamination, (3) sharp contact and lamination.
FIG. 6(b) Thin-section photographs (maginification 14.8 x ), core KS01: division I, horizontal lamination alternately rich and poor in sand-sized biogenic material; (2) division I'b, horizontal lamination of silt-sized biogenic material, less clear than (1); (3) division I'a, oblique lamination and abundant terrigenous material. White arrows show way up.
A
v
FIG. 6(c) Thin-section photographs (magnification 14.8 x ), core KS01 (1) division II, slight orientation of silt-sized grains; (2) division III, no sedimentary structures, F = foraminifer; (3) sharp contact between divisions I and III: (4~ erosive contact between divisions I'a and III.
Late Quaternary calcareous clayey-silty muds in the Obock Trough
aoojaJ0lAs
1>2~-::.:':'.'::.:.:.:.:...,:.'-:.'.51
~
~laqs ad01s
~
:
+
,
--
--
--
--
--!':-.'.
--
......... ++. . . . . . . . .
:"..'...'.-.:.'..:"..'!
.::.: .:-.'. :: . .: 9. :.... :' :.. :::-.::.
I~.}:.:.::::
~
............ ,. . . . . .
......
:':.
.
.1
LL 0
(J
15 : t a = 20.0 Inclination, declination following demagnetization to 23874 A/m.
.18 .40 .ll .13 .23 .33
h 8.4% 8.3 6.6 34.4 57.5 48.2 50.7 50.8 41.0
272
A.N. Shor, D.V. Kent and R.D. Flood
during deposition. However, four samples within Unit I of core RC 22-14 show alignments consistent with contour-current flow rather than turbidity currents. (5) Silt-laminated sequences in core RC22-02, which comprise the entire 6 m long core, show a magnetic fabric alignment which is consistent with contour-current flow. Silt layers in core RC22-02 are generally thicker and less closely spaced than in core RC22-14, and the core is located beneath the present core of a strong thermohaline current, whereas RC22-14 is near the upslope margin of this flow. (6) Samples from core RC22-15 do not show a clearly defined preferential alignment related to either downslope or alongslope flow conditions. At the present time, the strongest alongslope currents are restricted to depths greater than this core. While it is possible that some complex interaction of downslope and alongslope processes results in a 'smearing' of the directional information preserved in AMS, we prefer to leave interpretation of these data until such time as unambiguous results relating lithofacies and depositional processes in other settings provide a less subjective basis for discussing complexities. (7) Highest percent anisotropy values are observed in Unit III of the gravity core (downslope alignment) and in piston core RC22-15 (no clear preferential alignment) rather than in core RC22-02 which shows the strongest alongslope alignment. Anisotropy values are low in Unit 2 of core RC22-14 which shows signs of post depositional deformation.
Conclusions We believe that the anisotropy of magnetic susceptibility technique (AMS) holds considerable promise in differentiating downslope and alongslope depositional processes of fine-grained sediments on the Nova Scotia continental rise. Our present results from four cores in general support the hypothesis of Hollister (1967) that contour-following bottom currents have played a role in sediment redistribution on the continental rise during the glacial stages of the Pleistocene. AIongslope currents are also important depositional agents in the Holocene. However, our preliminary results demonstrate that downslope magnetic fabric alignments occur as frequently as alongslope alignments. ACKNOWLEDGMENTS: Support for AMS studies was provided by the US National Science Foundation (OCE80-12897); cores were collected through support from US Office of Naval Research (T0210-82-11111). Archive facilities at L-DGO are supported by NSF through OCE81-22083. We are grateful to both agencies for their support of this work. N. Iturrino assisted in collection of gravity cores, B. Tucholke collected piston cores. J. Kostecki and F. Hall carried out AMS and remanence measurements. J. Broda loaned the gravity corer. V. Kolla was instrumental in obtaining support for this project. E. Free and C. Elevitch typed the manuscript. This paper was reviewed by C. Hollister, N. McCave, J. Damuth, W. Ruddiman, D.A.V. Stow, A. Bouma and A.I. Rees. L - D G O Contribution//3621.
References AUFFRET, G.-A., SICHLER, B. & COLENO, B. 1981.
Deep-sea texture and magnetic fabric indicators of bottom currents regime. Oceanologica Acta, 4, 475-88. BULWNCH,D.L., LEDBETTER,M.L., ELLWOOD,B.B. & BALSAM, W.L. 1982. The high-velocity core of the Western Boundary Undercurrent at the base of the U.S. continental rise. Science, 215, 970-3. CRIMES,T.P. & OLDERSrIAW,M.A. 1967. Palaeocurrent determinations by magnetic fabric measurements on the Cambrian rocks of St. Tudwal's Peninsula, North Wales. Geol. J., 5, 217-32. ELLWOOD, B.B. 1980, Application of the anisotropy of magnetic susceptibility method as an indicator of bottom-water flow direction. Marine Geol., 34, M83-90. & LEDBETTER, M.T. 1979. Palaeocurrent indicators in deep-sea sediment. Science, 203, 1335-7.
-
-
GRANAR, L. 1958. Magnetic measurements of Swedish varved sediments. Arkiv Geo~vsik, 3, 1-40. HAMILTON, N. & REES, A.I. 1970. The use of magnetic fabric in paleocurrent estimations. In: Runcorn, S.K. (ed.), Paleogeophysics. Academic Press, New
York. 445-64. HOLLISTER,C.D. 1967. Sediment Distribution and Deep Circulation in the Western North Atlantic. Unpubl. doctoral dissertation, Columbia University in the City of New York, USA. -& HEEZEN, B.C. 1972. Geologic effects of ocean bottom currents: Western North Atlantic. In: Gordon, A. (ed.), Studies of Physical Oceanography 2. Gordon & Breach, London. 37-66. HORN, D.R., EWING, M., HORN, B.M. & DELACH,M.N. 1971. Turbidites of the Hatteras and Sohm Abyssal Plains, Western North Atlantic. Marine Geol., 11, 287-323.
Contourite or turbidite? KENT, D.V. 1973. Post-depositional remanent magnetization in a deep-sea sediment. Nature, Lond., 246, 32~,. - & LOWRIE, W. 1975. On the magnetic susceptibility of deep-sea sediment. Bull. geol. Soc. Am., 87, 321-39. KING, R.F. & REES, A.I. 1962. The measurement of the anisotropy of magnetic susceptibility of rocks by the torque method. J. geophys. Res., 67, 1565-72. LEDBETTER, M.T. 1979. Fluctuations of Antarctic Bottom Water velocity in the Vema Channel during the last 160,000 years. Marine Geol., 33, 71 89. -& ELLWOOD, B.B. 1980. Spatial and temporal changes in bottom-water velocity and direction from analysis of particle size and alignment in deep-sea sediment. Marine Geol., 38, 245-61. RE~, A.I. 1961. The effect of water currents on the magnetic remanence and anisotropy of susceptibility of some sediments. Geophys. J.R. astron. Soc., 5, 235-51. REES, A.I. 1965. The use of anisotropy of magnetic susceptibility in the estimation of sedimentary fabric. Sedimentology, 4, 257 71. --, BROWN, C.M., HAILWOOD, E.A. & RIDDY, P.J., 1982. Magnetic fabric of bioturbated sediments from the northern Rockall Trough: Comparison with modern currents. Marine Geol., 46, 161 73. -& FREDERICK, D. 1974. The magnetic fabric of samples from the Deep Sea Drilling Project, Legs I-VI. J. sed. Petrol. 44, 655-62. - VON RAD, U. • SHEPARD, F.P. 1968. Magnetic fabric of sediments from the La Jolla Submarine
Canyon and fan, California. Marine Geol., 6, 145-79. REES, A.I. & WODALL,W.A. 1975. The magnetic fabric of some laboratory deposited sediments. Earth planet. Sci. Lett., 25, 121-30. RICHARDSON, M.J., W1MBUSH, M. & MAYER, L. 1981. Exceptionally strong near-bottom flows on the continental rise off Nova Scotia. Science, 213, 887-8. SPIESS, F.N. & MUDIE, J. 1970. Small-scale topographic and magnetic features. In: Maxwell, A.E. (ed.), The Sea 4. John Wiley, New York. 205-50. STANLEY, D.J., SWIFT, D.J., SILVERBERG, N., JAMES, N.P. & SUTTON, R.G. 1972. Late Quaternary progradation and sand spillover on the outer continental margin off Nova Scotia, southeast Canada. Smiths. Contr. Earth Sci. 8. Smithsonian Inst. Press, Washington, D.C. 88 pp. STOW, D.A.V. 1979. Distinguishing between finegrained turbidites and contourites on the Nova Scotia deep water margin. Sedimentology, 26, 371-87. -& BOWEN, A.J. 1980. A physical model for the transport and sorting of fine-grained sediment by turbidity currents. Sedimentology, 27, 31-46. --& LOVELL,J.P.B. 1979. Contourites: their recognition in modern and ancient sediments. Earth Sci. Rer. 14, 251 91. TUCHOLKE, B.E. 1982. Origin of longitudinal triangular ripples on the Nova Scotia continental rise. Nature, Lond., 296, 735-7.
SHOR, D . V . K E N T , R . D . FLOOD, Lamont-Doherty Geological Observatory of Columbia University in the City of New York, Palisades, New York 10964, USA.
A.N.
273
Contourite facies of the Faro Drift, Gulf of Cadiz E.G. Gonthier, J.-C. Faug~res and D.A.V. Stow SUMMARY: The Faro Drift is an elongate sediment body (50 km long, 300 m thick) that parallels the northern margin of the Gulf of Cadiz south of Portugal. On the basis of location, morphology and seismic character of the drift together with bottom photographs, sediment distribution and the known regional oceanography, we can be certain that the Faro Drift was constructed by bottom currents related to the deep outflow of Mediterranean water. Detailed study of a closely-spaced suite of cores has therefore allowed characterization of the sediments into three contourite facies: sands and silts, mottled silts and muds, and homogeneous muds. The first is equivalent to the sandy contourites and the two others to the muddy contourites of earlier studies. These three facies are arranged in irregular vertical 'sequences' that reflect long-term variations in bottom current velocity. They arc distinctly different from typical fine-grained turbidite and hemipelagite facies. There is currently much interest in the effect of bottom currents on sediments in the deep ocean, in the complex interaction of bottom current, turbidity current and pelagic depositional processes, and in distinguishing between their respective deposits (e.g. Richardson et al. 1981 ; Shor et al., this volume). These studies are of particular importance for the understanding of deep-sea sedimentation and for the reconstruction of palaeo-oceans. Many of the major contourite drifts in the North Atlantic have now been cored and their sediments examined in detail. We have thus been able to construct a general picture of contourite facies, refining the earlier work of Hollister & Heezen (1972). Two main contourite types have been described (Stow & Lovell 1979: Faug6res et al. 1979; Stow 1982): muddy contourites are relatively homogeneous and extensively bioturbated, whereas sandy contourites occur as thin, irregular, bioturbated lag deposits or more rarely as clean, cross-laminated beds. Both commonly comprise a mixture of biogenic and terrigenous material. Gradations between these two types as well as coarser-grained gravel lag deposits (gral,elly contourites) have also been observed. However, most of the cores studied to date represent either isolated or very widely-spaced sites on drifts that are many hundreds of kilometres long and tens of kilometres wide. We thus have very little idea of the horizontal distribution or vertical sequences of sediments within such drifts. We also lack detailed information on the small-scale variability of contourite facies in response to the known dynamic variability of bottom currents. We are not yet, therefore, sufficiently confident to unambiguously identify contourites in ancient rock sequences, nor to use facies variability as an accurate indicator of the development and fluctuation of palaeo-bottom currents, although several interesting attempts
have been made in these fields (e.g. Pastouret et al. 1978; Auffret et al. 1981; Lovell & Stow 1981). For these reasons, a much more detailed study of a contourite drift was clearly necessary, preferably one sufficiently small to allow a close-spacing of core stations and in an area where the physical oceanography was well-known. Several relatively small contourite drifts have been identified on the south Iberian margin (Vanney & Mougenot 1981), and were almost certainly constructed by the deep Mediterranean outflow through the Straits of Gibraltar. One of these, the Faro Drift, was selected for detailed study (Fig. 1). The French oceanographic vessel, RV Noroit, was taken to the area in November 1982 and completed some 300 km of 3.5 kHz seismic profiling, occupied 24 sites for piston coring (with associated gravity coring) and 5 sites for bottom photography (Fig. 1). Subsequent laboratory analyses included X-radiography of centimetrethick slabs of core, impregnation and thin-sectioning, grain-size determination by sedigraph and sieving, compositional determination of the sand and clay fractions, and geochemical, palaeomagnetic and biostratigraphical studies. The preliminary results of this work have been reported by Faug6res et al. (in press). In this paper we describe in detail the nature of contourite facies, typical contourite "sequences' and their hydrodynamic interpretation.
Oceanography and bathymetry At the present day there is a deep outflow ofwarm (> 13 C), saline ( > 36.4%0), low-oxygen (4.1 to 4.6 ml/1) Mediterranean Water through the Straits of Gibraltar (Madelain 1970; Zenk 1975; Reid 1978; Ambar & Howe 1979). This watermass flows north and west along the southern margin of Spain and Portugal at depths of 400-1400 m (Fig. 1). At Cape St Vincent, part 275
276
E . G . G o n t h i e r , J . - C . FaugOres a n d D . A . V .
Stow
FIG. 1. Study area with Faro Drift in large stipple, showing 3.5 kHZ seismic track (dashed lines), core sites (solid circles), camera stations (open squares), known flow (heavy arrows) and inferred flow (light dashed arrow) of Mediterranean water, and location of seismic profile shown in Fig. 2 (heavy zig-zag). Contours in metre. Inset map shows passage of Mediterranean outflow water over sea floor and location of study area. (From Faug6res et al., in press.) spreads westward into the mid-ocean and part turns north forming part of a larger scale eastern boundary current. In the Gulf of Cadiz there appear to be three main flows, between 1200-1300 m, 700-900 m and 500-700 m, controlled largely by the bottom topography and partly interconnected by down-canyon flow. There is an overall increase in flow depth from west to east. Flow velocities of up to 300 cm/s have been recorded in the Straits of Gibraltar, decreasing to 180 cm/sjust west of the Straits, 30-40 cm/s in the vicinity of the Faro Drift, and 10-20 cm/s at the western end of the Gulf (Meli6res 1974; Reid 1978). There are local variations due to the influence of bottom topography, some meandering of the flow and counterclockwise gyres, but an overall decrease in velocity concurrent with a westward broadening and deepening of the water mass is observed. The general effects of the Mediterranean Outflow have been documented by a number of workers (Giesel & Seibold 1968; Heezen & Johnson 1969; Meli~res et al. 1970: Meli~res 1974: Vanney & Mougenot 1981). Significant erosion takes place in the Straits of Gibraltar as we~l as in some of the deep channels along the margin, such
as the Fossa Alvares Cabral and Fossa Diego Cao to north and south of Faro Drift. Non-deposition is evident in some areas, whereas the accumulation of a series of sediment drifts up to 300 m thick is noted in others. Acoustic and photographic evidence reveals areas of rocky and current-swept bottom and areas of smooth or rolling sediment-covered morphology. Currentcontrolled bedforms, particularly evident in channelized areas include sediment waves and ripple marks. There has also been a long history of chafing and corrosion of submarine telegraph cables in the Gulf and the Straits of Gibraltar closely related to the Mediterranean Outflow. The overall constructional aspect of the Gulf of Cadiz with its relatively shallow slope gradient (1 : 300) is probably also, in part, a result of the Mediterranean Outflow, although detailed study reveals a much more complex interplay of controis. A series of papers by Mougenot and coworkers on the south Iberian margin (Baldy et al. 1975: Mougenot et al. 1979; Mougenot & Vanney 1980, 1982; Vanney & Mougenot 1981) have shown a shelf 15-40 km wide, a narrow and moderately steep slope (1:6.5 to 1 : 25 gradient), a series of channels or deeps parallel to the margin interrupted by downslope canyons and four
Contourite facies of the Faro Drift, Gul[o/Cadiz
277
3~ E~
.s E eq'~
o"~
~o
9~ az
~az
278
E.G. Gonthier, J.-C. FaugOres and D.A.V. Stow
constructional mounds or drifts along the length of the margin near the foot of the slope (Fig. 1). The Faro Drift itself is 40-50 km long, 10-20 km wide, and up to about 300 m in thickness with a maximum present day relief of 160 m (Fig. 2). It appears to have been constructed over a late Miocene-early Pliocene erosional surface and to have prograded steadily to the north or northwest over a distance of about 10 km during the past 5-6 million years (Mougenot & Vanney 1982). The mode of growth appears closely analogous with lateral accretion in a meandering fluvial system or with constructional levee development in a deep-sea fan.
Sediment facies There is little doubt that over 90~ o of the sediments cored are true contourites, including virtually all of the Faro Drift cores, as borne out by the microphysiographic and bottom photographic evidence (Faug6res et al., in press). Whereas some 5-70/0 of the material in the canyon to the west, or on the channel floors and steep eroded channel slopes, occurs in distinct turbidite beds and slumped units or is an older more
consolidated sediment of unknown origin. In the following sections we describe only the contourites, for which we recognize three main facies: sands and silts, mottled silts and muds, and homogeneous muds. Sands and silts
The sands and silts form about 5~o of sediment and occur in irregular beds from a few centimetres to 20 cm in thickness (Fig. 3). They are almost always surrounded by the mottled silt and mud facies. Primary sedimentary structures The top and bottom contacts of the sand and silt beds are mostly irregular and may be sharp but relatively flat, clearly erosive or completely gradational. Erosional contacts appear to be slightly more common at the base of beds. It is often difficult to distinguish contacts that are of primary dynamic nature from those caused by bioturbation, and many of the contacts that change from sharp to gradational across the width of the core have probably been affected by secondary bioturbation (Fig. 4). There is almost no lamination remaining in any
FIG. 3. Photographs of core sections with mottled silt and mud, and homogeneous mud contourites. Sharp and erosive bed contacts shown by solid lines, gradational contacts shown by dashed lines.
Contourite facies of the Faro Drift, Gulf of Cadiz
279
Fro. 4. Thin sections photographs of impregnated slabs of sections of two cores, showing detailed structures of different contourite facies. Note, in particular, nature of contacts and bioturbation. Teichichnus-likeburrow indicated with 'b'. Mottled aspect at levels 1 is an artefact. of the beds cored, although the presence of ripple marks and other current bedforms on the presentday sea floor would suggest that many of the sandy layers were originally laminated. Rarely, there are discontinuous and indistinct clayey laminae within the sands, although their origin is unclear. Size grading within the sands and silts is often present but very variable in nature (Figs 5 and 11). Positive and negative grading occur equally commonly and may be relatively continuous through a bed or occur irregularly in rapid succession. Several of the thicker beds increase in grain-size from the base towards the middle, where there are pockets of coarser (often shelly) material, and then decrease in size towards the top.
Biogenic structures Bioturbation is the most common structure encountered, although in many of the sands it is only visible as an indistinct mottling (cm-scale) on X-radiographs (Figs 3 and 10). Bioturbation at
the top and bottom contacts of beds introduces pockets of sand into the adjacent sediment and wisps or streaks of mud into the sand. Large (up to 10 cm long) mud-filled or sand-filled burrows occur as straight isolated protrusions some of which appear to be Lophoctenium, or as a tortuous network of smaller diameter tubes.
Texture Although some of the silts and sands contain rare coarse (up to 5 mm) shell fragments, they are mostly relatively fine-grained with a mean grainsize between 30 and 50/zm (Figs 4, 5 and 6). They are mostly, therefore, medium-coarse grained silts with 20-40~ sand and less than 10~0 clay. More rarely, the fine sand fi'action is dominant. They are mostly moderately well sorted. The shape of the cumulative frequency grain-size curves (Rividre 1977) tends towards parabolic or to a combination of hyperbolic (coarse tail) and parabolic (fine tail). This relatively large fine tail is found even in the middle of the thicker beds, which suggests that it is not solely due to
280
E . G . Gonthier, J . - C . Faug~;res a n d D . A . V . S t o w
/
pa~JOS
~lJOOd
~
E~
I
[-...
_~_
PalJos
IlaM
I "--
CO I,l
.
> rr
_. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
N
E C~
. . . . . . . . . . . . . . . . . . . . .
9
(~
....
(i,)
=~ CO _
_
.
.
.
Z
.
8"L
rr (-9
"I:~-
-
:-:--'-
- ~-'k-
~-
~-
-I~- - ~ - - ~ -
~I~ .
.
.
.
.
.
.
.
.
-~-
-
-~
--I~ ~ - I ~ - ~ -
-~-
(I) ci. [D
_v)
_
_
L~ I-
t
90
9 Q
v
9
o
D fq.)
~
9
50
* D
O
t~
9
121
9
|
1
9
D" I:D
t
i
9 tt-
0
Y
O
9
,IF
z3
E
O
10
9
"IF
9
~3 'l-
99
9
1.0 9
,,
.5BX 9 7BX ~4G "2G
O "1~
7"
"
*9BX
0.1 Illl
70
t
50
t
I
I
I
I
I
I
30
I
I
10
Diameter (Fm) (a) FzG. 4. Cumulative frequency percent curves of the silt-fraction determined by the particle analyser (see text). Every fifth data point was utilized to construct the plots. (a) the western core transect; (b) the eastern core transect.
The sediment texture of contourites in Lake Superior
3os
m~ NG
99
,
#
..~ m , O
E 0
m
+, 0
m.
90
m.
O*
O. m, 9 . 9 9
0 9
(3) (D
50
.
-
"
O. 9
0
#
0 0
0 0 0 9
0
"
9
E
+
10
-)(-
0 -)(-
0
9
9 11 BX
o *
*
14G 016G 9
O.
1.0
-
9
@
9 21 B X
-)(..
o
0.1 Ill
70
I
I
50
i
I
I
I
1
I
3O
I
10
Diameter (p.m) FIG. 4 (b) by the continuous and smooth cumulative frequency plots from the electronic particle analyser (Fig. 4). Blaeser & Ledbetter (1982) and Bulfinch & Ledbetter (in press) published characteristic electronic particle size analyses of the carbonate-free silt fraction of the surface sediments underlying the Antarctic Bottom Water and the Western Boundary Undercurrent, respectively. Their results are analogous to the results of this study in that sediments underlying the core of the bottom currents are characterized by a relatively coarse grain-size. The marine studies, however, showed the skewness of the current-influenced sediments does not differ significantly from the uninfluenced sediments, but that the sorting does. This is contrary to our results. The differences between
the results of our study and the marine investigations can be explained by the magnitude of the change between the relative proportion of coarse and fine silt in the influenced versus the uninfluenced sediments. The marine investigations revealed a relatively smaller change than our study did in the relative proportions of coarse to fine silt. The size distribution, however, of the coarse and fine silt populations for both this study and the marine investigations are similar. The 3.5 kHz profiles illustrated by Johnson et al. (1980) clearly reveal that the coarse grain-size of the trough sediments results primarily from erosion of fne-grained material rather than from addition of coarse sediment. A typical histogram of the silt fraction of open lake sediment was changed theoretically by subtraction of fine silt
J.D. H a l / m a n and T.C. Johnson
302
i
LRTN 166 1 CN oDF-Z LJ (_0 Eli
>co Z
WCS~LLJ D_ LU > F-__J y-
18D
~2. O0
' 0
2'.O0
4~.00
B ~.00
6Rni N sIZE
8 ~.00
IO.O0 ~
12.00
1
. CIU
(PHI)
FIG. 5. Cumulative frequency percent curve determined by the pipette and sieve method for core 16G. This curve is typical of Lake Superior contourites. The curve can be broken down into three straight-line segments.
('erosion') and addition of coarse silt ('deposition') to see what proportion of the original material had to be removed or added to create a histogram that is characteristic of the trough sediment (Fig. 6). This exercise revealed that over 85 weight percent of the open lake sediment must be eroded to create the trough sediment. The results from the surface sediments of several cores recovered from the same core station often vary considerably because there is a patchy sand cover overlying the sandy mud. The lakeward margin of the eastern core transect is a good example (Fig. 2). The median grain-size of the silt fraction for 18BX is 30.61 #m (5.03 phi), whereas the median grain-size for the other cores are 15.52; 16.40 and 18.33 pm (6.01, 5.93, and 5.77 phi). The pipette and sieve analyses indicate similar inconsistencies. The X-radiographs of box cores that recovered anomalously coarser sediments than the adjacent gravity cores reveal an
atypical sandy layer overlying typical sediments for that core station. The textural analyses of 7BX (W.H. Busch 1982, unpub, data) at a depth below its sandy cap reveals a textural distribution similar to that in an adjacent gravity core (8G). Side-scan sonar records reveal crescentic depressions about 100 m in diameter that have hard reflectors on their floors (Flood & Johnson, in press). The seismic profiles sometimes indicate the presence of a hard reflector between overlapping acoustic echoes (Johnson et al. 1980; Johnson et al., in press). The evidence suggests that coarse lag deposits are preferentially winnowed in the localized depressions on the lake floor, and cause the patchy distribution in surface sediment grain-size. The median grain-size of the silt fraction in the trough sediments, revealed by the particle analyser, increases from the south-west to the northeast, contrary to the results of the pipette and
The sediment texture of contourites in Lake Superior
303
~-
NORMAL DISTRIBUTION
9
DISTRIBUTION 'DEPOSITED'
Q)
E
DISTRIBUTION
10-
'ERODED'
0
>
% %
>,, c
(1)
6-
IST (D LL E
2 -
E;
9
~_
9
I--1-11 60
[ 40
1
I
30
I
20
9
~
~
~
1 10
Diameter (/~m) FIG. 6. Hypothetical grain-size populations that were added to or subtracted from the average open lake grain-size distribution to determine how much deposition or erosion is required to create the trough sediments, Every fifth data point was utilized to construct the histograms. sieve data. The frequency histograms of the north-eastern trough reveal modes of sand to coarse silt in addition to the typical 2 phi mode for the sand population (Fig. 3(b)). The additional sand and silt modes, which are interpreted to represent material deposited due to the southwest to north-east decrease in current velocity, added coarse silt to the sediment distribution of the silt fraction to yield results inconsistent with the total sediment analysis. Down-Core
analysis
Cores 23P and 24P both reveal an increase in the median grain-size of the silt fraction with depth in the core to the postglacial/glacial contact (Fig. 7). The correlation coefficient (r) determined from a least-squares linear regression for core 23P is 0.42, which yields a confidence limit of over 99~. The down-core results can be interpreted in three ways: (1) The sediment source or distance to source changed due to a change in lake level. (2) The sediment source changed due to a change in the vegetation in the drainage basin. (3) The sediments in the Keweenaw area were influenced by a variable current velocity through time. Farrand (1969) and Drexler (1981) published results indicating that lake level did not fluctuate in the Keweenaw area after the last retreat of the
Laurentian Ice Sheet (c.9500 yrs BP). The area is close to the isobase of isostatic rebound passing through the controlling outlet for Lake Superior at Sault Ste. Marie (Saarnisto 1975). Palynological studies from the Lake Superior region indicate that the distribution of forest types since 9500 yrs BP has remained the same until the very recent (1880-1930 AO) deforestation by man (Wright 1969; Webb 1974; Maher 1977). The time required for first appearance of spruce trees for reforestation of a recently deglaciated area is about 40 yrs (Birks 1980). Therefore, the change in the median grain-size with depth is due to a change in the bottom current velocity. The results from the surficial sediment analyses Indicate that the fluctuation in the grain-size usually corresponds to the intensity of the regional bottom currents. Median grain-size is larger where the bottom currents are usually faster, although the patchy distribution of sandy caps offshore indicates that coarser size does not always imply stronger regional currents. It may reflect just a local perturbation caused by minor bathymetric features. The down-core trend from a smaller to larger grain-size is noisy, but the major deviations in the trend are consistent between 23P and 24P (assuming parallel sedimentation rates through the Holocene), perhaps due to a change in mean r
.
J.D. Halfman and T.C. Johnson
304
ELZONEMEDIAN GRAINSIZE (JlJM) 23P 22
24 P 20
+
=
18
20
+,
18
16
14 G 14
34 i
30 s
i
26 i
i
11
L
50
im
i' 50
100
Ir 100
i=
gE
1SG
F-
h
o
150 100
lb.
E 200
150 250
300
VARVES(-9500 YBP) 200
FIG. 7. The down-core fluctuations of the median grain-size of the silt fraction analyses (particle analyser) for cores 23P, 24P and 14G. The vertical line represents the average median grain-size for each core. bottom current intensity during the last 9500 years. The magnitude of the noise is less pronounced for 24P than 23P, which may correlate to the more distal location of 24P from the core of the bottom current. The noise is interpreted to reflect sporadic and perhaps more localized episodes of increased or decreased intensity of the bottom currents. The plots of the percent sand with depth in cores 23P and 24P reveal consistent results with the median grain-size plots derived from the particle analyser (Fig. 8), except at two depths in 23P where there are relatively large increases in the percent sand, yet the median grain size of the silt fraction decreases. The large quantity of sand indicates an episode of relatively strong bottom currents. The coarse silt was eroded away from this interval so the resulting grain-size distribution of the silt fraction appears finer than the sand fraction would predict. The difference between
the sand and silt analyses reveals the problem of relying on the silt fraction analysis alone to interpret palaeocurrent history. Core 14G reveals a decrease in the median grain-size of the silt fraction and the percent sand with depth in the core, contrary to what is observed in cores 23P and 24P. Unfortunately, a comparison between the grain-size stratigraphies of 14G and the parallel stratigraphies of 23P and 24P is difficult because chronostratigraphic markers for the three cores are lacking. Both 23P and 24P penetrated to glacial varves, so the complete postglacial record was recovered for these cores. The complete postglacial sequence was not recovered for 14G, however. The down-core trend for 14G may reflect a very localized development of a sandy cap as a crescentic depression formed near the core site in relatively recent times. Alternatively the two piston cores or all three cores may reflect very localized
The sediment texture o/contourites in Lake Superior
3o5
PIPETTE:SAND& SILT (WT. PERCENT) 23P 30 20 10
0
20
24P 10 0
40 r
14G 30 20 r
50 5~i
b.
100
p,
150
100l ul
0
U
Z
,5oi
F-
2oo~
~" 150
250 " ~ _ ~ i
!
I !
300
VARVE~ S>950Y 0BP/
-
200 FIG. 8. The down-core fluctuations of percent sand (dots) and silt (stars) by weight from the pipette and sieve analyses. The vertical line represents the average percent sand for each core.
accelerations of the bottom currents, and not reflect the history of regional bottom current strength in the study area. These results from the down-core analyses are not totally consistent with similar results of other studies in the Laurentian Great Lakes. Median grain-size of a core from Lake Michigan decreases with burial depth (Rea et al. 1980). The Lake Michigan core, however, like core 14G from this study, did not recover the complete postglacial record. The radiocarbon date from the base of the Michigan core was 3500-5000 yrs BP. A core recovered from Thunder Bay in northern Lake Superior reveals a noisy but recognizable trend of increasing grain-size with depth to the postglacial/glacial contact (Mothersill 1979), similar to cores 23P and 24P of this study. A Lake Huron core reveals a uniform grain-size with depth (Mothersill & Brown, in press). Slight increases of the grain-size are observable at the top of the core and between 8000 to 9500 yrs ~P.
The uniform grain-size is not surprising because the core was recovered from the centre of the Goderich Basin and probably is isolated from strong bottom currents. Other Lake Huron cores reveal a decrease in grain-size with burial depth (Graham & Rea 1980). Graham & Rea (1980) attribute the change in grain-size to a change in lake level. The grain-size stratigraphies from this investigation suggest that bottom current velocities were faster just after the retreat of the Laurentian Ice Sheet (about 9500 yrs BP) than in the recent past. The bottom currents in the study area are attributed to the downward extension of the northeastward-flowing, wind-driven Keweenaw Current that presumably is accelerated during storms (Johnson et al. 1980). Thus, a possible decrease in the storm intensity or frequency may have occurred between 9500 yrs BP and the recent past. This is the first evidence reported for the fluctuation of storm intensity or frequency with time in
3o6
J.D. Haljman and T.C. Johnson
the Lake Superior region. More studies such as this will have to be conducted before the 'palaeostorm' record is clearly defined, however, because some cores such as 14G may be indicating very localized rather than regional conditions.
Conclusions Size analysis of the silt fraction usually differentiated the sediments which are under the influence of bottom current activity at the base of the slope off the Keweenaw Peninsula in Lake Superior. The sediments from the trough at the base of the slope are characterized by a coarser median grain-size and an increased skewness, and are influenced by larger current velocities than the open lake sediments. Within the trough, the grain-size decreases and sediment thickness increases from the south-west to the north-east, which correlates to a parallel decrease in the bottom current velocity. The textural characteristics of the trough sediments probably result from the contour current winnowing as much as 85% by weight of the silt-sized sediment supply. Regional textural parameters are successful in delineating the lateral extent of the contour currents. The analysis of the silt fraction, however, occasionally yields results inconsistent with the total sediment analysis.
Uniform sandy layers are concentrated locally in crescentic depression along the lake floor. The textural parameters from the cores that presumably penetrate these sandy layers suggest that textural analysis of a single core to identify contourites might produce anomalous results. The down-core investigation revealed an increase in the grain-size with depth in the postglacial record in two of the three cores examined. The change in the grain-size may have resulted from a gradual decrease in the intensity of the bottom currents from 9500 years ago to the recent past. More studies are needed, however, to detect confidently the history of regional bottom current activity and distinguish it from local aberrations caused by shifting bedforms at the individual core sites. ACKNOWLEDGEMENTS: We thank E.B. Nuhfer, D.J.W. Piper, and C.F.M. Lewis for their suggestions and comments that greatly improved the manuscript. We thank the officers and crew of the R/V Laurentian. We thank R. Flood and W.H. Busch for their personal comments and use of their unpublished data. Contribution number 007 of the Limnology Program, University of Minnesota, Duluth, Minnesota. This work was funded by National Science Foundation grants OCE 8018339 and OCE 8109833.
References ALLISON, E. & LEDBETTER, M.T. 1982. Timing of bottom-water scour recorded by sedimentological parameters in the South Australian Basin. Marine Geol., 46, 131-47. BIRKS, H.J.B. 1980. The present flora and vegetation of the moraines of the Klutlan Glacier, Yukon Territory, Canada: a study in plant succession. Quat. Res., 14, 60-86. BLAESER, C.R. & LEDBETTER, M.T. 1982. Deep-sea bottom-currents differentiated from texture of underlying sediment. J. sed. Petrol., 52, 755-68. BOUMA, A.J. & HOLLISTER, C.D. 1973. Deep ocean basin sedimentation. In: Middleton, G. & Bouma, A.H. (eds), Turbidites and Deep Water Sedimentation. Soc. econ. PaleD. Min. Short Course, Anaheim, 79-118. BULFINCH, D.L. & LEDBETTER, M.T. 1982. Western Boundary Undercurrent delineated by sediment texture at the base of the North American Continental Rise. Deep Sea Res., in press. DREXLER, C.W. 1981. Outlet Channels Jor the PostDuluth Lakes in the Upper Peninsula of Mich(Tan. Ph.D. Thesis, University of Michigan, Ann Arbor. ELLWOOD, B.B. & LEDBETTER, M.T. 1977. Antarctic Bottom Water fluctuations in the Vema Channel:
-
effects of velocity changes on particle alignment and size. Earth planet. Sci. Lett., 35, 189-98. 1979. Palaeocurrent indicators in Deep-Sea sediment. Science, 203, 1335-37. EVANS,J.E., JOHNSON',T.C., ALEXANDER,E.C., LIVELY, R.S. & EISENREICH,S.J. 1981. Sedimentation rates and depositional processes in Lake Superior using 21~ geochronology. J. Great Lakes Res., 7, 299-310. FARRAND,W.R. 1969. The Quaternary history of Lake Superior. Proe. 12th Conf Great Lakes Res., International Assoc../'or Great Lakes Res., 181-97. FLOOD, R. & JOHNSON,T.C., in press. Side-scan sonar targets in Lake Superior---evidence for current transport of bottom sediments. Sedimentology. FOLK, R.L. 1974. Petrology of Sedimentao, Rocks. Hemphill Publishing Co., Austin. GRAHAM, E.J. & REA, D.K. 1980. Grain size and mineralogy of sediment cores from western Lake Huron. J. Great Lakes Res., 6, 129-40. HALFMAN, J.D. 1982. Textural Anah'sis o/" Lacustrine Contourites. M.S. Thesis, University of Minnesota, Minneapolis. HOLLISTER,C.D. & HEEZEN,B.C. 1972. Geologic effects of ocean bottom currents: Western North Atlantic. -
The sediment texture of contourites in Lake Superior In: Gordon, A.L. (ed.), Studies in Physical Oceanography, 2, 37-66. HUANG, T.C. & WATKINS,N.D. 1977. Contrast between the Brunhes and Matuyama sedimentary records of bottom water activity in the South Pacific. Marine Geol., 23, 113-32. JOHNSON, T.C., CARLSON, T.W. & EVANS, J.E. 1980. Contourites in Lake Superior. Geology, 8, 437-41. - - , HALFMAN,J.D., BUSCH,W. & FLOOD, R., in press. Effects of bottom currents and fish on sedimentation in a deep water, lacustrine environment. Bull. geol. Soc. Am. LEDBETTER, M.T. 1979. Fluctuations of Antarctic Bottom Water velocity in the Vema Channel during the last 160 000 years. Marine Geol., 33, 71-89. MAUER, L. 1977. Palynological studies in thc western arm of Lake Superior. Quat. Res., 7, 14-44. MCCAVE, I.M. & SWIFT, S.A. 1976. A physical model for the rate of deposition of fine-grained sediments in the deep sea. Bull. geol. Soc. Am., 87, 541-6. MOTHERSILL, J.S. 1979. The paleomagnetic record of the late Quaternary sediments of Thunder Bay. Can. J. Earth Sci., 16, 1016-23. & BROWN, H., in press. Late Quaternary stratigraphic sequence of the Goderich Basin: texture, mineralogy and palaeomagnetic record. J. Great Lakes Res. PIPER, D.J.W. & BRISCO, C.D. 1975. Deep water continental margin sedimentation. In: Hayes, D.E., Frakes, L.A. et al., Init. Repts. DSDP, Leg 28, US Govt. Print Off., Washington, DC. 727-55.
3o7
REA, D.K., BOURBONNIERE,R.A. & MEYERS,P.A. 1980. Southern Lake Michigan sediments: changes in accumulation rate, mineralogy, and organic content. J. Great Lake.; Res., 6, 321-30. SAARNISTO, M. 1975. Stratigraphical studies of the shoreline displacement of Lake Superior. Can. J. Earth Sci., 12, 300-19. SHII)ELER, A.L. 1976. A comparison of electronic particle counting and pipette techniques in routine analysis. J. Ned. Petrol., 46, 1017-25. STOW, D.A.V. 1979. Distinguishing between finegrained turbidites and contourites on the Nova Scotian deep-water margin. Sedimentology, 26, 371-87. & LOVELL, J.P. 1979. Contourites: their recognition in modern and ancient sediments. Earth Sci. Ret'., 14, 251-91. VAN ANDEL, J.H. 1973. Texture and dispersal of sediments in the Panama Basin. J. Geol., 81,434-57. VERNET, J.P., THOMAS, R.L., JAQUET,J.M. & FRIEDLI, R. 1972. Texture of the sediments of the Petit Lac. Eclogae Geol. Heh,., 65, 591-610. VISHER, G.S. 1969. Grain-size distribution and depositional processes. J. sed. Petrol., 39, 1074-106. WEBB, T. 1974. A vegetational history from Northern Wisconsin: evidence from modern and fossil pollen. Tile American Midland Naturalist, 92, 12-34. WRIGHT, H.E. JR. 1969. Glacial fluctuations and the forest succession in the Lake Superior area. Proc. 12th Conf Great Lakes Res., hTternational Assoc.jbr Great Lakes Res., 397-405. -
-
J.D. HALFMANl and T.C. JOHNSON, Limnology Program and Department of Geology, University of Minnesota, Duluth, Minnesota 55812, USA. I Present address: Duke University Marine Lab., Pivers Island. Beaufort, North Carolina 28516, USA.
Facies and sequence analysis of Nova Scotian Slope muds: turbidite vs 'hemipelagic' deposition P.R. Hill S U M M A R Y: Five fine-grained mud facies were identified in piston-cores obtained from the Nova Scotian Slope. The sedimentary structures of the muds indicated deposition under varying bottom-current conditions and rates of sediment supply. Two facies associations were recognized using Markov sequence analysis: (a) turbidite and (b) non-turbidite associations. The problems of recognizing turbidite, contourite and hemipelagic deposits in fine-grained sequences are discussed in the context of low-concentration, low-velocity flows. The dynamics of unconfined flows indicates that turbidity currents become unstable and are deflected by Coriolis to near contour-parallel flow. Dilution and deceleration of the flow may produce a transition to hemipelagic-type deposition. Transitions of this type within individual flows make recognition of discrete turbidite, contourite or hemipelagic processes very difficult in parts of the geological record. The shelfbreak marks an important transition between a shelf regime dominated by wave, tide and current processes (Johnson 1978) and the oceanic regime where waves and tides are much reduced in amplitude. Currents in the oceanic regime are controlled mainly by the large-scale circulation in the ocean, with the secondary effects of topography (e.g. internal wave refraction) often being important on the continental slope. Strong circulation-driven currents have been recognized on several continental margins, generally on the rise (Heezen et al. 1966; Stow & Lovell 1979). Sediment gravity-flows are also commonly associated with the margin seaward of the shelfbreak. Turbidity currents in particular may transport large volumes of sediment across the slope via canyons to the rise and abyssal plain. The continental slope is generally thought to be a zone of coarse sediment bypass and the dominant form of sedimentation is thought to be fine-grained hemipelagic settling (Doyle & Pilkey 1979). The term 'hemipelagic', although defined on the basis of composition (Rupke & Stanley 1974), is generally considered to mean deposition under near-zero current velocities. However, studies of bottom currents on slopes (Hill & Bowen 1983; McGregor, pets. comm.) indicate that velocities are rarely zero, even when there is very little net drift. The transport of fine sediment on the slope by bottom currents and turbidity currents is poorly understood, largely due to problems associated with distinguishing the products of different processes in the sedimentary record. This paper uses Markov sequence analysis to identify mud facies associations on the Nova Scotian Slope and discusses the intergradational nature of the process. The study area (Fig. 1) lies on the continental slope due south of Halifax, Nova Scotia, in a
region where the shelfbreak has a depth of approximately 250 m. Twenty-four piston-cores were collected and ten were used in the sequence analysis. The cores were split and X-radiographed, some sections in thin (0.5 cm) slabs for additional resolution.
Physiography A detailed acoustic survey of the study area revealed a complex slope morphology (Fig. 1, details in Hill 1981), related largely to massmovements on various scales. Two turbidity current channels have been identified in the area. The 'main channel' decreases in size downslope from > 1 km width and 70 m depth on the upper slope to < 200 m width and 15 m depth by 1000 m water depth. The lower reaches of the main channel cross a small depositionat lobe. The second small (20 m deep) channel (Fig. 1) in the west of the area appears to be discontinuous and traverses a small area of undisturbed slope. The remainder of the study area shows a very irregular morphology, with relief varying from over 100 metres to a few metres. This irregularity is related to late Pleistocene slumping on various scales (Hill 1981). The presence o f a 1 to 2 metre mainly Holocene mud drape over the whole area indicates that the features are largely pre-Holocene in age. The palaeo-oceanography of the Scotian Shelf during the late Wisconsin is not well defined, but there is evidence to suggest that a floating ice sheet may have been present over submerged parts of the shelf during this time (Mudie 1980). Sediment supply to the slope may have been influenced by the ice cover.
3II
312
P.R. Hill
FIG. 1. General location map showing
morphological features determined from GLORIA II sidescan and seismic profiles, from Hill (1981). Numbers refer to piston core locations.
Stratigraphy Core sequences show a very consistent division between (1) a lower red-brown to brown mud unit with associated silt, sand and gravel beds, and (2) an upper, olive-grey and dark yellow-brown mottled mud unit (Figs 2 and 3). These units are traceable for at least 300 km eastward on the slope (Stanley et al. 1972). The lower brown unit is highly variable in lithology, such that correlation of individual beds was not achieved between cores as closely spaced as 0.5 kin. In contrast, the upper unit shows a generally uniform sequence of interbedded olive-grey and dark yellow-brown beds which can be correlated between almost every core. Radiocarbon dating of dispersed organic material in the cores suggests that the age of the boundary between units 1 and 2 lies between 18 000 and 20 000 yrs BP (Fig. 3), although this may be an overestimation since the samples probably contain significant quantities of reworked 'dead' carbon (Hill 1981). The single date in core 58 and the more frequent preservation of
sedimentary structures within unit 1 suggests that the sedimentation rate decreased markedly above the unit l/unit 2 boundary.
Sediments Facies 1
Mottled mud
This facies includes a range of grain-sizes, from sandy-silt to silty-clay and is characterized by the presence of pervasive bioturbation, giving rise to a mottled or patchy appearance (Fig. 4(a)). The X-radiographs often reveal a variety of burrow structure, with Zoophycos being particularly common. Facies 2
Parallel-laminated mud
Generally finer-grained silty-clay, facies 2 muds show thin to very thin parallel laminations in X-radiographs (Fig. 4(b)). The laminae sometimes exhibit thinning-upward sequences on the
Facies and sequence analysis of Nova Scotian Slope muds
313
CORE II CM 0 T . '
2d
~live
Dark
2c
grey
kioturl~ated
sand)'
silt.
yello~,-I town bioturbated
mud.
I ' I . * . 0-- i
2b-
t)live
2o
Dark yellow-brown mud w i t h and bands of olive grey
grey
sandy" s i l t .
101.5
I00
cm,
red-brown
thin mud.
laminae
lamina.
Red b r o w n c l a y e y mud. N a i n l y s t r u c t u r e l e s s , with occasional fine laminae and red/grey brown colour banding near top.
Ib
I:rown,
mainly
structureless
mud.
200 Irm~n to yellm,-I mott 1 iny.
rown mud ~ i t h
sulphide
I
.-'ro~,n f:~ud ~ , i t h
fin{" p a r a l l e l
laminations.
FIG. 2. Type stratigraphic core from the Nova Scotian Slope defined by Hill (1981). This core was obtained outside the study area, but illustrates the main stratigraphic units which can be traced for 350 km along the Nova Scotian Slope.
scale of 2 to 3 cms, but more commonly show little organization and form units several tens of centimetres thick. Facies 3
Wispy-laminated mud
The laminae in Facies 3 are very thin and discontinuous, with an irregular wispy form (Fig.
4(c)). In places, climbing ripple-forms can be seen amongst the less regular wispy laminations. The facies includes a range of grain-sizes from clayeysilt to silty-clay. A similar lithology was recognized by Stow (1977) as thin intervals within turbidite units. Banerjee (1977) generated wispy laminae in a flume study under waning-current conditions.
SCOTIAN 54
55 d
d
GULF
55G d
c
C
2--
57
58
--23 el
9,920
_
2--
0
a I 8,860
tO0
(3
1117450
I
18,790 E
200 I t
"-rI(2_ I_lJ a
a
a I
al
1 28,000 300 i
400
I
500
FIG. 3. Carbon-14 dates (yrs BP) on cores from the Nova Scotian Slope. Date with asterisk is unreliable as it was sampled across a possible discontinuity between Units 1 and 2.
FIG. 4. Representative X-radiographs of the five fine-grained sediment facies, (a) Facies 1, mottled mud; (b) Facies 2, parallel laminated mud; (c) Facies 3, wispy-laminated mud; (d) Facies 4, homogeneous mud; (e) Facies 5, sandy-silt to mud graded couplets.
Facies and sequence analysis of Nova Scotian Slope muds Facies 4
Homogenous mud
t2)
Mud units in which no sedimentary structures are seen, even in X-radiographs, are included in this facies (Fig. 4(d)). Stanley (1981) has recognized a similar lithology in the Mediterranean, and has used the term 'unifite' to describe it. On the Scotian Slope, the facies generally occurs in thin ( < 10 cm) units.
Facies 5
315
\
4
Graded sandy silt to mud couplets
This facies consists of sharp-based thin silt or sandy-silt divisions which grade upwards into silty-clay divisions to form a couplet (Fig. 4(e)). These couplets may occur in bundles or be isolated within a muddy sequence. In places, the couplets may be separated by thin or very thin beds of silty-clay of a distinctly different colour, suggesting rapid deposition of the couplets between periods of more continuous mud deposition.
ORIGINAL
5
Facies associations
Apart from Facies 5, the facies generally have gradational upper and lower boundaries. To help define significant facies associations, a Markov sequence analysis was applied to ten core sequences. The method is identical to that used by Cant & Walker (1976). The final difference matrix (Table 1) highlights the statistically significant transitions between facies as positive values greater than 0.05 (Cant & Walker 1976). Fig. 5 shows the original, observed facies relationships in the form of a flow diagram and is compared to the preferred relationships derived from the Markov chain analysis. Although the total number of transitions is minimal for this method of analysis, two associations are usefully distinguished: (1) Turbidite association, Facies 2, 4, 5.
TABLE 1. Final difference matrix obtained from Markov analys&. Positive numbers greater than 0.05 indicate significant transition probability and are used in Fig. 5. Details of other steps in the analysis given in Hill (1981) 1
2
3
4
-0.34 0.32 -0.11 1 0-10 0.20 2 - 0.22 3 0.31 -0.30 0.43 4 - 0.42 0.31 -0.17 -0.13 5
5
PREFERRED
FIG. 5. Facies sequence analysis. Top: original observed transitions (numbers in brackets indicate number of observed transitions); bottom: preferred facies transitions from Markov analysis. Dashed line indicates insignificant transition probability. The analysis shows the close association between silt/mud couplets of Facies 5 and muds of Facies 2 and 4 (Fig. 5). The dominant sequence is a fining-upward sequence from the silt mud couplets (Facies 5), through parallel laminated mud (Facies 2) to homogeneous mud (Facies 4). The complete sequence is not always present but three typical sequences are recognized from visual examination of cores (Fig. 6): (a) Multiple silt/mud couplets. (b) Complete sequences of silt/mud couplets, parallel laminated mud and homogeneous mud. (c) Parallel laminated mud with slightly coarser basal silt laminae and homogeneous mud. Type (a) sequences are not recognized in the Markov analysis because only gradual transitions were included. Type (c) sequences often consist only of Facies 2 units, with the slightly coarser base allowing identification of individual units. Such distinctions can sometimes be made only in X-radiographs.
316
P.R. Hill
2
2
5
4
E3
!:i:'...
2
12 :.
EI
D
(a )
( b}
(c )
PIPER (1978)
FIG. 6. Typical turbidite mud sequences observed in Nova Scotian Slope cores. Numbers refer to facies defined in the study. Piper's (1978) model sequence for fine-grained turbidites shown in right hand column. The general sequence recognized here is very similar to the standard sequence proposed by Piper (1978) for turbidite muds and silts (Fig. 6). The basal part of Facies 5 is equivalent to a Bouma D division. Piper's model includes three subdivisions of the Bouma E division: E1 laminated mud, E2 graded mud and E3 ungraded mud. Subdivisions E1 and E2 are equivalent to Facies 2 of this study and E3 is equivalent to Facies 4. The sequence analysis suggests that the wispylaminated mud (Facies 3) might make up a fourth member of the association (Fig. 5). This facies shows a strong Markov dependence on the massive mud (Facies 4) and an insignificant dependence on the laminated mud (Facies 2). Stow (1977) in a detailed analysis of muddy turbidites, suggests that a wispy-laminated division occurs between divisions E 1 and E2 of Piper (1978), rather than between E2 and E3 as would be implied by this analysis. Experimental studies of fine sediment transport (Banerjee 1977) support Stow's results. Thus, it seems unlikely, in most cases here, that Facies 3 is part of the turbidite association. (2) Non-turbidite association, Facies 1 and 3. The transition analysis indicates a close twoway relationship between the mottled muds of Facies 1 and the wispy-laminated muds of Facies 3 (Fig. 5). There are generally no textural changes across these transitions; the main change is in the preservation of sedimentary structures. Facies 1 muds are clearly bioturbated to the extent that all primary structures have been destroyed, whereas intervals of undisturbed sediment usually reveal the characteristic wispy lamination of Facies 3. Bioturbated muds on continental slopes are generally interpreted to result from hemipelagic sedimentation (Doyle & Pilkey 1979) as they reflect relatively low rates of sedimentation and contain much biogenic sediment. The close two-
way association of these bioturbated muds with wispy-laminated muds suggest that their respective depositional processes are also related. The preservation of sedimentary structures is the only discernible difference between the two facies, suggesting that the rate of deposition may be the only difference in process. The facies in this association are not directly related to turbidity currents as indicated by the Markov analysis. Yet, the association is not typically hemipelagic, with the preservation of primary sedimentary structures indicating relatively rapid deposition under significant dynamic conditions. It is suggested that both facies were deposited under much the same kind of current regime, with only the rate of sediment supply causing differences in sedimentary structures. Extension of the term 'hemipelagic' to include the wispy-laminated facies is probably justified, rather than invent a new term.
Discussion
Mud turbidites have been successfully identified in this study by making use of the characteristic sequence of structures proposed by Piper (1978) and Stow & Shanmugam (1980). Furthermore, non-turbidite muds can be distinguished by the Markov sequence analysis despite the fact that they show sedimentary structures which are constituents of the Stow & Shanmugam (1980) sequence. It is not surprising, of course, that similar sedimentary structures are observed in muds deposited from different processes, because the range of current velocities from which mud deposits is relatively small. Hence, the near-bed mechanics of deposition may be very similar and concentration of sediment in the flow may be the only distinction between turbidite and non-turbidite deposition. Banerjee's (1977) experiments support this and also suggest that the rate of current deceleration is important to the type and sequence of sedimentary structures that develop. Where sand and silt interbeds are not present, it may be very difficult to identify discrete turbidite or non-turbidite intervals. Gradational bed boundaries may predominate and individual sedimentation units may vary greatly in thickness from a few millimetres up to tens of centimetres. This may be a source of annoyance to the geologist who wishes to neatly categorize the sediments into turbidite, contourite or 'hemipelagite' strait-jackets but it should be recognized that gradations between the three processes are probably common in the ocean. The following analysis attempts to illustrate this point. Consider a turbidity current flowing down a
Facies and sequence analysis of Nova Scotian Slope muds slope channel such as seen in the study area (Fig. 1). Komar (1969) has demonstrated that the Coriolis force is important in channelled flow and explains the difference in the height of channel levees by this effect. However, if the flow becomes unconfined at the end of the channel or by overflow of the channel walls, the flow will tend to be turned by the Coriolis force (to the right in the northern hemisphere and to the left in the southern). The alongslope acceleration is the sum of the Coriolis acceleration and the gravitational component as the flow spreads laterally. This can be expressed by: cqu
~h
where u and v are the alongslope (x) and downslope (y) velocity components, f is the Coriolis force, h the depth of the flow and g" = g ( p t - p ) / p where p, is the flow density, and p the density of sea water. For the latitude of Nova Scotia,f is in the order of 10 -4 s -~ so that for an initial flow velocity of 1 ms-~, the 'turning radius' would be of the order of 10 km. A faster current would have a correspondingly longer turning radius. The time-scale of this turning effect is given by T = 2g/f, i.e. in the order of 15 hours. The final direction attained by the flow (described by the angle 0) is given by: tan 0 -
fh Co V
Where Co is the drag coefficient and V the final velocity of the flow which is considered to be in a sheet-like steady-state. Assuming a reasonable value for Co (in the order of 10-3), the final flow direction is dependent on the ratio h/Vo. For a
317
flow of height 100 m and final velocity 1 ms-1, tan 0 = 10 or 0 ~ 90 ~ giving flow almost parallel to contours. Similarly a 10 m thick flow of the same velocity or a 100 m flow of 0.1 m s - 1 would result in tan 0 = 1 or 0 = 4 5 ~ giving flow oblique to contours. It is reasonable to assume that turbidity currents would vary considerably in flow thickness and velocity so that a complete spectrum of flow directions from downslope to contour-parallel is likely on the slope and rise. The flow velocity would decrease due to lateral spreading and loss of momentum until the flow is driven by the ambient currents only. Thus turbidity currents may both turn into 'contour-currents' if the ambient alongslope currents are strong, or supply sediment to the bottom boundary layer for deposition under 'hemipelagic' type of conditions. Under such a set of intergradational processes, it may be more useful to think in terms of sediment supply rates and flow concentrations as controls of sediment facies, rather than a discrete set of physical processes. Field identification of turbidite, contourite and hemipelagite sediments as such, is never likely to be possible due to the absence of unique distinguishing criteria and their gradational nature. More time-consuming methods such as the Markov analysis applied here, and current orientation by clay fabric analysis must be applied to allow such interpretations to be made.
ACKNOWLEDGEMENTS: My thanks go to Tony Bowen for discussions and calculations concerning the dynamics of unconfined turbidity currents. Jim Syvitsky, Dorrik Stow and Kevin Pickering read an earlier draft of the manuscript.
References BANERJEE,I. 1977. Experimental study on the effect of deceleration on the vertical sequence of sedimentary structures in silty sediments. J. sed. Petrol., 47, 771-83, CANT, D.J. & WALKER,R.G. 1976. 'Development of a braided-fluvial facies, model for the Devonian Battery Point-Sandstone, Qu6bec'. Can. J. Earth Sci., 13, 102-19. DOYLE,L.J. & PILKEY,O.H. 1979. Geology of continental slopes. Soc. econ. Palaeo. ?din. Spec. Pub. 27, 374 Pp. HEEZEN, B.C., HOLLISTER, C.D. & RUDDIMAN W.F. 1966. Shaping of the continental rise by deep geostrophic contour currents. Science, 152, 502-8. HILL, P.R. 1981. Detailed Morphology and Late Quaternary Sedimentation on the Nova Scotian Slope, South of Halifax. Unpubl. Ph.D. Thesis, Dalhousie Univ., Halifax, N.S., Canada. 331 pp.
--
& BOWEN,A.J. 1983. Modern sediment dynamics at the shelf-slope boundary offNova Scotia. In: The Shelf-Slope Boundary: a Critical Interface on Continental Margins. Soc. econ. Palaeo. Min. Spec. Pub. 33. JOHNSON, H.D. 1978. Shallow siliclastic seas. In: Reading, H.G. (ed.), Sedimentary Environments and Facies. Elsevier, New York. 207-58. KOMAR, P.O. 1969. The channelised flow of turbidity currents with application to Monterey Deep Sea Fan Channel. J. geophys. Res., 74, 4544-58. MUD1E, P.J. 1980. Palynology of Late Quaternary Marine Sediments, Eastern Canada. Unpubl. Ph.D. Thesis, Dalhousie Univ., Halifax, N.S., Canada. 464 pp. PIVER, D.J.W. 1978. Turbidite muds and silts on deepsea fans and abyssal plains. In: Stanley D.J. & Kelling G. (eds), Sedimentation in Submarine
318
P.R. Hill
Canyons, Fans and Trenches. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 163-176. - & NORMARK W.R., in press. Turbidite depositional patterns and flow characteristics, Navy submarine fan, California. Sedimentology. RUPKE, N.A. & STANLEY,D.J. 1974. Distinctive properties of turbidite and hemipelagic mud layers in the Alg6ro-Balearic Basin, Western Mediterranean Sea. Smith. Contr. Earth Sci., 13, 40pp. STANLEY, D.J. 1981. Unifites: structureless muds of gravity-flow origin in Mediterranean basins. GeoMarine Letters, 1, 77-83. --, SWIFT, D.J.P., SILVERBERG,N., JAMES, N.P. & SUTTON, R.G. 1972. Late Quaternary Progradation
and Sand 'Spillover' on the Outer Continental Margin off Nova Scotia, Southeast Canada. Smith. Contr. Earth Sci., 8, 88pp. STOW, D.A.V. 1977. Late Quaternary Stratigraphy and Sedimentation on the Nova Scotian Outer Continental Margin, Unpubl. Ph.D. Thesis, Dalhousie Univ., Halifax, N.S. Canada 360pp. - - & LOVELL,J.P.B. 1979. Contourites: their recognition in modern and ancient sediments. Earth Sci. Rev., 14, 251-91. --• SHANMUGAM,G. 1980. Sequence of structures in fine grained turbidites: comparison of recent deepsea and ancient flysch sequences. Sed. Geol., 25, 23-42.
P.R. HILL, Department of Geology, Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5. Present address: Atlantic Geoscience Centre, P.O. Box 1006, Dartmouth, Nova Scotia, Canada B2Y 4A2.
The role of canyons in late Quaternary deposition on the United States mid-Atlantic continental rise B.A. McGregor, T.A. Nelsen, W.L. Stubblefield and G.F. Merrill SUMMARY: The continental margin in the US mid-Atlantic region is dissected by many downslope-trending canyons, extending from near or at the shelf-edge out onto the rise. Sediments from the shelf and slope are transported seaward through the canyon system to the rise. Sand-size material from the shelf is introduced by means of spillover into the canyons and onto the upper slope. A dendritic gully system dissecting the entire slope also provides pathways for slope sediments, mainly silts and clays, to be introduced into the canyons. Distinct grain-size distributions of the shelf sands which are interbedded or mixed with the fine-grained sediments of the slope and rise can be used as tracers for transport pathways on the continental margin. Canyon erosion on the rise also may provide a local source of sediment which must be considered. Sand-size-sediment distribution and DSRV Alvin observations show that erosion and deposition take place periodically in the mid-Atlantic canyons and on the rise seaward of New Jersey and Delaware. The continental slope and rise seaward of New Jersey and Delaware are dissected by numerous submarine canyons. This study focuses on the slope and rise in the vicinity of Wilmington Canyon, a large canyon seaward of Delaware Bay (Fig. 1). The detailed morphology and canyon processes on the slope and rise can be evaluated by integrating narrow-beam bathymetric data, seismic reflection profiles, mid-range sidescan sonar data, piston cores, grab samples, and observations from DSRV Alvin. Initial study of the US mid-Atlantic continental slope morphology was by Veatch & Smith (1939), whose bathymetric survey indicated that the slope was dissected by numerous valleys (canyons). Energy and resource needs have renewed interest in the continental margin and resulted in new studies and research tools. Recent detailed bathymetric surveys (Kelling & Stanley 1979; Merrill & Bennett 1981; McGregor et al. 1982) and long-range sidescan sonar (GLORIA) images (Teleki et al. 1981; Twichell & Roberts 1982) have refined the interpretation of the slope morphology of Veatch & Smith in the vicinity of Wilmington Canyon. The slope is very dissected by canyons, and the walls of the canyons on the upper to midslope are cut by numerous gullies (Twichell & Roberts 1982). Seismic reflection profiles of the slope and upper rise show the depositional and erosional history of the margin during the Tertiary and Quaternary in the Wilmington Canyon area (Kelling & Stanley 1970; McGregor & Bennett 1981). Sediment samples collected to evaluate the activity of slope and canyon processes have resulted in differences in interpretation because of the varying objectives of the data collectors. After analysing a regionally distributed group of cores
collected as four core transects normal to the shelfbreak and spaced 35 km apart, Doyle et al. (1979) and Keller et al. (1979) concluded that the continental slope of the US mid-Atlantic is a region of deposition of fine-grained sediment and of little transport or deposition of sand-size material. Short-term, current-meter observations in the canyons also suggested that transport processes in the mid-Atlantic canyons are relatively inactive (Keller & Shepard 1978). In contrast, Kelling & Stanley (1970) and Forde et al. (1981), using seismic-reflection and sedimentologic surveys focused on specific canyons, suggested that the transport of sediment, including sand-size material, has taken place periodically in the mid-Atlantic canyons since the late Pleistocene. Data that are integrated in this study include mid-range sidescan images of the topography of the slope and rise (Fig. 2), which provide greater resolution of the gully system first observed on the GLORIA images, and details of canyon-floor morphology. The sidescan images were collected with a mid-range sidescan sonar system (Sea MARC I) as part of a cooperative study between the US Geological Survey and Lamont-Doherty Geological Observatory of Columbia University during August and September 1980. In September 1980, sidescan data were used in dives planned by the Marine Geology and Geophysics Laboratory of the National Oceanic and Atmospheric Administration (NOAA) in Miami. Features identified on the sidescan images were directly observed and sampled from DSRV Alvin. The canyon floors of Wilmington and South Wilmington Canyons were sampled in detail. A series of 6 m piston cores on the slope and rise and 25 grab samples on the shelf were 319
32o
B.A. McGregor et al.
OCEANOGRAPHER, ~ ~ 43 ~ 73 ~
42 ~ 74 ~
HUDSON
41 ~
i
75 ~
l
L
40 ~ 76 ~
~ [ LINS DP EE NN KC OEHRL _
A
,
RE
WlLM~NGTON~ !
BALTM I ORE
39 77
?,
38
36 ~ 79 ~
78 ~ 35 ~
77 ~
34 ~
76 ~
75 ~
33 ~
74 ~
FIG. 1. Index map of the US Atlantic coast showing location of study area outlined by heavy black lines.
Bathymetry is in metres. collected at sites selected by using detailed bathymetry and single channel seismic reflection profiles collected during 1975, 1977, and 1978 as part of a NOAA program to assess the sea floor stability of the continental margin between Wilmington and Lindenkohl Canyons (Fig. 2). The purpose of this discussion is to integrate these different types of data, each with different scales of resolution, in order to evaluate the
processes and transport pathways of the continental slope and rise.
Physiography The slope is dissected by canyons that can be grouped into two broad classes: those that indent the shelf-edge (e.g. Wilmington & Spencer) and
The role of canyons in late Quaternary deposition
32I
e~
r~ o
Q
o
3 o~ O.t~
.=..=
.~.=-
~
+,a 0
0
g i
s e.i
Fine-grained sediments associated with .fan lobes the thickest accumulation of these sediments in the world. These are dominantly turbidites and deep water hemipelagic mudstones (Natland & Kuenen 1951; Crowell et al. 1965). The early to middle Pleistocene deposits are conformable on the Pliocene and are characterized by the same facies types. During the middle Pleistocene, compression caused by the Pasadenan Orogeny produced folds and faults. This phase of compression rejuvenated the faults bordering Oak Ridge (north strikes) and produced strong vertical movements on the San Cayetano Thrust. The deformed Plio-Pleistocene beds dip at high angles to the south-east. Santa Paula Creek Section (Plio-Pleistocene) Within the Ventura Basin succession, the classical section of Santa Paula Creek (Natland 1933; Natland & Kuenen 1951) exposes the deformed Miocene sediments below the tilted succession of Plio-Pleistocene mainly turbiditic beds. The younger beds strike west-north-west and they were folded by the Pasadenan Orogeny. The dip is 50'-60 -J to the south-east and their stratigraphic thickness is about 3.6 km. Sedimentation was apparently continuous in a shallowing marine basin (1500 to 300 m in depth) (Natland 1933; Blake 1979). The sediments are mainly mudstones, siltstones and sandstones, containing deep-water foraminifera (Natland 1933), and secondarily, intervals of coarser-grained thickerbedded conglomerates and sandstones (Crowell et al. 1966; Johnson 1978) (Fig. 3). Johnson (1978) examined about half (1.8 kin) of this section and has documented seven megasequences of coarse-grained sediments that occur at irregular intervals in the hemipelagic mudstones and thin-bedded fine-grained turbidites that constitute the deep basin floor deposits of the ancient Ventura Basin. These seven megasequences, of varying thicknesses, are formed of sandstones and siltstones, together with thickbedded conglomerates and pebbly mudstones in mega-sequences II and VI. Mega-sequences composed of medium to very thick layers of conglomerates and sandstones were interpreted by Johnson as channel fill assemblages overlying interdistributary or fan margin mudstones and thin-bedded distal turbidites. Blake (1979) has examined the microfossil assemblages in the (hemipelagic) mudstones and suggest that most of the section measured by Johnson is early to middle Pleistocene with possibly some very late Pliocene in the lowest measured mega-sequence. Blake's correlations would suggest that the section described in this paper includes perhaps 600 000 yrs of record, all
4 21
younger than 1.2 my BP (the age of the Bailey ash in other sections). The total section measured by Johnson may extend back to almost 2 my BP SO that the first six mega-sequences he described averaged about 100000 yrs in separation, although they are not that regularly spaced. The gross accumulation rate is thus about 1 m/1000 yrs which is comparable to those in the contemporary Santa Barbara Basin (Soutar & Cril11977; Schwalbach 1982).
Sediments Above the coarse conglomeratic beds that constitute the lower part of mega-sequence VI (Johnson 1978) the upper portion of the Santa Paula Creek section includes four zones (Fig. 3). These include: (1) 185 m of fine sediments consisting of fine-grained thin turbidites and muds; (2) 47 m of coarser deposits composed of conglomerates and sands (base of mega-sequence VII of Johnson 1978); (3) 796 m of fine-grained thin turbidites with mudstones, and (4) 70 m of sands and conglomerates which form the uppermost deposits of the section and are unconformably overlain by later Pleistocene shallow water and subaerial deposits. The microfaunal data, based on about a dozen evenly spaced samples over the measured section, indicate that this time is one of general shallowing from about 1500 to less than 300 m (Blake 1979). Zone 1 The 185 m of sediments of Zone 1 (Fig 4) are composed mainly of mudstones containing foraminifera indicating a depth greater than 270 m, and of sands and fine sands in thin beds with classic turbidite structures (Bouma 1962). These sands and granule sands commonly appear in millimetre-thick beds showing primary distal lamination separated by millimetre-thick mud layers. More rarely these beds attain thicknesses of a few centimetres. The coarser sandy beds and granule sands often lie over a base of granulesized particles in successively obliquely stratified beds. These units have thicknesses of up to 10 cm and represent a rapid deposition on the interdistributary areas or on the basin floor. Locally, ripple marks appear on the upper surface of these thicker sands, It appears that the currents which deposited these sands maintained the same average direction during the period of deposition, but varied in intensity so that internal erosion periodically modified the general deposition of the cross-laminae. Mudstones, which appear homogeneous in outcrop and have thicknesses of up to 30 cm,
422
R. Bourrouilh and D.S. Gorsline
FIG. 3. Measured stratigraphic section at Santa Paula Creek, California. Roman numerals identify sequences cited in the text. (Johnson 1978; R. Bourrouilh & D.S. Gorsline). Ornament shows alternation of coarse-grained (stippled) and fine-grained (horizontal shading) turbidite facies.
Fine-grained sediments associated with fan lobes
423
FIG. 4. Photograph of Zone 1 sediments. Laminated fine to medium-grained sands alternating with laminated muds. when X-radiographed, commonly show oblique lamination in units with individual thicknesses of 1 mm or more. Parallel lamination is characteristic of the finer mudstones. We interpret the parallel lamination of the fine silts and muds to be mainly the product of climatic effects rather than solely due to deposition from distal turbidites. These alternations could have been regular as seasonally deposited beds, or could also be the result of more exceptional climatic events such as major floods every decade or so. Similar features have been observed in Santa Barbara Basin (Drake et al. 1972) and are the result of major floods at intervals of 10-30 yrs. Other exceptional events could have been the reworking of the narrow shelves of the northern source lands by storm waves (Emery 1960), or simply the product of normal nepheloid transport as observed in Santa Barbara Basin today (Drake 1972; Soutar & Crill 1977). In some silts and very fine sands the
oblique stratification appears to be the product of waning turbidity current deposition, characteristic of the outer fan or fan fringe. Weak bottom currents could also have been active. These contributions from the oxygenated shelf or slope contrast with the anoxic basinal environment indicated by dark laminated mudstones with no evidence ofbioturbation. These organic-rich beds may act as source rocks for hydrocarbons, as several oil seeps are found in the area of the section. Zone 2
Sandwiched between the two thicker fine-grained sequences (Zones 1 and 3) there are 47 m of thick-bedded ( > 1 m) conglomerates and sandstones which we have called Zone 2 (Figs 3 and 5). Overlying a weakly scoured basal contact we identify six main phases of deposition. These
424
R. Bourrouilh and D.S. Gorsline
(a)
(b)
Fine-grained sediments associated with .fan lobes
425
(c) FIG. 5. Photographs of Zone 2 sediments. (a) General view of massive sandstones. (b) Detail of sediments (upper part of Zone 2) showing dark mudstones with plant debris (above the coin), overlain by alternating silty mudstones and mudstones parallel lamination, and occasional ripple formation, locally deformed by thixotrophy. (c) Fine-grained sediments between thick turbidite sandstones. include: (1) an 9.5 m thick fining and thinning upward unit from conglomerates through sandstones with shale rip-up clasts to dark organicrich mudstones; (2) a 5 m thickening-upward sandstone sequence; (3) 6.25 m of coarse sandstones and conglomerates; (4) another thickening-upward sequence comprising 12.5 m of sandstones with slumps towards the top; (5) 4.75 m of finer-grained parallel-laminated sandstones; and (6) a 9 m thick fining and thinning-upward sandstone-mudstone sequence. We interpret the 47 m thick Zone 2 as the relatively distal part of a coarse-grained submarine lobe. There was a sharp incursion of material over the fine-grained basinal or fan fringe sediments of Zone 1, and then a rather more gradual return to the similar facies of Zone 3. This sudden increase in the rate of supply and the grain-size of the material may have been caused by tectonic activity, climatic change, lowered sea-level or a shift in position of the submarine lobe. Work in progress suggests that the main axis of the lobe was several kilometres west of the Santa Paula Creek section and we may be looking at fluctuation in fan dimension in a peripheral section. If so, then the expansion of the larger fan is likely to have been influenced by tectonic or climatic
factors rather than by simple lobe or channel avulsion. Nardin (1981) has described the expansion and contraction of fan zones in the Hueneme-Mugu Fan in Santa Monica Basin in the California Borderland during Wisconsin time. Similar changes with respect to sea-level fluctuations may be the forcing function in the Santa Paula Creek section. Shallowing of the basin floor as evidenced by foraminiferal data would suggest that sedimentation was more important than tectonism. In that case, climatic/sea-level changes may be the driving force. Zone 3
This zone includes 796 m of thin-bedded turbidites and mudstones (Figs 3 and 6). Although the change from the preceeding facies of Zone 2 was progressive, it nevertheless took place over a relatively short interval. Depth decreased from 600 to 270 m up section (G. Blake, pers. comm.). At this time the basin was probably anoxic, with deposition of very dark fine-grained deposits rich in organic debris and plant matter, but was open across its sills to the ocean. A precise indication of this connection was found in the discovery of a skeleton of the whale Megaptere, about 570 m
426
R. Bourrouilh and D.S. Gorsline
(a)
(b) FIG. 6. Photographs of Zone 3 sediments. (a) Parallel and oblique lamination in fine-grained sandstones and mudstones. (b) Climbing ripple lamination and coarse basal layer.
Fine-grained sediments associated with fan lobes above the base of this third zone. Analysis of the position of the skeleton shows that the whale sank rapidly after death and lay on the sea floor on its back. It is not badly disassociated which suggests an absence both of bottom scavenger activity and of strong currents. The sediment accumulation rate was fast enough so that the remains were apparently rapidly buried. The species was migratory and lived in the open ocean and shows no post-depositional transport. The basin was probably much like the contemporary borderland silled basins. More than 80% of Zone 3 sediments are mudstones with thin intercalated sandy and silty layers. The sands and silts exhibit the classic turbidity current depositional structures over a much reduced interval; graded bedding passes to cross lamination and then to parallel lamination. Thicknesses of the sand and silt beds range from a few millimetres to a few centimetres and these are separated by thick mudstones. The silts display a variety of oblique lamination types: (1) oblique laminae characteristic of the limit of laminar flow; (2) laminae formed by traction and erosion transport; (3) laminae accompanied by thixotropic structures; (4) laminae formed by fluctuating flow characteristics. This oblique lamination is probably formed either by reworking of the substrate by periodic bottom currents or slope currents, or by turbidity current overbank flows from channels across the adjacent levees. The mudstones are also laminated and some show oblique lamination in X-radiographs. At certain horizons, coarser sediments are intercalated (Fig. 7). For example, about 240 m above the boundary with Zone 2, up to 20 thick diamictites were deposited in beds averaging 10-20 cm in thickness (maximum 90 cm). Most of these beds contain gravels and soft sediment clasts; some have slump structures and thixotropic features. Our interpretation is that these sediments were deposited in a silled open basin with a diminishing sediment supply. The rare thicker diamictites may have been generated by slumps on the upper slope and resedimented as debris-flows and turbidity currents. Zone 4
Zone 4 comprises 70 m of conglomerates, pebbly mudstones and sandstones with minor siltstones and mudstones (Figs 8 and 9). The transition from Zone 3 mudstones is relatively abrupt (Fig. 3). We have divided the zone into five sequences FIG. 7. Detailed section of part of Zone 3 (240 m above base of zone) showing coarser-grained interbeds.
427
428
R. Bourrouilh and D.S. Gorsline
o N
.o
06
Fine-grained sediments associated with fan lobes
429
.=.
o
~o
:Z
r,g3
0
0
0
0
.E
0 N
Y. o 0 9
o
430
R. Bourrouilh and D.S. Gorsline
including: (1) a 12 m thickening-up and coarsening-up conglomeratic sequence, the last bed attaining a thickness of 5 m, overlain by a finer-grained thickening-up turbiditic sequence ending with two conglomerate beds; (2) a 12 m thickening-up conglomeratic sequence containing rolled pebbles with long axes of up to 60 cm; (3) 18 m of sandstone and conglomeratic turbidites, with some pebbly mudstones containing dispersed pebble layers; (4) 12 m of sandstoneconglomeratic turbidites (several over 3 m thick) with interbedded mudstones; and (5) a 17-18 m thickening-up conglomeratic sequence with a maximum bed thickness of 1.5 m. In detail, the beds of Zone 4 show a number of interesting characteristics. The conglomerates are mostly mud-supported or, more rarely, sand-supported. The pebbles are of heterogeneous composition from mixed sources, rounded and elongate (up to 60 cm) perhaps indicating a period of fluvial transport. Both turbiditic and debris-flow sandstones and siltstones are identified, with sand dikes, sand and mud volcanoes, flutes, ripple marks and 'hard-grounds'. Fine-grained turbiditic mudstones and thicker-bedded pebbly mudstones also occur. Many of the beds are organicrich with much terrigenous plant debris, and this appears to have matured sufficiently in some sections to yield hydrocarbons and oil seeps. It appears that these sediments have been deposited more rapidly than those of Zone 2. The thickness of beds, reduced slumping, high rates of sediment supply and benthic fossil evidence indicates a shallowing, low-gradient environment in which sedimentation has outpaced tectonic changes. Water depths appear to have been about 200-250 m during this period. Channels may have been meandering in form. Neptunian sand dikes and sand or mud volcanoes may be indicative of seismic activity and shock wave remobilization of the debris and mud flows.
Conclusions The Ventura Basin was a narrow east-west trough bordered to north and south by fault systems (Fig. 10). During Plio-Pleistocene time prior to the Pasadenan orogeny, both the northern and southern margins appear to have been tectonically active leading to basin and downwarping (Natland & Kuenen 1951; Crowell et al. 1966; Crowell 1976). Both tectonic effects and sea-level fluctuations were probably responsible for the frequent generation of mass-movement and turbidity current flows. Data presented in this paper show that the Ventura basin was probably in open communication with the ocean over shallow sills, and with other basins to the east. The characteristic background sedimentation comprises mudstones and fine-grained turbidites. Within these finer grained deposits, thick-bedded sandstones and conglomerates occur at irregular intervals forming mega-sequences up to 50 m thick. These are made up of smaller-scale sequences that are both fining-upward, coarsening-upward and irregular. The episodic occurrence of deep water sands in the section, the lack of well-defined channels and the evidence of rapidly-accumulated poorlysorted sediments indicates rapid shifts in depositional locale, relatively steep slopes and rapid transfer to the basin floor. This may have been the result of tectonic events or shifts in submarine lobes or variations in sea-level related to climate. Several canyon sources may have been active. Work on other parts of the Ventura Basin further to the west (Hsu et al. 1980) suggests that several canyon-fan systems were present at intervals along the long east-west northern margin of the basin, rather than one or two large points of influx. The section exposed in Santa Paula Creek is probably offset by a few kilometres from the main part of the lobe or fan axis. N
~ ~ O ~ S A N i
C~IEL
-'~
FAUL
T
S
( ~ NARROW EAST'WEST TROtK]V~ BORDERED TO NORTH A ~ SOUTH BY FAULT SYSTEM
COM'~UNICATION WITH OCEAN OVER SHALLOW SILLS|
Ir"=::E ;~ " " - : ~ Il ~ ' ~ I1 ~'~t~.~'~l~..',w.~,,.\ rI ~ ~ ~ ~ [ ,L.._
,
~
~
-~.--~=~i
'* " - , ~ _ . . , = . ~ ~ _ ~ ; ~ C ' ~ - - ~ ~ z ! ~ ! _ ~
~
il
i
i
, ]' I' I '
I '
/J,
I
I
' I
l
I
S0PPOSI~
~oqO~d~
BASEVENT ~ P I N G
~.
/
not to scale
FIG. 10. Interpretation of Ventura Basin tectonic setting and evolution.
Fine-grained sediments associated with .fan lobes In Pliocene time the Ventura Basin may have been similar in form and pattern to the presentday Santa Barbara Basin (Thornton 1981, Fig. 42). The presence of slumps in the megasequences, and the evidence for rapid deposition show that both canyon-fan and slope-centred systems were operative (Gorsline 1978). Fan development may have been mainly as lobes rather than as fully developed fan systems (Mutti 1979). The sand content of the adjacent river basins may have been high and Mutti's low-efficiency fan model may be most applicable. It is likely then that each sequence represents the progradation of a new lobe as either transport paths shifted or as new cycles of sea-level or tectonism occurred. The periodicity of roughly 100 000 yrs between lobes suggests that tectonic or climatic change (sea-level change) are more likely than lobe shifts which might be expected to occur with greater frequency. Certainly at times, fine-grained sedimentation dominated the section and thus indicates times of either reduced supply or lower gradient. The strong influence of sedimentation rate is well documented in the contemporary California Borderland basins where the surface circulation controls the suspended sedimentation (Gorsline 1980; Thornton 1981) aided by biological aggregation and pelleting (Gorsline 1983). In the areas of high sedimentation the deposits accumulate faster than the tectonic rates can deform them and lenses or nearly horizontal and conformable sands and muds form the basin fill. In the contemporary borderland, the basins seaward of the inner zone of high sedimentation rate typically contain warped layers indicating that tectonism is operating at rates comparable to the accumulation rates and may exceed those rates.
431
Santa Paula Creek section represents a passage from times of high accumulation and tectonic activity to times of tectonic quiet and decreased sedimentation rate. The influence of sea-level change may also be present. The range of frequencies of important climatic cycles probably goes up to 105 yrs while the important tectonic cycles probably begin there and extend into periods of 106 yrs and more for large-scale global cycles. The rapid shoaling of the basin at the top of the Plio-Pleistocene section may represent dominance of sedimentation but in a regime of decreasing rates of both sedimentation and tectonism. This was ended by the renewed orogenic activity of the Pasadenan Orogeny of mid-Pleistocene time (c. 500 000 yrs BP). The fine-grained sediments in the basin were deposited by hemipelagic processes, turbidity currents and associated mass-flows. Thinlylaminated dark organic-rich shales represent periods of sedimentation in an anoxic basin, perhaps recording seasonal cycles or major floods. Pebbly mudstones and obliquelylaminated siltstones are undoubtedly the product of distal turbidity current deposition and massmovements. The mudstones with intercalated fine sand and silt turbidites less than 1 cm in thickness may be intrafan deposits of the lower fan or basin floor. The section illustrates the multiplicity of fine-grained sediment types and the broad range of transport processes that deliver them to the basin floor, ACKNOWLEDGEMENTS: We thank the National Science Foundation and the D G R S T for financial support. Dorrik Stow reviewed and revised an earlier version of the manuscript.
References ALMAGOR, G. 1982. Marine geotechnical studies of continental margins: a review--Part II. Appl. Ocean Res., 4, 130-50. ANDERSON, D.L. 1971. The San Andreas Fault. Sci. Am., 225, 53-68. ATWATER, T. 1970. Implications of plate tectonics for
the Cenozoic tectonic evolution of western North America. Bull. geol. Soe. Am., 81, 3513-36. - - & MOLNAR, P. 1973. Relative motion of the Pacific and North American Plates deduced from sea-floor spreading in the Atlantic, Indian and south Pacific Oceans. In: Kovach, R.L. & Nin, A. (eds), Proc. Conference on Tectonic Problems of the San Andreas Fault System. Stanford University Pubs. Geol. Sci., 13, 136-48. BENNETT, R.H. & NELSEN, T.A. 1983. Seafloor characteristics and dynamics affecting geotechnical properties at shelf breaks. In: Stanley, D.J. & Moore,
G.T. (eds), The Shelf Break: Critical Interface on Continental Margins. Soc. econ. Paleo. Min. Spec. Pub. 33, 333-55. BLAKE, G.H. 1979. Palaeobathymetric and Depositional History of Basinal Sediments, Santa Paula Creek, California. Unpub. Research Paper, Univ. of So. Calif., Los Angeles, California. 14 pp. BLAKE, M.C., CAMPBELL,R.H., DIBBLEE, T.W., HOWELL, D.G., NILSEN, T.H., NORMARK,W.R., VEDDER, J.C. & SILVER,E.A. 1978. Neogene basin formation in relation to Plate-Tectonic evolution of San Andreas Fault System, California. Bull. Am. Ass. Petrol. Geol., 62, 344-72. BOUMA, A.H. 1962. Sedimentology of some Flysch Deposits. Elsevier, Amsterdam. 168 pp. CONREV, B.L. 1959. Sedimentary History of the Lower Pliocene in the Los Angeles Basin, CaliJornia. Un-
432
R. Bourrouilh and D.S. Gorsline
pub. Ph.D. Dissertation, Univ. of So. Calif., Los Angeles, California. 268 pp. CROUCH, J.K. 1979. Tectonic history of the outer part of the California Borderland. Geol. Soc. Am. Abst. Pros., 11, 407. CROWELL, J.C. 1974. Origin of late Cenozoic basins in California. In: Dickinson, W.R. (ed.), Tectonics and Sedimentation. Soc. econ. Paleo. Min. Spec. Pub. 22, 190-204. - 1976. Implications of crustal stretching and shortening of coastal Ventura Basin, California. In: Howell, D.G. (ed.), Aspects of the Geologic History of the California Continental Borderland. Pacific Section. Am. Ass. Petrol. Geol., Misc. Publ. 24, 365-82. --, HOPE, R.A., KAHLE, J.E., OVENSHINE, A.T. & SAMS, R.H. 1966. Deep-water sedimentary structures Pliocene Formation, Santa Paula Creek, Ventura Basin, California. Spec. Report 89, Cali[i Dir. Mines and Geol., 40 pp. DAMON, P.E. 1980. Continental uplift at convergent boundaries. In: Plateau Uplift: mode andmechanism. Lunar and Planetary Laboratory Topical Conference Proc. Flagstaff, Arizona. DAMUTH, J.E. & EMBLEY,R.W. 1979. Mass-wasting on the continental rise of eastern South America. Bull. Am. Ass. Petrol. Geol., 63, 438. DOYLE, L.J. & PXLKEY, O.H. 1979. Geology of Continental Slopes. Soc. econ. Paleo. Min. Spec. Pub., 27, 374 pp. DRAKE, D.E., KOLPACK, R.L. & FISCHER, P.J. 1972. Sediment transport on the Santa Barbara-Oxnard shelf, Santa Barbara Channel, California. In: Swift, D.J.P., Duane, D.B. & Pilkey, O.H. (eds), Shelf Sediment Transport. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 307-30. 1972. Distribution and Transport of Suspended Matter, Santa Barbara Channel, California. Unpub. Ph.D. Dissertation, Univ. of So. Calif., Los Angeles, California. 358 pp. EMBLEY,R.W. 1976. New evidence for the occurrence of debris flow in the deep sea. Geology, 4, 371-4. - 1980. The role of mass transport in the distribution and character of deep-ocean sediments with special reference to the North Atlantic. Marine Geol., 38, 23-50. EMERY, K.O. 1960. Sea Off Southern California. John Wiley, New York. 366 pp. FIELD, M.E. & EDWARDS, B.D. 1980. Slopes of the southern California Continental Borderland: a regime of mass transport. In: Colburn, I., Bouma, A., Field, M., Douglas, R. & Ingle, J. (eds), Quaternary Depositional Environments of the Pacific Coast. Pacific Section. Soc. econ. Paleo. Min. Symposium 4, 169-84. GORSLINE, D.S. 1978. Anatomy of margin basins. J. sed. Petrol., 48, 1055-68. -1980. Deep-water sedimentologic conditions and models. Marine Geol., 38, 1-21. - - & EMERY,K.O. 1959. Turbidity current deposits in San Pedro and Santa Monica Basins off southern California. Bull. geol. Soc. Am., 70, 279-88. HANER, B.E. 1971. Morphology and sediments of Redondo Submarine Fan, southern California. Bull. geol. Soc. Am., 82, 2413-32.
--
& GORSLINE, D.S. 1979. Processes and morphology of continental slope between Santa Monica and Dume Submarine Canyons, southern California. Marine Geol., 28, 77-87. HARTNETT, T.M. 1980. Vertical Sequence Analysis of Late Pliocene Pico Formation Sediments in Adams Canyon, Ventura County, California. Unpub. M.S. Thesis, Univ. of So. Calif., Los Angeles, California. 184 pp. HEIN, F.J. & WALKER,R.G. 1982. The Cambro-Ordovician Cap Enrage Formation, Quebec, Canada: a coarse-filled submarine channel. Sedimentology, 29, 309-30. Hsu, K.J., KELTY, K. & VALENTINE,J.W. 1980. Resedimented facies in Ventura Basin, California and model of longitudinal transport of turbidity currents. Bull. Am. Ass. Petrol. Geol., 64, 1034-51. INGLE, J.C. 1980. Cenozoic paleobathymetry and depositional history of selected sequences within the southern California Continental Borderland. Cushman Found. Spec. Publ., 19, 163-95. JACOBI,R.D. 1976. Sediment slides on the northwestern continental margin of Africa. Marine Geol., 22, 157-73. JENNINGS, C.W. & TROXEL, B.W. 1954. Ventura Basin. In: Jahns, R.H. (ed.), Geology of Southern California. Calif. Div. Mines and Geol. Bull. 170, Guide No. 2. 63 pp. JOHNSON, B.A. 1978. Vertical Sequence Analysis of a Deep-Sea Fan System, Santa Paula Creek, California. Unpub. M.S. Thesis Univ. of So. Calif., Los Angeles, California. 204 pp. KRUIT, C. 1975. Une excursion aux c6nes d'alluvions en eau profonde d'fige Tertiare pres de San Sebastian. International Sed. Congress. Guide pour l~xcursion Z-23, Nice, 75 pp. M1DDLETON, G.V. & BOUMA,A.H. 1973. Turbidites and Deep-water Sedimentation. Pacific Section, Soc. econ. Paleo. Min. Short Course Notes, 158 pp. M1DDLETON, G.V. & HAMPTON, M.A. 1972. SEDIMENT GRAVITYFLOWS:MECHANISMSOFFLOWANDDEPOSITION. In: Middleton, G.V. & Bouma, A.H. (eds), Turbidites and deep-water sedimentation. Pacific Section Soc. econ. Paleo. Min. Short Course Notes, 1-38. MUTTI, E. 1977. Distinctive thin-bedded turbidite facies and related depositional environments in the Miocene Hecho Group (south central Pyrenees, Spain). Sedimentology, 24, 107-31. - 1979. Turbidites et c6nes sous-marins profonds. In: Homewood, P. (ed.), Skdimentation Dbtritique (fluviatile, littorale et marine). 36me Cycle Romand en Sciences de la Terre, Univ. Fribourg, Switzerland. 353-419. & RICcI-LUccHI, F. 1972. Turbidites of the northern Apennines: Introduction to facies analysis. lnternl. Geol. Rev., 20, 125-66. -& WALKER, R.G. 1973. Turbidite facies and facies associations. In: Middleton, G.V. & Bouma, A.H. (eds), Turbidites and deep-water sedimentation. Pacific Section. Soc. econ. Paleo. Min., Short Course Notes. 119-58. NAGLE, H.E. & PARKER, E.S. 1971. Future oil and gas potential of onshore Ventura Basin, California. In:
Fine-grained sediments associated with .fan lobes Cram, I.H. (ed.), Future petroleum provinces of the U.S., Am. Ass. Petrol. Geol. Mem., 15, 254-97. NARDIN, T.R. 1981. Seismic Stratigraphy of Santa Monica and San Pedro Basins, California Continental Borderland." Late Neogene History of Sedimentation and Tectonics. Unpub. Ph.D. Dissertation. Univ. of So. Calif., Los Angeles, California. 295 pp. --, HEIN, F.J. & GORSL1NE, D.S. 1979. A review of mass movement processes, sediment and acoustic characteristics and contrasts in slope and base-ofslope systems versus canyon-fan-basin floor systems. In: Doyle, L.J. & Pilkey, O.H. (eds), Geology of Continental Slopes. Soc. econ. Paleo. Min. Spec. Pub. 27, 61-73. NATLAND, M.L. 1933. Depth and temperature distribution of some Recent and fossil Foraminifera in the southern California region. Bull. Scripps Inst. Oceanography, La Jolla, California. - & KUENEN, PH.H. 1951. Sedimentary history of the Ventura Basin, California and the action of turbidity currents. In: Turbidity Currents and the Transportation of Coarse Sediment to Deep Water. Soc. econ. Paleo. Min. Spec. Publ. 2, 76-107. NILSEN, T.H. 1980. Modern and ancient submarine fans: discussion of papers by R.G. Walker and W.R. Normark. Bull. Am. Ass. Petrol. Geol. Bull., 64, 1094-101. NORMARK, W.R. 1978a. Fan valleys, channels and depositional lobes on modern submarine fans: characters for recognitions of sandy turbidite environments. Bull. Am. Ass. Petrol. Geol., 62, 912-31. - 1978b. Turbidites, time and again. Bull. Am. Ass. Petrol. Geol., 62, 549.
433
PILGER, R.H. & HENYEY, T.L. 1979. Pacific-North American Plate interaction and Neogene volcanism in coastal California. Tectonophysics, 57, 189-209. PIPER, D.J.W. & NORMARK, W.R. 1971. Re-examination of a Miocene deep-sea fan and fan valley, southern California. Bull. geol. Soc. Am., 82, 1823-30. SCHWALBACH, J.R. 1982. A Sediment Budget ,/'or the Northern Region of the California Continental Borderland. Unpubl. M.S. Thesis, Univ. of So. Calif., Los Angeles, California. 212 pp. SLOSSON, J.E. 1958. Lithofacies and Sedimentary Paleogeographic Analysis of the Los Angeles Repetto Basin. Unpubl. Ph.D. Dissertation, Univ. of So. Calif., Los Angeles, California. 128 pp. SOUTAR, A. & CRILL, P.A. 1977. Sedimentation and climatic patterns in the Santa Barbara Basin during the 19th and 20th centuries. Bull. geol. Soc. Am., 8 8 , 1161-72. THORNTON, S.E. 1981. Suspended sediment transport in surface waters of the California Current off southern California: 1977-78 floods. Geo-Marine Letters, 1, 23-8. WALKER, R.G. 1978. Deep-water facies and ancient submarine fans: models for exploration for stratigraphic traps. Bull. Am. Ass. Petrol. Geol., 62, 932-66. WINN, R.D. & DOTT, R.H. 1979. Deep water fan channel conglomerates of late Cretaceous age, southern Chile. Sedimentology, 26, 203-28.
R. BOURROUILH,Laboratoire de Ge61ogie, Universit6 de PAU, Avenue Philippon, F.64000 PAU, France. D.S. GORSLINE, Department of Geological Sciences, University of Southern California, Los Angeles, California 90007, USA.
Origin of varve-type lamination, graded claystones and limestone-shale 'couplets' in the lower Cretaceous of the western North Atlantic A.H.F. Robertson SUMMARY: The petrography and sedimentary structures of pelagic and hemipelagic sediments in the early Cretaceous interval of the Blake-Bahama Basin reveal a subtle interplay between terrigenous source materials, plankton productivity and diagenesis. Three specific sediment types drilled at Site 534A (Leg 76) of the DSDP are: (1) persistent, fine parallel lamination resembling varves; (2) thin, size-graded nannofossil claystones and 'black shales', and (3) 'couplets', composed of pale burrowed and dark laminated nannofossil radiolarian carbonates. The varve-type lamination reflects some combination of fluctuations of terrigenous plant material plus fine clastic input and plankton productivity. Both possibly reflect short periodicity (10s to 100s of years) climatic variations. Lithification took longer in the clay- and organic-rich laminae. Slow compaction then accentuated the lamination. By contrast, purer nannofossil oozes lithified more quickly and now show less compaction. The graded claystones and 'black shales' were mostly redeposited by turbidites from within, or near, the oxygen-minimum zone on the upper continental slope. Turbiditic input reduces the need for bottom water anoxicity. The dark-light couplets reflect wetter periods on land averaging 20-60 000 years, during which abundant plant material entered the Atlantic. Contrary to some recent suggestions, the early Cretaceous Atlantic need not have been poorly circulated and weakly oxygenated. Instead, both short and longer periodicity climatic fluctuations controlled input of mostly land-derived material. High rates of organic matter input, partly turbiditic, produced sub-surface sediment anoxia below a circulating relatively fertile early Cretaceous Atlantic Ocean. Many recent studies of depositional processes in the oceans have concentrated on particular mechanisms, for example, attempts to distinguish distal turbidites from contourites (Stow 1979). On the other hand, it is becoming increasingly clear that sedimentation in the deep oceans must reflect a complex interplay of variables which include source, oceanography, climate and diagenesis. This is particularly true of the abyssal basins adjacent to passive margins. Here the role of depositional and diagenetic processes in the lower Cretaceous interval of Blake-Bahama Basin in the Western North Atlantic is examined. Particularly, the lower Cretaceous deposition at Site 534A, drilled on DSDP leg 76, is discussed along with evidence from related drill sites (Fig. 1). During early Cretaceous times (Berriasian-Barremian) Site 534A was located above the calcium compensation depth (CCD) at water depths of c.3.5 km (back-tracking; Sheridan et al., 1982). Sediments accumulated in an abyssal basin around 70 km east of a major carbonate platform, now the deeper unexposed parts of the Blake Plateau. Site 534A shows an interplay between pelagic accumulation and both terrigenous and bioclastic input from the continental margin (Sheridan et al. 1982; Sheridan et al. 1984). Attention is focused on the nature and origin of
three specific sediment types which are widespread in the lower Cretaceous Atlantic and have been noted elsewhere in the oceans and on land: (1) persistent fine parallel lamination resembling varves; (2) thin graded nannofossil claystones and black shales, and (3) cyclical alterations of pale burrowed and dark laminated nannofossil carbonates ('couplets'). Explanation of these features depends on a combination of fluctuations of surface productivity, elevated terrestrial input, and gravity redeposition, plus the effects of differential diagenesis. Contrary to some earlier views there may be little need to invoke a well stratified poorly oxidized ocean, at least in the early Cretaceous Atlantic.
Regional setting Throughout the north-western Atlantic the lithologies of latest Jurassic (Tithonian) to early Cretaceous (Hauterivian) age have been named the Blake-Bahama Formation, which is predominantly composed of pelagic carbonates (Jansa et al. 1979). Similar successions extend into the eastern North Atlantic (Cape Verde Basin) (Lancelot et al. 1978) and into the western Tethys as the Maiolica (Bernoulli 1972). During Leg 76, 392 m of the Blake-Bahama Formation were drilled, 437
438
A.H.F. Robertson
BLAKE BAHAMA FORMATION e ~
j
4~176
M .3~
/ ~
J
/
40~
. 387: / ~ ; BERMUDA
t BAH_A.MA / 11 "" \ BASIN ] , , / t ~ t-i.391[323,-335,,]---'~I- .I
"N
I
\
I
L,o,
30~
'
-20~
20 ~
Be I
7r I
6r I
FIG. 1. DSDP sites where early Cretaceous pelagic limestones have been drilled in the western North Atlantic. Thickness of the Blake-Bahama Formation shown in brackets. Horizon B correlates approximately with the top of the Blake-Bahama limestones: the eastern boundary indicates pinch-out against basement, near magnetic anomaly M-11, interpreted as Valanginian (modified after Jansa et al. 1978). of which 267 m were recovered. A fuller account of the sedimentology of the succession at Site 534A is given in the Initial Report for Leg 76, including clay mineralogy, major element chemistry and carbon-oxygen isotopic analyses (Robertson & Bliefnick 1984). The discussion in this paper centres on the Valanginian, Hauterivian and Barremian, excluding the Aptian-Albian higher interval of the lower Cretaceous. One advantage of studying the pre-Aptian lower Cretaceous is that anoxicity in the sediment is only incipient, so that the various competing processes of input and diagenesis are highlighted. The generalized descriptions of DSDP cores published in the Initial Reports often obscure important sedimentary details, particularly small-scale sedimentary structures. Close examination of core intervals
can reveal a complex interplay of competing sedimentary processes.
Lithologies Using DSDP nomenclature the following lithologies are recognized in the lower Cretaceous interval: (l) nannofossil limestones; (2) marly nannofossil chalk; (3) nannofossil claystones; (4) black organic-rich claystones and shales; (5) both bioclastic and terrigenous redeposited elastics. The relative abundance of each of these lithologies is summarized in Fig. 2. The lower part of the succession, of early Berriasian age, is dominated by nannofossil limestones, which decrease upwards and disappear by the end of late Valanginian time, while marly nannofossil chalks
Varve-type lam&ation, graded claystones and limestone-shale 'couplets'
CORE NUMBER 47
0
50
439
100%
49
5o
il Barremian
52 53 54 55 56
...
'
57 / / / "4/"~''-'~:'b~" 5958 5a .'.~.".~.~~ L.__ 60 .... 62 63 64
,
tl Hauterivian ' ~]
i 9 "
"'.i ." .' ".
6867 69 7O 5b 71 72 73 74 75 76
Early Va langinian
78 79 5c 8O 82
83 84 85 86 87 88 89 9o 91
Va langinian
Late Berriasian 5d
Early Berriasian ~
Marly nannofossilchalk (Undifferentiated) Palelaminated Dark laminated
Quartzose,siltstoneand sandstone Sandylimestone Nannofossillimestone
Claystone
Black shale
FIG. 2. Summary of relative abundances of lithologies in the early Cretaceous interval drilled at Site 534A. The quartzose siltstones and fine-grained sandstones and sandy limestones were deposited by turbidity currents. Note the relative abundance of dark organic-rich claystones ('black shales'). Marly nannofossil chalks, both burrowed and laminated give way to nannofossil limestones in early Berriasian. Compiled from shipboard Visual Description Sheets.
44 ~
A.H.F. Robertson
become more abundant. The marly nannofossil chalks range from massive, to highly burrowed, relatively well-cemented, to finely laminated, more argillaceous, and softer facies. The finely laminated marly nannofossil chalks are present both as paler and darker intercalations. The darker laminated intervals are richer in organic matter, shown by Habib (1979) to be mostly terrestrial in origin. The nannofossil claystones exist as thin size-graded units ( < 5 cm) reaching a relative maximum in late Valanginian time. The black carbonaceous claystones and 'black shales' are present throughout the succession but become most abundant above lower Berriasian, where they locally reach 25% of the sediment volume. The redeposited clastics are restricted to the late Valanginian to Barremian, becoming dominant in early Hauterivian (Fig. 2). Similar lithologies, including varve-type lamination and alterations of nannofossil, chalk, claystone and limestone, have also been encountered in the lower Cretaceous interval of the Western North Atlantic on the Lower Continental Rise Hills (Site 105, Ewing & Hollister 1972), on the Blake-Bahama Outer Ridge (Site 101, Ewing & Hollister 1972), on the Bermuda Rise (Site 387, Tucholke et al. 1979) and in the Blake-Bahama Basin at Site 391, located 22 km NW of Site 534A (Benson et al. 1978). Additionally, a very similar succession is seen in the Eastern North Atlantic in the Cape Verde Basin (Site 367, Lancelot et al. 1978). Comparisons show that carbonate-claystone 'couplets' and varve-type lamination extended across the lower Cretaceous Atlantic from the lower Continental Rise, towards the Mid-Atlantic ridge, and again appear in the Eastern Atlantic basin. These features are obscured where gravitydeposited clastics dominated the succession, for example off Morocco. Any explanation of the couplets and varve-type lamination must explain their wide distribution throughout the ocean not closely related to a particular continental margin. We now focus, in turn, on the origin of: (1) the fine parallel varve-type lamination; (2) the thinbedded nannofossil claystones and black shales; (3) the dark-light nannofossil chalk 'couplets', and (4) the role ofdiagenesis. The descriptions are based on detailed observation of cores from Site 534A in the Blake-Bahama Basin, supplemented by regional comparisons.
Varve-type lamination Ubiquitous fine lamination in the marly nannofossil chalks (Fig. 2) first appears in the Berriasian and persists to Aptian time at both Sites 534A and 391 in the Blake-Bahama Basin. The finer lamina-
tion in both the paler and darker facies (Fig. 3(a), (b), (c)) was examined petrographically. In thin section the lamination in the paler nannofossil chalks could rarely be resolved at all. Thicker (to 1 ram) paler laminations are sometimes composed of slightly flattened radiolarian shells replaced by calcite, set in a micritic matrix (Fig. 4(d), (e)). Lamination in the darker facies tends to be more clearly defined. At high magnification, individual laminations are typically seen to comprise numerous minute flecks of translucent brownish organic matter, plus occasional pyrite and dolomite in a micritic matrix. In detail, the parallel lamination often exhibits a micro-lenticular or streaky texture (Fig. 4(a)). In some cases the organic matter is seen to be associated with relative enrichment of mica and fine quartzose silt. Radiolarian shells may either follow individual fine laminations (Fig. 4(e)) or may be scattered randomly. The fine lamination is hard to resolve even with the scanning electron microscope, but a well-defined platy fabric of preferentially orientated calcareous nannoplankton can sometimes be seen (Fig. 4(b)). This fabric contrasts strongly with the near-random fabric seen in the intercalated nannoplankton claystones which are much less recrystallized (see below, Fig.
4(c)). The varve-like lamination has been noted in other DSDP cores particularly at Site 391, 22 km NW of Site 534 in the Blake-Bahama Basin (Benson et al. 1978), at Site 387 on the Lower Bermuda Rise Hill (Tucholke et al. 1979), at Site 101 at the south end of the Blake Plateau outer ridge (Ewing & Hollister 1972), at Site 535 in the Florida Straits (Buffler et al., in press), and in the eastern North Atlantic at Site 367 in the Cape Verde Basin (Lancelot et al. 1978). In addition, varve-type sediments have also been drilled by the DSDP in mid-upper Cretaceous successions elsewhere, including the South Atlantic (Bolli et al. 1978) and around Mid-Pacific Mountains and the Southern Hess Rise (Dean & Claypool 1981). The fine lamination at Site 387 (e.g. Cores 46, 47) on the Bermuda Rise is particularly important as this site was located close to the spreading ridge during early Cretaceous time. On the Lower Continental Rise Hills (Site 105) lamination is disrupted by extensive syn-sedimentary deformation not seen in oceanward sites. At Site 391 Freeman & Enos (1978) were able to count up to 67 laminations per 1 cm of sediment thickness, equivalent to one lamination per 6 years on average. At Site 534A only up to 30 laminations per centimetre could be detected, equivalent to 1 lamination per 15 years, assuming sedimentation rates of 21 cm/my (Sheridan et al. 1984).
Varve-type lamination, graded claystones and limestone-shale 'couplets'
441
FIG. 3. Photographs of early Cretaceous sediment cores from Site 534A, Blake Bahama Basin. (a) Highly bioturbated marly nannofossil limestone with subordinate intercalations of organic-rich varve-type nannofossil claystone. The limestones are much harder than the organic-rich laminae. Core 7, Section 4, 53-88 cm (Valanginian). (b) Smooth alternation between paler and darker varve-type laminated marly nannofossil chalk. The 1-2 cm thick massive units with some burrowing (e.g. lower) are graded turbiditic nannofossil claystones. Note the absence of burrowing in the varve-type laminations. Core 72, Section 5, 84-118 cm (Valanginian). (c) Graded nannofossil claystones turbidites (massive dark grey) alternating with subordinate finely laminated sediment. From the interval of Late Valanginian to Hauterivian high terrigenous input. Core 61, Section 3, 21-55 cm.
Mode of deposition
TurbMites
Ideally, fine lamination in the ocean could relate to one or a c o m b i n a t i o n of: (1) B o u m a Tb, Td divisions of classicial turbidites; (2) contourites; (3) nepheloid-flow and drift deposits; (4) suspension-cascading of lutites; (5) variations in terrestrially-derived clastics, organics or nutrients; (6) plankton productivity variations; (7) diagenesis. Taking these alternatives in turn:
A turbiditic origin is easily ruled out; in the Bouma sequence, the laminations could only belong to the Td division, but neither the interturbidite Bouma Te nor other divisions are seen. Regionally, there is no sign of any obvious proximal-distal variations related to any particular source. Similar lamination is seen in the lower Cretaceous on both sides of the Atlantic and on the palaeo-ridge flanks.
442
A.H.F. Robertson
FIG. 4. Photomicrographs and scanning electron micrographs of early Cretaceous facies at Site 534A. (a) Typical well cemented marly nannofossil limestone. Note that the radiolaria are replaced by calcite but not flattened. Many tests were partly replaced by pyrite early in diagenesis. Core 79, Section 2, at 74-76 cm. (b) Marly nannofossil chalk. Calcareous nannofossils heavily overgrown with blocky calcite crystals. Core 71, Section 3 at 80-82 cm (c) Nannofossil claystone turbidites. Note the excellent preservation and random orientation of calcareous nannofossils. This contrasts sharply with the platy fabric and greater recrystallization of the intercalated marly nannofossil chalks. Core 48, Section 2, 185-167 cm. (d) Marly nannofossil chalk. The radiolarian shells have been replaced by calcite and flattened. Core 68, Section 5,55-58 cm. (e) Typical 'varve'-laminated marly nannofossil chalk. The white blobs are slightly flattened radiolaria replaced by calcite. Core 48, Section 3, 50-52 cm. (f) Laminated quartzose silt of turbiditic origin. In contrast to the 'varves' this lamination is defined by quartzose silt and is associated with coarser grained turbidites. Core 69, Section 7, 7-10 cm.
Varve-type lamination, graded claystones and limestone-shale 'couplets' 443 Contourites
Bottom currents, or specifically contour-currents, might rework sediment to form a fine lamination. Indeed in the Atlantic lower Cretaceous at least some bottom water motion is implied by the preferential orientation of grains in the marly nannofossil chalks. Reworking by bottom currents to form the fine lamination is, however, unlikely: the laminations are regular and laterally continuous at least within cores. Ripples, crosslamination, scours or other sedimentary structures normally associated with current deposits (e.g. Stow 1979) are not seen. Unlike the Holocene contourites of the Blake Outer Ridge (Heezen et al. 1966; Hollister & Heezen 1972), the lamination is not restricted to the lower continental rise or any other specific local setting, but instead occurs over wide areas of the abyssal plain on both sides of the Cretaceous Atlantic. The fine lamination also contrasts with drilled contourite deposits, for example Site 533 (Leg 76) on the Blake Outer Ridge (Sheridan et al., 1984). This is an almost featureless pile of muddy nannofossil ooze with few visible sedimentary structures. Also, the seismic signature of current-drift deposits appears to be that of hummocky reflectors, seen for example, in the Blake Outer Ridge (Site 533; Sheridan et al. 1982). By contrast, reflectors in the lower Cretaceous are basin-levelling, typical of turbidites and pelagic sediments. Lutite flow
McCave (1972) developed a model in which lutite flows (low-density, low-velocity turbidites) are not able to penetrate the ocean water density structure, but instead reach a density interface, then flow out over clearer water finally losing particles by settling. This could produce an episodic 'suspension cascade' and so conceivably produce a fine lamination. One problem pointed out by McCave is that the Coriolis force would deflect such low-density turbid currents parallel to the slope and thus fail to reach the deep ocean basins. Also this mechanism might be expected to produce graded partings of similar composition rather than the sharp compositional variations seen.
Terrestrial input
Could the lamination have formed in response to seasonal or longer term pulses of terrestrial organic matter input? It was noted that terrestrially-derived organic matter defines the darker varve-type lamination, with an occasional corresponding increase in terrigenous silt. Habib (1979,
1984) has described a rich flora of pollen grains, fern spores, woody tissue (tracheids) and epidermal cuticles, plus several types of amorphous matter, mostly derived from delta back-swamp areas. Where sedimentation rates of terrestrialderived material were high, as in the Valanginian and Barremian-Aptian interval at Site 534A, the floral material is well preserved. Habib also confirms that the distribution of carbonized wood largely controls sediment blackness, consistent with the abundance of 'black shales' in the late Valanginian at Site 534A (Fig. 2). In general, during early Cretaceous time, the North Atlantic was ringed by major deltas (e.g. Jansa & Wade 1975) during a time of gradually rising sea-level (Vail et al. 1980). In a model in which pulsed input of terrestrial, organic matter produced the fine lamination, it would be assumed that finely divided organics floated out to sea, were dispersed by currents, then settled to form an organic-rich layer. On the other hand most of the terrigenous clastics would have been trapped in the delta, coastal and adjacent carbonate platform system. A modern analogue would be the much smaller Gulf of California where Calvert (1966) considered seasonal input to be the main factor forming varve-type lamination. This model for the Atlantic lower Cretaceous requires variations in terrestrial input over decades rather than annually (rare floods?). The organic matter would have had to drift up to 400 km, being dispersed by currents, then settled to form discrete laminations. An alternative suggested by Gardner et al. (1978), is that fluctuations in the input of terrestrially-derived nutrients influenced surface productivity, but the absence of any clear correlation of varve-type lamination with upwelling along the continental margin is a problem.
Productivity
The lamination is strikingly like that observed both in lakes and along continental margins influenced by upwelling. For example, Thornton (this volume) records similar varve-type lamination in the modern Santa Barbara Basin on the California continental borderland. In the Miocene Monterey Formation (Bramlette 1946), varved diatomaceous sediment is also attributed to upwelling (Soutar et aL 1981), although the nature of lamination is much more varied than in the Cretaceous Atlantic. In the early Cretaceous Atlantic progressive sea-level rise (Vail et al. 1980) may have stimulated increased marginal upwelling producing fine lamination when faecal pellets settled on the sea floor.
444
A.H.F. Robertson
The main problems with upwelling as the cause for varved sediment are: (1) the frequent correlation of the lamination with black plant material (micritic material, Habib, 1984); (2) why upwelling should have occurred at all away from the continental margins in the middle of the equatorial palaeo-Atlantic. First, the relatively low frequency of lamination could be partly an artifact of repeated erosion during deposition of the interbedded turbidites, both pelagic and terrigenous. Alternatively, the low periodicity could be due to surface blooms persisting for periods of up to decades (Funnel, In: McCave 1979). Inter-annual variation in solar activity coupled through the atmosphere to the oceans may also influence oceanic circulation patterns and thus upwelling. For example, the 'Southern Oscillation' of the atmosphere leads to inter-annual climatic variation in the tropical Pacific Ocean (Julian & Chervin 1978; Philander 1979). During the early Cretaceous the Atlantic was open to the Western Tethys (e.g. Bernoulli 1972). Whether or not a connection with the Pacific existed is uncertain. It is probable that westward surface drift issued from the Tethys. Palaeo-latitudes where varve-type sediment existed ranged simultaneously from 12-25~ outside a possible zone of upwelling due to equatorial convergence. One possibility is that flow to the Pacific was blocked and surface currents were deflected to form nutrient-rich gyres in the western North Atlantic. In summary, the fine varve-type lamination has to be attributed to some combination of variations of input of plant material and surface plankton productivity. At present it is not possible to evaluate the relative roles of these two processes. The wide distribution of the lamination including, by mid-Cretaceous, both North and South Atlantic and parts of the Mediterranean, and the presence of detectable plant material right out to the early Cretaceous MidAtlantic ridge, point to an important terrestrial control, as in the modern Gulf of California. On the other hand, the association of lamination in the paler (non-organic-rich) pelagic carbonates with calcareous nannoplankton, radiolaria and faecal pellets, plus the close similarity of the lamination with other upwelling areas, all point to upwelling variations also as an important control. In either case it seems unnecessary to invoke a stagnant, well stratified, or oxygen-depleted Atlantic ocean during early Cretaceous time. Ultimately, both plankton productivity and plant input may relate to climatic changes on land caused by fluctuations in solar activity. The two variables may thus be linked rather than operating independently.
Graded marly chalks and 'black shales' Occurring independently of the varve-type lamination are thin graded marly nannofossil claystones and 'black shales', which range in colour from greenish, greyish to black. Close examination of sedimentary structures and composition show that most of this material is not indigenous pelagic sediment but has been redeposited by turbidity currents probably from the upper continental slope, a fact which has important implications for the origin of other Atlantic Cretaceous black shales. As shown in Fig. 2 the nannofossil claystones appear in the early Berriasian and persist throughout early Cretaceous time, becoming most abundant in the late Valanginian-Hauterivian. Black claystones and 'black shales' appear in late Berriasian, and then reach a relative peak in the late Valanginian. In the cores, the claystones are typically massive thin beds mostly up to 5 cm thick showing subtle grading over the basal 1-2 cm, a massive interior, and a top often with small-scale burrowing (Fig. 3(a), (c)). Throughout most of the succession the nannofossil claystones occur separately from the coarser turbiditic elastic input. The black claystones and 'black shales' are unburrowed. In thin section the nannofossil claystones comprise argillaceous micrite with scattered radiolaria. The basal several centimetre-thick graded intervals contain calcite-replaced or pyritized radiolaria, rare benthic foraminifera, minute shell fragments, fish debris and minor fine-grained quartzose silt. Higher in the succession, where the claystones are interbedded with terrigenous elastics, the quartzose content increases. The scanning electron microscope reveals well preserved calcareous nannoplankton with an open porous texture and little recrystallization compared to the more lithified nannofossil chalks and limestones. Smectites and mixed-layer clays dominate the clay mineralogy with some chlorite and palygorskite (Fig. 4(c)). The 'black shales' and black claystones show similar sedimentary structures and petrography but contain a higher volume of organic matter (up to 2.8~), mostly carbonized debris of terrestrial origin (Habib 1979). Black organic-rich sediment is not restricted to the graded beds, but is also seen in the darker intervals of varve-type lamination described above. Redeposition from the oxygen-minimum zone
As recently summarized by Dean & Claypool (1981), anoxicity in the ocean could be promoted
Varve-type lamination, graded claystones and limestone-shale 'couplets' by one or a combination of: (1) additional surface water productivity and thus extra oxygen demand to combust organic material; (2) increased rate of terrigenous organic matter input; (3) reduced bottom circulation; (4) increased overall sedimentation rate; (5) reduced solubility of oxygen in warmer waters. Three main models have been used to explain the Cretaceous Atlantic black shales (Schlanger & Jenkyns 1976; Fischer & Arthur 1978; Dean et al. 1978). The first model involves formation of a stratified ocean with anaerobic bottom water similar to the modern Black Sea. The second model produces organicrich sediment where the oxygen-minimum zone intersects the topography, the continental slope, or oceanic edifices. The third model, which has received less attention, requires high organic matter input leading to a high oxygen demand and hence anoxicity within the sediment, while the ambient sea water remains oxidizing. All three models could potentially play a role in the Atlantic Cretaceous. A currently popular view is that during the lower Cretaceous a decreased thermal gradient between equator and pole could have reduced thermohaline circulation and the content of dissolved oxygen in the bottom waters. High productivity both on land and at sea could have widened the oxygen-minimum zone and allowed more organic matter to reach the ocean floor without combustion. In a sluggishly circulating, poorly oxidized ocean, organic matter could have had a greater chance to reach the abyssal plain and thus promote subsurface anoxicity. Habib (1979) shows that sediment colour of the black-shale facies closely follows the distribution of carbonized wood debris, which in turn is related to terrigenous input. The interpretation of Cretaceous 'anoxic events' (Schlanger & Jenkyns 1976; Fischer & Arthur 1978; Jenkyns 1980) has, however, relied heavily on the recognition of reduced in situ organic-rich layers over wide areas of the ocean floor. Thus, it is significant that the present study has shown that much of the organic-rich sediment is not indigenous but has been redeposited. The graded nannofossil claystones and black claystones occur independently of the varve-type lamination attributed above to surface productivity variations. The obvious explanation of the graded claystones is that they are relatively fine-grained distal turbidites. Deposition from nepheloid-flow would appear to be ruled out by the bed-thickness (to 5 cm) and the presence of some silt-sized material. There is however a complete contrast in sedimentary structures and composition from the coarser clastic turbidites interbedded in the Valanginian-Hauterivian interval. This shows that the thin graded beds are
445
not merely the distal tail of coarser turbidites, but instead originally consisted almost entirely of fine-grained material. The excellent nannofossil preservation is consistent with a source well above the CCD, while the absence of coarser terrigenous or bioclastic material points to a transport route separate from the claystone turbidites. The obvious site is the upper continental slope, a potential source area of much pelagic and hemipelagic sediment. The upper continental slope is normally within the oxygen-minimum zone and is bypassed by coarser clastic sediment which is transported through the canyon systems and issues onto the continental rise from pointsources. By contrast, sediment redeposited from the upper continental slope would have come from 'line-sources'. Paull & Dillon (1980) argue that the present steep profile of the Blake escarpment is due mostly to Tertiary erosion by contour-flowing currents. Although little is known about the Blake Plateau in the Cretaceous, Enos & Freeman (1978) found no evidence of a barrier reef edge on the Blake Nose, suggesting the existence of a broader, less steeply sloping margin than at present. This would have provided space for accumulation of large volumes of muddy pelagic sediments on the upper continental slope, as in the modern windward Bahama margin (Mullins & Neumann 1979). Finer grained material on the upper continental slope deposited within the oxygen-minimum zone would have been bypassed by coarser grained shelf-derived and terrigenous material, which accumulated separately on the continental rise and beyond. Along modern upwelling continental margins, the upper continental slopes are characterized by a well-developed oxygen-minimum zone. For example, along the Ivory Coast and Ghana, the intermediate waters are oxygen-depleted (about 1.5 ml/1) but not anoxic. Organic matter survives below 2000 m because of high input relative to rates of combusion (e.g. Demaison & Moore 1980). In the Bay of Bengal, despite the absence of high surface productivity an anoxic layer of intermediate water is well-developed (yon Stackelberg 1972). There, in the oxygen-minimum zone, olive-grey muds are deposited which contrast with light-brown muds deposited in deeper oxidizing waters. Thus, summarizing, the modern analogies and the sedimentary features suggest that much of the turbiditic claystone and 'black shale' was redeposited from qine-sources' within or close to the oxygen-minimum zone on the upper continental slope. To the extent to which the organic matter is redeposited, the need for anoxic bottom waters is reduced. Only development of subsurface anoxi-
446
A.H.F. Robertson
city is needed to preserve the organic matter. With high rates of input of terrestrial organic matter, burial takes the sediment rapidly from the sediment-water interface of aerobic bacterial decay into the zone of anoxic bacterial sulphate reduction (cf., Curtis 1980). Recognition of turbidity-redeposition of much organic matter still leaves the black intervals of varve-type sediment as a possible bottom water anoxicity indicator. In a sluggishly-circulating warm relatively poorly oxidized ocean, terrestrially-derived organic matter, periodically floating out to sea, dispersed by currents, then settling would indeed stand a good chance of reaching the sea floor in sufficient abundance to promote sub-surface anoxia. However, if organic matter input is sufficiently great, as appears to have been the case in the early Cretaceous, then black organic-rich layers may form in a more normal oxidizing circulating ocean. The early Cretaceous Atlantic was open to the Western Tethys and the abundance of radiolaria and nannoplankton point to active upwelling in opposition to a stable stratified ocean at this time.
Cyclical alternations: 'couplets' Two distinct types of 'couplet' are present in the late Berriasian to Hauterivian successions at Site 534A and 391: (1) alternations of paler and darker varve-type laminated marly nannofossil chalks, and (2) cycles of laminated marly nannofossil chalks alternating with harder burrowed nannofossil chalks and limestones (Fig. 3(a),(b)). Darklight laminated cycles are c.10-30 cm thick but may be completely intergradational, or comprise rapid alternations on a scale of several centimetres down to individual laminations. The relative percentage of the darker and paler laminated facies is shown in Fig. 2. The pale harder burrowed carbonates either appear gradationally, with progressive breakdown of varvetype lamination, or may appear quite sharply above either paler or darker laminated marly nannofossil chalks (see below). Overall, the burrowed intervals in the Berriasian tend to be harder and better cemented, and contain less clay and organic matter than those in Valanginian time. Burrowed intercalations are scarce around the Late Valanginian-Hauterivian boundary when clastic input was at a maximum (Fig. 2, Fig.
3(c)). To test for cyclicity the transitions between different sediment types were determined from core photographs where contacts were undisturbed. The clearly gravity-redeposited claystones and clastic turbidites were excluded. For
351 measurements, by far the most abundant transition was from dark to pale laminated marly nannofossil chalk (27%) and vice versa (25%). The next most common transition was from pale laminated marly nannofossil chalk to burrowed nannofossil chalk or limestone (15%) and vice versa (16%). The least abundant transition was from dark laminated marly nannofossil chalk to pale burrowed chalks and limestones (8.5%) and vice versa (7.4%). Time intervals
During Valanginian time, average sedimentation rates of 21 m/my suggest that the burrowed limestones appeared on average every 20000100 000 yrs and persisted for 10 000-60 000 yrs. Alternations of pale and darker laminated chalk are much more frequent, often less than 10000 yrs down to 100s of years. The two sediment types are often completely intergradational. Two alternative models can be proposed: (1) changes in bottom water chemistry; (2) long period fluctuations in terrestrial organic matter input. Changing water chemistry
McCave (1979) noted at Site 387 on the Bermuda Rise that the dark-light 'couplets' tended to be asymmetrical with a gradual increase in organic matter content then a sudden switch to oxidized burrowed sediment. He suggested that the darklight 'couplets' could relate to periodic renewal of more oxidizing bottom water. Such bottom water change could have been triggered by the 19 000 yr precession or the 41000 yr obliquity-periodicity. An immediate problem is that although bottom water change could relate to local physical barriers, oceanwide bottom water replenishment is harder to envisage, particularly since similar couplets occur on both sides of the palaeo-Mid Atlantic Ridge. Also, the transition analysis of the Site 534A succession shows that by far the most common change is from paler to darker more organic-rich laminated marly nannofossil chalks, with only rare burrowed paler intervals above the dark laminated marly nannofossil chalks (8.5% of transitions) as noted by McCave (1979). Terrestrial organic input
Dean et al. (1978) and Gardner et al. (1978) relate the cycles of pale organic matter-poor and darker organic matter-rich nannofossil carbonates to pulsing of terrestrial organic matter with a periodicity ranging from 30 000 to 50 000 yrs. Wetter
Varve-type lamination, graded claystones and limestone-shale 'couplets' periods produced more organic matter and drier times less organic matter. Such climatic variations are consistent with the hot-humid circumNorth Atlantic climate inferred by Chamley (1979). Jansa et al. (1979) and Arthur & Natland (1979) note that the same cycles tend to match the orbital periodicity. At Site 534A the transition analysis is consistent with fluctuations of organic matter on a time-scale down to hundreds of years but extending up to c.100000 yrs. On average every 20 000-60 000 yrs organic matter decreased sufficiently for bottom sediment to become oxidizing and an infauna to develop (see below), thus forming the paler burrowed carbonates. This is consistent with there being twice as many transitions of pale-to-burrowed chalk (and vice versa), than dark-to-burrowed chalk. The latter transition could result from more sudden climatic fluctuations on land.
Role of diagenesis Diagenesis has also played an important role in producing the laminated burrowed couplets. There is a difference in the degree of lithification and the preservation of microfossils between the laminated and burrowed alternations. In the better-cemented burrowed paler intervals, with less organic-matter, cementation clearly took place early in diagenesis, prior to significant compaction. The radiolarian shells are almost all spherical with little evidence of flattening (Fig. 4(a)). By strong contrast, radiolarian shells are often more flattened in the softer clay-rich and organic-rich intervals (Fig. 4(d)). The higher content of clay and organic matter has inhibited calcite recrystallization. By contrast, scanning electron micrographs (Fig. 4(b)) show an advanced state of recrystallization of calcareous nannoplankton in the harder burrowed chalks. It is thus possible to propose a model (Fig. 5) in which fluctuations of organic-matter input control the amount of dissolved oxygen within the sediment during early diagenesis. During times of lower input of organic-matter, bottom water remained sufficiently oxidizing to promote a rich infauna, turbating and virtually destroying an original varve-type lamination. Early diagenetic lithification preserved the radiolaria with little compaction. By contrast, as the organic matter input increased burrowing stopped and the fine varve-type lamination was preserved. Clay and terrigenous organic matter go together (Habib 1979) so that the reducing layers also tended to be those in which diagenetic cementation was retarded. The clay content inhibited the recrystal-
447
lization of calcareous nannoplankton and the sediment remained soft, causing radiolarian shells to be flattened during slow compaction.
Depositional model The various types of fine-grained sediments in the late Cretaceous of the Blake-Bahama Basin can be understood in terms of a complex interplay of climatic, depositional and diagenetic processes. The overall model is illustrated in Fig. 6. First, there is the effect of varying source material. During early Cretaceous time the climate was hot with fluctuating humidity (Chamley 1979) with high terrigenous input from the source a r e a s of Palaeozoic fold mountains of SE USA. Rising sea-level during the early Cretaceous is reflected in increased input of shallow water carbonate material presumably from the Blake Plateau during late Valanginian, Hauterivian and Barremian time. Both the terrigenous and bioclastic material was presumably transported through a conventional canyon system to issue on to the continental rise from 'point-sources'. By contrast, clay and organic matter-rich pelagic carbonates accumulated within, or close to, the oxygen-minimum zone on the upper continental slope and were then redeposited from linear source areas to form the thin graded layers of nannofossil claystone and 'black-shale'. The various turbiditic influxes reworked and destroyed some of the varve-type lamination thus reducing the apparent frequency. Secondly, there are the oceanographical factors. The abundance of radiolaria in the pelagic chalks points to relatively fertile surface waters and hence to upwelling. The reason for such upwelling is unknown but could relate to inflow of nutrient-rich surface water from the Western Tethys (Fig. 7), or from an accentuated upwelling along the margin during a time of sea-level rise. Ocean water instability would possibly also have been triggered by inter-annual climatic fluctuations. During early Cretaceous time (pre-Albian) bottom currents in the North Atlantic could conceivably have been sluggish and relatively oxygen-depleted, but there is no evidence that deep waters were ever really static or anoxic. Anoxicity did, however, develop in the subsurface during the many times of enhanced terrestrialorganic matter input. Conversely during times of lower input bottom faunas flourished. Thirdly, diagenesis has played an important role in forming the laminated burrowed alterations (couplets). Organic matter-poor nannofossil oozes lithified early, while additional clay plus
448
A.H.F.
Robertson
o.o
o . ~_
0
"13 o
. o _~
~."-~-
-- X ~- 0 I
r-
o
5.0
o
Q;
co
c,'F.
-~.~_
.~_
o; O0
w~
o
0 ~
0 I
I
0 ~"
I
I
I
->
l
~
~o~ ~
+~'o
....
I I I
~
o.~,
r
p.-,
"~ o ~
=~
"F_ .c.
o~•
.90 . _ ~
I I
Q;
o
"~ .~ m
T
0
"o--
-> ~'-'
J
0
I
c~
.~ =_ o
E~
o~~ \ m
o~o
o
o
o~
cO -&
~.=_ ~ "~''~u
0
~.~
~.~
0
. ~ Ob
r" "0 '~1) 0
0 ~ - ~~ - - ~
.0"0 --
..~
/
-~ ~. ~E_
O. cO
F--o
U9 . ~
(1)
~o
=
o
co ex
e-
(1) .-I
=o~ ~ ~.~
e- .-~
_~o
0
~
e~o ~~ =~
".~ ~
Varve-type lamination, graded claystones and limestone-shale 'couplets' 449 e.-
0 :~
o~
~
o:
u
oo ,,~
E
o
N E E
J
i%-,_.
.,x,X.x.0.30 m) between 3 m and 15 m thick are restricted to the base and middle parts of the succession. Individual sandstone beds are between 0.30 and 3.20 cm thick, the majority exhibit well-developed Bouma cycles. Sandstones form approximately 50% of the total sequence
(Fig. 2) and are repeatedly interbedded with mudstones and chalks (Figs 3 and 4). Sedimentation system: summary
Taken as a whole the Salir Formation sedimentation system represents deposition on a series of small submarine fans passing from inner-fan to mid-fan regions from east to west. The inner-fan area (proximal facies-association) consists of amalgamated conglomeratesandstone units deposited in braided channelized areas; in interchannel areas, sequences of interbedded fine sandstones, mudstones and chalks were deposited. To the west, the mid-fan environment (distal facies-association) consists of amalgamated sandstone units deposited in shallow distributary channels and bundles of thick-bedded sandstones deposited on a series of depositional lobes. Further details of the sedimentation system are documented elsewhere (Hayward 1982; Hayward & Robertson 1982). Sedimentology of the chalk beds
General Description Chalk beds are present throughout both the proximal and distal sedimentary sequences. White to cream chalk horizons are between 5 and 60 mm thick, laterally continuous and well indurated. Bedding surfaces are either parallel or show a wave-like form with a wavelength of 10-15 cm (Figs 4 and 5). Locally, chalks show relief over underlying mudstone or sandstone beds. Tops of the chalks are generally sharp. Bases are sharp (planar or irregular) or transitional. Rarely, small mudstone rip-up clasts are present towards the base of beds.
Composition In thin section, planktonic foraminifera and rare benthonic foraminifera along with debris from both are dispersed in a matrix ofmicrite. In some cases, planktonic foraminifera (dominated by Praeorbulina sp) account for 15% by volume of the chalks. Biogenic sand and silt-sized particles form up to 20% of the rock. Carbonate content ranges from 55-95%. The non-carbonate component comprises dispersed clay platelets and, in some beds, silt-sized particles. Dispersed carbonaceous material is sometimes abundant
(~ 5%). SEM studies show the micrite matrix to be composed of equigranular calcite crystals with abundant poorly preserved coccolith plates and planktonic foraminifera fragments. Dispersed
Hemipelagic chalks in a submarine fan sequence
457
EP-
,
A-
,
~'6,,
~~ =
~
E o
~ o
~ o
~ 8 ~~ -
o~ o
u
~
,~=o
-_~ ~ ' o.=.--
-
E
u
~
~ ~ ~ ~ .~
E,':-
E
u
~
"c. o_'z,
~.o
~z
~;~
E u co
E,..2
, ~ , l ~ "~1 o~ ,-~ .~,,~-~, ~
2%). sapropel-sapropelic variations of the type This latter may be emplaced either by hemipelagic observed in Fig. 3(b), (c) and (d) are frequently suspension settling or by some form of gravirecorded. ty--(and possibly bottom current) flow, or a In all the sequence variants described above, coupling of the two mechanisms (Stanley 1983). the sediment lithofacies comprising the sapropel In one extreme case, an episapropel sequence sequence are essentially hemipelagic deposits. contains only a portion of the sapropel layer, and These do not display structures in X-radiographs this may display varve-like lamination (see Fig. indicative of physical processes such as those 5(c), numbers 3 and 12). This lamination is the produced by sediment gravity-flows and bottom result of transport processes and not annual currents (Figs. 4(a)-(d) and 6(a), (b)). Good cycles. At the other extreme, this sapropel layer is examples of these sapropel sequences occur on extensively interrupted by non-sapropelic layers the Mediterranean Ridge south-west of Crete. (Fig. 5(c), numbers 6 and 13). Episapropel However, on the basis of examination of cores sequences generally incorporate different succesthroughout the eastern Mediterranean, we find sions of sediment types; some of these facies are that the above complete sapropel sequences not as well-defined as those in the sapropel account for less than 10% of organic-rich sequences. Moreover, lithofacies members of the sequences. sapropel sequence are missing, or poorly-developed, or repeated as a result of the above-cited transport processes which result in redeposition Episapropel sequences and/or erosion. Although colours in split cores Sequences consisting of essentially hemipelagic suggest typical sapropel development, X-radiodeposits tend to be an exception rather than the graphs reveal that such sequences are, in fact, (a)
(b)
(c)
(d)
Sapropels and organic-rich variants in the Mediterranean
503
FIG. 4 X-radiographs of sapropel sequences of the type illustrated schematically in Fig. 3. (a)= sapropel sequence; (b)-(d)= variants. The different sediment types comprising these sapropel sequences are indicated by a number corresponding to the legend in Fig. 3. Core locations are shown in Fig. 1. Discussion in text.
expanded by the presence of several redeposited layers. Good examples of episapropels occur in the central Hellenic Trench. Sapropelic sequences
Another important variant is the sapropelic sequence which displays the same succession of sediment types as a sapropel sequence, but with a significant difference: a sapropelic layer (organic carbon content between 0.5 and 2%) rather than a sapropel layer is developed. Three sapropelic sequence variants are depicted in Fig. 7(a), (b) and (c). The typical complete sequence (Fig. 7(a)) starts with a greyish-pale greenish mud passing upwards into an organic ooze which is separated from the sapropelic layer by a vaguely defined
contact. The sapropelic layer is succeeded by an organic ooze and, on top of this, an oxidized layer (Figs 7(a) and 8(a)). The sapropelic sequence shown in Figs 7(b) and 8(b) is similar to 7(a), but here an organic ooze above the sapropelic layer passes upward directly into well-oxygenated sediments (Fig. 8(b)). In a third case (Figs 7(c) and 8(c)), the sequence is comparable to (a) but the sapropelic layer is directly overlain by an oxidized layer. All sediment types which constitute the various sapropelic sequences described above (Figs. 7(a) and 8(c)), are hemipelagic-suspension settling deposits. Good examples of sapropelic sequences occur locally in the south-eastern Mediterranean on the periphery of the Nile Cone, and on topographic highs less than 1200 metres in depth.
504
G.C. Anastasakis and D.J. Stanley
SAPROPEL SEQUENCE EPISAPROPELSEQUENCE (a)
(b)
(c) 7 I.,~.,I~
Irll Illl-1~
6
GREY HEMIP.MUD 81~I~ I EPISAPROPELIC=Td (OF BOUMASEQUENCE)
2[~/7/~ ORGANICOOZE
9~ !
EPISAPROPELIC,"ret
aI
SAPROPEL
10~
EPISAPROPEL;Ta-Tc
4~
OXIDIZED LAYER1 1 ~
51TIT! SAPROPELIC 6~
(OF BOUMASEQUENCE)
{OF BOUMASEQUENCE)
EPISAPROPEL:Td (OF BOUMASEQUENCE)
12~-ZI (OF EPISAPROPEL"Tet BOUMA SEQUENCE)
TURBIDITICSILT 1 3 ~
TURBIDITICSAND
7~1 TURBIDITICMUD Episapropelic sequences Often, pronounced variations of sapropelic sequences are observed, particularly with respect to assemblages of sediment types. Many of these have been emplaced by gravity-controlled mechanisms. An episapropelic sequence is defined herein as a generally expanded sapropelic sequence that contains at least one sapropelic layer (which may, but need not be, deposited by hemipelagic-suspension settling), together with a series of other layers, for the most part deposited by gravity-flow and/or bottom current transport (Figs 7(d) and 8(d)). Thus, episapropelic sequences are expanded, and the associated lithofacies which form them may be well-defined, or poorly-developed, or missing or repeated. Good examples of episapropelic sequences occur in the Nile Cone and in several of the small Aegean Sea basins. Both episapropel and episapropelic sequences are extremely variable (Figs 5(c) and 7(d)). Among the enigmatic variants are the ones that display numerous vertical variations such as varve-like couplets (Fig. 8(d)), numbers 8, 9 and Fig. 6(c), numbers 9, 10, 11) and marked compositional variations within the sapropel-sapropelic lithofacies. These structures most likely result from fairly rapid emplacement involving rather short dispersal. Moreover, high amounts of organic material in these varve-like structures within the sapropel and sapropelic lithofacies probably modified the settling behaviour of particles. These structures likely result from downs-
FIG. 5. Schematic illustration of (a), (b) variants of the eastern Mediterranean sapropel sequence and (c) of an episapropel sequence. The latter is a hypothetical example showing possible sediment types and attributes that have been observed in a number of cores. Actual examples are shown by the core X-radiographs in Fig. 6.
lope reworking and segregation of the organicrich particles into distinct layers (Stanley 1983) rather than from seasonal input of alternating organic-poor and organic-rich sediments (cf. McCoy 1974). This hypothesis is supported by examination of stratigraphically synchronous sapropel and sapropelic layers in areas where there is good radio-carbon dated sample control. Episapropel (> 2% organic carbon) and episapropelic (0.5-2% organic carbon) layers (Figs 5(c), numbers 8-12; Fig. 7(d), numbers 8, 9, 14) display sedimentary structures visible on X-radiographs such as vertical grading and distinct size-segregated bands. These structures are indicative of gravity-flow rather than suspension settling. The vertical succession of lithofacies types varies markedly from core site to core site, showing the importance of local transport events in the development of both episapropel and episapropelic sequences.
Sapropel sequences as palaeoceanographic indicators In their review, Demaison & Moore (1980) indicate that organic-rich layers are deposited under stagnant and also in fully oxygenated brackish to lacustrine waters, as well as in open marine waters under a wide range of conditions. Most workers, including ourselves, tend to accept the original definition of sapropel, that is, a sediment rich in organic matter formed under
Sapropels and organic-rich variants in the Mediterranean
5o5
FI~. 6 X-radiographs of sapropel sequences (a),(b) and an episapropelic sequence (c) of the type illustrated in Fig. 5. Numbers refer to sediment types identified in Fig. 5. Core locations are shown in Fig. 1. Discussion in text.
reducing conditions in a stagnant water body (Wasmund 1930; Pontoni6 1937). As indicated in the first part of this paper, the term is now used to include a wide range of organic-rich lithofacies. There generally is no implication as to the specific origin of the organic matter. Insofar as the Mediterranean is concerned, most workers believe that sapropels record palaeoceanographic
conditions, for the most part associated with oxygen-deficient conditions (Olauson 1961; Sigl et al. 1978). Anoxic sea floor environments in open marine basins in general occur in (a) areas of intense upwelling and (b) below stratified waters, that is, along and below a distinct thermocline, pycnocline or halocline (temperature, density and salinity boundaries, respectively) or combina-
G.C. Anastasakis and D.J. Stanley
506
SAPROPELIC SEQUENCE
(a)
(b)
(c)
EPISAPROPELIC SEQUENCE (d)
7 14 ,9 -6
--... ?
28 \1
2~
ORGANICOOZE
e~
TURBIDITIC MUD EPISAPROPELIC:Td
~.~
OXIDIZED LAYER9 ~
EPISAPROPELIC:Te'
sliT[- ~ SAPROPELIC 6~
13~
TURBIDITIC SILT 1 4 ~
(OF BOUMA SEQUENCE) {OF BOUMA SEQUENCE)
TURBIDITICSAND EPISAPROPELI BOUNA I SEQUENcE)T(I C,(oF -TC
tions thereof. The eastern Mediterranean sapropels likely were deposited below stratified waters as suggested by Bradley (1938), along and below a pycnocline-halocline boundary (Stanley 1978). Together, the structures, texture and composition of sapropel and sapropelic lithofacies should serve as reliable indicators of physical and chemical conditions under which they were deposited. An assessment of the development of the Mediterranean sapropel sequence is attempted by reviewing the succession of lithofacies types. We believe that the initiation of the stagnation phase is marked by deposition of the greyish mud associated with a decrease in the amount of the dissolved oxygen in the bottom waters. As this condition continues, life on the sea floor becomes unfavourable for most organisms. Only the most adaptable forms, including some benthonic foraminifera, remain; the enhanced preservation of organic matter in sediments (Fig. 2), due to reduced decay, attracts sediment-grazing (perhaps worm-like) organisms (see facies 2 in (a) and (b), Fig. 4). This is expressed by the transition of the greyish mud to grey ooze which always displays the greatest degree of bioturbation of any sediment type forming the sapropel sequence. Sometimes it appears that the ooze progressively grades upwards to a sapropelic layer (the protosapropel layer depicted by Maldonado & Stanley 1976). Closer inspection, however, reveals that a contact always separates the ooze from the overlying sapropelic layer (with organic carbon >0.5%). The absence, or scarcity, of bioturbation in the ooze immediately below the contact suggests that the concentration of dissolved oxygen in the bottom waters began to reach the tolerance limits for even the worm-like organisms
FIG. 7. Schematic illustration of sapropelic sequences (a--c) and episapropelic sequence (d) from the Eastern Mediterranean. The latter is a hypothetical example showing possible sediment types and attributes that have been observed in a number of cores. Actual examples are shown by the core X-radiographs in Fig. 8.
(upper part of facies 2 in (a),(b),(c) in Fig. 4, and (a) and (b) in Fig. 6). This attribute of the ooze thus signals the gradual establishment of anoxic conditions. The marked increase in organic carbon content in the sapropel and sapropelic lithofacies marks the phase of complete oxygen depletion. Marked oxygen minimum variations occurred laterally and with time during a single stagnation event. This is recorded by pronounced organic carbon variations within a sapropel lithofacies at any one location, and the concurrent deposition of both sapropel and sapropelic sequences in different parts of the basin. This latter is demonstrated by good stratigraphic control using numerous radiocarbon-dated samples. The transformation ofa sapropelic layer to a sapropel layer in time and space indicates that their deposition resulted from dissolved oxygen fluctuations rather than from differences in surface waters productivity. This conclusion is based on analysis of numerous time-equivalent organic-rich samples which show no marked time-space variation in the ratio of the total organic matter substance (including amorphous organic matter) to the organic matter of recognizable remains, mostly dinoflagellates visible in SEM photographs. The deposition of time-equivalent sapropel and sapropelic sequences at the same water depths in different parts of a basin probably indicate the presence of variable oxygen content within the oxygen-depleted water column. The accumulation of an ooze above the sapropel or sapropelic layer records a gradual return to oxic waters; the deposition of an oxidized layer directly above a sapropel or sapropelic layer would, on the other hand, indicate a more abrupt return of oxygenated waters. There may have been an increase
Sapropels and organic-rich variants in the Mediterranean
507
FIG. 8 X-radiographs of sapropelic (a)-(c) and episapropelic sequences (d) of the type illustrated in Fig. 7.
Numbers refer to sediment types identified in Fig. 7. Core locations are shown in Fig. 1. Discussion in text.
in organic productivity (perhaps up to three times that in normal oxic Mediterranean waters) during stagnation episodes in the eastern Mediterranean. This is suggested by the enhanced organic carbon content in the oxidized layer deposited just after the above stagnation events. Increased productivity alone, however, could not have induced oxygen depletion sufficiently to produce stagnation. The correlation among episapropel and episapropelic sequences and palaeoceanographic events is not clear, particularly in cases where sequences largely result from gravity-flow deposition. Those sequences which contain primarily hemipelagic lithofacies and only one or a few deposits emplaced by high energy mechanisms (gravity-flow, bottom currents) can shed some light on palaeoconditions. We recall that it is the hemipelagic lithofacies which are of most use in
interpreting palaeoceanography. The degree of relation between adjacent members is noted on the organic carbon versus carbonate carbon/organic carbon diagrams (Fig. 2). In some cases the episapropel and episapropelic layers (Fig. 5(c), numbers 5, 9, 12) show no sequence relation to the adjacent lithofacies members (Fig. 5(c), numbers 7, 13) of a sequence. In other cases, there may be good sequence continuity (Fig. 5(c), numbers 10, 11; Fig. 7(d), numbers 5, 9, 14). When, as in the case of sapropel sequences, there is continuous deposition of hemipelagic lithofacies (Fig. 2(a)), all samples plot nearlinearly and continuously within quadrant 1 (all sapropel and sapropelic types) and quadrant 4 (organic ooze, oxidized layer, grey mud). In cases where gravitative mechanisms have emplaced some layers in expanded episapropel and episapropelic sequences, some samples plot in other
508
G.C. Anastasakis and D.J. Stanley
portions of the graph. For example, the samples which plot to the right of a vertical line drawn perpendicular to the carbonate carbon/organic carbon ratio at 7 indicate gravity-emplaced episapropelic layers (quadrant 2, Fig. 2(b),(c). Samples which plot to the left of this line (at 7) and below the 0.5~ organic carbon line (quadrant 3, Fig. 2(c), also indicate gravity-transported layers interbedded within episapropel and episapropelic sequences. Organic carbon contents overlap, especially in adjacent members within a sapropel sequence, but carbonate carbon/organic carbon ratios of each specific lithofacies are clearly distinguished (Fig. 2 quadrants 1 & 4). As plotted the relation between organic carbon and carbonate carbon is most revealing because the carbonate-rich sediment that settles through the water column is directly influenced by two contrasting environments, the well-oxygenated water above the thermocline-halocline boundary and anoxic water below (Sigl et al. 1978). This results largely because of differences in oxygen and hydrogen sulphide content in the water column which results in dissolution of the carbonates as particles settle through the column below the thermocline-halocline boundary. The petrological-chemical classification presented in this study helps us to better interpret the organic-rich deposits which accumulated during the last major stagnation episode in the eastern Mediterranean at approximately 8000 yrs. BP (Ryan 1972). During this event a variety of sapropel-sapropelic and episapropel-episapropelic sequences of the type shown in Figs 3, 5 and 7 were deposited. There are sites in the basin where the upper-most and best-defined Holocene sapropel (or sapropelic) sequence is represented by only a few centimetres of sapropel (or sapropelic) layer. Time-equivalent episapropel (or episapropelic) sequences contain up to several metres of organic-rich sediment. Thus, a few centimetres of sapropel (or sapropelic) layer accumulated during the same stagnation time-span as the episapropel (or episapropelic) layer. The latter may be a hundred times thicker due to numerous interbedded episapropelic and non-sapropelic layers, some very fine-grained and without obvious sedimentary structure. In conclusion, recognition of sapropel-sapropelic and correlative episapropel-episapropelic sequences in the deeper basins of the eastern Mediterranean can be used to more precisely assess the causes and development of stagnation. Moreover, continuing investigation indicates that the Quaternary sapropels and their variants described in this study are comparable to older (upper Miocene to Quaternary) sequences reco-
vered during DSDP Leg 42A in this basin. From preliminary observations it would appear that systematic study of Mediterranean organic-rich deposits and their relation to oxic-anoxic cycles may bear on the interpretation of some open marine black shales in the rock record.
Summary (I) Detailed examination of late Quaternary organic-rich layers in radiocarbon-dated cores from the eastern Mediterranean show remarkable petrologic variations. The present study indicates that the term sapropel sequence s e n s u s t r i c t o in this basin should be reserved for the following succession (from base up) of lithofacies: grey hemipelagic mud, organic ooze, sapropel, organic ooze and oxidized layer. (2) The succession of types forming the sapropel sequence records the progressive oceanographic changes that influenced the basin: diminution of oxygen in the deeper water masses; further deterioration of oxygenation; complete anoxia throughout the basin; restoration of some oxygen in the water column; and rapid return of fully oxygenated conditions above the sea floor. (3) The five basic lithofacies types constituting the sapropel sequence, essentially of suspension settling origin, can be distinguished on the basis of physical and biogenic structures, textures and composition, including the organic carbon content. (4) The sapropel lithofacies is characterized by a high (over 2%) organic carbon content and low carbonate carbon/organic carbon ratio; the other three types (organic ooze, oxidized layer and grey hemipelagic mud) have lower organic carbon content ( 2%) and episapropelic (organic carbon, 0.5%-1.9%) are identified. These latter
Sapropels and organic-rich variants in the Mediterranean sequences, also correlative with true sapropels, generally occur on or near slopes and are expanded sequences that include flowemplaced layers such as mud turbidites. (7) Examination of open marine organic-rich layers throughout basins such as the Mediterranean reveal a large number of variants, as characterized by composition and sequence development, that lend themselves to classification. This investigation suggests that the petrologic-chemical approach is applicable to the study of dark organic-rich sediments in the adjacent Black, Adriatic, Marmara and Red Seas. (8) Recognition of the many variants and definition of their sequence development can be used to better correlate organic-rich layers throughout a basin and interpret the palaeoceanographic conditions (oxic-anoxic cycles) that lead to their origin. Facies analysis of these deposits may serve to better understand some open marine black shales in the ancient record.
509
ACKNOWLEDGMENTS: This study is based on cores available in the Smithsonian collection supplemented by core materials from the University of Rhode Island (1976 R/V T R I D E N T cruises 171 and 172), Woods Hole Oceanographic Institution (1975 R/V C H A I N cruise 119), Lamont-Doherty Geological Observatory of Columbia University (1958 R/V V E M A cruise 14 and 1965 R/V C O N R A D cruise 9; supported by ONR grant N00014-67-A-0108-0004 and NSF grant GA-35454; 1981 B A N N O C K cruise), and University of Paris VI (1976 N/O A R I A N E cruise). Our sincere appreciation is expressed to these organizations. We thank Mssrs. H. Sheng, and J. Nelen, who assisted in most laboratory operations and Dr. R. Stuckenrath for providing numerous radiocarbon dates. Drs. S.E. Calvert and J.W. Pierce reviewed the manuscript. This investigation, part of the Mediterranean Basin (MEDIBA) Project funded by Smithsonian Scholarly Studies grant 1233S201, was completed while the first author held a Smithsonian Postdoctoral Fellowship.
References ANASTASAKIS, G. 1981. Sedimentation and Shallow Structure of the South Cretan-Karpathos sector of the Hellenic Trench System. Unpubl. Ph.D. thesis. Keele University, England. 392 pp. BLANC, F., BLANC-VERNET, L., LAUREC, A., LECAMPION, J. & PASTOURET,L. 1975. Application palOo6cologique de la m6thode d'analyse factorielle en composantes principales: Interpr6tation des microfaunes de foraminif6res planctoniques Quaternaires en M6diterranOe. II. Etude des esp~ces de M6diterran6e Orientale. Palaeogeo., Palaeoelimac, Palaeocol., 18, 293-312. BRADLEY, W.H. 1938. Mediterranean sediments and Pleistocene sea-levels. Science, 88, 376-9. CITA, M.D., CHIERIC~, M.A., CIAMPO, G., ZEI, M., D'ONOFRIO, S., RYAN, W.B.F. & SCORZIELLO, R. 1973. The Quaternary record in the Tyrrhenian and Ionian Basins of the Mediterranean Sea. In: Ryan, W.B.F. & Hsii, K.J. et al., Init. Repts OSDP, 13. Natl. Science Found., Washington, DC. 1263-339. & PODENZANI, M. 1980. Destructive effects of oxygen starvation and ash falls on benthic life: A pilot study. Quat. Res., 13, 230-41. DEMAlSON,G.J. & MOORE,G.T. 1980. Anoxic environments and oil source bed genesis. Bull. Am. Ass. Petrol. Geol., 64, 1179-209. DOMINIK, J. & MANGINI, A. 1979. Late Quaternary sedimentation rate variations on the Mediterranean Ridge, as results from the Th-230 excess method. Sed. Geol., 23, 95-112. -& STOFFERS, P. 1979. The influence of Late Quaternary stagnations on clay sedimentationin the Eastern Mediterranean Sea. Geol. Rundsch., 68, 302-17.
-
-
HUANG, T.-C. & STANLEY,D.J. 1972. Western Alboran Sea: sediment dispersal, ponding and reversal of currents. In: Stanley, D.J. (ed.), The Mediterranean Sea--A Natural Sedimentation Laboratory. Dowden, Hutchinson & Ross, Stroudsburg, Pa.. 521-59. HERMAN, H. 1972. Quaternary eastern Mediterranean sediments: micropaleontology and climatic record. In: Stanley, D.J. (ed.), The Mediterranean Sea--A Natural Sedimentation Laboratory. Dowden, Hutchinson & Ross, Stroudsburg, Pa.. 129-47. KIDD, R.B., CITA, M.D. & RYAN, W.B.F. 1978. The stratigraphy of Eastern Mediterranean sapropel sequences recovered by DSDP Leg. 42A and their paleoenvironmental significance. In" Hsfi, K.S. Montadert, L. et al., Init. Repts. DSDP, 42. Part 1. Natl. Science Found., Washington, DC. 421-43. KULLENBERG, B. 1952. On the salinity of the water contained in marine sediments: Medd. Oceanogr. Inst. Goteborg, 21, 1-38. MALDONADO,A. & STANLEY,D.J. 1976. The Nile Cone: Submarine fan development by cyclic sedimentation. Marine Geol., 20, 27-40. --, KEELING, G. & ANASTASAKIS, G. 1981. Late Quaternary sedimentation in a zone of continental plate convergence--The Central Hellenic Trench System. Marine. Geol., 43, 83-110. MANGINI, A. & DOMIN1K, J. 1979. Late Quaternary sapropel on the Mediterranean Ridge: U-budget and evidence for low sedimentation rates. Sed. Geol., 23, 113-25. McCoY, F.W., JR. 1974. Late Quaternary Sedimentation in the Eastern Mediterranean Sea. Unpubl. Ph.D. Thesis, Harvard University, USA. 123 pp.
5IO
G.C. Anastasakis and D.J. Stanley
MULLINEAUX, L.S. & LOHMAN, G.P. 1981. Late Quaternary stagnations and recirculation of the Eastern Mediterranean: changes in the deep water recorded by fossil benthic foraminifera. J. Foram. Res., 11, 20-39. MUERDTER, D.R. 1984. Low salinity surface water incursions into the Strait of Sicily during sapropel intervals in the Late Quaternary. Marine Geol., (in press). NESTEROFF,W.D. 1973. Petrography and mineralogy of sapropels. In: Ryan, W.B.F., Hsii, K.J. et al., Init. Repts. OSDP, 13, Natl. Science Found., Washington, DC. 712-20. OLAUSSON, E. 1961. Studies of deep sea cores. Repts. Swedish Deep-Sea Exped., 8, 337-91. PARKER, F.L. 1958. Eastern Mediterranean foraminifera. Repts. Swedish Deep-Sea Exped. 1947-1948, 8, 217-83. PASTOURET, L. 1970. Etude s6dimentologique et pal6oclimatique de carottes pr61ev6es en M6diterran6e orientale. Tethys, 2, 227-66. PONTONII~,R. 1937. Die nomenklature der unterwasserablagerungen. Jb. preuss, geol. Landesanst. Bergakad., 58, 426-8. ROSSIGNOL-STRICK, M., NESTEROEF, W., OLIVE, P., VERGNAUD-GRAZZIN1, C. 1982. After the deluge: Mediterranean stagnation and sapropel formation. Nature, (Lond.), 295, 105-10. RYAN, W.B.F. 1972. Stratigraphy of Late Quaternary sediments in the Eastern Mediterranean. In: Stanley, D.J. (ed.), The Mediterranean Sea--A Natural Sedimentation Laboratory. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 149-69. SIGL, W., CHAMLEY,H., FABRICIUS,F., D'ARGOUD, G., MI~LLER, J. 1978. Sedimentology and environmental conditions of sapropels. In: Hsfi, K.J., Montadert, L. et al., Init. Repts. DSOP, 42A. Natl. Science Found., Washington, DC. 445-65.
STANLEY, D.J. 1978. Ionian Sea sapropel distribution and Late Quaternary palaeoceanography in the Eastern Mediterranean. Nature, (Lond.), 274, 149-52. 1983. Fine parallel laminated deep-sea muds and coupled gravity flow-hemipelagic settling in the Mediterranean. Smith. Contr. Marine Sci., 19, 27 PP. -& BLANPIED, C. 1981. Late Quaternary water exchange between Eastern Mediterranean and the Black Sea. Nature, (Lond.), 285, 537-41. & MALDONADO,A. 1979. Levantine Sea-Nile Cone lithostratigraphic evolution: Quantitative analysis and correlation with paleoclimatic and eustatic oscillations in the Late Quaternary. Sed. Geol., 23, 37-65. -& MALDONADO,A. 1981. Depositional models for fine-grained sediment in the western Hellenic Trench, Eastern Mediterranean. Sedimentology, 28, 273-90. THUNELL, R.C., WILLIAMS,D.F. & KENNETT,J.P. 1977. Late Quaternary paleoclimatology, stratigraphy and sapropel history in Eastern Mediterranean deep sea sediments. Mar. Micropaleont., 2, 371-88. VERGNAUD-GRAZZINI,C., RYAN, W.B.F. & CITA, M.B. 1977. Stable isotope fractionation, climatic change and episodic stagnation in the Eastern Mediterranean during the Late Quaternary. Mar. Micropaleont., 2, 353-70. WASMUYD,E. 1930. Bitumen, sapropel and gyttja. Geol. For. Stockh. Forh., 52, 315-50. WILLIAMS, D.F. & THUNELL, R.C. 1979. Faunal and oxygen isotopic evidence for surface water salinity changes during sapropel formation in the Eastern Mediterranean. Sed. Geol., 23, 81-93.
-
-
G.C. ANASTASAKISAND D.J. STANLEY, Division of Sedimentology, NMNH, Smithsonian Institution, Washington, DC, 20560, USA.
The sedimentology of the Swedish Alum Shales A. Thickpenny SUMMARY: The middle and upper Cambrian (and locally lowermost Tremadoc) Alum Shale Formation of Scandinavia and the Baltic Sea area outcrops from Finnmark in north Norway to Bornholm in the south, and from Estonia westwards to the Oslo region. The Formation is generally underlain by shallow water sandstones but is locally unconformable on Precambrian continental basement; it is overlain by shallow marine glauconitic, phosphatic limestones. Detailed examination of borehole cores and some quarry sections from several of the Swedish Lower Palaeozoic outliers indicates the occurrence of five lithofacies: (1) Black laminated mudstone, rich in organic carbon, suggesting extreme anoxic conditions; (2) organically banded mudstone showing more variable organic input: (3) grey mudstone, laminated or unlaminated and lacking significant organic matter, suggesting more oxic conditions; (4) bituminous, fossiliferous limestone concretions also indicative of more oxic conditions, possibly above wavebase; and (5) glauconitic siltstones and sandstones, graded and transported, and quartz rich siltstones, mostly transported. Pyritic laminae may also occur within any of the above mudstone lithofacies. Sedimentation rates for some of the above facies are estimated and show values comparable with modern pelagic sedimentation, the lowest occurring within the bituminous limestones and progressively increasing to interlaminated black and grey mudstones. The sedimentation rates in combination with the lithological variation suggest less extreme reducing conditions in the middle Cambrian compared with the upper Cambrian, a factor important in the control of Uranium and other trace element anomalies found within the upper Cambrian. The Alum Shale Formation of Scandinavia and the Baltic Sea area (Fig. 1) is of early middle Cambrian to early Tremadoc age. Interest has been shown in the deposit for some 300 years since the salt alum was first mined. This century, the deposit has been opencast mined for oil and uranium which occur in high concentrations, and it is presently being assessed for trace metal extraction. The formation is part of a Cambrian and lower Ordovician relatively shallow marine succession deposited unconformably on Precambrian crystalline basement forming the western part of the Baltoscandian platform (Thorslund 1960). Outcrops can be conveniently separated into those of the Southern Swedish outliers where little structural deformation has occurred, and those of the Caledonide front and range, where structural style is dominated by thrust and nappe tectonics. Over much of the area, however, the stratigraphy of the Cambrian succession shows relatively little variation. Figure 2 is a generalized stratigraphic section for the Billingen-Falbygden area (Dahlman & Gee 1977). The Mickwitzia and Lingula sandstones are shallow marine deposits and the latter is locally conglomeratic, glauconitic and phosphatic in its upper part. They thicken southwards and westwards from central Sweden and in places are overlain by thin grey-green siltstones and shales that show variable transitions into the overlying Alum Shale. The limestones overlying the Alum Shale are sandy and
normally glauconitic and phosphatic. Thus, units above and below the Alum Shale are demonstrably of relatively shallow marine origin, and by inference so is the Alum Shale. The main Alum Shale lithology is black to dark brown organic-rich mudstone, with subordinate fossiliferous bituminous limestone concretions ('orsten' in Swedish). Minor intercalations of grey mudstone occur, mainly in the middle Cambrian, together with localized glauconitic sandstones and thicker packets of grey siltstones. The distribution of limestone concretions is highly variable although at two horizons in southern Sweden there are more continuous marker bands known as the Great Stinkstone and the Exporrecta Stinkstone (Fig. 2). Apart from local occurrences within the mudstones in SkS_ne (Scania), fossils are confined to the concretions. Using the concretions, a detailed biostratigraphy has been developed by Westergfird (1922, 1940, 1942, 1944a and b, 1947, 1948) with modification by Henningsmoen (1957). Figure 3 shows the zonation scheme, mainly after the SkS,ne section, adopted for the Alum Shale in several of the southern Swedish outliers. Because trilobites are preferentially preserved within discontinuous limestone concretions, their absence in any one section is not necessarily indicative of non-deposition or erosion, but probably due to their low preservation potential. However, the biostratigraphic control is generally good, allowing reliable sedimentolo5II
512
A.
Thickpenny
FIG. 1. Distribution of autochthonous and parautochthonous Cambrian deposits in Scandinavia and the Baltic area.
The sedimentology of the Swedish Alum Shales
FIG. 2. Generalized lower Palaeozoic stratigraphy of the Billingen area.
513
A. Thickpenny
5 I4
ALUM
SHALE
BIOSTRATIGRAPHY GEOGRAPHICAL AREA RANGES zll,I
FOSSIL
ZONES
oZ o~, .E~~
:0
--.,I "z
I~
mlk
m :O
T
t
Ceratopyge
Diclyonema flabelhforme Acerocare
q_
Peltura scarabaeoides
i
Peltura minor Prolopeltura praecursor
I
Leptoplastus ~ Eurycare Parabobna spinu&sa & Orusia Olenus & Hypagnoslus obesus Agnostus pisiformis Lejopyge laevigata Solenopleura brachymefopo (,3
Ptychognostus lundgreni d Goniognostus nothorsti Ptychognostus punctuosus
u) :5
.E
Hypagnostus parvifrons 01
Tomagnostus fissus PLy_c_b_~nostus atavus Ptychagnostus gibbus
~h
Eccoparodoxides pinus Eccoparadoxides insuloris FIG. 3. Middle and upper Cambrian Biostratigraphy.
_L_
e
:o
The sedimentology of the Swedish Alum Shales gical and geochemical correlation over large geographical areas. Previous geochemical studies have been confined mainly to the Billingen-Falbygden, Nfirke and Osterg6tland areas where a high Uranium peak has been identified in the P. scarabaeoides zone (Fig. 3) of the upper Cambrian (Dahlman 1962; Andersson 1974, 1977, 1979, 1980a, band c; Armands 1972; Edling 1974). Recent unpublished data of Sveriges Geologiska Unders6kning show similar uranium peaks together with other trace element anomalies in the J~imtland area of the Caledonide front (Fig. 1).
Stratigraphy Southern Swedish outliers
The Alum Shale in southern Sweden is wellknown stratigraphically and is relatively constant in thickness, generally 15-25 m, thickening in Skgme to 80-100 m. Variations in thickness and lithology are summarized in Fig. 4, which also includes representative columns from the Caledonides. In SkS.ne, P. paradoxissimus mudstones rest on lower Cambrian sandstones, locally silty or calcareous. Mudstones, locally sandy near the base of the formation and with rare stinkstones continue up into the Tremadoc, to be overlain by glauconitic phosphatic limestones. Organic carbon contents within the mudstones are lower than elsewhere in southern Sweden (Dahlman & Eklund 1953). The outliers ofV/isterg6tland (Halleberg-Hunneberg, Kinnekulle and Billingen-Falbygden) possess similar stratigraphy (Fig. 2)~ Billingen being the most studied area (Dahlman & Gee 1977). Deposition of Alum Shale commenced in P. paradoxissimas times (as in SkS,ne); the shale overlies lower Cambrian sandstones, conglomeratic at the top. The stinkstone content is much higher throughout, compared with SkS,ne, and thin Dictyonema shales are locally present at the top of the Alum Shale. A more detailed description is given in Andersson et al. (1983). (3sterg6tland, although in close proximity to Billingen, exhibits three major stratigraphic differences. The formation thins eastwards (Thorslund 1960) from 25 to 14 m, the P. paradoxissimus stage is developed as grey mudstone rather than Alum Shale and the Dictyonema shales are thicker and of greater extent. As the formation thins, stinkstones become more common, more conglomeratic and more phosphatic. In Nfirke the Alum Shale, 12-19 m thick, is further reduced in time span since the P. forchhammeri stage is
515
dominated by stinkstone and overlies grey and grey-green mudstone, and Ordovician limestones directly overlie the P. scarabaeoides zone mudstones. On (3land, the Alum Shale thins to < 1 m and is entirely up.per Cambrian in the North, while in southern (31and it reaches 24 m in thickness and about half is Tremadocian. The oldest stinkstone is of P. forchhammeri age and overlies non-Alum Shale lithologies (Westerghrd 1940). Several zones are apparently missing in the upper Cambrian, which contains progressively more stinkstone northwards. Recent drilling in central Gotland has identified Tremadoc and upper Cambrian black mudstones at depths of around 400 m (D.G. Gee, pers. comm. 1982). Stinkstones and some sandstones are subordinate lithologies in the sequence, which is up to 4.5 m thick. Swedish Caledonides
The stratigraphy of the Lower Palaeozoic succession in the Caledonides, compared to the southern Swedish outliers, is complicated by complex deformation especially in the incompetent Alum shales. However, both autochthonous and allochthonous sequences can be recognized (Gee et al. 1978) along the mountain front. Autochthonous late Precambrian and Lower Palaeozoic sediments are preserved along the eastern margin of the Caledonides between basement and overthrust allochthon. The Alum shale normally contains bituminous limestone concretions and varies slightly in thickness and lithology from north to south (Gee & Zachrisson 1979). In the southernmost Swedish Caledonides it rests on lower Cambrian shallow marine sandstones and is overlain by mid-Ordovician shales (Tegengren 1962). Northwards the sole thrust cuts out the Alum Shale as far as central J/imtland (Fig. 1) where an upper Cambrian black shale overlies middle Cambrian grey-green shales (Thorslund 1960). Uppermost Cambrian zones are missing suggesting a possible hiatus, since they are overlain by lower Ordovician phosphatic calcarenites. Local basement highs active during deposition have also been recognized (Thorslund 1940) on the basis of local thinning of the mudstone. In northern J/imtland about 15 m of middle and upper Cambrian black shales are preserved beneath the sole thrust. In south V/isterbotton, Lid6n (1910) recognized 40 m(?) of autochthonous upper Cambrian black shales overlying middle and lower Cambrian grey and green silts and shales. In northernmost Sweden (Tornetr/isk, Fig. 1) the autochthon reoccurs and is referred to as the 'Hyolithes' zone. Kullings (1964) estimate is of 40 m of probably middle Cambrian black
516
A. Thickpenny
d
,o ...a 0
.....,
E
o
,4
The sedimentology of the Swedish Alum Shales shales, lacking stinkstones, in the Tornetr/isk area. In Finnmark, northern Norway, Reading (1965) has described about 200 m of black shale of upper Cambrian or Tremadoc age, although further information as to detailed palaeontology and sedimentology is at present lacking. This is the thickest succession in Scandinavia and may imply either greater subsidence during deposition, or a small basinal area. Allochthonous late Precambrian and Lower Palaeozoic deposits are preserved in a relatively undisturbed state in northern and central JS.mtland, where a fairly detailed lithostratigraphy has been produced (Gee et al. 1974) including the Alum Shale equivalent, the Fj~illbr/inna Formation. Sediments immediately underlying the formation have yielded lower and middle Cambrian fauna and are shallow marine quartzites and shales. Black shales of the Fj/illbr~inna Formation are 50-55 m thick, thicken slightly westwards, and the base is often marked by a conglomerate (up to 1 m thick) with pyritized shale clasts (Gee et al. 1974). Stinkstones are present, particularly in the upper part, and non-bituminous limestones are locally present near the top. Tremadoc black phyllites are associated with pillow lavas and dolerite dykes, further west at the Norwegian border (Gee 1981). Black phyllites with the characteristic Alum Shale geochemistry are also found as far west as TommerS, s in central Norway and similar rocks, of probable Cambrian age, occur locally in the middle Allochthon which has been thrust from west of the lower Allochthon (Gee & Zachrisson 1979). Sedimentation rates in the Alum Shale
An important facet of the depositional environments for the Alum Shale is the relative sedimentation rates in the various successions. The method used for estimation of sedimentation rates is similar to that of Churkin et al. (1977). The SkS,ne section was selected to determine zonal lengths within the middle and upper Cambrian because it shows the least lithologic variability and contains fauna within the mudstones. The total time period of the Alum Shale (P. p a r a d o x i s s i m u s to D i c t y o n e m a ) is estimated to be c.23-25 my (mainly after Cowie & Cribb 1978) although the disparity in ages given by Gale et al. (1979), Harland et al. (1964), McKerrow et al. (1980) and Ross et al. (1978) for the Cambrian time-scale testify to the poor radiometric age control. An average thickness for each zone in Skhne was calculated from palaeontologically zoned boreholes (Westerggtrd 1944a) and the Alum
5I 7
Shale is assumed to have an almost constant rate of sedimentation since lithology varies little and there are no apparent breaks in succession: thus zone thickness is proportional to duration (Fig. 5). Measurements of the compaction ratio of mudstone to stinkstone at Ranstad Quarry Billingen, show that the mudstone is now about 20% of its original thickness (assuming that stinkstones are effectively uncompacted). Hence the original thickness of the mudstone may be estimated. Using the zonal durations derived from SkS,ne, sedimentation rates have been calculated for the two cores illustrated in Fig. 6, and the results are given in Table 1. Where possible, different lithofacies have been separated; the detailed palaeontology allowing this, is not shown in Fig. 6. The figures calculated, though by no means exact due to unknown variables, do give an estimate of sedimentation rates within the deposit.
Lithofacies Detailed examination of hand specimen and thin section material of borehole cores shows several lithofacies. Five lithotypes are recognized throughout the Alum shale, and two borehole core logs that show typical vertical distributions of the lithotypes are shown in Fig. 6. Black laminated mudstone
This lithotype consists of dark brown to black mudstone (grain-size < 10/~m) with a high content of total organic carbon (TOC), ranging from 5-7% in the middle Cambrian to 20~ locally in the upper Cambrian (Armands 1972). The organic matter is fine-grained and presumably of algal origin; it is present as compressed and discontinuous laminae separated by less organic rich areas. The characteristic fine parallel lamination is caused mainly be slight changes in organic matter content. Trilobites are almost invariably absent and there is no disturbance of laminae due to burrowing. Trace fossils are lacking. The mineralogy has been estimated by Armands (1972) from a borehole in the Billingen area. The dominant silicate minerals are illite (30~o in the upper Cambrian, 40% in the middle Cambrian), quartz (variable around 25%) and K-feldspar (5-15%). Chlorite, found in small quantities near the base of the formation is absent higher up the succession. The other main constituent is pyrite, forming 5-15% of the rock. The clay minerals form a fabric with the plates flattened parallel to lamination and constitute the
518
A. Thickpenny
NAME OF BOREHOLE
THICKNESS IN METRES
ANDRARUM GISLOVSH. TOSTERUP AVERAGE S. SANDBY ANDRARUM 1 2
ZONE D. flabelliforme Acerocare Peltura Lep toplas tus P. spinulosa Olenus A. pisiformis P forchhammeri P. paradoxissimus
9.6 7.6 20.6 1.3 7.5 12.7 6.1 12.8 11"7
7.4 2.3 10"5 11.4 4.5 5.1 10.8
8"7 8"7 11"5 2"2
16"5 4"6 20"4 1-1 5"5 6"8 3"9 3"8 12 "1
7.7 18"0 1.4 9.1 9"8 1"5 5"0 15"3
13 "1 8"0 17"6 1"7 8"2 10"2 4"0 6"7 12"4
DURATION (my)
3.7 2-2 5.0 0.5 2.3 2.9 1.1 1.9 3.5 ,,,
F1G. 5. Middle and upper Cambrian zonal thickness in Sk~ne (Scania) and estimated periods of duration.
Estimation o f sedimentation rates for different lithologies in two boreholes f r o m Ostergotland and Billingen
TABLE 1.
L YCKHEM BH Zone
Thickness now Original thickness
Dictyonema Total: 2.4 m
Sedimentation rates
Total: 11.2 m Orsten: 0.2 m Mudst.: 11 m % Orsten: 1.8
Mudst. + Orsten 3 m/my
Orsten: 0.2 m Mudst.: 2.2 m % Orsten: 8
Upper Cambrian (11.8 my)
Total: 5.7 m Orsten: 2.6 m Mudst.: 3.1 m % Orsten: 46
Total: 18.1 m Orsten: 2.6 m Mudst.: 15.5 m % Orsten: 14
Total: 1.3 mm/103 Great Stinkstone: 0.15 mm/103 yr
Middle Cambrian (5.4 my)
Total 6.1 m Orsten: 0.5 m Mudst.: 5.6 m % Orsten 8
Total: 28.5 m Orsten: 0.5 m Mudst.: 28 m ~ Orsten: 1.8
Mudst. + O r s t e n 5.3 mm/103 yr
(3.7 my)
= 3 mm/103 yr
Remaining mudst.: 3.4 mm/103 yr
DBH 16/74 Using the same method as outlined as above, the following results were obtained: Zone
Sedimentation rate
Peltura Great stinkstone A. pisiformis P. forchhammeri P. paradoxissimus
Mudstone + Orsten: 7.5 mm/10 3 yr 0.27 mm/103 yr 3 mm/10 3 yr Mudstone + Orsten: 16 mm/103 yr Mudstone + Orsten: 10 mm/103 yr
The sedimentology of the Swedish Alum Shales
519
FIG. 6. Sedimentological logs of the Alum shale in two bore-hole cores from Osterg6tland and Billingen.
520
A.
Thickpenny
dal in shape and up to 2 m in diameter. Often bedding is continuous from the concretions into adjacent shale beds, and may converge towards the margins (Henningsmoen 1974) due to concretion growth during sediment compaction. Concentric growth rings are also apparent in some examples (Henningsmoen 1974). Central parts of concretions are normally of arenitic grain-size, comprising uncompressed, fragmented trilobite remains and calcite crystals. The margins are often recrystallized to fibrous cone-in-cone calcite crystals, up to 10 cm in length. In many cases, the nodules appear to have amalgamated laterally and may show preferential growth along certain horizons giving a ridged appearance (Henningsmoen 1974). At several horizons, conglomeratelike layers have developed, though clasts are angular and do not appear to have been transported far. Organically-banded mudstone Concretions generally contain more than 80% calcium carbonate (Henningsmoen 1974). The The dark brown organically-banded mudstone has a similar mineralogy to the black laminated carbonate contains organic matter, which immudstone. However, there are lower TOC con- parts a brown colour and bituminous odour. The tents, always < 10% and often closer to 5%. Slight remaining matrix is of normal mudstone compovariation in TOC content causes colour banding sition. In thin-section, the majority of the carbonate consists of rounded sand-sized grains of in hand specimen. random orientation set in a poorly laminated matrix. Fossil fragments are distinguishable together with filamentous growths sub-parallel to Grey mudstone bedding, which also occur unattached in mudGrey mudstone occurs dominantly in the lower stones, as crystalline growth. At the margins there part of the Alum Shale formation (for example, are fewer carbonate grains and they appear to see Fig. 6). Two types of grey mudstone can be infill pores in the mudstone. The large fibrous recognized. The first is less common, occurring crystals have grown roughly perpendicular to the locally in the Tremadocian Dictyonema shales, concretion margins, and have apparently and consists of pale grey, non-calcareous deformed the surrounding mudstone as replacive laminated mudstone, with gradational but paral- cone-in-cone growth. lel upper and lower boundaries. The second type occurs mainly near the base of the Alum Shale and consists of grey-green mudstone, commonly Sandstones and siltstones with calcareous laminae, and is generally ex- Thin beds and laminae of sandstone or siltstone tremely friable and poorly laminated. Contacts occur locally near the base of the Alum Shale, and with interbedded darker laminae are generally may be glauconitic. 1-2 cm thick glauconitic uneven. A shelly fauna including bivalve and sandstone beds occur in core DBH 16/74; the brachiopod fragments may be present, for exam- glauconite is associated with quartz and both ple in the Lyckhem core (Fig. 6) where the minerals show a rounded appearance. Glauconite lithology grades downwards into grey-green grains are discrete and well-formed showing no mudstone lacking all darker beds. evidence of intermediate stages in formation. Beds are often graded from sharp, slightly erosive bases and show good sorting. The matrix is Bituminous limestone (stinkstone or orsten) mudstone. Other detrital minerals such as rutile, Bituminous limestone concretions occur chiefly staurolite and plagioclase are present in small as discontinuous or semi-continuous lenses in all amounts, and pyrite nodules are common. One parts of the Alum Shale, but more commonly in bed in core DBH 16/74 shows convolute-laminathe upper Cambrian. Locally, in southern tion of glauconitic sand and darker mudstone and Sweden, two continuous 'bands' occur; the Great may be a slide horizon. Several thin silty horizons are also visible in Stinkstone and the Exporrecta Stinkstone. The concretions are generally roughly ellipsoi- thin-section, consisting of quartz laminae, a few
extremely fine-grained groundmass. Quartz is also fine-grained and is distributed randomly throughout the sediment. K-Feldspar is partly detrital and partly authigenic, the latter form becoming dominant upwards through the formation. Fine-grained pyrite is ubiquitous in the groundmass but pyrite also occurs as scattered blebs and as laminae. Laminae often occur in 'packets', mainly in the upper Cambrian. Laminae are 1-2 mm thick and in packets 2-5 cm thick. The laminae generally show a gradational contact with surrounding mudstone, often showing pyrite concentrated towards the base and decreasing gradually upwards. In thin-section, the pyrite shows a fine-grained framboidai form. Diagenetic calcareous laminae, of similar thickness, may be associated with the pyrite laminae.
The sedimentology of the Swedish Alum Shales grains in thickness and with occasional glauconite grains, generally slightly coarser than the quartz. Thin non-glauconitic beds and laminae also occur near the base of the Alum Shale. Two types of laminae can be recognized. The first consists of pale < 1 cm thick quartz-rich laminae showing sharp contacts and a poorly-defined lamination, and has only been observed in core DBH 16/74. The second type is common throughout the lower part of the Alum Shale and consists of quartzrich, siltier layers, one to two grains thick, forming parallel lamination in the mudstone and only visible in thin section. Vertical and lateral distribution of lithofacies
Typical vertical variation of sediment types is shown by the sedimentary logs in Fig. 6, which shows a progressive transition from grey mudstone with fauna via interlaminated mudstones to black mudstone, probably reflecting a transition from oxic to anoxic conditions. The change is better shown in the Lyckhem borehole, where sedimentation rates can also be seen to roughly halve, from the grey mudstone dominated middle Cambrian to the predominantly black mudstone of the upper Cambrian. A similar, though less marked trend is visible in core DBH 16/74, with the additional features of glauconitic sandstones and quartz-rich siltstones. Bituminous limestone concretions occur predominantly in black mudstones but rarely in grey mudstone. A similar vertical facies change is reflected in many other cores from the southern Swedish outliers. Large-scale lateral facies changes are also present (Fig. 4). The upper Cambrian part of the formation varies least: slight thickness changes and the presence or absence of limestone concretions being the most common variables. The upper part of the middle Cambrian comprises mostly black mudstone, though conglomeratic limestones also occur, together with interlaminated black and grey mudstone. This is a common lithofacies association occurring as mmscale alternations of dark brown or black mudstone and medium-pale grey mudstone. Contacts between the two are variable, often sharp and straight at both boundaries, but lower contacts of the grey mudstone may be uneven and partly mixed with underlying black mudstone. Parallel lamination occurs locally within the grey mudstones and there is no obvious grading. Calcareous pods and laminae are more common where black and grey mudstone are interlaminated and where nodules of pyrite are present, they are generally confined to the grey laminae. The lower part of the middle Cambrian, however, shows considerable lateral facies variation with only
52I
local development of black mudstone (e.g. Skgme), while elsewhere greater thicknesses of non-Alum Shale lithology are developed (e.g. Gland). Tremadocian Dictyonemashales are variably developed in thickness and character; bituminous limestone may or may not be present and locally grey mudstone and siltier mudstone may occur. Where the Dictyonernashale is lacking, it is not clear whether it was eroded prior to subsequent limestone deposition, or whether it was never deposited. Evidence of erosion is also lacking in areas where parts of the upper Cambrian are not present (e.g. Nfirke).
Discussion Facies interpretation
The fine grain-size of the black mudstone and its lack of current induced sedimentary structures suggest deposition in quiet water below wavebase. High organic carbon contents, syngenetic pyrite and the absence of a fauna or trace fauna suggest anoxia at the sediment-water interface. The calculated sedimentation rates of 3 to 8 mm/103 years are similar to modern pelagic oozes (Berger 1974) although the depositional environment of the Alum Shale was on a shelf; a situation with no present-day analogues. The laminated mudstone facies was probably deposited below the aerobic/anaerobic boundary within the water column where the anoxic layer impinged on a large shelf. Thickness variations are slight in the upper Cambrian, where this lithofacies is dominant, suggesting there was no significant ponding. Slight variations in organic matter content are on a time-scale too large to be explained by annual variations and must be due to longer-term climatic changes. The occurrence of pyrite in laminae, mainly within the black mudstone facies, suggests they are primary sedimentary deposits, formed in a similar way to the mechanism described by Berner (1970). Silt laminae in the silt/sandstone facies have not been preferentially pyritized so it is unlikely that the pyritic laminae are related to grain-size variation. Controlling factors in pyrite formation are amounts of organic matter, reactive iron, sulphate and elemental sulphur. However, in the case of the Alum Shale, none of these factors have conclusive supporting evidence for controlling the production of pyrite laminae. The fact that pyritic packets are more common in the thinner sequences of Alum shale in Sweden may suggest slower sedimentation and/or decreased terrigenous input. Thus, the pyrite-producing reactions could take place with a continuing
522
A. T h i c k p e n n y
supply of organic matter, but in the absence of dilution by a terrigenous component. In addition, variations in elemental sulphur content may have been controlled by slight vertical movements of the oxic-anoxic boundary. The depositional environment of the dark brown mudstones was similar to that of the black mudstone. The variations in organic matter may be explained either by a variable input of organic matter and/or variations in detrital terrigenous input. For the Lyckhem borehole (Fig. 6), the estimated sedimentation rate (including stinkstones) is 5.3 mm/103 years for the middle Cambrian, where this lithology is most common, while the rate is only 3.4 mm/103 years for the upper Cambrian (excluding stinkstones). This suggests a less stable environment during the middle Cambrian. The poorly laminated grey-green mudstone (lithofacies probably indicates relatively normal oxic conditions. Poor lamination and uneven contacts are the result of burrowing, and the shelly fauna indicate the presence of macroscopic invertebrates requiring >1 ml O2/1 (Rhoads & Morse 1971). This lithology underlies the Alum Shale sensu stricto in the Lyckhem borehole, but interdigitates in the lower part of the 16/74 core, testifying to short periods of oxic conditions during transition to the typical Alum Shale environment. The isolated grey laminated mudstone occurrences within the Dictyonema shale indicate a variable organic carbon supply, without the establishment of fully oxic conditions. The gradational boundaries suggest that the process was not a sudden event and that variation occurred on the scale of hundreds of years, according to the calculated sedimentation rates (Table 1). The characteristic fine-scale alternations of the black and grey interlaminated facies association suggest variability in the input of organic matter and/or terrigenous material. Burrowing, evident at some levels in the grey mudstone shows that the bottom waters were not always toxic for a small soft-bodied fauna. This is supported by the local occurrence of calcareous macrofossil fragments within the mudstone. The interlaminated association suggests a dysaerobic environment (Rhoads & Morse 1971; Cluffet al. 1981) with alternating anoxic and more oxic conditions. Oxygen content was probably controlled by variations in the supply of organic matter. The origin of the bituminous limestone concretions has been explained as the remains of a once continuous limestone bed that was subsequently dissolved (Bjorlykke 1973), and as early stage concretions (Henningsmoen 1974). The evidence presented here suggests that the concretions are
similar to the early formed diagenetic concretions of Raiswell (1971), and not the remains of a previous limestone bed. The concretions appear to originate by nucleation on fossiliferous layers, possibly on the sea floor, followed by continued growth during early stages of compaction. This interpretation poses two major problems: (1) the concretions are highly fossiliferous and yet adjacent mudstones are apparently devoid of any fauna or traces of it, and (2) up to three trilobite zones occur in succession within the Great Stinkstone bed, indicating development over very long time periods. Hallam's model (1981) for the formation of Liassic concretions may explain the first problem, whereby slight undulations of the sea floor (on the horizontal scale of 10's of kilometres and the vertical scale of 10's of metres) occur on a broad epicontinental shelf. Intra-basinal highs are proposed as the sites of concretion formation, such that they penetrated through the probable permanent anoxic-oxic boundary within the water column allowing colonization by trilobite faunas, possibly adapted to low dissolved oxygen contents (Henningsmoen 1957). Winnowing and reworking could have concentrated the organic/ skeletal remains, although for the most part current structures are lacking. The extremely slow sedimentation rates would favour local lithification by calcite precipitation around concentrations of trilobite debris. Slight fluctuations of the oxic-anoxic boundary and/or subsidence of the highs would have destroyed the living trilobite fauna and brought the carbonate into more acidic conditions, allowing dissolution of much of the trilobite shell material except where concentrated or lithified. Growth and reprecipitation could have occurred around these nucleation sites giving the present fossil-rich concretions and the adjacent shale which is devoid of fossils. The second problem may be explained by considering the above model operating over long time periods where the environment was almost starved of sediment influx. Advanced lithification and coalescence of the incipient concretions may have occurred and later the whole horizon may have been modified by growth during compaction. That faunas are apparently in succession within the Great Stinkstone suggests progressive lithification upwards and low current and wave activity. However, the conglomeratic horizons in the Exporrecta stinkstone, for example, may suggest reworking, perhaps due to uplift to wavebase or to occasional storm action. They also testify to very early lithification. The most important consequence of this model is to show that conditions were oxic or semi-oxic for long periods of time over varying extents of
The sedimentology o f the Swedish Alum Shales the Alum Shale basin during the Upper Cambrian. Sedimentary evidence indicates that the coarser sandstone lithofacies and probably the siltier beds too, are transported material, possibly slumped in some cases. Odin & Matter (1981) suggest that glauconite generally occurs towards continental margins or on submarine highs where slow sedimentation rates or non-deposition and an ambient non-oxidizing micro-environment occur. Slight topographic highs with little terrigenous input probably developed in the Alum Shale depositional areas and could have favoured the accumulation of glauconite. Berner (1981) suggests glauconite is characteristic of the postoxic, non-sulphidic environment where low contents of organic matter were present, inhibiting sulphate reduction and pyrite production. The non-glauconitic siltstones of this lithofacies imply increased terrigenous input, although the processes by which this occurred are difficult to define. Wind transport and current activity (indicated by cross-lamination) are possible processes.
Alum shale depositional environment Discontinuous outcrops of Alum Shale at the present day suggest an original Alum shale depositional area of up to 2000 km north-south by 800 km east-west. Shortening by thrusting is estimated to be up to 500 km (Gee 1975), and several of the nappes contain black phyllites which are probable Alum Shale equivalents (Gee 1975). Therefore the basin may originally have extended west of the Norwegian coast. In the highest nappes, no black phyllites are recorded and metamorphosed sediments may, in part, be deeper water equivalents of the Alum Shale, tectonically transported eastwards (Gee 1975). Over the entire basinal area, the Alum shale is apparently overlain and underlain by shallow marine deposits with little variation in lithology and this evidence is used to suggest that the Alum Shale is also a relatively shallow water deposit. The size of the outcrop area suggests deposition in an epicontinental sea environment, which lacks modern analogues. Hallam (1981) has summarized likely oceanographic conditions in such a sea. Where water depth is shallow ( < 200 m) there is a tendency for stagnation away from the open ocean. This appears to be the environment of the Alum Shale, particularly as lithologies are remarkably constant over large areas and for long periods of time. General stratigraphic considerations suggest that contemporaneous sea-levels were high (Leggett et al. 1981). Consistent slow sedimentation in
523
a shallow water environment lends support to this idea, since it implies restricted detrital supply, one cause of which might be flooding of source areas. Redeposited sediment is rare, suggesting gentle sea-floor topography and the dominant process appears to have been sedimentation from suspension. Characteristically, the deposits of an epicontinental shelf sea show great lateral persistence (Hallam 1981). However, the lithofacies recognized here show variations in space and time. The vertical variation of lithofacies previously described, reflects a progressive cutting-off of sediment supply and expansion and stabilization of anoxic conditions in the Alum Shale sea. However, the occurrence of bituminous limestone concretions testifies to the likely presence of long ranging intra-basinal highs, penetrating the oxic layers of the upper water column. The sedimentation rates of the fossiliferous precursor sediments of the concretions is the slowest of all the lithofacies, and probably involves considerable pauses in sedimentation. Vertical facies variations not only indicate more variable sediment supply within lower parts of the Alum Shale, but also more rapid variations in the oxygen content of deeper waters. Lateral variation of lithofacies indicates a slow expansion of anoxic conditions through the middle and upper Cambrian reaching a peak roughly in the P. scarabaeoides zone. The overlying Dictyonema shales show a waning of these conditions since deposition appears to have become more localized and the sediments may contain silty laminae, indicating increased detrital supply. Since much of the original Alum Shale has probably been eroded it is not clear whether intra-shelf basins existed. However, if present, they would be candidates for more localized anoxic conditions. Similar, but thicker and less extreme facies trends also occur in Britain (Leggett 1980 and references therein), SE Newfoundland, Cape Bretton Island and New Brunswick (North 1971 and references therein). Anoxic conditions with the development of black mudstones often with concretionary limestones occur at certain preferred periods within the middle and upper Cambrian. Black or dark grey mudstones are generally interspersed with periods dominated by bioturbated mudstones and siltstones, with local development of sandstones. These facies variations demonstrate that the area of the probable SE margin of the Iapetus Ocean intermittently affected by anoxic conditions was of greater extent than the present Alum Shale outcrop. ACKNOWLEDGMENTS: I would like to acknow-
A. Thickpenny
524
ledge the help of the Swedish Geological Survey, in particular D r David Gee, and Aktiebolaget Svensk Alumskifferutreckling, in allowing examination and discussion of borehole core material. I would also like to a c k n o w l e d g e receipt of research assistantship support from Chalmers
University of Technology, G o t h e n b u r g during 1981-82. Finally, I w o u l d like to t h a n k Drs Jeremy Leggett, Kevin Pickering and D o r r i k Stow for critically reading the manuscript and A m a n d a H a r t l a n d for typing it.
References ANDERSSON,A. 1974. Unders6kningar av den 6verkambriska alunskiffer i fern b6rrkarnor fr~n Sydbillingens h6gplatfi. Ranstad Skifferaktiebolaget, Report TPM-IR-826, 1-12. 1977. Analys av alunskifferformationen i DBH 65/77 fr~n Ranstadsverkets dagbrott. Ranstad Skifferaktiebolaget, Report TPM-U-976, 1-9. 1979. Analyser av hela alunskifferformationen i DBH 16/74 fr~n Sydbillingen. Ranstad Skifferaktiebolaget, Report TPM-I106, I-3. 1980a. Borrningar 1975 av 15 k~irnor genom 6verkambrisk alunskiffer i Sydbillingen-Ranstadomr~tdet. Koordinatf6rtforteckning mfiktigheter och analyser. Ranstad Skifferaktiebolaget, Report TPM-1283, 1-5. 1980b. N/irkes tillg~nger av alunskiffer. Aktiebolaget Svenstk Alunskifferutveckling, Report ASA T9/80, 1-6. 1980c. Billingen-Falbygden-omr~tdets tillg~ngar av alunskiffer. Aktiebolaget svensk Alunskifferutveckling, Report ASA T58/80, 1-10. , DAHLMAN,B. & GEE, D.G., 1983. Kerogen and Uranium resources of the Cambrian Alum shales of the Billingen-Falbygden and N/irke areas, Sweden. Geol. F6r. Stockh. F6rh., 104, 197-209. ARMANDS, G. 1972. Geochemical study of Uranium, Molybdenum and Vanadium in a Swedish Alum shale. Stockh. Contr. Geol., 27, 1-148. ASKLUND,B. 8~;THORSLUND,P. 1935. Fj/illkedjerandens Bergbyggnad i Norra J/imtland och Angermanland. Sver. geol. Unders., ser. C 382, 1-110. BERGER,W.H. 1974. Deep sea sedimentation. In: Burk, C.A. & Drake, C.L. (eds), The Geology of Continental Margins. Springer-Verlag, New York. 213-41. BERNER,R.A. 1970. Sedimentary pyrite formation. Am. J. Sei., 1-23. 1981. A new geochemical classification of sedimentary environments. J. sed. Petrol., 51,359-65. BJORLYKKE,K. 1973. Origin of limestone nodules in the Lower Palaeozoic of the Oslo region. Norsk geol. Tiddskr., 53, 419-31. CHURKIN, M., CARTER, C. & JOHNSON, B.R. 1977. Subdivision of Ordovician and Silurian time scale using accumulation rates of graptolitic shale. Geology, 5, 452-6. CLUFF, R.M., REINBOLD,M.L. & LINEBACK,J.A. 1981. The New Albany Shale Group of Illinois. Illinois State Geol. Surv., cir. 518, 1-83. COWIE, J.W. & CgmB, S.J. 1978. The Cambrian system. In: Cohee, G.V., Glaessner, M.F. & Hedberg, H.D. (eds), Contributions to the Geological Time Scale. Am. Ass. Petrol. Geol. Studies in Geology 6, 355-62.
-
-
-
-
-
-
2 6 8 ,
-
-
DAHLMAN,B. 1962. Unders6kningar av Overkambrium reed hfinsyn till uranfordelning och sediment i mellersta och s6dra Sverige utforda vid Sveriges geologiska unders6kning. Sver. geol. Unders., internal report. Unpubl. DAHLMAN, B. & EKLUND, J. 1953. Sveriges uranf6rande alunskiffrar. Sver. geol. Unders., Unpubl. report. & GEE, D.G. 1977. Oversikt 6ver BillingenFalbygdens geologi. In: Betiinkande av Billingeutredningen, Billingen 4, exampel. Statens offentliga utredningar 1977: 47, Industri departementet, Stockholm. EDL1NG, B. 1974. Distribution of Uranium in Upper Cambrian Alum Shale from Ranstad, Billingen, V~istergotland. Publications from the Palaeontological Institute of the University of Uppsala, special volume 2, 1-118. GALE, N.H., BECKINSALE,R.D. & WADGE, A.W. 1979. A Rb-Sr whole rock isochron for the Stockdale Rhyolite of the English Lake District and a revised mid-Palaeozoic time-scale. J. geol. Soc. Lond., 136, 235-42. GEE, D.G. 1975. A tectonic model for the central part of the Scandinavian Caledonides. Am. J. Sci., 275, 465-515. 1980. Basement-cover relationships in the central Scandinavian Caledonides. F6r. Stockh. Fgrh., 102, 455-74. -1981. The Dictyonema-bearing phyllites at Nordaunevoll, eastern Trondelag, Norway. Norsk geol. Tiddskr, 61, 93-5. --, KARIS, L., KUMPULAINEN,R. 8r THELANDER,Y. 1974. A summary of Caledonian front stratigraphy, northern Jfimtland/southern VSsterbotten, central Swedish Caledonides. Geol. F6r. Stockh. F6rh., 96, 389-97. --, KUMPULAINEN,R. 8r THELANDER,T. 1978. The Thsj6 d6collement, Swedish Caledonides. Sver. geol. Unders., set. C742, 1-35. ZACHRISSON, E. 1979. The Caledonides in Sweden. Sver. geol. Unders., set. C 769, 5-48. HALLAM, A. 1981. Facies Interpretation and the Stratigraphic Record. Freeman, Oxford. HARLAND, W.B., SMITH, A.G. & WXLCOCK,B. (eds.). 1964. The Phanerozoic Time Scale. Q. J. geol. Soc. Lond., 120S. HENN1NGSMOEN,G. 1957. The trilobite family Olenidae, with description of Norwegian material and remarks on the Olenid and Tremadocian Series. Skr Norske Vidensk-Akad, Oslo. Mat. naturv. Kl., l, 1-303. 1974. A comment. Origin of limestone nodules in
-
-
-
-
The sedimentology of the Swedish Alum Shales the Lower Palaeozoic of the Oslo region. Norsk. geol. Tiddskr., 54, 401-12. KULLING, O. 1964. Oversikt 6ver norra Norrbottensfjfillens kaledonidberggrund. Sver. geol. Unders., ser. Ba 19, 1-166. LAMBERT, R. ST.J. 1971. The pre-Pleistocene Phanerozoic time-scale--a review. In: Harland, W.B. & Francis, E.H. (eds). The Phanerozoie Time-scale: a supplement. Spec. Publ. geol. Soc. Lond., 5, 9-31. LEGGETT, J.K. 1980. British Lower Palaeozoic black shales and their palaeo-oceanographic significance. J. geol. Sor Lond., 137, 139-56. LIDI~N, R. 1910. Kalkstensf6rekomster utefter inlandsbanan mellan Str6ms Vattudal och Pite /ilf. Sver. geol. Unders., ser. C 235, 1-45. MCKERROW, W.S., LAMBERT, R. ST. J. & CHAMBERLAIN, V.E. 1980. The Ordovician, Silurian and Devonian time scales. Earth planet. Sci. Lett., 51, 1-8. NORTH, F.K. 1971. The Cambrian of Canada and Alaska. In: Holland, C.H. (ed.), Cambrian of the New World. John Wiley, London. ODIN, G.S. & MATTER, A. 1981. De glauconarium origine. Sedimentology, 28, 611-43. RAISWELL, R. 1971. The growth of Cambrian and Liassic concretions. Sedimentology, 17, 147-71. READING, H.G. 1965. Eocambrian and Lower Palaeozoic geology of the Digermul peninsula, Tanafjord, Finnmark. Norg. geol. Unders., 234, 167-91. RHOADS, D.C. & MORSE, J.W. 1971. Evolutionary and ecologic significance of oxygen-deficient marine basins. Lethaia, 4, 413-28. Ross, R.J., NAESER,C.W., IZETT, G.A., WHITTINGTON, H.B., HUGHES,C.P., RICKARDS,R.B., ZALAS1EWICZ, J., SHELDON, P.R., JENKYNS,C.J., COCKS, L.R.M.,
525
BASSETT, M.G., TOGHILL, P., DEAN, W.T. & INGHAM, J.K. 1978. Fission-track dating of Lower Palaeozoic bentonites in British stratotypes. In: Zartman, R.E. (ed.), Short papers of the 4th International conference, Geochronology, Cosmochronology and Isotope Geology. US geol. Surv. Open-file Rep. 78-701,363-5. TEGENGREN, F.R. 1962. Vassbo blymalmfyndighet i Idre och dess geologiska inramning. Sver. geol. Unders., ser. C 586, 1-61. THORSLUND, P. 1940. On the Chasmops series of J~imtland and S6dermanland (Tv~iren). Sver. geol. Unders., ser. C 436, 1-191. 1960. The Cambro-Silurian. In: Magnusson, N.H., Thorslund, P., Brotzen, F., Asklund, B. & Kulling, O. Description to accompany the map of the pre-Quaternary rocks of Sweden. Sver. geol. Unders., ser. Ba 16, 69-110. WESTERG.~RD, A.H. 1922. Sveriges Olenidskiffer. Sver. geol. Unders., ser. C 18, 1-189. 1940. Nya djupborrningar genom/ildsta Ordovicium och kambrium i Osterg6tland och N/irke. Sver. geol. Unders., ser. C 437, 1-72. 1944a. Borrningar genom Sk'~nes Alunskiffer 1941-42. Sver. geol. Unders., ser. C 459, 3-38. 1944b. Borrningar genom Alunskifferlagret p'~ ()land och i Osterg6tland. Sver. geol. Unders., ser. C 463, 1-45. -1947. Supplementary notes on the Upper Cambrian trilobites of Sweden. Sver. geol. Unders., ser. C 489, 1-34. 1948. Non-agnostidean trilobites of the Middle Cambrian of Sweden. I. St, er. geol. Unders., ser. C 498, 1-32. -
-
-
-
A. THICKPENNY,Department of Geology, Royal School of Mines, Imperial College, London SW7 2BP.
Models for the deposition of Mesozoic-Cenozoic fine-grained organic-carbon-rich sediment in the deep sea M.A. Arthur, W.E. Dean and D.A.V. Stow SUMMARY: The widespread occurrence of organic-carbon-rich strata ('black shales') in certain portions of Jurassic, Cretaceous and Cenozoic sequences has been well-documented from Deep Sea Drilling Project sites in the Atlantic and Pacific Oceans and from sequences, now exposed on land, originally deposited in the Tethyan ocean. These ancient black shales usually have been explained by analogy with examples of modern deep-sea sediments in which organic matter locally is preserved by (1) increasing the supply of organic matter, (2) increasing the rate of sedimentation, and/or (3) decreasing the oxygen content of the bottom water. However, detailed examination of many black shales reveals characteristics that cannot be explained by simple local models, including: their approximate coincidence in time globally; their occurrence in a variety of different environments, including open oxygenated oceans, restricted basins, deep and shallow water; their interbedding with organic-carbonpoor strata which often dominate a so-called black shale sequence; their deposition by pelagic, hemipelagic, turbiditic and other processes; and the variations in type and amount of organic matter that occur even within the same sequence. A more complex model for the origin of black shales therefore appears most appropriate, in which the cyclic preservation of organic matter depends on the interplay of the three main variables, namely supply of organic matter, sedimentation rate, and deep-water oxygenation, each of which varies independently to some extent. The variation and relative importance of these parameters in individual basins and widespread black shale deposition in general are linked globally and temporally by changes in global sea-level, climate and related changes in oceanic circulation. An important and often overlooked factor for the supply of organic matter to deep-basin sediments is the frequency and magnitude of redepositional processes. The interplay of these variables is discussed in relation to the middle Cretaceous and Cenozoic organic-carbon-rich strata, in particular, which show marked differences in the relative importance of the different variables. Results of analyses of Deep Sea Drilling Project (DSDP) cores show that strata containing relatively high concentrations of organic matter (more than 2% organic carbon) are common at different stratigraphic horizons and in many parts of the world ocean. These strata are often loosely described as 'black shales' although they usually consist of interbedded rocks with varying contents of carbonate and/or biogenic silica and differ mainly in colour and/or concentration of organic matter. The most common lithologies are interbedded black or dark-olive shale or claystone, and lighter greenish-grey shale or claystone (i.e. interbedded 'black' and 'green' argillaceous rocks). The oldest occurrences of organic-carbon-rich strata in the present ocean basins are upper Jurassic (Callovian-Oxfordian) black claystones recovered at two DSDP Sites in the South Atlantic and one in the North Atlantic, and Tithonian-Neocomian deep-water marlstones and limestones that have been recovered from several DSDP sites in the North Atlantic. Blackshale deposition was widespread in the North and South Atlantic during the Barremian-Albian (100-115 my BP) and Cenomanian-Turonian (86-100 my ~p) at both continental-margin and
deep-basin sites (Arthur & Natland 1979; Jenkyns 1980; Weissert 1981) and locally during the Coniacian-Santonian (75-86 my BP). Cretaceous organic-carbon-rich strata in the Pacific are much more restricted in both time and space than in the Atlantic. They have been recovered from seven DSDP sites on the flanks of elevated volcanic plateaus and seamounts, mainly within very restricted stratigraphic intervals (Schlanger & Jenkyns 1976; Dean et al. 1981; Thiede et al. 1982). Organic-carbon-rich strata in the Pacific all occur within the same general middle Cretaceous time interval, but they are not strictly synchronous and occur in strata that range in age from 86-120 my SP. Middle Cretaceous strata recovered from deep basins of the Pacific generally are not rich in organic carbon, but a thin layer containing 5?/0organic carbon was recently recovered at the Cenomanian-Turonian boundary within a turbidite sequence in the eastern Mariana Basin (DSDP Site 585; Moberly et al. 1983). Organic-carbon-rich sediments of Eocene and Miocene age have been recovered at a few sites in the Atlantic (Arthur & von Rad 1979; McCave 1979a; Dean & Gardner 1982), but these strata are much more restricted in both time and space than those of Cretaceous age. In the Pacific, 527
M.A. Arthur, W.E. Dean and D.A.V. Stow
528
however, extensive organic-carbon-rich Miocene strata occur in both offshore and onshore localities. In this paper we first provide an outline of the main modes of preserving organic matter in modern deep-sea sediments, then describe the lithology, distribution and organic-carbon content of ancient Mesozoic and Cenozoic black shales, and finally discuss the elements of models for black shale deposition. Our emphasis throughout is on deep-water and fine-grained organic-carbon-rich sediments.
matter occurs during the dominantly oxic conditions of diagenesis in most pelagic deep-sea sediments (e.g. Heath et al. 1977; Demaison & Moore 1980). There are three main ways of preserving relatively high concentrations of organic matter in deep-sea sediments: the first is by increasing the supply of organic matter, the second is by increasing the rate of sedimentation, and the third is by decreasing the oxygen content of the water mass overlying the sediment. These processes usually but not always act in consort in modern marine environments.
Preservation of organic matter in modern deep-sea sediments
Organic matter supply
Most modern deep-sea environments are characterized by sediments that are relatively depleted in organic carbon ('
o ~, ~
01
O
rr 9
, o.oi
W.BALTIC
o PERU 9 OREGON
oool Ol
1
lO
10o
SEDIMENTATION
lO0O
RATE
o ARGENTINE BASIN
[cm(loooy)-I ] ,
i JJtllllt
L illqlll
ill~,.u
9 N.W. A F R I C A
o CENTRAL PACIFIC i ~Jl.Jl
A BLACK SEA
PRIM. PROD. RATE (gC m-2y -1)
300 t B 150 O-.J
, r2.. """
~ ") *
I
~mo
ORGANIC CARBON % 01 ! ................................. 01 1 10 100 1000 SEDIMENTATION RATE
[cm0oooy)-l]
FIG. 2. Relations among bulk sedimentation rate and (A) percent organic carbon, (B) primary productivity, and (C) organic carbon accumulation rates for modern marine sediments (Miiller & Suess 1979). The scatter about the best-fit regression line represents the effects of dysaerobic or anoxic conditions which further enhance preservation (e.g. Black Sea points), as well as no correction for differences in depth of deposition among other problems.
53o
M.A. Arthur, W.E. Dean and D.A.V. Stow
recovered at DSDP sites. The main reason for this relationship is probably that high sedimentation rates effectively bury organic matter more rapidly, removing it from zones of bioturbation, oxic decomposition, and sulphate reduction. Although organic-carbon contents and accumulation rates may be higher in high-sedimentationrate sequences, the organic matter itself may be somewhat more poorly preserved losing a greater proportion of H, P, N than that deposited under anoxic conditions. Hydrocarbon source potential is determined not only by the organic carbon content but also the characteristics of the organic matter.
Bottom-water oxygenation Oxygen concentrations of less than 0.5 ml/l and certainly 0.2 ml/l inhibit the activity of benthic metazoans (e.g. Rhodes & Morse 1971). Lack of bioturbation in turn limits the residence time of organic matter at the sediment/water interface and, therefore, the state of its oxic decomposition. Low-oxygen concentrations occur in both midwater oxygen-minimum zones and in silled stratified basins, and, where lowest (i.e. 1%orgCi (good preservation,} high Hydrogen ~ Indices) amount } depends on supply i d oA ~
Few primary sedimentary structures preserved
TEXTURE
(homogeneous to bioturbate) Generally < 1% orgC oxidized/refractory (structured to dark amorphous, with Hydrogen Index < 100 mg HC/gC) [May vary with productivity/sed rate]
9 "6 / [~
SEDIMENT
SEDIMENT ORGANIC CARBON CONTENT SEDIMENT
Reduced SIC < 0.4 (generally) (may vary with sedimentation rate) sulphate-reduction only below Sed/H20 interface
REDUCEDSULPHUR CONTENT
Shades of black,-shades of grey s brown, olive green- grey ~ olive-green ~ to tan and red - C o l o u r depends on CaCO3 content, Corg content (and type), presence of acid volatile sulfides, etc. GENERAL
CRITERIA
BOTTOM-WATER LEVELS
IN M U D D Y
FOR
RECOGNITION
DISSOLVED MARINE
SEDIMENT COLOUR
OF
OXYGEN
ENVIRONMENTS
It
certain species of polychaetes (e.g. Calvert, 1966) and/or calcareous or arenaceous benthic foraminifers (e.g. Phleger and Soutar, 1973) may inhabit surficial environments where benthic metazoans are excluded.
2~
degree of lamination in part depends on seasonal variations in clastic input a l t e r nating with higher productivity intervals, and on sedimentation rate.
3~
see text.
FIG. 5. Oxygen content at the sediment/water interface and generalized expected relationships between types of benthic organisms, primary sedimentary structures, sediment chemistry and mineralogy (sulphides), organic content and type (data and concepts from numerous sources).
M . A . Arthur, W.E. Dean and D . A . V . Stow
534
~E o "5 E
% v-
4 PERU 3
P 57
uJ I-.,< I:E z
2
P 39
o I._I
3 P.~ 1
1 n."
o ,10. o a.
//~1462
~CIR 189B
~
S . W ~
/ /.,ACK SEA
0.5
l/
(UNIT A)
CIR 17713. '~CIR 179B DSDP
1470
t ~1432
r 12302--1
1233e--4
12327-4 12337-4
/N.W. AFRICA/ 12329-4 12328-4 12347.4 12345-4 tDSDP 0
T 5
"1 10
T 15
20
~ 25
30
35
40
45
ORGANIC CARBON ACCUMULATION RATE (1() s rnol cm 2 ~1)
FIG. 6. Plot of total phosphorus accumulation rate vs. organic carbon accumulation rate for Holocene sediments from the Peruvian slope, the south-west African shelf (Namibia), the Black Sea abyssal plain and slope, the north-west African slope, and for average mid-Cretaceous 'black shales' from selected Atlantic DSDP Sites (see Glenn & Arthur, in press, for sources of data and calculations). Letters and numbers indicate individual cores. with coastal divergences with well-developed midwater oxygen-minimum zones. Note that the organic carbon and total phosphorous accumulation rates for upwelling settings are much higher than for the Black Sea, which is characterized by an anoxic water column but moderate to low productivity. Seasonal wind-driven upwelling, high nutrient supply and high rates of organic carbon production and deposition over the relatively shallow shelf off SW Africa (e.g. Calvert & Price 1971) probably are much more important than oxygen levels in controlling the accumulation of organic carbon because bottom water
oxygen levels rarely fall below 1 ml/1 on either the shelf or slope (with the exception of a few local areas of < 0.1 ml/1), and because the organic-carbon-rich sediments are not laminated. Lack of lamination may also be due to a lack of strong seasonal variations in supply of sediment components other than diatoms and organic carbon. On the Peruvian margin, some sediments with higher organic-carbon accumulation rates are associated with deposition under low-oxygen conditions ( < 0.02 ml/l) within the oxygen-minimum zone. There are some differences in phosphorus and
Deposition of organic-carbon-rich sediment
535
of supply of organic matter probably offset some of the effects of an oxygenated environment. Transfer of organic carbon to sediments is largely by faecal pellets (Bishop et al. 1978), and marine lipids are the only ones present off the arid coast of Namibia. Concentrations of dissolved oxygen in bottom waters off Peru are significantly lower than in bottom waters of most areas off Namibia, and this must aid in preservation of organic carbon. The higher phosphorus accumulation rates of several Peruvian cores (P-36 and P-42, Fig. 6) may indicate a supply of phosphorus in addition to that associated with organic carbon
organic carbon accumulation rates between offshore Peru and SW Africa (Namibia). Surface water nutrient supply and productivity are nearly the same in both settings (Koblentz-Mishke et al. 1970; Huntsman & Barber 1977). The results shown in Fig. 6 for SW Africa all came from sediment samples from the upper 50 cm or less in cores from shallower than 134 m on the Namibian shelf; the results for Peru are from cores at depths of 186-645 m on the outer shelf or upper slope. The accumulation rates of organic carbon are not too different in either setting. The shallow depth of the Namibian examples and the high rate
_M C
KEY M - MAESTRICHTIAN
(M)
c - 1 % - 30~
~.
_
_
~'~ (T,R)
~7
C - CAMPANIAN S - SANTONIAN
mat 3 - 2 0 %
~_
c
s
~cT " I - 3%(M'T'R)
C - CONIACIAN
c
u
M'~
T - TURONIAN
A ~](M)
C - CENOMANIAN A - APTIAN
A A
H
'
H - HAUTERIVIAN
v
I
V - VALANGINIAN
B
B - BERRIASIAN
L
L JUR- LATE JURASSIC
, IUF
~ GENERAL CORE
AI
k
BI
MI?
I
1-
A
(M,T)
-6
V
VI
SITE 364
JR
357 c.~ ]:::.: ~%
(M) 23% (M)
V
A
-j
HI
c,li:
SITE
356
COVERY
D- 6%
A
C l : u: "r (T,R)
V .-g
'~' _..A 1-19%
--,,m
-1%
.~wis
I
SITE
m II
C
B
B - BAFIREMIAN
T
i
TI:A
A
A - ALBIAN
IIII =~,
c
cmF
C
<M)
Z-~c ~
I
530 EDIMENT INDICATING SITE 363
t-.i ~.,i ?
.2: A il
OR MUDSTONE " LAMINATED M
TO SLIGHTLY BIOTURBATE
c -g-
2%=ORGANIC CARBON CONTENT ~ N T
B
-fi-
"7"
Corg TYPE
(M)= MARINE AMORPHOUS
uR ~
(R) =OXIDIZED RESIDUAL A A
SITE 511
E
~
3~
330
0.5%
/ ,,J
,so
"7" c A , I% (M)
A
H
B
SITE
c --
-g-,
9
-~
3,,
~-
s/"
4% (M) 92%
V
r-
M
) a~, ') j" ~j
"T
(T) = TERRIGENOUS
-5% (M,T)
(~ 1 - 3% (T,R)
?"
v i L JUR SITE 327
FIG. 7. Stratigraphic and core recovery columns for DSDP Sites from South Atlantic Ocean basins for the Mesozoic. Diagram illustrates intervals of 'black shale' deposition, generalized contents and types of organic matter.
536
M.A. Arthur, W.E. Dean and D.A.V. Stow
cW
-~ C
HI
I
T'' A
I
C T~ T ul ~ sl
M
~~ T C T T u ~s C
C
M - -
S
A ,~
A
i
--
T
T
u
m
S
C A
A
~-g-
B
T
C
__~ ~ B~
H__ v
H
S
~A
C u S T
AIZ~o~
~__~.~___L~'
organic matter it was not necessarily the only cause for accumulation of organic-carbon-rich strata. An increase in the relative amount of organic debris deposited was equally important. At different sites, there were differences in the relative amounts of organic debris derived from areas of increased productivity within the basin and preserved in sediments under oxygen-minima on continental margins and organic debris from areas with increased terrigenous supply. In many places, organic matter was redeposited to basinal sites by turbidity currents; in other places, there was relatively rapid accumulation of pelagic organic matter. The main variables that may have
02;>0 ~
elln
Oz=O
_'y_ . . . . . . . . . . . . . . . . . . . . . . . . . .
\
L a t e l i b Y a n to E a r l y
Slope'~ ~
C. . . . . . i . . . . d Coniacian
8....
A
02>0 ~
T rb dr U
~
"~s
~
~~
I
0
o,I I
.......
,org. . . . . . . ,oo> 2 ~ o , - - ~ \ \
? T r O,od.....
.--. o..o
' . ' . o . . . . . . '--
551
....o . . . . . . ,,,~
Gf...... ~ _ ~ (.......... 09) / I black sediment ~ .// -4000m I
'J
I
FIG. 11. Cartoons illustrating the interplay between organic carbon fluxes from terrigenous sources and from surface water productivity, and changes in their relative importance in sedimentary deposits as the result of changes in climate, sea-level, upwelling and oceanic fertility, and deep-water oxygen contents. Also emphasized are pathways for organic matter deposition, including by downslope movement (e.g. turbidites) from river mouths (terrigenous organic matter) or from within oxygen-minimum zones (well-preserved marine organic matter), pelagic settling (e.g. faecal pellets and/or marine snow), and settling from nepheloid-layers (both marine and terrigenous organic matter). (A) Late Albian-early Cenomanian or Coniacian-Santonian in North and South Atlantic: low to moderate productivity with 'normal' mid-water oxygen-minimum zone, oxygenated (i.e. > 1.0 ml/1) deep-water masses. (B) Late Cenomanian-early Turonian or possibly middle Aptian-middle Albian in South and North Atlantic. Higher rate of supply of organic carbon and greater extent of oxygen depletion in deep-water masses (after Dean et al., in press).
552
M.A. Arthur, W.E. Dean and D.A.V. Stow
operated during a period of time when less organic matter was accumulating and during a period of time when more organic matter was accumulating are summarized in Fig. 11. The widespread distribution of black shale and related organic-carbon-rich facies occurred within relatively restricted time periods during the Cretaceous. The term 'Oceanic Anoxic Events' (OAE) has been applied to these periods (Schlanger & Jenkyns 1976). Unfortunately, this term has been construed by others to indicate that anoxic conditions existed throughout the water column in all ocean basins at precisely the same time. The original intent of the term was to point out that there were periods of time (BarremianAlbian, Cenomanian-Turonian and possibly Coniacian-Santonian) when organic-carbon-rich marine facies were much more widespread in epicontinental seas and open ocean basins than they are today. Although single black shale beds probably cannot be correlated from basin to basin, packages of organic-carbon-rich facies, no matter what their amount and type of organic matter, generally are time equivalent. This emphasizes the possibility that the occurrences of organic-carbon-rich facies in different basins might be linked in some way, even though they may occur in different lithologies and depositional settings. We favour fluctuations in sea-level and climate as the dominant causes for variations in amount of organic matter supplied and preserved (e.g. Fischer & Arthur 1977). Higher global sea-level and warm, equable climates probably influenced rainfall, production of terrestrial vegetation and runoffto the oceans, thereby increasing the flux of terrigenous sediment and organic matter to ocean basins, particularly the relatively narrow North and South Atlantic Oceans during the Hauterivian through Albian (e.g. Sites 398; 400; 402; Habib 1979). The rapid rate of supply of terrigenous organic matter to a basin that was already poorly-oxygenated (but not necessarily anoxic) probably led to enhanced oxygen deficits in some basins, and allowed preservation of what marine organic matter was produced and supplied to basin deeps. Locally, particularly in shallow epicontinental seas and restricted ocean basins, high runoff periodically may have led to less saline surface waters and stable stratification of the water column, which, in turn, produced deeper water oxygen deficits (e.g. Ryan and Cita 1977; Thierstein & Berger 1978; Arthur & Natland 1979). This would be particularly true if the supply of terrigenous organic matter was high, or if autochthonous productivity increased, perhaps stimulated by nutrients supplied by runoff. McCave (1979b) and Hochuli & Kelts (1980)
suggested that periodic (every 50 ky or less) high productivity intervals could have produced the interbedded organic-carbon-rich and organiccarbon-poor facies so typical of the Cretaceous deep-water black shale sequences. In general, however, sea surface productivity during the Cretaceous probably was low overall in comparison to that of today (Roth 1978; Berger 1979; Thierstein 1979), and only in certain regions (e.g. offN.W. Africa) can high sea surface productivity be called upon to produce Hauterivian-Albian deep-water organic-carbon-rich facies. Marine organic matter appears to be more prevalent in strata of mid-Aptian to early Albian age, and of Cenomanian age in many AtlanticTethyan sequences. This might indicate periods of overall higher sea surface fertility and intensification of midwater oxygen minima related to higher sea-level stands coupled with more rapid production of warm, saline bottom water (Southam et al. 1982; Brass et al. 1982) or some other mechanism (Tucholke & Vogt 1979; Summerhayes 198 lb). A high sedimentation rate may have been a factor in the preservation of organic matter in some regions (e.g. Ibach 1982), but some of the lowest sedimentation rate intervals, which occur during higher sea-level stands (e.g. de Graciansky et al. 1981), have the greatest enrichment of organic matter. Figure 12 illustrates a compilation of organic-carbon contents (core averages) for the western North Atlantic Cretaceous sequences plotted against interval sedimentation rates (compare with Fig. 2, after Miiller & Suess 1979). The dotted line represents the approximate Miiller & Suess (1979) relationship for sediments from modern marine environments corrected for compaction of the older Cretaceous strata. The scatter of points is high, and many of the points lie above the line, suggesting that the rate of organic carbon supply and/or preservation under lowoxygen conditions was at times more important than bulk sedimentation rate in causing enhanced organic carbon accumulation in Cretaceous ocean basins. Even so, most Cretaceous deepwater organic carbon accumulation rates were, on average, much slower than modern shallower water organic carbon accumulation rates (Fig. 6). Factors in the origin of Cenozoic organic-carbonrich strata
The oceanographic and climatic conditions under which the Cenozoic black shales were deposited in the Atlantic are very different from those in the Cretaceous. The ocean basin was wide and unrestricted by the Eocene so that basin floor deoxygenation is not likely and there are no character-
Deposition of organic-carbon-rich sediment 7-
553
SEDIMENTATION RATE vs Corg AVERAGE CRETACEOUS WESTERN NORTH ATLANTIC
SITE O SITE 3 8 6 [] SITE 105
s
A r~
FZ ILl 0 nLU n
SITE
101A
V SITE 391 SITE 387 [] SITES 417D.418A&B
4
_~p2
3 0
0 Lu .r w u.i ,r
m 2
0
**
[]
.
IB O
O o
a
0
[]
v
9
.
M-S relationship assuming ~ 50% compaction ~
,
=
9
*V
[] []
V
A
v
r~2
7 I
"7
A I
~
v I
10
,o
l
go
INTERVAL SEDIMENTATION RATE (mm 1000y-')
F16. 12. Plot of interval sedimentation rate (mm 1000 y - l) versus average organic carbon content by core (or core section) in DSDP Sites from the western North Atlantic. The points represent compilation of data from the Hauterivian through Cenomanian of Sites 386, 387, 101, 105, and 417, 418 (data in Arthur & Dean, in press). The dashed line is the 'best-fit' regression line of Mfiller & Suess (1979) for recent marine sediments adjusted for 50~ compaction of Cretaceous strata. Note large scatter of points (see text for discussion).
istics of the organic-carbon-rich facies that would suggest that such an event occurred. Global climate had begun to deteriorate near the beginning of the Tertiary so that by the late Eocene there was a fairly marked temperature differential between the poles and equator. There is considerable evidence now to infer that modern deep thermohaline circulation began at this time, with the formation of cold dense water at high southern latitudes and its northward flow into the Atlantic as Antarctic Bottom Water (Berggren & Hollister 1977; Kennett 1977; Tucholke & Vogt 1979; Shor & Poore 1978; Moore et al. 1978). Initially upwelling of older nutrient-rich waters forced by the deep circulation changes would have led to increased productivity, particularly along the continental margins, and hence to an increased supply of organic matter to the sediments. Redeposition of this material to the deeper
ocean basins (e.g. at DSDP Sites 367, 370 and 386) was most likely by turbidity currents and other mass-flow processes. In contrast to the above possibility, Brass et al. (1982) suggested that the production of warm saline bottom water was linked to eustatic sea-level rise and transgression of shelf areas. An early to middle Eocene sea-level highstand may have increased production of bottom water and increased deep-water turnover rates thereby stimulating biologic productivity, which could have led to intensification of oxygen-minimum zones (Southam et al. 1982). Kelts & Arthur (1981; and references within) further suggested that the widespread Eocene cherts of the deep North Atlantic ('Horizon A') may be explained by increased productivity of siliceous organisms on the margins and subsequent downslope redeposition. The middle to late Miocene to Pliocene
554
M.A. Arthur, W.E. Dean and D.A.V. Stow
organic-carbon-rich facies (mainly biogenic siliceous sediments) in the Atlantic (Diester-Haas & Schrader 1978; Gardner et al., in press) and Pacific (Ingle 1981; Summerhayes 1981a) also appear to be related to vigorous ocean circulation, upwelling and prolific diatom productivity. This was in response to the continued deterioration of global climate and mid-Miocene buildup of the Antarctic ice cap (Savin 1977; Kennett 1977). Antarctic Bottom Water, Arctic Bottom Water and Norwegian Sea Overflow Water all provided deep vigorous circulation at this time (e.g. Shor & Poore 1978). Organic-carbon-rich sediments accumulated under upwelling zones in the open, oxygenated north-east and south-east Atlantic, and all around the margin of the north Pacific which apparently was more fertile than the Atlantic because of an estuarine-type circulation (Berger 1970). A widespread period of tectonism in the circum-Pacific led to the formation of silled marginal basins which became deoxygenated where the midwater oxygen-minimum zone impinged on the bottom at sill depth (Ingle 1981). Lowered sea-level also may have helped to lower the base of the oxygen-minimum zone (Summerhayes 1981a), and coastal divergences may have moved seaward, off the shelf edge, and provided a more direct source of organic matter to deeper oceanic settings (e.g. Gardner et al., in press).
Conclusions From the above discussion it is apparent that the origin of interbedded more- and less-reduced lithologies with variable amounts of organic matter and variable amounts of pelagic, hemipe-
lagic, and terrestrial sediment is complex and probably is not due to any one simple process although events in different ocean basins may be linked by a common factor, such as changes in global sea-level. The main variables are supply of organic matter from land and from surface water productivity, sedimentation rate, and oxygen concentration in bottom waters. These variables are greatly influenced by climatic, oceanographic, geographic and tectonic factors. The supply of organic matter in turn determines the thickness and intensity of a midwater oxygen-minimum zone. Surface water productivity is determined, at least in part, by the intensity of upwelling of nutrient-rich water from within the upper part of the oxygen-minimum zone. Another important factor in the accumulation of deep-water black shales is the frequency and magnitude of sediment redepositional events. The middle Cretaceous black shales owe their origin to an interplay of these three main variables. At this time, the oceans were poised at relatively low oxygen levels which periodically tipped in favour of organic matter preservation, largely due to increased supply of organic matter from surface productivity or terrigenous input, commonly aided by downslope resedimentation. By contrast, the Cenozoic oceans witnessed an increasingly vigorous thermohaline circulation, possibly higher overall fertility, widespread upwelling and enhanced productivity. Organicmatter was preserved both in restricted marginal basins and in open ocean sites; redeposition was locally important. Each region and even each bed within a black shale sequence is likely to have its own particular combination of factors that led to organic-matter preservation, so that generalized models may not always be applicable.
References M.A. 1979. North Atlantic Cretaceous black shales: the record at Site 398 and a brief comparison with other occurrences. In: Sibuet, J.-C., Ryan, W.B.F. et al. (eds), lnit. Repts. DSDP, 47, part 2. US Govt. Print. Off., Washington, DC. 719-51. & N A T L A N D , J.H. 1979. Carbonaceous sediments in North and South Atlantic: the role of salinity in stable stratification of early Cretaceous basins. In: Talwani, M., Hay, W.W. & Ryan, W.B.F. (eds),
reef-reservoired giant oilfields: Bull. Am. Ass. Petrol.
ARTHUR,
- -
Deep Drilling Results in the Atlantic Ocean: Continental Margins and Paleoenvironment. Maurice
--
Ewing Series 3, Am. Geophys. Union, Washington, DC. 375-401. & SCHLANGER, S.O. 1979. Cretaceous 'oceanic anoxic events' as causal factors in development of
Geol., 63, 870-85.
--
- -
& PREMOLI SILVA, I. 1982. 2. Development of widespread organic carbon-rich strata in the Mediterranean Tethys. In: Schlanger, S.O. & Cita, M.B. (eds), Nature o f Cretaceous carbon-rich facies. Academic Press, London. 7-54. & VON R A D , U. 1979. Early Neogene base-of-slope sediment at Site 397, DSDP Leg 47A: sequential evolution of gravitative mass transport processes and redeposition along the northwest African passive margin. In: von Rad, U., Ryan, W.B.F., et al. (eds), Init. Repts. DSDP, 47A. US Govt. Print. Off., Washington, DC. 569-614. , VON RAD, U., CORNFORD, C., McCoY, F. &
Deposition of organic-carbon-rich sediment SARNTHEIN, M. 1979. Evolution and sedimentary history of the Cape Bojador continental margin, Northwest Africa. In: von Rad, U., Ryan, W.B.F., et al. (eds), Init. Repts. DSDP, 47A. US Govt. Print. Off., Washington, DC. 773-815. BARKER, P.F., DALZIEL,I.W.D. et al. 1977. Init. Repts. DSDP, 36. US Govt. Print. Off., Washington, DC. 1079 pp. BARRON, E.J. & WASHINGTON,W.M. 1982. Cretaceous climate: a comparison of atmospheric simulations with the geologic record: Palaeogeogr., Palaeoclimatol., Palaeoecol., 40, 103-33. BARTON, E.D., HUYER, A. & SMITH, R.L. 1977. Temporal variation observed in the hydrographic regime near Cabo Corviero in the northwest African upwelling region, February to April, 1974. Deep Sea Res., 24, 7-24. BENON, W.E., SHERIDAN, R.E. et al. 1978. Init. Repts. DSDP, 44. US Govt. Print. Off., Washington, DC. 1005 pp. BERGER, W.H. 1970. Biogenous deep-sea sediments: fractionation by deep-sea circulation. Bull. geol. Soc. Am., 81, 1385-1402. -1976. Biogenous deep-sea sediments: production, preservation and interpretation. In: Riley, J.P. & Chester, R. (eds), Chemical Oceanography, 5, 2nd Edition. Academic Press, London. 265-388. 1979. Impact of Deep Sea Drilling on paleoceanography, In" Talwani, M., Hay, W.W. & Ryan, W.B.F. (eds), Results of Deep Drilling in the Atlantic Ocean: Continental Margins and Paleoenvironment: Maurice Ewing Series 3, Am. Geophys. Union. 297-344. , DIESTER-HAASS, L. & KILLINGLEY, J.S. 1978. Upwelling off northwest Africa: The Holocene decrease as seen in carbon isotopes and sedimentologic indicators. Oceanologica Aeta, 1, 3-7. -& VON RAD, U. 1972. Cretaceous and Cenozoic sediments from the Atlantic Ocean. In: Hayes, D.E., Pimm, A.C. et al., Init. Repts. DSDP, 14. US Govt. Print. Off., Washington, DC. 787-953. BERGGREN, W.A. & HOLLISTER, C.D. 1977. Plate tectonics and paleocirculation-commotion in the ocean. Tectonophysies, 38, 11-48. BERNER, R.A. 1977. Stoichiometric models for nutrient regeneration in anoxic sediments. Limnol. Oeeanog., 22, 781-6. 1978. Sulfate reduction and the rate of deposition of marine sediments. Earth planet. Sci. Lett., 37, 492-8. BERNOtJLLI, D. 1972. North Atlantic and Mediterranean Mesozoic facies: a comparison. In: Hollister, C.D., Ewing, J.I. et al., Init. Repts. DSDP, 11. US Govt. Print. Off., Washington, DC. 801-72. & JENKYNS, H.C. 1974. Alpine, Mediterranean, and central Atlantic Mesozoic facies in relation to the early evolution of the Tethys. In: Dott, R.H., Jr. & Shaver, R.H. (eds), Modern and Ancient Geosynclinal Sedimentation. Soc. econ. PaleD. Min. Spec. Pub. 19, 129-60. BISHOP, J.K., KETTEN,D.R. & EDMOND,J.M. 1978. The chemistry, biology and vertical flux of particulate material from the upper 400 m of the Cape Basin in
-
-
-
-
555
the southeast Atlantic Ocean, Deep Sea Res., 25, 1
1
2
1
-
6
1
.
BOLLI, H.M., RYAN, W.B.F. et al. 1978. Init. Repts. DSDP, 40. US Govt. Print. Off., Washington, DC. 1079 pp. BRASS, G.W., SOUTHAM,J.R. & PETERSON,W.H. 1982. Warm saline bottom water in the ancient ocean. Nature, 296, 620-3. BREMNER, J.M. 1980. Physical parameters of the diatomaceous mud belt off South West Africa. Marine Geol., 34, M67-M76. BRONGERSMA-SANDERS,M. 1971. Origin of major cyclicity of evaporites and bituminous rocks: an actualistic model. Marine Geol., 11, 123-44. BURNETT, W.C., VEEH, H.H. & SOUTAR, A. 1980. U-series oceanographic, and sedimentary evidence in support of recent formation of phosphorite nodules off Peru. In: Bentor, Y.K. (ed.), Marine Phosphorites. Soc. econ. PaleD. Min. Spec. Pub. 29, 61-71. CALVERT, S.E. 1964. Factors affecting distribution of laminated diatomaceous sediments in Gulf of California. In van Andel, T.H. & Shot, G.G., (eds), Marine Geology of the Gulf of California. Am. Ass. Petrol. Geol. Mere. 3, 311-30. 1966a. Accumulation of diatomaceous silica in the sediments of the Gulf of California. Bull. geol. Soc. Am., 77, 569-96. 1966b. Origin of diatom-rich, varved sediments from the Gulf of California. J. Geol., 74, 546-65. & PRICE, N.B. 1971. Upwelling and nutrient regeneration in the Benguela Current, October 1968. Deep Sea Res., 18, 505-23. CORNFORD, C. 1979. Organic deposition at a continental rise: organic geochemical interpretation and synthesis at Site 397, Eastern North Atlantic. In: von Rad, U., Ryan, W.B.F. et al., Init. Repts. DSDP, 47A. US Govt. Print. Off., Washington, DC. 503-10. DEAN, W.E., A~TnUR, M.A. & STOW, D.A.V. 1984. Origin and geochemistry of Cretaceous deep-sea black shales and multicolored claystones, with emphasis on DSDP Site 530, southern Angola Basin. In: Hay, W.W., Sibuet, J.C. et al., Init. Repts. DSDP, 75. US Govt. Print. Off., Washington, DC, in press. & GARDNER, J.V. 1982. Origin and geochemistry of redox cycles of Jurassic to Eocene age, Cape Verde Basin (DSDP Site 367), continental margin of northwest Africa. In: Schlanger, S.O. & Cita, M.B. (eds), Nature and Origin of Cretaceous Organic Carbon-Rich Facies. Academic Press, London. 55-78. - - , GARDNER,J.V., JANSA, L.F., CEPEK, P. & SEmOLD, E. 1977. Cyclic sedimentation along the continental margin of northwest Africa. In: Lancelot, Y., Seibold, E. et al., Init. Repts. DSDP, 41. US Govt. Print. Off. Washington, DC. 965-86. --, TH1EDE, J. & CLAYPOOL, G.E. 1981. Origin of organic carbon-rich mid-Cretaceous limestones, Mid-Pacific Mountains and southern Hess Rise. In: Vallier, T.L., Thiede, J. et al., Init. Repts DSDP, 62. US Govt. Print. Off. Washington, DC. 877-90. DEGENS, E.T. & MOPPER, K. 1976. Factors controlling
-
-
-
-
-
-
-
-
556
M.A. Arthur, W.E. Dean and D.A.V. Stow
the distribution of early diagenesis of organic matter in marine sediments. In: Riley, J.P. & Chester, R. (eds), Chemical Oceanography, 6, 2nd Edition. Academic Press, New York. 59-113. -& Ross, D.A. (eds) 1974. The Black Sea-Geology, Chemistry and Biology. Am. Ass. Petrol. Geol. Mem., 20, 633 pp. -& STOFFERS,P. 1976. Stratified waters as a key to the past. Nature, 263, 22-7. DEMAISON, G.J. & MOORE, G.T. 1980. Anoxic environments and oil source bed genesis. Bull. Am. Ass. Petrol. Geol., 64, 1179-209. DEROO, G., HERBIN, J.P., ROUCACHE, J. & T1SSOT, B. 1980. Organic geochemistry of Cretaceous sediments at DSDP Holes 417D (Leg 51), 418A (Leg 52), and 418B (Leg 53) in the Western North Atlantic. In: Donnelly, T.W., Francheteau et al., Init. Repts. DSDP, 51-53, part 2. US Govt. Print. Off., Washington, DC. 737-46. , HERBIN, J.P., ROUCACHE, J., BOUDON, J.P., ROBERT, P., JARDINE, S. & MARESTANGE,P. 1982. Geochemistry and optical study of organic matter in some Pleistocene and Pliocene sediments from DSDP/IPOD Site 474 to 481 of Leg 64, Gulf of California. In: Curray, J.R., Moore, D.S. et al. lnit. Repts. DSDP, 64. US Govt. Print. Off., Washington, DC. 855-64. DEUSER, W.G. 1974. Evolution of anoxic conditions in the Black Sea during the Holocene. In: Degens, E.T. & Ross, D.A. (eds), The Black Sea-Geology, Chemistry and Biology. Am. Ass. Petrol. Geol. Mem., 20, 133-6. DE GRACIANSKY, P.C., AUFFRET, G.A., DUPEUBLE, P., MONTADERT, L. & MULLER, C. 1979. Interpretation of depositional environments of the Aptian/Albian black shales on the north margin of the Bay of Biscay (DSDP Sites 400 and 402). In: Montadert, L., Roberts, D.G. et al., Init. Repts. DSDP, 48. US Govt. Print. Off., Washington, DC. 877-907. DE GRACIANSKY,P.C. et al. 1982. Les formations d'fige Cr&ac6 de l'Atlantique nord et leur mati&e organique: palaeog6ographie et milieux de d6pft. Rdv Inst. Francais Pdtrdle, 37, 275-336. & CHENET, P. 1979. Sedimentological study of Cores 138 to 56 (upper Hauterivian to middle Cenomanian): an attempt at reconstruction of paleoenvironments. In: Sibuet, J.-C., Ryan, W.B.F. et al., Init. Repts. DSDP, 47B. US Govt. Print. Off., Washington, DC. 403-18. DIESTER-HAASS, L. & SCHRADER, H.J. 1979. Neogene coastal upwelling history of northwest and southwest Africa. Marine Geol., 29, 39-53. DONNELLY, T.W., FRANCHETEAU,J. et al. 1980. Init. Repts DSDP, 51-53, part 1. US Govt. Print. Off. Washington DC. 351 pp. DOUGLAS, R.G. & SAVIN, S.M. 1975. Oxygen and carbon isotope analyses of Tertiary and Cretaceous microfossils from Shatsky Rise and other sites in the North Pacific Ocean. In: Larson, R.L., Moberly, R.L. et al., Init. Repts. DSDP, 32. US Govt. Print. Off., Washington, DC. 509-21. DSDP LEG 76 SCIENTIFICPARTY 1981. Challenger drills at sites off east coast. Geotimes, 26, 23-5. DUNBAR, R.B. & BERGER,W.H. 1981. Fecal pellet flux -
-
to modern bottom sediment of Santa Barbara Basin (California) based on sediment trapping. Bull. geol. Soc. Am., 92, 212-8. EINSELE,G. & KELTS, K. 1982. Pliocene and Quaternary mud turbidites in the Gulf of California: Sedimentology, mass physical properties and significance. In: Curray, J.R., Moore, D.G. et al., Init. Repts. DSDP, 64. US Govt. Print. Off., Washington, DC. 511-28. - & WIEDMANN, J. 1975. Faunal and sedimentological evidence for upwelling in the Upper Cretaceous coastal basin of Tarfaya, Morocco. lOth International Congress on Sedimentology, Nice, Theme 1. 67-72. EMBLEY, R.W. 1976. New evidence for the occurrence of debris flow deposits in the deep sea. Geology, 4, 371-4. & MORLEY, J.J. 1980. Quaternary sedimentation and paleoenvironmental studies of Namibia (SW Africa). Marine Geol., 36, 183-204. EMERY, K.O. 1960. The Sea off Southern California: A Modern Habitat o f Petroleum. John Wiley, New York. 366 pp. FISCHER, A.G. & ARTHUR, M.A. 1977. Secular variations in the pelagic realm. In: Cook, H.E. & Enos, P. (eds), Deep Water Carbonate Environments. Soc. econ. Paleon. Min. Spec. Pub. 25, 19-50. GARDNER, J.V., DEAN, W.E. & WILSON, C. 1983. Carbonate and organic-carbon cycles and the history of upwelling DSDP Site 532, Walvis Ridge, South Atlantic Ocean. In: Hay, W.W., Sibuet, J.-C., et al., lnit. Repts. DSDP, 75. US Govt. Print. Off., Washington, DC., in press. GASKELL, S.J., MORRIS, R.J., EGLINTON,G. & CALVERT, S.E. 1975. The geochemistry of a recent marine sediment off northwest Africa. An assessment of source of input and early diagenesis. Deep Sea Res., 22, 777-89. GILBERT, D. 1984. Organic facies variations in the Mesozoic South Atlantic. In: Hay, W.W., Sibuet, J.-C. et al., lnit. Repts. DSDP, 75. US Govt. Print. Off., Washington, DC., (in press). GILBERT, D. & SUMMERHAYES,C.P. 1981. Distribution of organic matter in sediments along the California continental margin. In: Yeats, R.S., Haq, B.U. et al., lnit. Repts. DSDP, 63. US Govt. Print. Off., Washington, DC. 757-61. -1982. Organic facies and hydrocarbon potential in the Gulf of California. In: Curray, J.R., Moore, D.S. et al., Init. Repts. DSDP, 64. US Govt. Print. Off., Washington, DC. 865-70. GLENN, C.R. & ARTHUR, M.A. 1983. Sedimentary and geochemical indicators of productivity and oxygen contents in modern and ancient basins, I. The Holocene Black Sea as the "type" anoxic basin. Chemical Geol., (in press). GOLDHABER, M.B. & KAPLAN, I.R. 1974. The sulfur cycle. In: Goldberg, E.D. (ed.), The Sea, 5, Marine Chemistry. John Wiley, New York. 569-655. GRASSHOFF,K. 1975. The hydrochemistry of landlocked basins and fjords. In: Riley, J.P. & Chester, R. (eds), Chemical Oceanography, 2, 2nd Edition. Academic Press, New York. 456-597. HABIB, D. 1979. Sedimentary origin of North Atlantic Cretaceous palynofacies. In: Talwani, M., Hay, W. -
-
Deposition of organic-carbon-rich sediment
557
continents to oceans: J. geol. Soc., London, 137, 171-88. KELTS, K. & ARTHUR, M.A. 1981. Turbidites after ten years of Deep Sea Drilling-wringing out the mop?. In: Warme, J.E., Douglas, R.G. & Winterer, E.L. (eds), The Deep Sea Drilling Project: A Decade o f Progress. Soc. econ. PaleD. Min. Spec. Pub. 32, 91-1127. KENDR1CK, J.W. 1979. Geochemical studies of black clays from Leg 43, Deep Sea Drilling Project. In: Tucholke, B. Vogt, P. et al., Init. Repts. DSDP, 43. US Govt. Print. Off., Washington, DC. 633--42. KENNETT, J.P. 1977. Cenozoic evolution of Antarctic glaciation, the circum-Antarctic Ocean, and their impact on global paleoceanography. J. geophys. Res., 82, 3843-60. KIDD, R.B., C1TA, M.B. & RYAN, W.B.F. 1978. Stratigraphy of Eastern Mediterranean sapropel sequences recovered during DSDP Leg 42A and their paleoenvironmental significance. In: Hsfi, K., Montadert, L. et al., Init. Repts. DSDP, 42, part 1. US Govt. Print. Off., Washington, DC. 421-43. KOBLENTZ-MISHKE, O.J., VOLKOVINSKY, V.V. & KABANOVA, J.G. 1970. Plankton primary production of the world ocean. In: Wooster, W.S. (ed.), Scientific Exploration of the South Pacific. National Academy of Science, Washington, DC. 183-93. LANCELOT, Y., HATHAWAY, J.C. & HOLLISTER, C.D. 1972. Lithology of sediments from the western North Atlantic. Leg 1l, Deep Sea Drilling Project. In: Hollister, C.D., Ewing, J.I. et al., Init. Repts. DSDP, l l . US Govt. Print. Off., Washington, DC. 901-50. - - , SEIBOLD,E. et al. 1977. Init. Repts. DSDP, 41. US Govt. Print. Off., Washington, DC. 1259 pp. - - , WINTERER,E.L. et al. 1980. Init. Repts. DSD P, 50. US Govt. Print. Off., Washington, DC. 868 pp. LEVENTHAL, J.S. 1983. Relationship between organic carbon and sulfide sulfur in euxinic sediments. Geochim. cosmochim. Acta, 46, 133-8. MCCAVE, I.N. 1979a. Diagnosis of turbidites at Sites 386 and 387 by particle-counter size analysis of the silt (2-4/~m) fraction. In: Tucholke, B., Vogt, P.R. et al., Init. Repts. DSDP, 43. US Govt. Print. Off., Washington, DC. 395-405. - 1979b. Depositional features of organic black and green mudstones at DSDP Sites 386 and 387, 159-80. western North Atlantic. In: Tucholke, B.E., Vogt, JANSA, L.F., GARONER, J.V. & DEAN, W.E. 1977. P.R. et aL, lnit. Repts. DSDP, 43. US Govt. Print. Mesozoic sequences of the central North Atlantic. Off., Washington, DC. 411-6. In: Lancelot, Y., Seibold, E. et al., Init. Repts. McCoY, F.W. & ZIMMERMAN,H.B. 1977. A history of DSDP, 41. US Govt. Print. Off., Washington, DC. sediment lithofacies in the South Atlantic Ocean. In 991-1031. Perch-Nielson, K., Supko, P. et al., Init. Repts. , ENOS, P., TUCHOLKE, B.E., GRADSTEIN, F.M. & DSDP, 39. US Govt. Print. Off., Washington, DC. SHERIDAN, R.E. 1979. Mesozoic-Cenozoic sedimen1047-79. tary formations of the North Atlantic basin; western North Atlantic. In: Talwani, M., Hay, W.W. & MEYERS, P.A., BRASSELL, S.C. & HUE, A.Y. 1984. Geochemistry of organic carbon in South Atlantic Ryan, W.B.F. (eds), Deep Drilling Results in the sediments from Deep Sea Drilling Project Leg 75. In: Atlantic Ocean: Continental Margins and PaleoenHay, W.W., Sibuet, J.-C. et al., Init. Repts. DSDP, vironment. Maurice Ewing Series 3, Am. Geophys. 75. US Govt. Print. Off., Washington, DC. (in Union, Washington DC. 1-57. press). JENKINS, J.A. & WILLIAMS, D.F. 1983. A model for Pleistocene anoxic events in the Eastern Mediter- MOBERLY, R.L. & SCHLANGER, S.O. et al. 1983. The Mesozoic superocean. Nature, 302, 381. ranean. Bull. geol. Soc. Am., (in press). MOORE, J.C., JR., VAN ANDEL., T.H., SANCETTA,C. & JENKYNS, H.C. 1980. Cretaceous anoxic events: from
& Ryan, W.B.F. (eds), Deep Drilling Results in the Atlantic Ocean." Continental Margins and Paleoenvironments. Maurice Ewing Series 3, Am. Geophys. Union, Washington DC. 420-37. 1982. Sedimentary supply origin of Cretaceous black shales. In: Schlanger, S.O. & Cita, M.B. (eds), Nature and Origin of Cretaceous Carbon-rich Facies. Academic Press, London. 113-28. HALLAM, A. & BRADSHAW, M.J. 1979. Bituminous shales and oolitic ironstones as indicators of transgressions and regressions. J. geol. Soc. Lond., 136, 157-64. HAY, W.W., S1BUET, J.-C., et al. 1982. Sedimentation and accumulation of organic carbon in the Angola Basin and on Walvis Ridge: preliminary results of Deep Sea Drilling Project Leg 75. Bull. geol. Soc. Am., 93, 1038-50. HEATH, G.R., MOORE, T.C. & DAUPHIN, J.P. 1977. Organic carbon in deep-sea sediments. In: Anderson, N.R. & Malahoff, A., (eds), The Fate of Fossil Fuel C02 in the Oceans. Plenum Press, New York. 605-25. HsO, K.J., MONTADERT, L. et al. 1978. Init. Repts. DSDP, 42A. US Govt. Print. Off., Washington, DC. 1249 pp. HULSEMANN, J. & EMERY, K.O. 1961. Stratification in recent sediments of Santa Barbara Basin as controlled by organisms and water character. J. Geol., 69, 279-90. HUNTSMAN, S.A. & BARBER, R.T. 1977. Primary production off northwest Africa: the relationship to wind and nutrient conditions. Deep Sea Res., 24, 25-33. HUYER, A. 1976. A comparison of upwelling events in two locations: Oregon and northwest Africa. J. mar. Res., 34, 531-46. IBACIq, L.E.J. 1982. Relationship between sedimentation rate and total organic carbon content in ancient marine sediments. Bull. Am. Ass. Petrol. Geol, 66, 170-88. INGLE, J.C. 1981. Origin of Neogene diatomites around the North Pacific rim. In: Garrison, R.E., Douglas, R.G., Pisciotto, K.E., Isaacs, C.M. & Ingle, J.C. (eds), The Monterey Formation and Related Siliceous Rocks of California. Pacific Section, Soc. econ. PaleD. Min., Spec. Pub.
558
M.A. Arthur, W.E. Dean and D.A.V. Stow
PISIAS, N. 1978. Cenozoic hiatuses in pelagic sediments: Micropaleontology, 14, 113-38. MORRIS, K.A. 1979. A classification of Jurassic marine shale sequences: an example from the Toarcian (Lower Jurassic) of Great Britain: Palaeogeogr., Palaeoclimatol., Palaeoecol., 26, 117-20. MULLER, P.S. & SUESS,E. 1979. Productivity, sedimentation rate, and sedimentary organic carbon content in the oceans. Deep Sea Res., 26, 1347-62. NARDIN, T.R., EDWARDS,B.D. & GORSLINE,D.S. 1979. Santa Cruz Basin, California Borderland: Dominance of slope processes in basin sedimentation. In: Doyle, L.J. & Pilkey, O.H., Jr. (eds), Geology of Continental Slopes. Soc. econ. PaleD. Min. Spec. Pub. 27, 209-22. NATLAND, J.H. 1978. Composition, provenance, and diagenesis of Cretaceous clastic sediments drilled on the Atlantic continental rise off southern Africa, DSDP Site 361. In: Bolli, H., Ryan, W.B.F. et al., Init. Repts. DSDP, 40. US Govt. Print. Off., Washington, DC. 1025-62. NESTEROFF,W.D. 1973. Petrography and mineralogy of sapropels. In: Ryan, W.B.F., Hsii, K.J. et al., lnit. Repts. DSDP, 13. US Govt. Print. Off., Washington, DC. 713-20. PARRISH, J.T. & CURTIS, R.L. 1982. Atmospheric circulation, upwelling, and organic-rich rocks in the Mesozoic and Cenozoic eras. Palaeogeogr., Palaeoclimatol., Palaeoecol., 40, 31-66. , ZIEGLER, A.M. & SCOTESE, C.R. 1982. Rainfall patterns and the distribution of coals and evaporites in the Mesozoic and Cenozoic. Palaeogeogr., Palaeoclimatol., Palaeoecol., 40, 67-10 I. PETERS, K.E. & StMONEIT, B.R.T. 1982. Rock-eval pyrolysis of Quaternary sediments from DSDPIPOD Leg 64, Sites 479 and 480, Gulf of California. In: Curray, J.R., Moore, D.S. et al., Init. Repts. DSDP, 64. US Govt. Print. Off., Washington, DC. 925-32. PHLEGER, F.P. & SOUTAR, A. 1973. Production of benthic foraminifera in three east Pacific oxygen minima. Micropaleontology, 19, 110-5. PISCIOTTO, K.A. & GARRISON, R.E. 1981. Lithofacies and depositional environments of the Monterey Formation, California. In: Garrison, R.E., Douglas, R.G., Pisciotto, K.E., Isaacs, C.M. & Ingle, J.C. (eds), The Monterey Formation and Related Siliceous Rocks of California. Soc. econ. PaleD. Min. Spec. Pub., 97-122. RHOADS, D.C. & MORSE, J.W. 1971. Evolutionary and ecological significance of oxygen-deficient marine basins. Lethaia, 4, 413-28. RHODES, D.C. 1973. The influence of bottom-feeding benthos on water turbidity and nutrient recycling. Am. J. Sci., 273, 1-22. RITTENBERG, S.C., EMERY, K.O. & ORR, W.L. 1953. Regeneration of nutrients in sediments of marine basins. Deep Sea Res., 3, 23-45. Ross, D.A., NEPROCHNOV, S. et al. 1978. Init. Repts. DSDP, 42, part 2. US Govt. Print. Off., Washington, DC. 1244 pp. ROSSIGNOL-STRICK, M., NESTEROFF, W., OLIVE, P. & VERGRNAUD-GRAZZINI,C. 1982. After the deluge:
Mediterranean stagnation and sapropel formation. Nature, 295, 1105-1110. ROTH, P.H. 1978. Cretaceous nannoplankton biostratigraphy and oceanography of the northwestern Atlantic Ocean. In: Benson, W.E., Sheridan, R.E. et al., Init. Repts. DSDP, 44. US Govt. Print. Off., Washington, DC. 731-52. RYAN, C.B.F. 1972. Stratigraphy of late Quaternary sediments in the eastern Mediterranean. In: Stanley, D.J. (ed.), The Mediterranean Sea--A Natural Sedimentation Laboratory. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 149-69. RYAN, C.B.F. & CITA, M.B. 1977. Ignorance concerning episodes of ocean-wide stagnation. Mar. Geol., 23, 197-215. SAVRDA, C.E., BOTTJER, D.J., & GORSLINE,D.S. 1982. Refinement of benthic basin biofacies models: biogenic sedimentary structures and macrofauna from Santa Monica Basin, California Continental Borderland. Geol. Soc. Am. Prog. Abst., 14, 608. SCHLANGER, S.O. • JENKYNS, H.C. 1976. Cretaceous oceanic anoxic events-causes and consequences. Geologic en Mijnbouw, 55, 179-84. SCHOLLE, P.A. & ARTHUR, M.A. 1980. Carbon isotopic fluctuations in pelagic limestones; potential stratigraphic and petroleum exploration tool. Bull. Am. Ass. Petrol. Geol., 64, 67-87. SCHRADER, H. et al. 1980. Laminated diatomaceous sediments from the Guaymas Basin Slope (Central Gulf of California): 250,000-year climate record. Science, 207, 1207-9. SHANKS, A.L. & TRENT, J.D. 1980. Marine snow: sinking rates and potential role in vertical flux. Deep Sea Res., 27, 137-44. SHERIDAN, R.E., GRADSTEIN, F.M. et al. 1982. Early history of the Atlantic Ocean and gas hydrates on the Blake Outer Ridge: results of the Deep Sea Drilling Project Leg 76. Bull. geol. Soc. Am., 93, 876-85. SHIMKUS, K.M. & TRIMONIS, E.S. 1974. Modern sedimentation in the Black Sea. In: Degens, E.T. & Ross, D.A. (eds), The Black Sea: Geology, Chemistry, and Biology. Am. Ass. Petrol. Geol. Mem. 20, 249-78. SHOR, A.N. & POORE, R.Z. 1978. Bottom currents and ice rafting in the North Atlantic: interpretation of Neogene depositional environments of Leg 49 cones. In: Luydendyk, B.P., Cann, J.R. et al. Init. Repts. DSDP, 49, US Govt. Print. Off., Washington, DC. SIESSER, W.G. 1980. Late Miocene origin of the Benguela upwelling system off northern Namibia. Science, 208, 183-285. SOUTAR. A., JOHNSON, S.R. & BAUMGARTNER, T.R. 1981. In search of modern analogs to the Monterey Formation. In: Garrison, R.E., Douglas, R.G., Pisciotto, K.E., Isaacs, C.M. & Ingle, J.C. (eds.), The Monterey Formation and Related Siliceous Rocks of California. Soc. econ. Paleo. Min., Pacific Section, Spec. Pub., 123-48. SOUTHAM, J.R., PETERSON,W.H. & BRASS,G.W. 1982. Dynamics of anoxia. Palaeogeogr., Palaeoclimatol., Palaeoecol., 40, 183-98.
Deposition of organic-carbon-rich sediment
-
Sxow, D.A.V. 1984. Turbidite facies, associations and sequences in the Angola Basin. In: Hay, W.W., Sibuet, J.-C. et al., Init. Repts. DSDP, 75. US Govt. Print. Off., Washington, DC., (in press). & DEAN, W.E. 1984. Middle Cretaceous Black Shales at Site 530 in the southeastern Angola Basin. In: Hay, W.W., Sibuet, J.-C. et al., lnit. Repts. DSDP, 75. US Govt. Print. Off., Washington, DC., (in press). SUESS, E. 1980. Particulate organic carbon flux in the oceans-Surface productivity and oxygen utilization. Nature, 288, 260-3. SUMMERHAYES,C.P. 1981a. Oceanographic controls on organic matter in the Miocene Monterey Formation, offshore California. In: Garrison, R.E., Douglas, R.G., Pisciotto, K.E., Isaacs, C.M. & Ingle, J.C. (eds), The Monterey Formation and Related Siliceous Rocks. Pacific Section, Soc. econ. Paleo. Min. Spec. Pub., 213-9. 1981b. Organic facies of middle Cretaceous black shales in the deep North Atlantic. Bull. Am. Ass. Petrol. Geol., 65, 2364-80. --, BORNHOLD, B.D. & EMBLEY, R.W. 1979. Surficial slides and slumps on the continental slope and rise off South West Africa: A reconnaissance study. Marine Geol., 31,265-77. --, MASRAN, P. 1983. Organic facies of Cretaceous and Jurassic sediments from DSDP Site 534 in The Blake Bahama Basin, Western North Atlantic. In: Sheridan, R.E., Gradstein, F. et al., Init. Repts. DSDP, 76. US Govt. Print. Off., Washington, DC., 469-81. THIEDE, J., DEAN, W.E. & CLAYPOOL, G.E. 1982. Oxygen deficient depositional paleoenvironments in the mid-Cretaceous tropical and subtropical central Pacific Ocean. In: Schlanger, S.O. & Cita, M.B. (eds), Nature and Origin of Cretaceous Organic Carbon-Rich Facies. Academic Press, London. 55-78. -t~ VAN ANDEL, T.H. 1977. The paleoenvironment of anaerobic sediments in the late Mesozoic South Atlantic Ocean. Earth planet. Sci. Lett., 33, 301-9. THIERSTEIN, H.R. 1979. Paleoceanographic implications of organic carbon and carbonate distribution in Mesozoic deep sea sediments. In: Talwani, M., Hay, W.W. & Ryan, W.B.F. (eds), Deep Drilling Results in the Atlantic Ocean." Continental Margins and Paleoenvironment. Maurice Ewing Series 3, Am. Geophys Union, Washington, DC. 249-74. & BERGER, W.H. 1978. Injection events in Earth history. Nature, 276, 461-6. THUNELL, R.C., WILLIAMS, D.F. KENNETT, J.P. 1977. Late Quaternary paleoclimatology stratigraphy and sapropel history in eastern Mediterranean deep-sea sediments. Mar. Micropaleont., 2, 371-88. TISSOT, B. 1979. Effects on prolific petroleum source rocks and major coal deposits caused by sealevel changes. Nature, 277, 463-5. , DEMAISON, G., MASSON, P., DELTEIL, J.R. & COMBAZ,A. 1980. Paleoenvironment and petroleum potential of middle Cretaceous black shales in -
-
-
-
-
559
Atlantic basins. Bull. Am. Ass. Petrol. Geol., 64, 2051-63. - - , DEROO,G. & HERBIN,J.P. 1979. Organic matter in Cretaceous sediments of the North Atlantic: contributions to sedimentology and paleogeography. In: Talwani, M., Hay, W.W. & Ryan, W.B.F. (eds), Deep Drilling in the Atlantic Ocean: Continental Margins and Paleoenvironment. Maurice Ewing Series 3, Am. Geophys. Union, Washington, DC. 362-74. --, DURAND, B., ESPITALII~ • COMBAZ, A. 1974. Influence of nature and diagenesis of organic matter in formation of petroleum. Bull. Am. Ass. Petrol. Geol., 58, 499-506. & WELTE, D.H. 1978. Petroleum Formation and Occurrences. Springer-Verlag, New York. 538 PP. TOTH, D.J. & LERMAN,A. 1977. Organic matter reactivity and sedimentation rates in the ocean. Am. J. Sci., 277, 465-85. TtJCHOLKE, B.E. & VOGT, P.R. 1979. Western North Atlantic: sedimentary evolution and aspects of tectonic history. In: Tucholke, B.E., Vogt, P.R. et al., Init. Repts. DSDP, 43. US Govt. Print. Off., Washington, DC. 791-825. VON DER DICK, H., RULLKOTTER, J. & WELTE, D.H. 1983. Content, type, and thermal evolution of organic matter in sediments from the eastern Falkland Plateau, Deep Sea Drilling Project, Leg 71, In: Ludwig, W.J., Krasheninnikov, V. et al., Init. Repts. DSDP, 71. US Govt. Print. Off., Washington, DC. 1015-1132. VON RAD, U., CEPEK, P., YON STACKELBERG, U., WISSMAN, G. & ZOBEL, B. 1979. Cretaceous and Tertiary sediments from the northwest African slope (dredges and cores supplementing DSDP results): Marine Geol., 29, 213-312. WEISSERT, H. 1981. The environment of deposition of black shales in the Early Cretaceous: an ongoing controversy. In: Warme, J.E., Douglas, R.G. & Winterer, E.L. (eds), The Deep Sea Drilling Project: a Decade of Progress. Soc. econ. Paleo. Min. Spec. Pub. 32, 547-60. - - , MCKENZIE,J. & HOCHULI,P. 1979. Cyclic anoxic events in the Early Cretaceous Tethys Ocean. Geology, 7, 147-51. WELTE, D.H., CORNFORD,G. & RULLKOTTER,J. 1979. Hydrocarbon source rocks in deep sea sediments. Proceedings Offshore Technology Conference, Houston. 457-61. WIEDMANN, J., BUTT, A. d~. EINSELE, G. 1978. Vergleich von Marokkanischen Kreide-Kustenuafschlussen und Tiefsee bohrungen (DSDP): Stratigraphie, Paleoenvironment and Subsidenz, an einem passiven Kontinentalrand. Geologisches Rundschau, 67, 454-508. WILDE, P. & BERRY,W.B.N. 1982. Progressive ventilation of the oceans-potential for return to anoxic conditions in the post-Paleozoic. In: Schlanger, S.O. & Cita, M.B. (eds), Nature and Origin of Cretaceous Carbon-Rich Facies. Academic Press, London. 209-24. WILLIAMS,D.F., THUNELL,R.C. & KENNETT,J.P. 1978. Periodic fresh-water flooding and stagnation of the
560
M.A. Arthur, W.E. Dean and D.A.V. Stow
eastern Mediterranean Sea during the late Quaternary. Science, 201, 252--4.
WYRTKI,K. 1962. The oxygen minima in relation to ocean circulation. Deep Sea Res., 9, 11-23.
M.A. ARTHUR1, Department of Geology, University of South Carolina, Columbia, South Carolina 29208, USA. W.E. DENY,US Geological Survey, Denver, Colorado, 80225, USA. D.A.V. STOW, Grant Institute of Geology, University of Edinburgh, West Mains Road, Edinburgh, Scotland EH9 3JW. 1 Present address: Graduate School of Oceanography University of Rhode Island, Narragansett, R.I. 02882, USA.
Plasticity and compaction characteristics of the Quaternary sediments penetrated on the Guatemalan Transect DSDP Leg 67 R.W. Faas S U M MARY: The plasticity characteristics of the Quaternary sediments of the Guatemalan continental margin were determined from five sites drilled on Leg 67 of the Deep Sea Drilling Project. Sixty-four samples were analysed from various marine environments, including the Cocos oceanic plate, Middle America Trench, and the trench lower slope to mid-slope of the Guatemalan continental slope. Quaternary sediments on the Guatemalan continental slope are generally mixed siliceous and calcareous hemipelagic silty-clays and clayey-silts, frequently interbedded with volcanic ash. In the Middle America Trench, similar sediments alternate with sandy turbidites. Siliceous hemipelagic silty-clays and clayey-silts form the uppermost 170 metres of the Cocos Plate. The sediments are generally classified as 'organic clays of medium to high plasticity, containing micaceous sands and silts,' with 14~oclassed as 'inorganic clays of medium to high plasticity.' Plasticity (measured by the Plasticity Index, It,) decreases downslope to the trench, then increases sharply in the Cocos Plate sediments. Organic matter is greatest highest on the slope (Site 496) and decreases seaward, averaging 1.14~o at Site 495 on the Cocos Plate. Sediment compaction varies, according to position along the traverse and to the dominant sediment type. Greatest compaction (44.5~) occurs at Site 494 at the base of the lower trench slope. Least compaction (18.6%) is found at Site 495 on the Cocos Plate. Compaction of the trench and mid-slope sediments ranges between 31-38.5%. High sedimentation rates in Quaternary sediments are due, in part, to sediment gravity-flows which depend upon rheological properties, i.e. sediment plasticity. Mudflows and cohesive debris-flows appear to be significant downslope transport mechanisms in these highly plastic sediments. The goals of the Guatemalan Transect were to study a convergent margin where accretion and imbrication may have been continuous during most of Tertiary time; to strengthen the tie between onshore and offshore geology; and to study modern sediment sequences in the Middle America Trench and continental slope. The objective of this research was to evaluate the usefulness of the Atterberg limits of the Quaternary sediments in predicting modes of downslope subaqueous sediment transport; and to determine the gravitational compaction characteristics of the sediments accumulating on the Guatemalan continental slope, Middle America Trench, and on the Cocos oceanic plate. (Fig. 1). Gravitational compaction is defined as the expulsion of pore fluids and pore volume thereby decreases in a sedimentary column as a result of normal and shear-compressional stresses due to the overburden load (Riecke I I I & Chilingarian 1974).
Methods Quaternary sediments---general The Quaternary sediments of the Guatemalan continental margin form a seaward thinning
mantle of hemipelagic muds, containing significant quantities of siliceous and calcareous fossil remains. Sediment thickness varies from greater than 200 m (Site 4 9 6 ~ u p p e r trench slope), to approximately 100 m (Site 498--lower trench slope). Within the Middle America Trench, sediments are interbedded pelagic diatomaceous muds and fine sand, and muddy turbidites. Sediment thickness varies, ranging from about 80 m (Site 500) to 120 m (Site 499). On the oceanic plate, diatomaceous hemipelagic muds thin to about 60 m at Site 495. Sedimentation rates vary from 300 m/my in the trench (Site 499) to 37 m/my on the oceanic plate (Site 495). The greatest accumulation occurs in the trench as sediment is transferred to it from higher elevations on the slope. Dominant clay minerals are illite, montmorillonite, and kaolinite; the latter two are most abundant. Illite is missing from Site 495 on the ocean plate. As more interest is focused on active continental margins and their deposits, it is obvious that many are areas of downslope sediment transport. Such transport may be rapid and episodic, slow and pervasive, or some combination of the two. Lowe (1979) has indicated the importance of rheology in distinguishing the mass transport
563
564
R. W. Faas
~,~
........... ~,..-i-':i;..
91~oo' '.. .'
9 '
9 .
9
. .'.
9o~oo' 9
San
/,
.
.
~
.
. .
.
I
/
..,. ~
' '.'.::::!....;
""'"--
I I
,----
....
I /
.
/"
,,'" I
Jose'
,..
-
"'"",i7.'.;'.':....
"
" ' ' : ' " : " {.J..,7;;:.7.-..!:.... ;...
I
I I
\ \
N I I /
\\
I I I
I
/
13
:.~...'....
\
~ 00'
0o 03
Middle
America
Trench
Axis
I~ 03
~D 03
FIG. 1. Location map of Guatemalan continental margin and location of all D S D P Leg 67 drill sites.
565
Plasticity and compaction characteristics mechanisms as fluidal flows or debris-flows by their fluid and plastic behaviour respectively. If this concept is valid, the initial step in the study of submarine mass transport mechanisms should be the determination of the expected rheological behaviour of the sediments. This can be readily accomplished by analysing the sediments for their Atterberg limits.
Atterberg limits Atterberg limits (Casagrande 1948) are based on the concept that a sediment is a two-phase system of inorganic particles and water. The rheological behaviour state of such a mixture changes from fluid to plastic to solid as the mixture loses water (Fig. 2). The water content at which a sediment/ water mixture ceases to behave as a liquid is its liquid limit (Wc). Similarly, plastic behaviour of such a system ceases at its plastic limit (Wp) and it passes through a semi-solid state, determined by its shrinkage limit (ws). Each limit is the empirical measure of the water content of the mixture at that point where its behaviour changes from one state to another. The plasticity index (Ix) is a measure of the range of water content through which the mixture exhibits plastic behaviour
Analytical procedure The primary source of data for this study was provided by wet bulk density profiles of the cores retrieved during Leg 67 drilling. Wet bulk density is measured continuously on all cores, using the Gamma Ray Attenuation Porosity Evaluator (GRAPE) (Evans & Cotterell 1970). Each core is slowly passed through the GRAPE and its wet bulk density is recorded. Prior to each analysis, the GRAPE is calibrated to a distilled water (1.00 Mg/m 3) and aluminum (2.60 Mg/m 3) standard. A description of the GRAPE method and results has been presented by Boyce (1975). Atterberg limit tests were performed on board the Glomar Challenger immediately after the cores were sampled. A sediment sample was placed in a preweighed aluminum moisture dish, weighed on a triple-beam balance, and placed in the drying oven at 110~ for drying overnight. At the same time, liquid limits were determined. Half of the sample was used to determine its water content; the other half was used for the plastic limit analysis. Liquid limits were performed according to the simplified procedure described by Lambe (1951) and calculated as follows
Z~=wL-w,,
0.121
w
N
CONSI5TENCY 5 TA TE5 LIQUID wL Liquid limit
PLASTIC
Wp Plas'fic limit SEMI-SOLID wS Shrinkage limit
SOLI O
FIG. 2. Sediment consistency states and relation to Atterberg limits.
where wN=the water content of the sediment sample which closes the groove in N blows in the standard liquid limit cup. Values of liquid limit (Table 1) represent an average of three separate determinations of the same sample. The plastic limit is defined as the water content of the remoulded sediment as it passes from a plastic to a brittle condition. The test involves rolling the sediment over a frosted glass plate and forming a thread 1/16" in diameter. When the thread no longer holds together, the plastic limit has been reached, and the water content is determined. The procedure is fully discussed in Lambe (1951). Samples were also taken for organic carbon analysis and particle size distribution. Some of the carbon analyses were performed on ship, others were done later at the University of Oklahoma. All were analysed in a Leco combustion apparatus. Grain-size analysis was carried out at the DSDP laboratory in La Jolla, California. Shear strength was measured perpendicular to the bedding at one-metre intervals, on the split core with the hand-held CL-600 Torvane. The Torvane was rotated at a rate designed to reach failure in about 10 seconds with constant loading.
566
R. W. Faas
.+~
.~o 0 0 0 0 0 0 0 0
9,,,.+
-
.~._~..~._~
~ ~
0 0 0 0
000000000
~ .=
-~+.~.~
~
oo
lleqopue ezooe~=uoq,eoJ
pnw sno~oewo)e!a
/
•l••qo
@
.=-
II
II
~. o
azoo eieuoq~e~
pn~ snoeoemme!a
IIIt-qOUai/ }
~lleq0
8
c~
r
to
s~.uatu!pesII!pqoueJ/
_1
I
pnm o!6elad!UJaH o
sJ.uelu!pesiI!}-qoueJ.L
lPnw o,,elad!'",H I
c~ o I
pnw i!.otouueu pue ,,,m.e!Q
I o~
pues pue pnlN
i
.n,,,,.o,ouu.u.u.,-o.,o
O
pn,,, I!SSojouueupue ,-m.e!a
]
pnm ^pu.s
~
~
I
r
....
. , . I pmu ApueS
pmu l!SSOjOUUeUpue wole!Q I
I
I
I
I
. i
~
1
(w) ql.deO mo~oq-qn$
l
I
I
I
576 :s 9
4
R. W. Faas
3"0 2-5 2"0 1"5 1.0 0-5
9 ---..-.._. 9 ,-...._.__
I00
.o
80
0
9
!
|
/.7
I 9
/J
i
/
/o
II
/
60
$ (66"4)
S/j
9
(81.6)
9 (68.1)
40
(57"7)
20
(58-5)
I 496
1 494 + 494A
I 497
Slope
Inner trench wall
I 499 4500
I 495
Trench axis
Ocean plate
F16.9. Plasticity index and Total Organic Content (TOC) profile along the Guatemalan transect. Site 495, its elevated plasticity may reflect this lack of illite. This, plus the fact that Site 495 averages almost 60% clay-sized particles (Table 1) may be the reason why Site 495 sediments possess such high plasticity. Site 495 is located on a horst block approximately 2000 m above the floor of the Middle America Trench. It shows overcompaction in the upper 50 m and undercompaction from 50 m to 200 m. This may reflect a direct lithologic control inasmuch as Pleistocene hemipelagic silty muds extend from the mudline to 57 m sub-bottom, followed by Pliocene and lower middle Miocene water-rich diatomaceous silty mud to 180 m sub-bottom.
Summary and conclusions The study of the physical properties of sediment cores obtained by rotary drilling techniques is always hampered by the obvious problems associated with core disturbance, particularly in the younger, unlithified sedimentary section. H o w -
ever, it is possible to obtain useful information about depositional processes through the use of Atterberg limits which yield information on the rheological characteristics of the sediments while severely disturbed (remoulded). It is also possible to evaluate the various states of compaction in these disturbed sediments. A technique which utilizes the relationship between overburden pressure versus depth for actual marine sediments and compares the compaction curves with theoretical compaction curves for various types of normally compacting, commonly occurring marine sediments has proven to be useful in assessing the compaction states of the sediments drilled on the Guatemalan Transect on DSDP Leg 67. This evaluation is qualitative and should not be considered equivalent to consolidation tests performed on undisturbed cores of high quality. Continental slope sediments show the same degree of compaction as those in the Middle America Trench. However, they tend to follow a compaction curve intermediate between pelagic clays and diatom ooze. This pattern may arise from the presence of bacterially generated methane, or from decomposition of a gas hydrate zone
Plasticity and compaction characteristics of unknown thickness which occurs at about 300 m (von Huene et al. 1980). The overall effect is to reduce the lithostatic load in the upper 300 m and create zones of undercompaction in the relatively impermeable fine-grained sediments. The trench sediments follow a compaction curve intermediate between pelagic clays and terrigenous sediments. This results from the mixing of pelagic sediments with terrestrially-derived turbidites. The compaction behaviour of the Cocos Plate sediments is controlled by sedimentation rate, sediment type, and tectonic events. Slowly accumulating hemipelagic sediments appear overcompacted. The water-rich siliceous oozes below the impermeable hemipelagic clays show relative undercompaction due to impeded drainage. Plasticity of the Quaternary sediments decreases from the upper continental slope to the lower trench slope and reaches its lowest value in the trench-fill turbidites within the Middle America Trench. Plasticity increases significantly in the
577
hemipelagic sediments of the oceanic plate. Sedimentation within the trench is high (300 m/my) and indicates that significant downslope massmovement is occurring. The high plasticity of the upper slope and ocean plate sediments suggests that the dominant processes are mudflows and cohesive debris flows as these processes depend upon plastic internal mechanical behaviour and can move long distances over very low angle slopes. ACKNOWLEDGEMENTS: I thank Bill Harrison and Joe Curiale for providing organic matter analyses of many of the samples. Grain-size and additional organic matter analyses were done at the DSDP Laboratory in La Jolla, California. Various drafts of the manuscript were reviewed by Richard Bennett, Kate Moran, Jim Booth and Paul Carlson and I have benefited from their comments. Finally, I thank the National Science Foundation for its continued support of the Deep Sea Drilling Program.
References BISHOP, R.S. 1979. Calculated compaction states of thick abnormally pressured shales. Bull. Am. Ass. Petrol. Geol., 63, 918-33. BOYCE, R.E. 1975. Definitions and laboratory techniques of compressional sound velocity parameters and wet water content, wet-bulk density, and porosity parameters by gravimetric and gamma ray attenuation techniques. In: S.O. Schlanger & E.D. Jackson et al., Init. Repts. DSDP, 33. US Govt. Print. Off., Washington, DC. CASAGRANDE,A. 1948. Classification and identification of soils. Trans. Am. Soc. cir. Eng., 113, 901-31. EVANS, H.B. & COTTERELL, C.H. 1970. Gamma ray attenuation density scanner. In: M.N.A. Peterson et al., Init. Repts. DSDP, 2. US Govt. Print. Off., Washington, DC. FAAS, R.W. 1982. Gravitational compaction patterns determined from sediment cores on the Guatemalan Transect: Continental slope, Middle America Trench and Cocos Plate, on D.S.D.P. Leg 67. In: J. Aubouin., R. von Huene, et al., Init. Repts. DSDP, 67, US Govt. Print. Off., Washington, DC. 617-38. FIELD, M.E. 1981. Sediment mass-transport in basins: Controls and patterns. In: R.G. Douglas, Colburn, I.P. & Gorsline D.S. (eds), Depositional System of Active Continental Margin Basins. Pacific Section, Soc. econ. Paleo. Min., Short course notes, San Francisco. 61-83. HAMILTON,E.L. 1960. Ocean basin ages and amounts of
original sediments. J. sed. Petrol., 30, 3, 370-9. 1976. Variations of density and porosity with depth in deep sea sediments. J. sed. Petrol., 46, 2, 280-300. HEIN, F.J. & GORSLINE,D.S. 1981. Geotechnical aspects of fine-grained mass flow deposits: California Continental Borderland. Geo-Marine Letters, 1, 1-5. LAMBE, T.W. 1951. Soil Testing for Engineers. John Wiley, New York. LOWE,D.R. 1979. Sediment gravity flows: their classification and some problems of application to natural flows and deposits. Soc. econ. Paleo. Min. Spec. Pub. 27, 75-82. RIEKE, III, H.H. & CHILLINGARIANG.V. 1974. Compaction of Argillaceous Sediments. Elsevier Scientific Publishing Co. 424 pp. SEED, H.B., WOODWARD,R.J. & LUNDGREN,R. 1964. Clay mineralogical aspects of the Atterberg limits. J. Soil Mech. Div., A.S.C.E., 90, SM4, 107-31. YON HUENE, R., AUBOUIN,J. et al. 1980. Leg 67: The Deep Sea Drilling Project Mid-America Trench transect off Guatemala. Bull. geol. Soc. Am., 91, l, 421-32. WAGNER, A.A. 1957. The use of the Unified Soil Classification system by the Bureau of Reclamation. Proceedings of the 4th International Conference on Soil Mechanics and Foundation Engineering (London), 1. 125 pp. --.
R.W. FAAS,Department of Geology, Lafayette College, Easton, Pennsylvania, USA.
Fabric of muds and shales: an overview C.F. Moon and C.W. Hurst S U M M A R Y : This contribution essentially constitutes an attempt to update a previous review of clay sediment microstructure written over a decade ago. Recent developments in this field are considered and a scheme has been developed in which the fabric of argillaceous sediments is traced from its inception during sedimentation and its subsequent modification at and below the depositional interface. It is held throughout that the primary control in the determination of clay fabric is the chemical environment that prevails during and after the deposition of the sediment. Consideration of the evidence from a variety of sources suggests that fissility in mudrocks is the result of an anisotropic microfabric in which clay aggregates are preferentially aligned. This microstructure is thought to result from the dispersing action of organic compounds common in certain sedimentary environments; a brief discussion of these compounds is included. Other factors are also taken into account which have been offered as the cause of differing mudrock fabrics.
The microfabric of clay sediments and mudrocks has for long been regarded as the 'b6te noire' of sedimentary geology. This is due, in no small part, to the inherent problems associated with their investigation and foremost among these is the extremely small size of clay particles. Clarke (1924) estimated that argillaceous sediments account for 80% of the global rock column and if this is so, it is rather surprising that more attention has not been paid to their microscopical characteristics (cf. sandstones). In the past twenty years, various authors have made concerted attempts to unravel the nature and origins of the fabric of clay sediments. The results of a number of thorough investigations from a variety of disciplines have revealed that certain common factors exist in the genesis of microfabric. In contrast, these same studies have cast doubt on a number of traditional assumptions that have previously been adhered to. There is little doubt that clay sediments are complex materials. Sub-microscopic particle size is merely one aspect of their nature; the complexity is greatly increased by the electrochemistry of clay particles. Their behaviour is difficult to predict in sedimentological terms and has largely been realized by deduction; it is only now that a pattern of behaviour of the origin and modification of clay microstructure is becoming clearer. This has been made much easier in recent years by use of the scanning electron microscope (SEM). A decade ago, one of the authors was responsible for a review (Moon 1972) which outlined the studies and conclusions that had emerged on the topic of clay microstructure to that date. This work is now sadly lacking since much relevant information has appeared in the interim period. The present contribution is an attempt to update the earlier paper and to include various observa-
tions made by the authors which have a direct bearing on more recent published discussions. This contribution traces the development of microfabric throughout the history of the sediment from its deposition through burial. Initially, the structure ofsedimenting clays is discussed and this is followed by a study of the fabric of uncompacted sediments. Compacted clays are then investigated, with particular attention to the controversial topic of fissility and a review of the effects of shear forces on argillaceous deposits. It will be observed throughout this discussion that much evidence for microstructural aspects has been supplied by sources in engineering, ceramics and pedology in addition to sedimentology. The authors have strived to assimilate the various approaches and synthesize the results into a coherent scheme from which clay microstructure can be seen to develop in geological environments.
The microfabric of sedimenting clay Under this heading are considered dilute clay suspensions in which clay mineral particles are settling, i.e. moving predominately under the influence of gravity towards the depositional interface. As a direct result of the practical difficulties involved in the observation of suspended clay particles, this topic is open to the widest speculation. Few authors have reported any specific findings but certain generalities do emerge from their work to date. Due to the extremely complex electrochemistry of clay particles and their immediate surroundings, the chemistry of the depositional environment is critical in deciding the spatial arrange579
580
C.F. Moon and C. W. Hurst
ment that the particles will assume during settling and deposition. The major chemical controls are pH, Eh and the presence or absence of surface active ions (Goodwin 1971). In addition, it would seem to be logical that the volume concentration of clay is an important factor. It is widely recognized that clay particles will form aggregates in electrolytic solutions by the mechanism of flocculation (Van Olphen 1963). Whitehouse et al. (1960) in an exhaustive study of clay deposition write: 'Clay minerals are not transported and do not settle as single solid grains or in irreversible states of aggregation in saline water. The settling unit is a thermodynamically reversible assembly of solid grains or threads, in association with occluded water and water of hydration, and has been termed a 'coacervate'.' This then, suggests that although flocs may initially form when clays are introduced into a saline environment, they may dissociate themselves by dispersion if the environment alters; a change in environment may occur at, or above the depositional interface. Whitehouse et al. (1960) conclude that the flocculation characteristics of a sediment depends on its mineralogy. However, this has subsequently been challenged by Kranck (1975) who attaches more importance to the particle size mode. Sherman (1953) has shown that flocs are able to form in freshwater environments. Probably Van Olphen (1963) has contributed more to the understanding of clay particle association than any other author and his hypothesis suggested a number of possible particle arrangements. All of these involve the edge-to-edge (EE), edge-to-face (EF) and face-to-face (FF) interactions of platy-clay crystals, and it appears that electrolyte concentration decides which of these modes will result. Subsequently, Rieke & Chilingarian (1974) have claimed to have observed four modes of aggregation in 'thick' clay-water suspensions. In two of these, their 'flocculateddispersed' (sic) and 'flocculated with EF contacts' models are not single particle structures but are packets of particles with FF interactions. Unfortunately, the authors do not explain h o w these have been observed, neither do they mention the volume concentration of clay which is critical in this context. An important step in the study of suspended clay aggregates was made by Schweitzer & Jennings (1971) who, by means of a complex lightscattering technique indirectly observed montmorillonite associations in dilute suspensions. In a detailed analysis of their results they conclude that the most likely arrangement is of the EF
cardhouse type which compares most favourably with one of Van Olphen's models. As a rider to this conclusion, Flegmann et al. (1969) have noted that at values above 7 the pH of a clay suspension may give rise to EE flocculation (pH of sea-water ~ 8). It does seem that the available evidence, although limited, suggests that single-plate, cardhouse-type structures are present in sedimenting clays under normal conditions ofpH and salinity. These structures have perhaps a dozen individual particles arranged in a simple EF array. In recent years, an interesting development has emerged which demonstrates that aggregation may not be solely due to electrochemical bonding. Ernissee & Abbott (1975), Zabawa (1978) and McCall (1979) have shown by electron microscopy that suspended sediments of clay size may be agglomerated by faunal secretions. The significance of this in microstructural studies cannot be underestimated.
The microfabric of freshly sedimented clay In this context the expression 'freshly sedimented' refers to that material which has immediately settled and has not been subjected to any degree of compaction by superincumbent sediments; it lies at the depositional interface (or mudline). The fabric of a sedimenting clay is difficult to observe directly and the same is true of a fresh sediment, but by studying material that has suffered only slight burial, certain inferences can be made and conclusions drawn by the extrapolation of results. Many models of clay microstructure have emerged from various sources, notably from the study of clays by soil engineers who have contributed much in this direction. It must be stated that observations made on engineering clay soils are, in many cases not directly applicable to a study of fresh clay sediments but if treated with caution, certain common factors emerge. Karl Terzaghi (1925) suggested a variety of clay microfabrics which he termed 'strukturarten'. Foremost among these was the 'honeycomb' structure which was later adopted by Casagrande (1932) although the latter author admitted: 'We have not succeeded, so far, in reproducing this natural structure in the laboratory'. It is indeed a tribute to Terzaghi that more recent authors have directly observed fabrics bearing a strong resemblance to his model (Mattiat 1969; Sergeyev et al. 1980). Rosenqvist (1953) thought of marine clay as having a very open structure due to flocculation
Fabric o f muds a n d shales." an overview and chemical influences were taken up by Lambe in 1958. Lambe considered that clay sediment microstructure is primarily dependent upon electrolyte concentration. Three fabric types emerge from his thesis: salt-flocculated, non salt-flocculated and the dispersed structure. The first two of these are characterized by a predominance of EF particle contacts and were nominated 'cardhouse' structures. These were considered three-dimensionally by Tan (1959). The cardhouse structure assumed great significance in later years and was rigorously adopted by Rosenqvist (1962). A certain amount of support for the existence of the cardhouse structure was provided by the author via transmission electron microscopy (TEM), but the 'salt leaching theory' of quick clay behaviour has now been largely abandoned (see Conlon 1966; Cabrera & Smalley 1973; Moon 1979). A modified version of the cardhouse structure was postulated by Von Engelhardt & Gaida (1963) in their 'aggregate structure' which is held to form in electrolyte-rich solutions. A significant development in microstructural studies emerged from the Australian authors Aylmore & Quirk who introduced the domain concept. In 1960 they wrote: 'A domain was envisaged as a microscopic or submicroscopic region within which the clay particles are in parallel array. These groups of oriented crystals, or domains~ are randomly placed with respect to one another throughout the clay matrix; that is, the domains are in turbulent array.' Domains in this context are also referred to by the authors as turbostratic groups and two major types have been identified by subsequent workers (Fig. 1). Aylmore & Quirk's conclusions emerged by way of their TEM studies using carbon replicas and it was at this time that the scanning electron microscope (SEM) was being taken up in earnest by sedimentologists and soil researchers. The direct observation of fine texture in threedimensions radically changed the whole approach to the investigation of clay microstructure. Smart (1967, 1969) was one author in the vanguard of SEM investigations and produced a comprehensive account of particle associations. His observations demonstrated that both singleplate and turbostratic structures could be present
I
|
~
|
a
I
•
b
I
I
I
I
I
c
Fro. 1. Domain structures. (a) 'bookhouse' (Sloane & Kell 1966). (b) and (c)'stepped face-to-face variants' (Smalley & Cabrera 1969).
581
in freshly sedimented clay. Another authority who carefully investigated microstructure at this time was O'Brien (1971) who has contributed a wealth of information by his research. By way of an SEM study of laboratory prepared, flocculated clays, this author observed randomly orientated domains which could perhaps, be termed a random domain structure. The domains exhibited the stepped FF structure of Smalley & Cabrera (1969) (Fig. 1). A study by Mattiat (1969) utilized freeze-dried laboratory prepared kaolinite and illite sediments for an SEM investigation. Among the fabrics recognized was the 'kartenhausgefuge' which does appear to be composed of domains in cardhouse arrangement. The evidence therefore, would seem to point to the existence of domains in freshly deposited, uncompacted clays and this is given some support by Bowles et al. (1969) who studied natural sediments (by TEM) and identified domains in a clay which had been subjected to a low overburden pressure of -,~50kPa. The conclusion here is that domains are an original feature of the clays and hence present before compaction. Similarly, Barden (1972) offered the view that single plate arrays are only present in dilute clay suspensions; in a dense clay system, only domain structures in some arrangement exist. The years following the studies of O'Brien and of Barden represent something of a hiatus in clay microfabric research. It was not until much later in the decade that an excellent and highly comprehensive work by Bennett and his co-workers (1977, i 981) provided a detailed investigation of previous work, techniques and the results of their own observations on Mississippi Delta and DSDP samples. Of paramount importance is the conclusion that sediments subjected to very shallow burial (1.4 m): 'reveal numerous randomly arranged domains in EE and EF contact producing a relatively high porosity sediment.' In general terms the authors showed that the microfabric of the DSDP samples approximated to that proposed by Pusch (1966) for quick clays and which consists of flocs with linking chains. The Mississippi Delta deposits, however, were similar to the flocculated (i.e. random domain) model of Moon (1972). The former of their conclusions echoes somewhat the words of Burnham (1970) who identified domains in a Recent marine alluvium sampled at a depth of only 1.6 m. Both natural and artificial clay sediments have been closely investigated by Osipov & Sokolov (1978) and the microstructure of Recent uncompacted deposits was found to be a type of honeycomb arrangement: ' . . . of foliated clay particle microaggregates
582
C.F. Moon and C. W. Hurst
interacting face-to-edge, less frequently faceto-face...' i.e. the random domain structure. Other types of natural sediments from Lake Maracaibo have been the subject of study by Hyne et al. (1979). It must be admitted that their results are to a certain extent enigmatic in that non-marine delta lev6e samples exhibit domainlike fabrics but the authors claim that higher salinity lake sediments have a 'cardhouse or honeycombed microstructure'. The manner of deposition would appear to have significance in fabric studies and yet this topic has received scant attention. An example is provided by O'Brien et al. (1980) who made a careful SEM examination of the clay microstructure of turbidites and hemipelagic silts. Their observations reveal that the hemipelagites have a greater degree of preferred orientation of clay particles than the turbiditic sediments. This is attributed to a slower sedimentation rate in the dispersed state as opposed to a more rapid and flocculated deposition which results from turbidity flows. On reflection it emerges that domain structures are usually present in freshly deposited clay sediments; the weight of the available evidence is in favour of this. If the conclusions of the previous section are taken into account then there must be a transition from single-plate to domain structures at, or very near to, the depositional interface. It is presently uncertain why this change occurs; possibly it is due to a sudden change in the clay mineral volume concentration as more particles are brought into closer contact. Whatever the reason, one feels that it must be an electrochemical mechanism that is operative.
The microfabric of compacted clay: beyond the mudline Under this heading some discussion of terminology is necessary. The term 'shale' is firmly entrenched in the geological literature and has been so for many years (Tourtelot 1960). A cursory glance at a number of geological texts reveals that even today, some confusion surrounds the precise definition of the word. The main contention involves the use of the descriptors 'laminated' and 'fissile', i.e.: shale: ' . . . has well-marked bedding plane fissility (primarily due to the orientation of the clay particles parallel to the bedding planes).' (Whitten & Brooks 1972) shale: ' . . . laminated and fissile.' (Pettijohn 1975) shale: ' . . . laminated clayey rock.' (Tourtelot 1960).
In an attempt to rationalize the terminology, Potter et al. (1980) insist on specificity in stating that shale must have > 33% clay size material and have laminae < l0 mm thick. On the subject of fissility, this is defined by the authors simply as a type of parting having a thickness between 0.5 and 1.00 mm. Hence, it is possible for a non-fissile shale to exist. Spears (1980) has considered this very problem of classification. In this discussion, the authors will be content to regard a shale as an indurated clay rock readily splitting into thin laminae along the bedding planes. Consolidated clay rocks not fitting this description will be referred to as mudstones. The consideration of compacted clay presented here attempts to trace the influence of increasing axial pressure on a clay deposit and consider the associated aspects of diagenesis. In connection with diagnetic changes, one must include, of course, the chemical transformations that occur. As this review concerns only microstructural changes it is not proposed that any more than a mention be made of this. The subject has been dealt with at length by von Engelhardt & Gaida (1963), Weaver & Beck (1971) and Chilingarian et al. (1973). It is appreciated here that there is a complex interrelationship between the chemical and mechanical changes that take place during the diagenesis of clays and muds. An important point for discussion concerns the reason why some clay sediments become lithified to shale, whilst others are converted to mudstone. Certain authors have assumed that the development of fissility is merely a function of a high overburden pressure; Baldwin (1971) has suggested that fissility develops in clays at 350 m depth. The evidence available is against this premise, since many sediment cores sampled at depth show fissile shales to be found a b o v e non-fissile mudrocks (White 1961). Another mechanism other than pressure must operate in the formation of shale as opposed to a mudstone and it emerges that environmental factors are relevant in this context. The domain structure has been repeatedly observed in clays subjected to compaction and among the first authorities to identify these structural units were Sloane & Kell (1966) whose 'bookhouse' structure was readily adopted in the literature (Fig. 1). Much later work has proved with little doubt the existence of domains in clay deposits compacted both naturally and artificially (Barden et al. 1969; Barden & Sides 1970, 1971). A pertinent point has been made by Blackmore & Miller (1961) who found that as compaction of a montmorillonite clay proceeds, or as overburden depth increases, a greater number of particles is incorporated into each
Fabric o f muds and shales: an overview domain. In contrast to these findings, however, McConnachie (1974) found that this was not the case in kaolinite slurries under test. Hesling (1970) in direct contrast, appears to consider only single-particle structures in compacted clays and maintains that compactive pressure increases particle parallelism. From his specific surface measurements the author concludes that at about 1000 m depth there is a dramatic decrease in specific surface which he attributes to particle fusion. Of prime concern in this section is the reason for the development of fissility. An obvious assumption would be that the existence of an anisotropic fabric would produce weak planes of splitting as in slate. The temptation, therefore, is to conclude that fissility in shales is due to the alignment of clay flakes in a preferred direction: 'fissility in shales is usually associated with a parallel arrangement of the micaceous clay p a r t i c l e s . . . ' (Ingram 1953). If this is so, this must by necessity be modified in the light of the domain model. Examination of clay shales by SEM is an easy matter in comparison with sediments having a high moisture content. Since the moisture content of indurated rocks is low there is a minimal chance of fabric disruption due to drying or water replacement. O'Brien (1968a, 1968b) has observed well-developed particle alignment in addition to the identification of domains. The present authors have similarly noted a high degree of preferred orientation in all shale samples studied (Fig. 2). Non-fissile mudstones do not appear to display any degree of preferred orientation (Fig. 2(e)) and so it does seem that fissility is related to particle (domain) orientation. It must be admitted that domains are not easy to recognize in highly compacted mudrocks, especially shales; the very process of compaction pushes domains so close together that they merge into one another. Logically, domains should be present in compacted clays even though difficult to recognize essentially because they are present in fresh, uncompacted clay sediments. Smart (1967) has mapped domain distributions in an artificially prepared kaolinite sediment which was consolidated under a pressure of ,-~550kPa. In addition, McConnachie (1974) identified domains in all of the kaolinite samples which were compressed to a maximum of 105kPa and comments: ' . . . in the experiments described here, domains were present from the outset.' Another comprehensive study of artifical compaction, both isotropic and anisotropic, has been undertaken by Krizek et al. (1975)in which kaolinite slurries were prepared with flocculating
58 3
and dispersing additives. Among the conclusions is the observation that anisotropic compaction (i.e. with lateral restraint) tends to produce a 'highly oriented fabric' and that dispersion produces greater orientation than flocculation. There was a suggestion of domains in some samples. By careful assimilation of the various conclusions above, it is apparent that: (1) there is a distinct connection between fissility and clay particle orientation; (2) clay particles are orientated to form domains which display parallel alignment in shales but not so in mudstones; (3) there is a suggestion that for some clay mineral species, increasing pressure produces an increase in the number of particles per domain. The above speculations have been discussed by Byers (1974) who offers an alternative explanation for the absence offissility in some mudrocks. This author, in making a careful study of a Devonian argillaceous horizon, found a lateral transition from shale to mudstone. Close inspection revealed that the mudstone was strongly bioturbated and Byers' conclusion is that fissility is due simply to a lack of benthic infauna. In essence, mudstones display no fissility because they have been re-worked and hence a biogenic origin is envisaged for their microstructure rather than a physico-chemical mechanism. As a rider, the author adds that the azoic shales did display an orientated fabric which is presumably responsible for fissility. The role of bioturbation has also been given some importance in clay sediments lacking fissility by Barrows (1980). In any study of fissility and mudrock microstructure reference must be made to Spears' (1976) study of Coal Measures sediments. Following an investigation of weathered and unweathered shales from the same sequence, Spears concluded that fissility is primarily due to the weathering of organic-rich laminations. The explanation offered is that on weathering, the laminations absorb a large quantity of water which causes swelling and induces splitting along these laminations. In support of this, it has been reported that organic-rich layers in mudrocks are characterized by strong preferred orientation (Odom 1967). To some extent, the importance that Spears places upon weathering is a development of the conclusions of Ingram (1953). This constitutes a reasonable approach to the problem but logically one would expect a fissile structure to result from the weathering of a rock with a parallel, orientated microfabric basically because weathering will proceed more rapidly in the plane of particle orientation. The higher weathering rate is the
584
C.F. Moon and C. W. Hurst
(a)
(b)
(c)
(d)
(e) FIG. 2. (a) Fissility of non-marine shale, Lower Coal Measures. Scale bar lmm. (b) Microstructure of shale shown in Fig. 2. Scale bar 20/~m. (c) Microstructure of Ludlovian (upper Silurian) shale. (d) Microstructure of Stockdale Shale, Llandoverian, (lower Silurian). Note disruption of fabric by authigenic framboidal pyrite. Scale bar 20#m. (e) Microstructure of mudstone (Ashgillian, upper Ordovician). Scale bar 20pm.
Fabric o f muds and shales." an overview result of a much greater permeability in this direction. The weathering hypothesis was later taken up by Curtis et al. (1980) who observed, as perhaps would be expected, that equant quartz grains in mudrocks have a disruptive effect on microstructure. On the subject of fissility, however, they state rather paradoxically: ' . . . it may therefore be that clay orientation has nothing to do with fissility.' This is in direct contrast to the conclusions of a great variety of other authorities (e.g. O'Brien 1968; Barrows 1980).
The influence of non-vertical forces: the effects of shear In some environments, sediments may be subjected to shearing forces prior to complete consolidation, an obvious example being gravity sliding on submarine slopes. Some investigations have been made to assess the influence of shear and the development of shear planes upon the microstructure of clays. These studies have in the main, been performed by workers in the field of soil mechanics who have sheared plastic clays (i.e. at moisture contents below the liquid limit) to failure and have studied the induced failure planes microscopically in an attempt to gain a better understanding of the failure mechanism of clay soils. Many of these investigations have immediate relevance to the study of fine-grained sediments and possibly to tectonic mechanisms. A brief review of the more pertinent literature is given below and this is augmented by certain of the present authors' observations. A large number of investigations have successfully demonstrated that in a random domain fabric, shear strains of varying magnitude cause an alignment of particles parallel or sub-parallel to the plane of shear. At very large strains more radical effects on fabric are seen, but in all cases varying degrees of rotation of domains result. Weymouth & Williamson (1953) studied the effect of extrusion on a plastic clay by forcing the material out of a large diameter cylinder into a smaller one arranged concentrically. The fabric was studied by the optical microscopy of peels and amongst their conclusions is found: 'Diagonal shear joints crossed the fabric of the extruded cylinder. They contained clay layers having (001) at a small acute angle to the shearing surfaces.' (Presumably, since optical methods were employed, it was domains rather than particles which were identified.) A comprehensive account of shear-induced
585
microstructural changes has been provided by Morgenstern & Tchalenko (1967). After subjecting laboratory-prepared kaolin samples to direct shear, a variety of features were apparent. Among these were the Reidel shears, or tension fractures noted by Weymouth & Williamson above. Also recognized were kink-bands containing zones of orientated domains and 'compression textures' along shear surfaces in which two bands of domains orientated parallel to the shear plane enclose another aligned zone with particles/ domains set at 40~ ~ to these bands (Fig. 3). Tchalenko (1968) recognized similar structures in natural samples of London Clay. The effect of shear has also been studied by Clark (1970) who, in accordance with other authors maintains that distortion as opposed to simple compression is more significant in the development of preferred orientation. Clark suggested that preferred orientation resulting from the shear stresses is primarily due to the breakdown of flocs. This may have some importance in that it is possible that domains are destroyed by shear. Extensive use of the SEM has been made by Barden et al. (1969) whose results lent much support to the conclusions of Morgenstern & Tchalenko, especially in that they observed shear zones to have a tripartite structure, i.e. the compression texture. Similar observations have been made by Foster & De (1971) and a more recent, quantitative analysis has been undertaken by Tovey & Wong (1980). In addition to parallel alignment, other features may be identified which seem to be produced at large shear strains. Included here are compression textures and also
FIG. 3. Diagrammatic representation of 'compression texture' (Morgenstern & Tchalenko 1967).
586
C.F. Moon and C. W. Hurst
o
,.o
0
0"~ N.~
"~_, ~1
~'~
6~
Fabric of muds and shales." an overview 'fabric rolls'. The authors have identified both of these features in sheared clays and they are shown in Fig. 4. Both particle arrangements are reminiscent of shear-induced megascopic features familiar in structural geology. It may be reasonable to consider the compression texture to be a form of imbricate or schuppen structure whilst the fabric rolls could be regarded as drag folds (Badgley 1965). Apart from the work of Tchalenko (1968) few authors have paid attention to the identification of shear structures in natural sediments. Foremost in this context are the more recent investigations by Bohlke & Bennett (1980) in relation to a study of pro-delta crusts. The crusts are typified by a high shear strength and a TEM study showed them to have a dense domain microfabric with a high degree of preferred orientation quite unlike the sub- or superjacent sediments. The authors suggest that the existence of the crusts and the origin of their fabric can be explained by shear deformation resulting from mud flow sliding and slumping on an unstable pro-delta slope. As a direct corollary to the above, a rather radical approach has been advanced by Davies and his co-workers (Davies & Cave 1976; Davies 1980). The postulate is that primary clay flake orientation is held responsible for cleavage in slates and by way of an elegant argument, Davies (1980) attributes slaty cleavage to a parallel clay mineral fabric produced by shear slippage in unconsolidated clays. The shear component is the result of the viscous flow of sediment on a slope (or due to an 'uneven superincumbent load'). The author writes: '. . . . . . the restriction of commercial quality slate to a few, comparatively thin stratigraphic horizons (in North Wales) could imply that the clay structure on these horizons became particularly meta-stable and more susceptible to shear and consequent reorientation of the clay flakes.' It can be inferred from the foregoing that to a certain extent, in high water content sediments, parallel alignment is easily produced by shear forces. At lower moisture contents below the liquid limit, other more complex microstructural features are the result. As a postscript to this section, mention is made of the effect of shear forces upon sedimenting muds. Stow & Bowen (1978) have invoked 'shear sorting' as a mechanism in the development of lamination in fine-grained turbidites. During turbid flow, clay flocs are intermixed with silt particles of similar diameter and hence similar settling velocity. Upon reaching the flow boundary layer, however, shear stresses cause the flocs to disaggregate which in turn reduces their set-
587
tling velocity and segregation of clay and silt produces a laminated sediment. In this context, one is reminded of Clark's (1970) conclusions.
Synthesis and conclusions The previous sections attempt to trace the origin and development of particle arrangements in clays. It remains to marshal the available information into a logical scheme and provide some explanation of the observations and conclusions made to date. It is apparent that several authors have observed a parallel fabric in fissile mudrocks. Conversely, a random domain structure is seen to be present in non-fissile clay sediments. There appears a distinct and recognizable correlation between fissility and domain orientation. Preferred orientation of platy particles produces an anisotropic fabric which is not uncommon in other rocks, e.g. flaggy sandstones and slates, which are typified by strength anisotropy manifested by planes of splitting. The authors can see little reason why this does not apply to mudrocks, at least, in the majority of cases. If this is accepted, then a mechanism must be found in which preferred orientation is developed in some mudrocks and not in others. It is the authors' contention that this is mainly the result of the environment of deposition. Especially, the chemical conditions that exist during and immediately after, the sedimentation of an argillaceous deposit. The detailed chemistry involved is at present obscure but the variables concerned must include electrolyte concentration, pH and Eh. Rheologists (e.g. Flegmann et al. 1969; Goodwin 1971) have demonstrated that the manner in which clay particles associate depends to some extent upon pH. Colloid chemistry has highlighted the equal importance of electrolyte concentration (Van Olphen 1963) but perhaps it is oxidation potential that has not received the attention it otherwise deserves. The basic premise here is that in organic-rich conditions (marine or non-marine) there will exist the required dispersing agents which allow particles to settle individually and produce a parallel, orientated microfabric. This in turn will result in an anisotropic strength responsible for fissility. It is not assumed that this is the sole reason for fissility but the authors consider it to be one of great importance. One must also bear in mind the speculations of Spears (1976) and Hallam (1980) who attribute fissility to localized bands of organic material. Various authors have noted the association between organic content and shales (Ingram
C.F. M o o n a n d C. W. H u r s t
588
1953; O'Brien 1968b; Leventhal & Shaw 1980). The organic material in these cases manifests itself in the dark colour of the rock and Ingram (1953) has observed that shales tend to be dark whilst mudstones are usually of a light colour. Indeed, black shales are commonplace rocks and regarded as a distinct facies in their own right. It is well known that certain organic ions in very low concentrations are capable of dispersing (or peptizing) clay minerals (Van Olphen 1963) and a number of these organic compounds are common in a variety of depositional environments. Probably the most studied of these is the anoxic basin in which bottom waters become oxygen-depleted due to the aerobic decomposition of planktonic organisms and produce a reducing environment. Classic examples are
l]ords (Strem 1955; Grasshof 1975), fjord-like basins such as the Saanich Inlet (Brown et al. 1972) and other ocean deeps for example the Santa Barbara Basin (Degens et al. 1961) and the Cariaco Trench (Deuser 1975). In addition to these is the barred basin typified by the Black Sea (Degens & Stoffers 1980). An examination of the hydrochemistry of these areas (Deuser 1975) indicates that an anoxic zone exists at a depth which is rich in organic compounds. The transition from oxic to anoxic conditions may coincide with the mudline or it may be located within the overlying water column. It proposed that clay sediments entering the anoxic zone have the charges on their particle surfaces neutralized by organic compounds and a dispersed (domain?) deposit results. Following compaction and induration, a parallel orientated
o
+
.
z / o
_./
I ~ /l . . . .m. .u d - -J-- - ~~ - ~k._i___ 20,urn
--22000-C: Chor~drites T:Teichichnus P: Planolites S:
Z:Zoophycos
Scolicia
FIG. 4. Ichnofaunal changes due to variations in sediment texture, Core 347, N.W. African margin, 2576 m depth. See text for discussion.
602
A. Wetzel
are replaced by Scolicia traces which preferentially burrow in the coarser sediments. These changes in ichnofauna occur within the Zoophycos ichnofacies. The adaptation of infauna to a coarser grainsize needs a certain time interval. In the case of adequate time for adaptation, ichnofauna may change in correspondence with the environmental variations. Off N W Africa, a criterion for fully adapted ichnofaunal change is that the organisms are able to maintain about 100~ bioturbation.
Consequently, the threshold rate of change in sediment parameter, below which the fauna can fully adapt to this change, can be evaluated if the degree of bioturbation is slightly, but definitively lowered after the environmental changes have taken place (Fig. 5). In the deep sea o f f N W Africa, the ichnofaunal change occurs within 3-6 cm of sediment at an average sedimentation rate of about 20 cm/1000 yrs (Fig. 4). Considering the vertical mixing effect by Scolicia-producing organisms (Wetzel 1981)
FIG. 5. Partially unburrowed aeolian dust deposits off NW Africa, Core 239 (119-134 cm depth) X-radiograph radiograph (negative) scale 3 cm; (a) Scolicia, (b) Helminthopsis, (c) Zoophycos, (d) undisturbed silt layer.
Bioturbation in deep-sea fine-grained sediments these changes have been taking place in an absolutely shorter (about 1/3) time interval. Thus, the coarsening of sediment texture due to aeolian dust input might have taken place in less than 200 yrs at an average sedimentation rate of about 20 cm/1000 yrs. During this time the median grain-size (of non-carbonate components) increases from 5 #m to 35 #m (Fig. 4). As shown in Fig. 5 such aeolian dust layers are not always completely bioturbated. Thus, the threshold rate for environmental and ichnofaunal changes appears to have been reached. Consequently, the time interval necessary for adaptation to a coarser-grained substratum can be estimated at 100-200 yrs.
Turbidites The influence of turbidites on grain-size and other parameters is markedly different from that of aeolian dust in that: (1) turbidites are deposited very rapidly; (2) a comparatively thick sediment layer is built-up during a short time; (3) turbidites normally contain a coarser and/or wider grainsize spectrum than aeolian dust deposits, and (4) a turbidity current might have eroded the surface sediment layer more or less effectively. Therefore, in a biotope influenced by turbidite deposition, rather different behavioural and physical adaptations of infaunal organisms are found. The biogenic traces occurring in turbidites and associated sediments appear to be preferentially adapted to fine-grained sediments, to a wide grain-size spectrum, or to coarse-grained deposits. The relationship between these three burrow categories within an ichnocoenose is determined by the recurrence time of turbidity current events and the non-turbiditic sedimentation rate.
Burrows adapted to fine-grained sediments The deep-burrowing forms such as Chondrites, Planolites, Teichichnus, and Zoophycos are all adapted to fine-grained sediments. So too are the graphoglyptid traces that show a highly-developed feeding behaviour and characterize the Nereites ichnofacies (Seilacher 1967). These burrows are commonly mucus-lined hollow smalldiameter tube systems. Highly-developed forms such as Palaeodictyon, have multiple openings at the sediment surface for ventilation and are therefore restricted to the upper layer. Because graphoglyptids are hollow, they can easily be destroyed by compaction and therefore require special conditions for preservation (Seilacher 1977 a). They are commonly restricted to areas where the erosional effects of turbidity currents are low, or where bioturbation and
603
tiering of organisms within the sea floor sediments is reduced. Besides their infilling by turbidites, in slightly compacted sediments two cases can be identified: (1) The non-turbiditic sediment accumulation is low; in which case the sediment build-up between two turbidite events will be too small to establish a second bioturbational level below the top level (about 3-5 cm, Berger et al. 1979; Wetzel 1979) and graphoglyptids may predominate. (2) The non-turbiditic sediment accumulation rate is high; in which case graphoglyptids will co-occur with biogenic traces of deeper levels and, in some cases, those adapted to a wider grain-size spectrum. The sedimentation rate is sufficiently high (>40 cm/1000 yrs) that the underlying levels are only incompletely burrowed and hence biogenic surface traces are also found deeper within the sediment. However, observations on such a deep-sea ichnocoenose are still lacking.
Burrows adapted to a wide grain-size spectrum In sedimentary sequences where turbidites are common, biogenic traces occur that are adapted to a wider grain-size spectrum, for example, Miinsteria, Phycosiphon, Scolicia, and Thalassinoides (~ Granularia). Because these forms are normally lacking in continuously accumulated fine-grained deep-sea sediments, they may represent forms transported from the shelf or the slope to deeper environments (Crimes 1977; Wetzel 1981). A significant occurrence of these burrows mainly depends on the recurrence time of turbidite events rather than on the spacing between such layers. This has beendeduced from comparisons between the deposits off NW Africa and the Sulu Sea. Off NW Africa, thick (2-36 cm) turbidites are found separated by non-turbidite intervals of 3-20 cm thick that represent periods of at least 1500 yrs (assuming a sedimentation rate of about 2 cm/1000 yrs). Burrows adapted to a wide grain-size spectrum are sparse. This can be also estimated by calculation of population density based on the relation between frequency of burrows and the sediment volume studied (Wetzel & Werner 1981). In the Sulu Sea Deep, 2-4cm thick turbidites are more frequent and the spacing between them is 10-60 cm. Based on absolute age determinations (Erlenkeuser; Kiel, unpubl) the recurrence time of turbidity currents is 100-700 yrs, or about 170 yrs on average. Within these deposits, Miinsteria, Phycosiphon, Thalassinoides, and Zoophycos frequently occur (~>1 burrow m-Z; Fig. 6). However, they do not occur so commonly where the turbidites have a rela-
6o4
A. Wetzel
8~
eq'~
S~
~o
I~
,,~,,,~ " , ~
.-~
s
~g .~ 9 ~,.~
Bioturbation in deep-sea fine-grained sediments tively long recurrence interval as in the examples off N W Africa. With respect to these observations, two aspects have to be considered in some more detail: establishing and maintaining of a formerly exogenous fauna adapted to a wide grain-size spectrum. The critical population density to establish an exogenous fauna in the deep-sea may be reached when the imported organisms could reproduce in the deeper environment. Because a turbidity current derives animals from a small source area and spreads them over a large depositional area, only a low population density of exogenous organisms can be established by any one turbidity current. Therefore, the probability of reproduction will be raised when turbidity currents repeatedly occur during the life-time of the imported organisms. Assuming a life-time of 50-100 yrs for deep-sea macrobenthos, as Turekian et al. (1975) determined for some deep-sea animals, a turbidity current frequency of 20-200 events/1000 yrs seems to be adequate to establish a reproducing, but formerly exogenous, fauna adapted to a wide grain-size spectrum. On the other hand, to maintain a fauna as discussed above, a lower frequency of turbidity currents may be sufficient. Based on the observations on the Sulu Sea deposits, a frequency of 0.5 to 2 turbidity currents/1000 yrs seems to be adequate if the sedimentation rate of finegrained, non-turbiditic material does not exceed 50 cm/1000 yrs.
605
Environmental changes and ichnofauna diversity As shown above, the recurrence time of turbidity currents and the non-turbiditic sedimentation rate are two important controls in the composition of ichnofauna in deep-water environments. Besides a quantification of the boundary conditions for the various groups of burrows, the effects of these parameters on biogenic trace communities should be expressed by diversity values. Sediments o f f N W Africa and offthe Sulu Sea will be compared and, to complete the environmental and bioturbational trends from the slope to deep sea, deposits of the Central Pacific Ocean are also discussed (Fig. 1 and 7). In upper slope sediments of both areas, diversity is high, and biodeformational structures are replaced by more specialized biogenic traces (Wetzel 1981). In middle-slope to rise sediments a nearly constant number of ichnogenera has been observed, although in the Sulu Sea the ichnofauna within this depth range is less diverse than o f f N W Africa. This may be due to the low oxygen content and/or to the less variable climatic conditions in the Holocene sediments of the Sulu Sea. Towards the deep sea, both coastal upwelling and fluvial and/or aeolian sediment input are less and the biotope becomes increasingly uniform with distance from coast. In such an environment an optimal fractionation into ecological niches can take place due to a long-term adaptation of
wafer degree of biofurbotion (%) number of biogenic depth NW-Africo/PQcific Sulu Sea trace types preserved (km) 0 50 100 0 50 100 0 5 10 15 |11111111 :::::::::::::::::::::::::::::
1.0
!ii!iiiiiiii!iiiiiiiiiiiii
1.5
iiiiiiiiiiiiiiiiiiiiii
2.0 2.5
I
3.0 3.5 /,.0 4.5 5.0 m
~
I
/ Pocifi/
Sulu Sea
FIG. 7. Degree of bioturbation and diversity of preserved biogenic trace types in sediments off NW Africa, in the Central Pacific Ocean, and in the Sulu Sea Basin. See text for discussion.
6o6
A. Wetzel
the fauna (Seilacher 1977a). In fact, abyssal surface sediments contain a comparatively high number of ichnogenera if grapholyptids occur (Kitchell 1978; Ekdale 1980). However, biogenic trace diversity may be secondarily reduced due to destruction of these shallow burrows by deeper penetrating ones. In this way a monotonous low-diversity ichnofauna may reach the fossil record, dominated by Chondrites, Planolites, (Teichichnus), and Zoophycos, and sometimes with 'composite burrows' of Helminthoida/Helminthopsis (Chamberlain 1975). This type of ichnocoenose has been observed in DSDP cores (e.g. Chamberlain 1975; Ekdale 1977) as well as in modern sediments (e.g. Ekdale 1978; Berger et al. 1979), and corresponds to the Zoophycos ichnofacies (Seilacher 1967). It is typical for continuously accumulating deep-sea deposits without major changes of environment and sediment texture. It probably also occurs in rise to slope sediments accumulating under similarly uniform environmental conditions. Although the morphology of Zoophycos varies widely, a depth zonation of different forms has not been observed (Wetzel & Werner 1981). In areas of extremely slow accumulation, the bioturbated zone becomes very thin as organic matter cannot be buried deeply within the sediment (M/iller & Suess 1979). In this case, Zoophycos may also be absent, because there is no food available in the sediment depth where Zoophycos normally occurs (Wetzel & Werner 1981). On the other hand, Zoophycos is also lacking in sediments deposited more rapidly in water depth above 2000-1000 m off NW Africa and the Sulu Sea. Thus, in both extremely slowly accumulating and upper slope deposits the index burrow of the Zoophycos ichnofacies is absent making an ichnological interpretation difficult. In the Sulu Sea Deep, the high frequency of turbidites prevents complete bioturbation and steady state sedimentation so that shallow-penetrating biogenic traces may also be preserved and a high-diversity ichnofauna may reach the fossil record. Such highly diverse, graphoglyptid-bearing ichnocoenoses belong to the Nereites ichnofacies which therefore is typical for discontinuous sedimentation due to turbidites. This comparison shows that different ichnofacies may occur in the same range of water depth. Their occurrence seems to depend mainly on the presence or absence ofturbidites. Thus, the use of ichnofacies for palaeobathymetric analyses is restricted in abyssal plain sediments. Assuming sufficient supply of benthic food and oxygen, medium burrow diversities (10_+ 5 biogenic trace types) occur abundantly in slope to rise sediments, with the exception of the upper
slope and canyons. In the deep sea, both extremely low diversity values (~< 5 biogenic trace types) occur in totally bioturbated deposits and very high values ( >~ 15 biogenic trace types) occur in incompletely reworked sediments containing turbidites. Variations or organic matter and/or oxygen content, and of sediment texture may influence these values.
Conclusions (1) Biogenic trace assemblages can be best interpreted using a tier model with a variable number of distinctive bioturbation levels. Thus, the preservation potential of a biogenic trace mainly depends on the burrowing rate at deeper levels within the sediment where buried shallow burrows are replaced by deeper-penetrating burrows. (2) The response of ichnofauna to environmental changes requires a certain although very short time interval. An adaptation to a coarser sediment texture due to input of aeolian dust (coarse silt) takes place within < 100-200 yrs in rise to abyssal plain sediments. This value also seems to be applicable to other environmental changes, if the 'new' ichnofauna occurs latently. (3) Highly-developed near-surface biogenic traces, so-called graphoglyptids, can best reach the fossil record in turbidite environments. Infauna adapted to a wide grain-size spectrum seems to be imported from shelf to slope sediments, because these deposits represent a similar grain-size spectrum even if turbidites are intercalated. It can be assumed that an abundant occurrence of these burrows in the deep sea depends on recurrence time of turbidites rather than of the spacing between individual coarsegrained layers. For a significant and permanent occurrence of new, perhaps exogenous, (ichno-) faunal elements, an autochthonous reproduction of these seems to be necessary. This can be best achieved if repeated 'import' events occur during the life-time of such organisms. To establish a fauna adapted to a wide grain-size spectrum, a value of 20-200 turbidites/1000 yrs has been assumed. To maintain this fauna 0.5-2 turbidites/1000 yrs. seem to be adequate. (4) Biogenic trace diversity profiles from slope to the abyssal plain show medium diversities for middle slope to rise sediments, 10_+5 burrow types; variations in the intensity of coastal upwelling or of fluvial or aeolian sediment input, may raise or lower this number. In abyssal plain sediments, extreme values of ~~15 burrow types have been found. They correspond to two basic types of bioturbation: (a) Low-diversity Zoophycos ichnofacies (slope to abyssal plain);
Bioturbation in deep-sea fine-grained sediments under stable environmental conditions, slowly and continuously-accumulating sediments are completely reworked by organisms. Thus, the fossil record is d o m i n a t e d by deeply-penetrating biogenic traces, such as Chondrites, Planolites, (Teichichnus), and Zoophycos. (b) High-diversity Nereites ichnofacies (rise to abyssal plain?); in turbidite-dominated sequences the continuous bioturbation process is interrupted by coarse sediment layers preventing total bioturbation. Thus, near-surface biogenic traces may also reach the fossil record. (5) Consequently, the use of biogenic traces for bathymetric interpretation of the deep-sea sediments is restricted.
607
ACKNOWLEDGEMENTS: The investigations on the sediment cores have been carried out in the laboratories of the Geologisches Institut der UniversitS.t Kiel. E. Seibold (now Bonn) and F. Werner (Kiel) advised and contributed valuable help and stimulating discussion during investigations. P. Ballance (Auckland) and M. Pye (Britoil, Glasgow) critically reviewed the manuscript, and D. Stow (Edinburgh) made helpful c o m m e n t s on the text and carefully improved the English. Financial support for an earlier phase of these investigations was given from the Deutsche Forschungsgemeinschaft. All these contributions are gratefully acknowledged.
References BERGER, W.H., EKDALE,A.A. & BRYANT, P.P. 1979. Selective preservation of burrows in deep-sea carbonates. Marine Geol., 32, 205-30. BLANPIED, C. & BELLAICHE,G. 1981. Bioturbation on the Pelagian platform: ichnofacies variations as paleoclimatic indicators. Marine Geol., 43, M49-M57. BROMLEY,R.G. & ASGAARD,U. 1975. Sediment structures produced by a spatangoid echinoid: a problem of preservation. Bull. geol. Soc. Denmark, 24, 261-81. CHAMBERLAIN,C.K. 1975. Trace fossils in DSDP cores of the Pacific. J. Paleontol., 49, 1074-96. CRIMES,T.P. 1977. Trace fossils of an Eocene deep-sea sand fan, northern Spain. In: Crimes, T.P. & Harper, J.C. (eds), Trace Fossils 2. Geol. J., Spec. Iss., 9, 71-90. EKDALE,A.A. 1977. Abyssal trace fossils in the worldwide Deep Sea Drilling Project cores. In: Crimes, T.P. & Harper, J.C. (eds), Trace Fossils 2. Geol. J., Spec. Iss., 9, 163-82. -1980. Graphoglyptid burrows in modern deep-sea sediments. Science, 207, 304-6. -& BERGER, W.H. 1978. Deep-sea ichnofacies: modern organism traces on and in pelagic carbonates of the western equatorial Pacific. Palaeogeogr., Palaeoclimatol., Palaeoecol., 23, 263-78. EXON, N.F., HAAKE,F.-W., HARTMANN,M., KOGLER, F.C., M(ILLER, P.J. & WHITICAR,M.J. 1981. Morphology, water characteristics and sedimentation in the silled Sulu Sea, Southeast Asia. Marine Geol., 39, 165-95. FUCHS, T. 1895. Studien fiber Fukoiden and Hieroglyphen. Denkschr. Akad. Wiss. Wien, 62, 369-448. HANTZSCHEL, W. 1965. Vestigia Invertebratorum et Problematiea; Fossilium Catalogus I." Animalia pars 108. W. Junk, s'Gravenhage. 142 pp. -1975. Trace fossils and problematica. In: Teichert, C. (ed.), Treatise on Invertebrate Paleontology; Part
W. Miscellanea, Supplement 1. Geol. Soc. Am., New York and Univ. Kansas Press, Lawrence, XXXI, 269 pp. K1TCHELL, J.A., K1TCHELL, J.F., JOHNSON, G.L., & HUNKINS, K.L. 1978. Abyssal traces and megafauna: comparison of productivity, diversity and density in the Arctic and Antarctic. Paleobiology, 4, 171-80. KOOPMANN, B. 1981. Sedimentation yon Saharastaub im subtropischen schen Nordatlantik wfihrend der letzten 25.000 Jahre. 'Meteor' Forsch.-Erge., C, 35, 23-59. MULLER, P.J. 1975. Diagenese stickstoffhaltiger organischer Substanzen in oxischen und anoxischen marinen Sedimenten. 'Meteor' Forsch.-Erg., C, 22, 1-60.
--
& SUESS,E. 1979. Productivity, sedimentation rate and sedimentary organic matter in the oceans. I. Organic carbon preservation. Deep-Sea Res., 26, 1347-62. SARNTHEIN, M., THIEDE, J., PFLAUMANN,U., ERLENKEUSER, H., F1]TTERER, D., KOOPMANN,B., LANGE, H. & SEIBOLD, E. 1982. Atmospheric and oceanic circulation patterns offNorthwest Africa during the past 25 million years. In: von Rad, U., Hinz, K., Sarnthein, M. & Seibold, E. (eds), Geology of the Northwest African Continental Margin. SpringerVerlag, Berlin, Heidelberg. 545-604. SEIBOLD, E., DIESTER-HAASS,L., F~TTERER, D., HARTMANN, M., KOGLER, F.-C., LANGE, H., MULLER, P.J., PFLAUMANN,U., SCHRADER,H.J. & SUESS, E. 1976. Late Quaternary sedimentation off the western Sahara. An. Acad. bras. Cient., 48, supl., 287-96. SEILACHER,A. 1967. Bathymetry of trace fossils. Marine Geol., 5, 413-28. 1977 a. Pattern analysis of Paleodictyon and related trace fossils. In: Crimes, T.P. & Harper, J.C. (eds), Trace fossils 2. Geol. J., Spec. Iss., 9,289-334. - 1977 b. Evolution of trace fossil communities. In:
6o8
A. Wetzel
Hallam, A. (ed.), Patterns of evolution. Developments in Paleontology and Stratigraphy, 5. Elsevier, Amsterdam. 359-76. SUESS, E. 1980. Particulate organic carbon flux in the oceans--surface productivity and oxygen utilization. Nature, 288, 260-3. TAIT, R.V. 1971. Meeres6kologie. Thieme, Stuttgart. 305 pp. TUREK1AN,K.K., COCHRAN,J.K., KHAKAR,D.P., CERRATO, R.M., VAISNYS, J.R., SANDERS, H.L., GRASSLE,J.F. & ALLEN,J.A. 1975. Slow growth rate of a deep-sea clam determined by 228Ra chronology. Proc. Nat. Acad. Sei. USA, 72, 2829-32. WERNER, F. 8/; WETZEL, A. 1982. Interpretation of biogenic structures in oceanic sediments. Bull. Inst. Geol. Bassin d'Aquitaine, 31, 275-88. WETZEL, A. 1979. Bioturbation in spdtquartiiren Tiefwasser-Sedimenten vor N W Afrika. Ph.D. Thesis, Univ. Kiel. 111 pp.
-
-
-
-
- -
- -
1981. Okologische und stratigraphische Bedeutung biogener Gef/ige in quart/iren Sedimenten am NW-afrikanischen Kontinentalrand. 'Meteor' Forsch.-Erg., C, 34, 1-47. 1983. Biogenic sedimentary structures in a modern upwelling area: the NW African continental margin. In: Thiede, J. & Suess, E. (eds), Coastal Upwelling: and Its Sediment Record, Part B, Sedimentary Records of Ancient Coastal Upwelling. Plenum Press, New York. 123-44. • WERNER, F. 1980. Biogene Sedimentgef/.ige. In: Werner, F. Fahrtbericht "Meteor'-Reise 53 C/DOstatlantik vor Marokko. Unpubl. Report, Geologisch-Palfiontologisches Institut der Universit/it Kiel. 20-2. & WERNER, F. 1981. Morphology and ecological significance of Zoophycos in deep-sea sediments off NW Africa. Palaeogeogr., Palaeoclimatol., Palaeoecol., 32, 185-212.
A. WETZEL, Geologisches Institut der Universit/it, Sigwartstrasse 10, D7400 T/ibingen, West Germany.
Deep-water fine-grained sediments: facies models D.A.V. Stow and D.J.W. Piper S U M M A R Y: Based on a large amount of published data and stimulated by the papers and discussion at the International Workshop on Fine-Grained Sediments held in Halifax, Canada in August 1982, we have attempted a synthesis of deep-water fine-grained sediment facies. Three main facies groups related to depositional processes can be identified: turbidites, contourites and pelagites/hemipelagites. There is a continuum between the different processes and hence a continuum between facies. Nevertheless, it is possible to define several distinct facies models within each of these groups on the basis of sedimentary structures, texture and composition, and to provisionally interpret these in terms of depositional hydrodynamics. Patterns of horizontal and vertical facies distribution can be related to depositional subenvironments. There is much variability within and departure from the facies models we propose, and many interesting and problematic areas of research remain in the quest for better understanding of deep-water fine-grained sediments.
Only five years ago it was fair to say that fine-grained sediments were grossly understudied by comparison with, for example, sands and gravels. However, in the last few years, a large amount of data has appeared on both modern and ancient clays, muds and silts and their biogenic equivalents (e.g. Potter et al. 1980). A variety of fine-grained facies types have been recognized on the basis of size, grade and internal organization of beds (e.g. Piper 1978; Stow 1984a; in press), and new questions have emerged regarding their origin, processes and environments of deposition. However, there is still need for a general synthesis or organization of these data into composite facies models. Such models may then serve as a framework to guide description, a standard against which to compare new information, a predictive tool in new geological settings and as a basis for hydrodynamic interpretation (Walker 1975). There have been several recent attempts to construct facies models for deep sea sediments, including turbidite muds and silts (Piper 1978; Stow & Shanmugam 1980), 'homogeneous' muds (Stanley 1981), carbonate turbidites (Hesse 1975; Stow, Wezel et at., this volume), muddy contourites (Stow & Lowell 1979; Stow 1982; Faug6res et al. 1984; Gonthier et al. this volume), hemipelagities and pelagites (Hoffert 1980; Einsele 1982; Jenkyns, in press; Thornton, this volume; Hill, this volume). There have been similar attempts to collect together the diverse data on more shallowwater fine-grained sediments (e.g. tidal deposits, Ginsburg 1975; deltaic deposits, Coleman 1976; shelf deposits, Shepard et al. 1960): these are not considered further here. In this paper we attempt, therefore, to collate information on deep-water fine-grained sediments from our own studies, from a growing body of literature, and from the papers and
discussion at the International Workshop on Fine-Grained Sediments (Halifax, Canada, 1982) from which this volume has been edited. We recognize and describe various facies models within three main facies groups: fine-grained turbidites, muddy contourites, and pelagites/ hemipelagites; and briefly discuss their interpretation in terms of depositional processes. We then consider the distribution of these facies in different environmental settings and some of the post-depositional changes that they undergo. Our discussion highlights some of the interesting and problematic areas of interpretation. There is by no means a concensus view on the description and interpretation of fine-grained sediments. Many readers will feel we are premature in defining some of the facies models and are guilty of generalization in the discussion of their horizontal and vertical distribution in the deep sea. However, we believe that some degree of synthesis and simplification is necessary and that the conclusions are useful. Certainly, they should be the subject of rigorous scrutiny by future research.
Processes and facies There appears to be a continuum of processes operating in the deep sea (e.g. Walker 1978; Stow 1984a). These include resedimentation (mass gravity) processes, normal bottom currents and pelagic settling (Fig. 1). There is close interaction and overlap between these processes both during transport and during the final stages of deposition. Any single resedimentation event, for example, may be initiated by the slumping of unstable slope sediments and then, by mixing with seawater, evolve into a debris flow, a high concentration turbidity current and, finally, a low-concent6II
D.A.V. Stow and D.J.W. Piper
612 TIME AND/OR SPACE
ROCKFALL
CREEP
SLUMP
DEBRITE
TURBIDITES + ASSOCIATED FACIES Collr l e medium fine
CONTOURITE
PELAGITE HEMIPELAGITE
ROCKFALL
~ SL,DE~~
CREEP
SLUMP REMOLDING SPILLOVER ~
LIOUEFACTION
---.___
1 SHELF S USPE NSION
~
FLUID TURBUL EN C E
L~176176
SURFACE-WATER PRODUCTIVITY AEOLIAN INPUT
"'2EP20~;2rA::: E" + SURFACE SUSPENSION ~, CURRENTS
FLOW
INITIATION
LONG-DISTANCE
DEPOSITIONAL
FACIES
MODELS
TRANSPORT
FIG. 1. Diagram showing the main processes of flow initiation and long distance transport in the deep sea and the depositional facies models related to each distinct process. In nature, there is a continuum of processes and facies. (Modified after Walker 1978). ration turbidity current. Deflection of the lowvelocity dilute tail of a turbidity current by a regional contour-following bottom current will cause the downslope flow to grade imperceptibly into an alongslope bottom current. Pelagic sediment settling vertically through the water column in the open ocean may be deflected slightly by the action of very weak bottom currents. The mechanisms involved in erosion, transport and deposition of fine-grained sediments within this process continuum are discussed more fully in Section 1 of this volume (papers by McCave, Gorsline, Kranck, Eittreim, all this volume). We recognize that there is also a facies continuum in the deep sea but that it is nevertheless possible to distinguish facies types that are the result of a distinct turbidity current, bottom current and pelagic processes. Our facies models are, therefore, closely related to the depositional process (Fig. 1). We have not attempted to include facies models for slumps, sediment creep deposits or debris-flow deposits ('debrites'), although these commonly involve fine-grained sediments (but see Naylor 1980, 1981; Thornton, this volume; Stow 1984 a,b, in press.). Several other facies types have been proposed in the literature, resulting from processes such as turbid layer flows (Moore 1969; Stanley and Maldonado 1981), suspension cascading
(McCave 1972), unifite flows (Stanley 1981), nepheloid layers (Biscaye & Eittreim 1977), canyon currents (Drake et al. 1978) and so on. These do not, however, differ significantly either in character or in the depositional process from the broader categories that we suggest and do not appear valid to us as distinct facies. They do serve to indicate the range of mechanisms that may exist within our general process groups. Black shales may be deposited by a variety of processes and so are not considered as a separate facies in this context. However, several of the papers in this volume describe organic-rich sediments of different types and deposited by different processes (e.g. Crevello et al.; Isaacs; Anastasakis & Stanley; Thickpenny, all this volume). Arthur et al. (also this volume) have attempted to synthesize a more generalized model for the black shale sedimentation. Neither do we include a detailed discussion of the acoustic character of deep-sea sediments and the recognition of acoustic facies: useful summaries are given by Damuth (1975, 1978), Jacobi (1982) and Nardin et al. (1979).
Fine-grained turbidites Fine-grained turbidites are made up of material
Deep-water fine-grained sediments. facies models dominantly in the silt and clay size grades (i.e. over 50~ less than 63 /~m grain size). They are widespread in the deep sea and, volumetrically, the most important of our facies groups (Piper 1978). They occur as very thin to very thick beds that have been deposited rapidly (from a few hours to a few days) from a single resedimentation event. The chief criteria that can be used to distinguish them from other facies in the deep sea which they may resemble, include: (1) a regular vertical sequence of sedimentary structures commonly associated with a positive grading; (2) the presence of sedimentary structures indicating rapid deposition, with bioturbation restricted to the tops of beds; (3) compositional, textural or other features which indicate that they are exotic to their depositional environment. With high resolution acoustic profiling systems, fine-grained turbidites show good penetration and appear well stratified, in contrast to the hard, often irregular, reflective bottom found in coarser turbidites. Fine-grained turbidites occur in a variety of environments that can be distinguished in seismic reflection profiles, such as levees and ponded basin plains. They may pass laterally into coarser turbidites that often fill channels and appear as discontinuous strong reflectors in seismic reflection profiles. In Bouma's (1962) classical sequence for sandmud turbidites, all fine material was classed as a featureless E division. Later work distinguished between turbiditic and pelagic mud (Kuenen 1964; Van der Lingen 1969) and Piper (1978) further subdivided the turbidite mud into El, E2 and E3 structural divisions. Fine-grained turbidites that occur alone are, in some cases, the distal equivalents of thicker-bedded, coarser-grained turbidites, but they may also occur without the proximal sand-mud 'parents'. On the basis of grain size, internal organization and composition we recognize four distinctive facies within this group: silt turbidites, mud turbidites, biogenic turbidites and disorganized turbidites. Each of these can best be described in terms of a separate facies model. Silt turbidites
6I 3
Well-documented examples include those of the Aleutian Trench (Piper 1973), the Indus fan (Jipa & Kidd 1974), the Nile cone (Maldonado & Stanley 1976), the Antarctic continental rise (Piper & Brisco 1975), the Zaire fan (van Weering & van Iperen, this volume) and the south-east Angola Basin (Stow 1984c). Similarly, many ancient slope, fan and basin plain successions comprise interbedded siltstone turbidites and mudstones (e.g. Lundegard et al. 1980; Pickering, this volume; Stow, Wezel et al., this volume). Facies model (Fig. 2)
Silt turbidites commonly exhibit the same suite of structures (Piper 1978; Kelts & Arthur 1981) as the thicker-bedded, classical sandy turbidites described by Bouma (1962). A complete sequence from top to bottom would be (Fig. 2): F hemipelagic or pelagic sediment, biogenic, bioturbated E mud, graded, commonly bioturbated D fine silt, parallel-laminated alternating silt and minor clay, often with synsedimentary deformational structures, graded C medium silt, cross-laminated, rarely convolute, graded
PELAGITE
TURBIDITE MUD ___ graded + laminated
D
TURBIDITE SILT graded, fine, parallel-laminated
C
.graded, medium cross-laminated
B
_+ graded, medium parallel-laminated
In distal turbidite environments, silt beds ( > 70~ I massive, medium-coarse, silt-sized particles) are more abundant than sands E A poor or no-grading and commonly occur as thin or medium-bedded ~' sharp -F scoured base turbidites. Early work demonstrated a change from sands to silts moving basinwards from fans to abyssal plains (e.g. Horn et al. 1971; Normark FIG. 2. Silt turbidite facies model. Structural divi& Piper 1972), and there have since been many sions follow Bouma (1962) as modified by van der descriptions of silt turbidites from the deep sea. Lingen (1969)
6I 4
D.A.V. Stow and D.J.W. Piper
FIG. 3. Photographs of fine-grained turbidite facies. Width of sections approximately 7 cm. (a) Silt turbidites, upper Cretaceous, DSDP Site 530A, Angola Basin. (b) Silt and mud turbidites, upper Jurassic, Brae slope apron, North Sea. (c) Mud and ooze turbidites (3 core sections), Plio-Pleistocene, DSDP Site 530B, Angola Basin. (d) Calcirudite-calcilutite turbidite, upper Cretaceous, DSDP Site 530A, Angola Basin. (e) Disorganized silty-mud turbidite, Pleistocene, DSDP Site 530B, Angola Basin. (f) Silt laminated mud turbidites, Cambro-Ordovician, Halifax Formation, Nova Scotia (g) Calcarenite-calcilutite turbidite, upper Cretaceous, Scaglia Rossa Formation, Italy.
D e e p - w a t e r f i n e - g r a i n e d sediments." f a c i e s m o d e l s B medium silt, parallel-laminated, graded A medium-coarse silt or sandy silt, massive, poor or no grading, some floating clasts, minor scouring at base The interpretation and origin of this sequence, in terms of flow decline during deposition from a single turbidity current event, we suggest is similar to that proposed for the Bouma sequence in sandy turbidites (Harms & Fahnestock 1965). However, the complete sequence is very rarely found (e.g. Fig. 3). More commonly, base-cut-out sequences occur comprising CDE and DE divisions (graded sorted silts and laminated silts respectively of Piper, 1978) in beds from 1 to 10 cm in thickness. Thicker-bedded and coarser silt top-cut-out (AB and B) or mid-cut-out (AE) sequences occur more rarely (the ungraded massive silts of Piper, 1978).
615
Facies model
From this large amount of data it is possible to synthesize an ideal facies model for mud turbidites (Fig. 4). Piper (1978) proposed the subdivision of Bouma's (1962) E division in to three parts, giving from top to bottom: F hemipelagic or pelagic sediment E3 ungraded mud E2 graded mud E1 laminated mud D laminated sand and silt Apart from the topmost E3 division, mud turbidites commonly show slight but distinct positive grading in both grain size and composition. Textural grading is best observed as a progressive decrease in maximum or mean size through successive silt laminae (El division) (Piper 1972b), or as a decrease in silt to clay ratio through the E2 division. A wide range of compositional grading has been observed, including Mud turbidites an upwards increase in micas, various clay Deep-sea terrigenous successions are commonly mineral species and organic carbon, and upward dominated by muds and, in many settings, decrease in heavy minerals, quartz and foraminibetween 50% and 80% of this is ofturbidite origin fera (Rupke & Stanley 1975). Carbonate is known (e.g. Hesse 1975; Piper 1978). The features that to show either an increase or decrease depending characterize mud turbidites are subtle and fre- on the type and grain size of the carbonate and quently overlooked, especially when the turbi- associated components. Colour changes comdites are interbedded within a thick monotonous monly mirror the textural and compositional hemipelagic mud sequence. However, clear exam- grading and provide the simplest method for ples have been described from both the east and visual identification of mud turbidites. Stow (1977) and Stow & Shanmugam (1980) west Mediterranean (Rupke & Stanley 1974; Bartolini et al. 1975; Got, this volume), the have shown that there is a further hierarchy of Cascadia Channel (Griggs & Kulm 1970) and structures within the graded laminated mud and Astoria fan (Nelson 1976) in the eastern Pacific, silt of Piper's scheme, many of which can be the Angola Basin (Stow 1984c), the Northwest helpful as diagnostic criteria in describing mud Atlantic Mid-Ocean Channel (Chough & Hesse turbidites. The complete sequence from top to 1980), the Laurentian Fan (Stow 1981) and the bottom is as follows (Fig. 4): Cap Ferret fan in the north-east Atlantic (Cremer P pelagite or hemipelagite, bioturbated 1981, 1983). T8 turbidite (+ part pelagite), microbioturIn ancient sequences exposed on land, weatherbated ing and fracturing often make it impossible to T7 ungraded mud, occasionally with silt discern characteristic structures in mudstones pseudonodules and shales. Where preservation has been good T6 graded mud, often with dispersed silt lenses and rocks have been suitably smoothed, as on Ts wispy convolute silt laminae in mud wave-cut platforms, the same features as deT4 indistinct, discontinuous silt laminae in scribed for modern mud turbidites may be even mud more clearly demonstrated. Examples range from T3 thin, regular, continuous parallel silt the Precambrian slope facies of northern Norway laminae in mud (Pickering 1982a, this volume), through succesT2 thin, irregular, slightly lenticular silt sions from the Cambro-Ordovician (Piper 1972a; laminae in mud, often with low-amplitude Stow, Wezel, et al., this volume), Silurian (Piper climbing ripples 1972b), Devonian-Carboniferous (Hall & StanTI thick mud layer, often with thin convolute ley 1973; Lundegard et al. 1980), Triassic (Hicks silt laminae 1981) and Cretaceous (Ingersoll 1977) to the To thick, basal, lenticular, silt lamina, often well-known Cenozoic turbidite formations of the with fading-ripple top, microlaminated inItalian Apennines (e.g. Mutti 1977; Mutti et al. terior and scoured, load-cast base 1978; Ricci Lucchi 1978, 1981). The complete sequence is interpreted as a single
616
D.A.V. Stow and D.J.W. Piper
P
PELAGITE/HEMIPELAGITE
1-8
bioturbation, microbioturbation
F --
E3
Tr
E2
T6
UNGRADED TURBIDITE MUD
,~
GRADED TURBIDITE MUD
~ h,~
-I- silt lenses
~ . . . . .. .~176
,,
,~176
... ..~176 .~176176
T~
9. o - .
.~
..........
wispy silt. laminae
o...-o.
indistinct silt laminae
la3
.~176 ........
El
T3
regular parallel silt laminae
T2
irregular~ thin lenticular silt laminae
Tt
convolute silt laminae
To
thick basal lenticular silt lamina
5,,}-r'~a3 .~cr (_9t--
FIG. 4. Mud turbidite facies model. Structural divisions after Piper (1978) and Stow (1977). depositional unit from a large fine-grained turbidity current (e.g. Stow & Bowen 1980). Internal lamination in the basal layer (To) and migrating ripple lamination indicate periods of tractional movement during deposition of the coarser silt grains. The silt ripples with muddy troughs (fading ripples) are deposited from turbidity currents containing a high proportion of claysized material. This commonly also gives rise to a thick mud layer (TI), with convolute silt laminae formed either by loading into the soft mud or by incipient ripple development. At a lower current velocity and with less silt available the very thin laminae of low-amplitude, long wavelength ripples (T2) are formed. Continued waning of the current velocity as the flow passes results in the overall positive grading observed (T3-T4) in which the alternation of silt and mud laminae is believed to be due to the depositional sorting of silt grains from clay flocs caused by increased shear in the bottom boundary layer (Stow & Bowen 1978, 1980; but see Hesse & Chough 1980). The more homogeneous mud unit (T7) at the top of the sequence comprises the finest silt and clay which
was not so effectively sorted into distinct laminae because of the lack of silt and the very fine grain size.
Variability Although a standard structural sequence can be identified, a wide range of variations is possible and, as with the Bouma sequence for sandy turbidites, the complete set of divisions is rarely present in any one bed (Figs 3 and 5). In many cases the beds are relatively thin ( < 10 cm), averaging 2-5 cm, and comprise only the upper parts of the sequence (EzE3 or T4-Ts, base-cut-out units), or lower parts of the sequence E 1 or T0-T4, top-cut-out units). In the extreme base-cut-out case we have a mud or clay turbidite with no trace of silt lamination, whereas the top-cut-out case is gradational to a silt turbidite. In other cases, only the middle parts of the sequence occur (E2 or Tz-Ts), or the middle parts are cut out to give an E IE3 (T07s) bed analogous to a Bouma AE turbidite. Repetition of these different bed types in any one area can lead to the development of a
Deep-water fine-grained sediments." facies models .-..-
617
+..'.:.. : : . . : . . - : . : : - .
-. ~...
.......
i"'
- .
.
.
.
.
.
T6
.
. . .....
.
..., .....,
.. .. ....
9. . , ,
"5""''57....
.......... ,. .....
...
..........
TZ,Z;. . . . . . .
I2
FACIES3-
... . . . . . . . .
FACIESl
_.--
~ ' J ~b~O~t~qctc~
T~
IDEAL SEQUENCE
""~" 17
CO?,
FACIES4 - ~ ~ c)UT S,o
T6
r-7 _.___.
T3
FIG. 5. Examples of variability in mud turbidites: typical base-cut-out and top-cut-out sequences (after Stow, Alam & Piper, this volume). distinctive facies, such as thick lenticularly laminated silts and thin parallel-silt-laminated mud, in which individual turbidite units may not be readily distinguished. Thin-bedded turbidites of these various kinds have been widely reported from both modern and ancient successions (e.g. Piper 1978; Nelson et al. 1978; Stow & Shanmugam 1980; Kelts & Arthur 1981; Hill 1981 and this volume; Chough, this volume; van Weering & van Iperen, this volume). In some cases, the thick basal laminae of a graded laminated unit is preceded by one or two thinner, finer grained silt and mud laminae. These appear to be precursors of the main depositional event. In other cases, there is variation in the sequence of To to T8 divisions through a single unit, perhaps related to an instability in the flow conditions. Both medium and thick-bedded mud turbidites have also been described, commonly 10-50 cm thick but ranging up to several metres (Fig. 3) (e.g. Rupke & Stanley 1974; Piper 1978; Blandpied & Stanley 1980; Stanley 1981; Stow 1984c). These show the same sequence of structures (El-E3, T0-Ts) but each division is vertically expanded. They probably result either from large muddy turbidity currents derived entirely from fine-grained sediments (e.g. from a slump on a
muddy upper slope), or from the ponding of a muddy turbidite tail in an enclosed basin (Wezel 1973; Bowen et al. 1984). Stanley (1983) suggests that slow (hemipelagic) settling from detached turbidity currents in a well stratified basin is an important process in the accumulation of beds that he describes as unifites. Similar slow settling from thick dilute turbidity currents may be important elsewhere (e.g. Stow & Bowen 1980). However, there are generally no diagnostic features that allow recognition of depositional processes involving turbidity current detachment. We recognize that the upper parts of mud turbidites may have been deposited by a variety of turbidity-current-related processes.
Biogenic turbidites Biogenic pelagic sediments are very widespread in the open ocean and, where terrigenous input is minimal, along the continental margin. In areas of topographic relief and/or tectonic activity, such as mid-ocean ridges, seamounts and other submarine highs, resedimentation of pelagic siliceous and calcareous oozes occurs via slumping, debris flows and turbidity currents (e.g. Kelts & Arthur 1981). Off carbonate platforms and reef margins, resedimented biogenic material derived
618
D.A.V. Stow and D.J.W. Piper
from shallow water is also common. Calcirudites, calcarenites, calcidebrites and carbonate slump deposits have been described from many slope and basinal systems (e.g. McIlreath & James 1979). As with terrigenous mud turbidites, finegrained carbonate turbidites are not always readily distinguished from associated pelagic and hemipelagic facies, except where deposition has occurred below the carbonate compensation depth (Hesse 1975). Several authors besides Hesse have encountered similar problems in making clear facies distinctions (e.g. Wilson 1969; Carrasco 1977; Cook & Taylor 1977; Enos 1977; Reinhardt 1977; Homewood & Winkler 1977; Stow, Wezel et al. this volume; Faugeres et al. this volume; Heath & Mullins, this volume). In other cases, more definitive interpretations of carbonate turbidites have been made for a variety of Palaeozoic to Recent limestones (e.g. Van Andel & Komar 1969; Thomson & Thomasson 1969; Davies 1977; Anatra et al. 1980; Kennedy 1980; Kelts & Arthur 1981; Faugeres et al. 1982). Siliceous turbidites rich in diatoms, radiolarians, sponge spicules and other siliceous organisms are less well known than the carbonate equivalents. However, they have been encountered at several DSDP sites in the North Atlantic (Beall & Fisher 1969; Peterson et al. 1970;
Kagami 1979; McCave 1979) and in the Gulf of California (Curray et al. 1980).'A few ancient examples have also been documented from the Mediterranean area (Nisbet & Price 1974; Kalin et al. 1979), Japan (Imoto & Fukutomi 1975) and North America (Folk & McBride 1978). There is not always a clear compositional distinction between the different biogenic turbidites so that all mixtures of carbonate, silica and clay materials can occur. In addition to relatively pure carbonate and siliceous ooze turbidites, Stow (1984c) describes marl, 'sarl' (siliceousclayey) and 'smarl' (siliceous-calcareous-clayey) biogenic turbidites from the south-east Angola Basin (terminology of Dean et al. 1984). Facies m o d e l
Because there are close similarities between the numerous descriptions of calcareous and siliceous fine-grained turbidites and because many of them comprise various admixtures of these components, it seems appropriate at this stage to propose a unified facies model for biogenic turbidites irrespective of composition. We have used a modified Piper (1978) sequence for mud turbidites, including an E/F division to emphasize the extremely gradational transition from turbidite to pelagite that commonly occurs (Fig. 6).
PELAGITE
E/F
BIOGENIC TURBIDITE-PELAGITE reverse-graded bioturbated
E3
BIOGENIC MUD ungraded bioturbation increases upward
E2
FIG. 6. Biogenic turbidite facies model. Structural divisions introduced here, modified from Piper (1978) and Stow, Wezel et al., (this volume).
BIOGENIC MUD slightly-graded isolated burrows
BIOGENIC MUD-SILT graded parallel diffuse lamination broad scoured sharp base
Deep-water fine-grained sediments: facies models The detailed structural divisions of the Stow (1977) sequence have not yet been recognized in biogenic turbidites, whereas the twofold unifitehemipelagite division of Stanley (1981) appears too simple. The complete sequence from top to bottom is: F hemipelagic or pelagic sediment, bioturbated E/F mixed turbidite and pelagite, reversegraded, extensively bioturbated E3 ungraded, very fine-grained, very homogeneous, isolated burrows E2 slightly graded, fine-grained to very finegrained, otherwise homogeneous, rare isolated burrows E1 graded, laminated, laminae horizontal and indistinct or diffuse, alternation of coarser (biogenic) and finer (clay-biogenic) material, thicker basal lamina of sandy-silty biogenic debris with sharp scoured base. Textural grading is most evident through the graded laminated division (El), but is very slight through the E2 and absent in the E3 divisions, which are commonly very fine-grained. The background pelagic or hemipelagic sediment, comprising large unbroken forams, diatoms and other planktonic organisms, is often coarser grained than the E2-E3 turbidite divisions. The gradual upward transition from turbidite to pelagite, therefore, shows a marked reverse grading. Where the components are of mixed types, compositional grading is the most striking. There has been little experimental work on the relative hydraulic equivalence of different biogenic particles (e.g. Berger & Piper 1972) but empirical data from descriptions of biogenic turbidites suggests that forams and shell debris alternate with terrigeneous clay-rich laminae in the lower divisions, whereas diatoms and the finest carbonate material (e.g. nannofossils) are concentrated towards the top. This sequence of structures and associated grading can be interpreted in terms of deposition from a waning turbidity current, in the same way as for terrigenous mud turbidites. The coarsest material deposits first as a thin layer at the base. Shear-sorting together with a waning current velocity results in the graded laminated division (Piper 1972b; Stow & Bowen 1978, 1980), in which the rather diffuse lamination indicates a poor separation of silt-sized biogenics from claysized material. This may be, in part, a function of the floc composition, with floc strength less than in clay-rich sediment. The absence of structures and poor or absent grading in the E2-E3 divisions, the extended transitional division, E/F, and
619
the extent of burrowing into the turbidite, all appear to suggest that there is an even less clear distinction between the processes responsible for deposition of the turbidite and pelagite facies in biogenic than in terrigenous systems (Heath & MuUins; Crevello et al., both this volume). Stow, Wezel et al. (this volume) suggest that finegrained 'pure' carbonate carried into a basin by a turbidity current is more likely to disperse into the water column than the equivalent terrigenous material, perhaps because it is less prone to forming strongly-bound flocs. Thus, the finegrained low-concentration tails of carbonatecharged turbidity currents mix with surface and mid-water suspensions derived in situ or from the basin margins and settle slowly through the water column. The biogenic turbidite will therefore grade imperceptibly upwards into pelagite. Whether or not a similar reasoning applies to siliceous biogenic turbidites is not certain. However, it seems less common that both the background pelagic sedimentation and the resedimenting turbiditic material are of equivalent siliceous composition, so that there is mostly a marked compositional distinction between the pelagite and turbidite facies. Variabilit),,
Examples of variability of biogenic turbidites are illustrated in Fig. 3. Thick-bedded (commonly > 1 m), fine-grained, biogenic turbidites showing the complete structural sequence (E 1-E/F) occur in two main settings. They are found as relatively proximal deposits on platform slopes and ridge flanks where the source material is entirely finegrained (e.g. Stow 1984c). Thinner-bedded basecut-out turbidites (E2-E/F, E3-E/F and E/F) occur more distally (e.g. Bartolini et al. 1975). They also occur as more distal deposits in small ponded basins (e.g. Wezel 1973; Stanley 1981) where the original turbidity current may have already deposited its coarser load. There are many examples of thick fine-grained caps to even thicker (up to 10 m) coarse-grained biogenic debrites and turbidites. Complete gradational sequences do occur (e.g. Faug6res et al. 1982; Gonthier et al. 1982; Stow 1984b,c), but more commonly the basal units are variously missing so that a thin to thick fine-grained bed overlies a much reduced biogenic sandy or gravelly layer. The contact between the two is often sharp or apparently erosive. The other main variable besides structural sequence and bed thickness is composition. Carbonate mud turbidities may be rich in nannofossils (e.g. Kennedy 1980), resedimented periplatform ooze of mixed composition (e.g. Crevello &
D.A.V. Stow and D.J.W. Piper
620
Schlager 1980; Heath & Mullins, this volume), or mixed planktonics and indeterminate lime mud (Stow, Wezel et al., this volume). Every gradation of mixed carbonate-elastic mud turbidites occurs (see examples in Cook & Enos 1977; McIlreath & James 1978), with the characteristic features of the facies changing accordingly. Gradations between siliceous and calcareous biogenic turbidites are also well-known (e.g. Kelts & Arthur 1981; Stow 1984c). Almost pure diatomaceous turbidites have been described from the Gulf of California (Curray et al. 1979), radiolarian chert turbidites from Greece (Nisbet & Price 1974), and mixed siliceous turbidites from many parts of the world (e.g. Kelts & Arthur 1981; Chough, this volume).
Disorganized turbidites Disorganized turbidites (Fig. 7) show only poor diffuse positive grading with bioturbation, if present, concentrated towards the tops of beds. Otherwise they are structureless. They may be dominantly silt grade, clay grade or mixed and have variable admixtures of biogenic and terrigenous material. However, separation into distinct laminae of different grain size or composition has not occurred and the sorting is therefore
~Z~'~
~
poor. They have a sharp, often scoured base and a gradational, indistinct top, occurring in beds from 5 cm to several metres in thickness. Disorganized fine-grained turbidites of this sort are relatively widespread, occurring both as the dominant turbidite type in an area and as isolated examples within a suite of more organized thin-bedded turbidites (Fig. 3). However, their interpretation is not always clear. In some cases they may be a part of one of the other turbidite sequences, such as the massive silts and ungraded muds described by Piper (1978). In other cases they may result from the ponding of large turbidity currents in small basins (e.g. Wezel 1973; Stanley 1981). In proximal slope or shelfedge settings, disorganized turbidites may be deposited by turbidity currents that have not matured sufficiently during flow to allow some internal sorting of their loads (e.g. Ballance et al., this volume; Hill, this volume). In more distal settings, where the associated turbidites are wellorganized (e.g. van Weering & van Iperen, this volume), very rapid deposition perhaps due to a local topographic effect may cause deposition of isolated disorganized beds. Chough & Hesse (1980) interpret such thin poorly-sorted layers, intercalated with thinly laminated fine-grained turbidites on the levees of the Northwest Atlantic Mid-Ocean Channel, as resulting from spill-over of the head rather than body of a turbidity current.
PELAGITE
Contourites u
poor or diffuse grading bioturbefion near top sfrucfureless poor sorting
rarefloatingclasts sharp+ scouredbase
~r A
m FIG. 7. Disorganized turbidite faciesmodel (this paper).
Contourites are a volumetrically significant facies in the present-day deep sea, where they can occur in enormous sediment drifts that are equivalent in size to some larger submarine fans, but they have proved extremely difficult to identify in the ancient record. This is perhaps because they have a low preservation potential, since almost all are developed on oceanic crust. Contourites are sediments deposited or reworked by currents that are persistent in time and space and flow mainly along-slope in relatively deep water. Deposition is controlled by bottom currents driven by thermohaline circulation or, more rarely, by wind-forced surface currents, rather than by gravity-driven turbidity currents. Johnson et al. (1980) and Halfman & Johnson (this volume) have extended the definition of the facies to include lacustrine contourites from Lake Superior. The chief sedimentary criteria that can be used to distinguish contourites from other deep sea facies include: (1) an irregular vertical arrangement of facies types and structures with both negative and
Deep-water fine-grained positive grading, but no regular structural sequence; (2) evidence for more or less continuous bioturbation that has kept pace with deposition, but with the 'ghosts' of current-induced structures remaining; (3) compositional, textural or other features that indicate a combined in situ and exotic origin. For present-day examples there are a range of non-sedimentary criteria that provide important additional evidence for a contourite interpretation. These include the measurement of bottom currents and nepheloid layers, the identification of large and small-scale current-induced morphological features, and certain seismic and stratigraphic characteristics. The acoustic character of contourites using high-resolution profiling systems is dependent on grain size variation and on the presence of intermediate-scale bedforms. In seismic reflection profiles, both acoustically transparent and well-laminated contourite drifts have been described (e.g. Scrutton & Stow 1984). The nature of drift growth and the positive accumulation features (large-scale mud waves, sediment drifts) are diagnostic. There are other bottom currents in the deep sea, such as the up and down-canyon currents measured by Shepard et al. (1979), and these may deposit sediments with very similar characteristics to the contourites we describe. There have been very few descriptions of sediments that can be related clearly to bottom currents of this type, so we have not considered them further in this paper. According to Heezen et al. (1966) and Hollister & Heezen (1972), the thin laminated silt and fine sands on the continental rise off eastern North America are the typical contourite facies. Subsequent work on this margin has shown that the distinction between contourites and interbedded facies is not so clear cut (Stanley et al. 1979; Stow 1979a; Shor et al., this volume). Piper & Brisco (1975) describe a different set of characteristics for silty contourites on the Antarctic margin. Based largely on evidence from coring and drilling the major sediment drifts in the North Atlantic, Stow & Lovell (1979) and Stow (1982) have characterized two main contourite facies: (1) muddy contourites, that result from deposition by bottom currents; and (2) sandy contourites, that result from a combination of deposition and reworking by bottom currents. To these we might add a third, less common facies type: (3) gravel-lag contourites, that result from erosion and winnowing of fines by powerful bottom currents (Carter & Schafer 1983).
sediments." facies models
621
Faug~res et al. (1984) and Gonthier et al. (this volume), from a study of the Faro Drift contourires in the Gulf of Cadiz, point out that there is in fact a continuum of contourite facies from the very fine clay grade to the coarse sand and gravel grades. They recognize an additional intermediate mottled silt and mud contourite facies as well as a silty-sandy facies (rather than a strictly sandy facies). We describe two contourite facies models in this paper, one for muddy contourites and one for silty or fine sandy contourites, and then discuss their occurrence in irregular vertical sequences.
Muddy contourites Muddy contourites include a range of deposits at the fine end of the grain size spectrum, and appear to be the dominant contourite facies in the deep sea, making up over 75% of the main contourite drifts. They have been described from some 15 separate drifts in the North Atlantic (see reviews by Stow & Lovell 1979; Stow & Holbrook, this volume) and show a consistent set of sedimentary characteristics (Fig. 8) that are similar in many ways to hemipelagic and pelagic sediments. There have been no specific identifications of muddy contourites in ancient rocks, although several authors have interpreted ancient slope-rise sequences where they believe bottom currents to have been active (e.g. Bein & Weiler 1976; Anketell & Lovell 1976). On first appearance the facies is monotonous, homogeneous and structureless, but closer inspection reveals important structural features. There are no very distinct beds, but both positive and negative grading between more clayey and more silty-sandy sediment is commonly present over a few centimetres to a few tens of centimetres. The contacts between clay-rich and siltrich parts are mostly gradational, less commonly sharp and erosive. Primary lamination of either silt or clay is rare, and may be distinct and planar or indistinct and wavy. Where not laminated, the muds are thoroughly bioturbated. The silt fraction is often concentrated in irregular pockets and lenses that show some horizontal alignment, but that appear to have been broken up by bioturbation (Faug~res et al. 1984). Texturally, these mud contourites are siltyclays with up to about 10 or 15% sand fraction and a mean size that varies from less than 5 ~tm to about 40 ~m. They are mostly poorly sorted. The shape of the grain size curves (Rivi~re 1977) varies from logarithmic with slight hyperbolic tendency for the finer-grained samples to parabolic for the more silty samples (Gonthier et al., this volume).
D.A.V. Stow and D.J.W. Piper
622
d
~ - ~,r ~'ll ~ _ ~
""
_
Structure
'~ - -
9
,.-w~.)~ ]
_ 6 ~i|
irregular horizontal lense% layers + wispy
|
rare parallel lamination
--G-
_
[email protected]~Ts 9 E)
w Ior
D oI.-
bedding poor or absent
"_~
. ~
,,,.
.
,, ,~,'-.;
, ' . . " ,~,,--.~ ,' . ~ : ,~.~
9 -. ~ -
"~."""~E~-, ~'. ,.
Z tJ
cm
Texture dominantly silty mud 0-15% sand sized poor to moderate sorting irregular vertical variation
__~
,. ,,. L.....-"v ) I0
mostly homogeneous Jr thoroughly bioturbated~
~..~.'~.. "9- ' . " ~ ' - c . - ~.. ~/
r 1
Composition mixed biogenic-terrigenous mixed fragmented-whole biogenics mixed planktonics-deep benthonics
o
FIG. 8. Muddy contourite facies model (after Stow & Lovell 1978; Stow 1982).
In terms of composition, many muddy contourites are mixtures of biogenic and terrigenous material in variable proportions depending on the distance from land, source of supply and so on (Laughton et al. 1972; McCave et al. 1980; Stow & Holbrook, this volume). The biogenic fraction includes calcareous and siliceous planktonic and benthonic forms. Some are broken and ironstained. The benthonics are usually m situ deeperwater forms rather than shallow-water derived species. The terrigenous fraction is commonly dominated by fine quartz and clays. Other contourites may have a more or less pure pelagic composition (Shor & Poore 1978; Kidd, pers. comm. 1983) and some appear to be dominated by volcanogenic material from mid-ocean ridges and seamounts (Taylor et al. 1975; Van Stackelberg et al. 1979).
Silty-sandy contourites Silt and fine sand contourites appear to be less abundant in the present deep sea as distinct beds. Many of the muddy contourites described from contourite drifts contain intervals that are more sandy and silty, some of which might be termed sandy muddy silts. In other cases, cleaner sandy and silty beds have been documented (e.g. McCave et al. 1980; Faug~res et al. 1979, 1984; Gonthier et al., this volume). The origin of the sharply-defined silt laminae and cross-laminated fine sands on the continental rise off North America remains enigmatic (e.g. Hollister & Heezen 1972; Stow 1979; Shor et al., this volume), and so we have not included the features of these deposits in our synthesis of a silty-sandy contourite facies model (Fig. 9). Most of the ancient
623
Deep-water fine-grained sediments."facies models n
M UD
I ~'~'---~:='I I -'~ ~"~-C'~'~
bioturbated laminated
a w
cross-laminated
SILT
c~
bioturbated
silt mottles + lenses irregular horizontal alignment
MOTTLED SILT + MUD
w --> I--
bioturbated
CO
0(3_
massive
irregular sandy pockets
SANDY SILT
bioturbated contacts
sharp to gradational
silt mottles~ lenses + irregular layers bioturbated contacts variable
MOTTLE O SILT -I- M U D
----
tm LI.I a
l-
oE
'Ot
MUD
-F . ~-- ~_~a~--
,~, ~
-
1
W
bioturbated
0 FIG. 9. Typical arrangement of contourite facies in negatively to positively graded couplets (after Gonthier et al., this volume; Faug6res et al., 1984).
sandy contourites that have been described have been identified on the basis of the Heezen-Hollister criteria (see reviews in Stow & Lovell 1979; Lovell & Stow 1981), and may in fact not be contourites, Silty or sandy contourites occur in irregular beds from 1 cm to about 20 cm in thickness. The top and bottom contacts of beds may be sharp and relatively flat, erosional or completely gradational. In many cases they display no primary structures apart from irregular concentrations of coarser material and slight positive or negative grading. In other cases, primarycross-lamination and parallel lamination has been preserved, but no consistent sequence of structures has been observed. Secondary bioturbation, continuous throughout the beds, is the most common feature, ranging from large distinct burrows to small irregular mottling (e.g. Chough & Hesse, in press), Medium to coarse-grained silts with up to 4 0 ~ sand and less than about 10% clay are the most common, but fine sandy contourites are also
known. They are mostly moderately well-sorted but with a distinct fine tail evident on grain size curves. According to Gonthier et al. (this volume) the shapes of the curves tends towards parabolic or to a combination of hyperbolic (coarse tail) and parabolic (fine tail). Compositionally, silty-sand contourites are similar to muddy contourites in comprising mixed biogenic-terrigenous material (Faug6res et al. 1984). The larger biogenic grains are commonly fragmented and iron-stained, and there is clearly less clay material. More or less pure foraminiferal contourite sands have also been described (Faug6res et al. 1979; Shor et al. 1980).
Contourite 'sequence' There is no regular sequence of structures within contourites as there is in the various turbidite models described. However, a distinctive feature ofcontourite successions appears to be the presence of both negatively-graded sequences in
624
D.A.V. Stow and D.J.W. Piper
which the grain size increases, and positivelygraded sequences in which the grain size decreases upwards (Gradstein et al. 1982; Faug6res et al. 1984; Gonthier et al., this volume). Both sequences may be from about 10 to 100 cm thick and occur separately or as a combined negativepositive unit (Fig. 9) which shows, from top to bottom: - - homogeneous mud facies --mottled silt and mud facies - - mottled facies with silt layers - - silt-sand facies --mottled facies with silt layers --mottled silt and mud facies --homogeneous mud facies There is considerable variability in this sequence as it is observed in different contourite successions, particularly in terms of its thickness, its completeness and its symmetry. Such variations in grain size of the facies and in the associated structural and compositional characteristics can be interpreted in several different ways. Faug6res et al. (1984) relate the Faro Drift sequences to long-term variation in velocity of the transporting current. A complete negative-positive sequence represents a gradual increase, a maximum and then gradual decrease in the average current velocity at a given site. The time-scale for deposition of the sequence would be of the order, say of 1000 to 30 000 yrs, and the mean velocities might vary between about 5 and 25 cm/s (Gonthier et al., this volume). Alternatively, such sequences might reflect variation in the grain size and/or biogenic content of material supplied to the system. We emphasize, again, both the subtlety of the sedimentary features observed in contourites and their variability (Fig. 10a-c). On the one hand they are similar to those of some indistinct fine-grained turbidites, and on the other hand they are almost indistinguishable from hemipelagite characteristics. This fact underlines the existence of a process-continuum in which flow velocity, concentration and frequency can all vary, together with sediment supply.
Pelagites and hemipelagites The third major facies group of the deep sea comprises the pelagic and hemipelagic sediments (Hsu & Jenkyns 1974; Jenkyns 1978, in press; Einsele & Seilacher 1982). These are widespread throughout the world's oceans and widely recognized in both modern and ancient successions. They have been deposited primarily by slow settling through the water column in the absence of any s u b s t a n t i a l bottom current or turbidity
current activity. However, many of the processes proposed for the accumulation of hemipelagites involve current-induced slow advection of suspended sediment (Drake et al. 1978). Hemipelagic sediments accumulate on continental margins and in other settings not far removed from terrigenous sediment sources. They comprise a mixture of indigenous biogenic material and silt and clay size terrigenous detritus. They accumulate slowly and are thus intensely bioturbated. True pelagic sediments accumulate in the open ocean and comprise principally skeletal parts of plankton with some admixture of very fine silt and clay, much of which has reached the open ocean by aeolian transport. The proportion of terrigenous material may be increased by dissolution of the biogenic components. Bioturbation is ubiquitous except in anoxic basins. Textural and compositional criteria can be used to distinguish four types ofhemipelagite/pelagite facies (modified after Berger 1974): (l) pelagic ooze, with > 75% biogenics; (2) muddy pelagic ooze ('arl'), with 25-75% biogenics and a terrigenous component predominantly of clay; (3) pelagic clay, with 60% clay in the terrigenous fraction; and (4) hemipelagites, with >5% biogenics and a terrigenous component with > 40% silt. Other minor facies include the purely chemogenic sediments that are composed almost entirely of authigenic minerals, such as ferromanganese nodules and phosphorites. These are not considered further here. The chief distinguishing features of pelagites and hemipelagites include: (1) evidence for low or very low rates of sedimentation and continuous bioturbation (except in anoxic basins); (2) no primary sedimentary structures or other evidence of current-controlled deposition; (3) a mainly uniform composition within any one succession, that may show a regular cyclicity related to climatic or other controls; (4) a variable biogenic component mainly of planktonic tests, a very fine grained, often far-travelled, terrigenous component, and commonly a significant authigenic component. Seismic reflection profiling shows a more or less uniform, acoustically transparent sediment drape over bottom irregularities although there may be a thickening in basins (Moore 1969). Stratigraphically, pelagites and hemipelagites commonly show continuous deposition, although accumulation rates may be as low as one metre per million years and hiatuses may be present.
Deep-water fine-grained sediments."facies models
625
FIG. 10. Photographs of contourite, pelagite and hemipelagite facies. Section widths about 7cm. (a) Silty contourite, Pleistocene, Faro Drift, Gulf of Cadiz. (b) Mottled silt and mud contourite, Pleistocene, Faro Drift, Gulf of Cadiz. (c) Muddy contourite, Plio-Pleistocene, Blake-Bahama Outer Ridge, western North Atlantic. (d) Bioturbated pelagite-hemipelagite, upper Cretaceous, Scaglia Rossa Formation, Italy. (e) Organic-carbon-rich siliceous pelagite (diatomite), Miocene, Monterey Formation, California. (f) Alternating black shade and marlstone, mainly hemipelagic (part turbiditic), Cretaceous, DSDP Site 530A, Angola Basin. (g) Bioturbated siliceous/calcareous pelagite, Plio-Pleistocene, DSDP Site 532, Walvis Ridge.
626
D . A . V . S t o w a n d D . J . W . Piper
Pelagic ooze
Pelagic oozes are most typical of the open ocean basins far from a terrigenous source. They have been the subject of much study over the past 125 years and hence their sedimentary characteristics are well known. Some important syntheses are to be found in the volumes by Hsu & Jenkyns (1974) and Cook & Enos (1977), and in papers by Arrhenius (1963) and Jenkyns (1978 and in press). Specific examples in this volume where pelagic sediments are described as part of the succession, include those of the western and central North Atlantic (Robertson), the north-western Mediterranean (Monaco & Mear), the Miocene of southern Turkey (Hayward) and the Cretaceous-Tertiary of the Apennines (Stow, Wezel et al.).
N r-l-
.J-I
o
I
massive no primary structures bioturbated sandy-silty-clayey bedding poor or absent > 7 0 % calcareous tests
~la-.-L_ _~-I I
-', '~L
I "L- --'-~'---,~
massive no primary structures except rare ash layers bioturbated silty-clay < I 0 % biogenics
Facies model (Fig. 11) One of the chief characteristics of pelagic oozes is their very slow rate of accumulation, commonly from less than 1 mm to 10 mm/1000 yrs, although this can be an order of magnitude higher under zones of upwelling. They are usually, therefore, thoroughly homogenized by bioturbation, and without any primary current-induced structures. A variety of burrow types may be preserved with different assemblages or ichnofacies dependent on different environmental factors, such as water depth, grain size, sedimentation rate and redox conditions (Seilacher 1967; Werner & Wetzel 1982; Wetzel, this volume). Some of the main diagnostic trace fossils are Zoophycos, Chon-
drites, Planolites, Scolicia, Trichichnus, Teichichnus and Lophoctenium. Several tiers of trace fossil assemblages are commonly superimposed on one another (Werner & Wetzel 1982). The grain size of pelagic oozes is largely dependent on the composition of the biogenic fraction. Coccolith plates are very small (clay size), whereas some foram-rich or diatom-rich oozes may have a mean grain size that is in the silt range. The terrigenous component is mainly clay-sized. Full grain size analyses, however, are rarely carried out on pelagic oozes as the hydraulic equivalence of biogenic particles is not wellknown and hence interpretation of the grain size distribution would be difficult. Pelagic oozes are composed dominantly (> 75%) of the tests of planktonic organisms, either calcareous (coccoliths, forams, pteropods) or siliceous (radiolarians, diatoms, silicoflagellates) or a mixture of both (Berger 1974). The other components (Lisitzin 1972) can include very fine-grained terrigenous material (principally quartz, feldspars and clays), volcanogenic debris (palagonite and derived clay minerals),
ferromanganese nodules massive no primary structures bioturbated sandy-silty-clayey bedding poor or absent
> 70%
FIG. 11. Pelagite facies models. Schematically interbedded calcareous ooze, red clay and siliceous ooze, with ferromanganese nodules developed at ooze-clay contact.
authigenic minerals (such as phosphate, barite, zeolites, ferromanganese nodules and coatings), and rare extraterrestrial material. Under normally oxic conditions the organic-carbon content is extremely low, but under anoxic conditions pelagic black shales can contain over 20% organic carbon (Isaacs 1981; this volume; Arthur et al., this volume). The characteristics, composition and distribution of the different types of pelagic ooze are dependent on a number of interacting variables which we will not discuss at length (but see Berger 1970; Lisitzin 1972; Broecker 1974). These include: the water depth and the corresponding carbonate compensation depth; the source and supply of terrigenous and volcanogenic material; surface water productivity and the supply of biogenic material; surface currents and bottom circulation patterns; climate and basin physiography; and physiochemical conditions. Pelagic sediments are perhaps more affected by this range of
D e e p - w a t e r f i n e - g r a i n e d sediments." f a c i e s models different controls than are turbidites or contourites because of their two principal attributes: (1) they are composed mainly of biogenic CaCO3 and/or SiO2 both of which are soluble in sea water; and (2) they settle relatively slowly through the water column and are buried only slowly, and hence are exposed to external factors for a relatively long time period, either in the water column and/or on the sea-floor. The actual processes of settling as single grains or as larger flocs and pellets are discussed at more length in papers by Gorsline, McCave, and Eittreim (all this volume).
Muddy pelagic ooze There is a continuum of facies from pelagic ooze with > 75% biogenic material to pelagic clay with < 25% biogenics. Muddy pelagic ooze (or 'arl', terminology of Dean et al. 1984) is a relatively common intermediate sediment type, with characteristics intermediate between an ooze and a clay. It differs from true hemipelagic sediment in having a dominantly clay-sized rather than siltsized terrigenous component, and in being an open-ocean rather than continental margin facies.
627
Clay minerals are the dominant components, whereas quartz, feldspar and other terrigenous materials are very minor (Arrhenius 1963). Authigenic components include the zeolites, ferromanganese minerals (goethite, micronodules) together with some clays and feldspars. Biogenic material is often very scarce, but a complete gradation exists with muddy pelagic oozes ( > 25% biogenics). Volcanogenic material, essentially palagonite, is present in very variable amounts. In certain environments especially close to mid-ocean ridges or immediately overlying ocean floor basalts, pelagic clays can be highly enriched in a variety of metals and trace elements. Hoffert (1980) has identified four types of pelagic clay based on slight differences in their mineralogical and chemical compositions. These are associated with (1) siliceous oozes, (2) calcareous oozes, (3) volcanogenic material, and (4) none of the above, but with a mixed composition. It is by alteration and dissolution of the different biogenic or volcanogenic components of these other oceanic sediments, together with authigenesis and diagenesis of new minerals, that pelagic clays are formed in a process somewhat analogous to pedogenesis on land.
Hemipelagites Pelagic clays Pelagic clays, also known as red clays, brown clays and abyssal clays, accumulate in the deepest, and most remote parts of the ocean basins. They are particularly well represented in parts of the Pacific Ocean, resulting from dissolution of biogenics and aeolian transport of dust. Their sedimentary characteristics are relatively well known (e.g. Arrhenius 1963; Griffin et al. 1968; Hsu & Jenkyns 1974; Jenkyns 1978; Hoffert 1980). Ancient examples have also been documented (e.g. Audley-Charles 1965).
Facies model (Fig. 11) Pelagic clays have one of the lowest sedimentation rates of all the pelagic/hemipelagic sediments, commonly less than 1 ram/1000 yrs but ranging up to about 7.5 mm/1000 yrs. The rate of growth of ferromanganese nodules and crusts, however, is one or two orders of magnitude lower. As with pelagic oozes, pelagic clays are usually well-oxygenated and thoroughly bioturbated. The trace fossil assemblages are those adapted to the deepest water, and finest grain size. Texturally, they are very fine-grained, clay or fine-silt sized, and poorly to moderately well sorted with an even distribution of grain sizes over a small size range.
Hemipelagic sediments are more typical of marginal oceanic settings where there is a ready supply of terrigenous material. In published descriptions of sediment sequences they are frequently referred to as the background, normal, ubiquitous or interbedded facies, and are often not described in any great detail although they may comprise the greater part of a given sequence. At high latitudes, for example in the Arctic Ocean and Baffin Bay, ice-rafting is a major contributor to fine-grained hemipelagic sediments. Amongst the large volume of literature that simply refer to hemipelagic sediments in this vein there are, however, some authors who have documented the facies characteristics in rather more detail. Particularly useful descriptions of modern hemipelagites include those of Rupke (1975) and Stanley & Maldonado (1979) from the western and eastern Mediterranean respectively, Stanley et al. (1972) and Hill (1981) from Nova Scotian margin, Moore (1974) and Kolla et al. (1980) from the Indian Ocean, and various contributions by Gorsline and colleagues (e.g. Gorsline 1978, 1981) from the basins of the California Borderland. In this volume there are a number of papers dealing, in part, with hemipelagic facies in the modern deep sea (e.g. Monaco & Mear, Chough, Krissek, Thornton, Gorsline et
628
D.A.V. Stow and D.J.W. Piper
al., Faug6res et al., Isaacs, and Auffret et al.; all
this volume). Although more difficult to identify in ancient rocks there have been a number of papers describing hemipelagites from inferred slope and basinal settings (e.g. Hesse 1975; Piper et al. 1976; Ingersoll 1978; Hicks 1981; Pickering 1982b). Papers in this volume by BaUance et al., Bourrouilh & Gorsline, and Pickering also describe ancient hemipelagites. Facies model (Fig. 12)
We are therefore able to summarize the chief sedimentary characteristics ofhemipelagites (Fig. 12). They are commonly homogeneous and structureless, with bedding poorly defined or absent except when it has been accentuated by burial and diagenesis (see below). There are no primary current-induced sedimentary structures such as lamination, ripples or erosional contacts, although a depositional lamination may be preserved under anoxic conditions (e.g. Isaacs, this volume; Thornton, this volume).
Under normal oxic conditions, however, bioturbation is ubiquitous and thorough, often resulting in a completely homogenized sediment with a mottled aspect. Burrow traces are also often preserved, with the same major trace fossil assemblages represented as for the pelagic oozes described above. These similarly depend on water depth, grain size, sedimentation rate and redox state (Werner & Wetzel 1982; Wetzel, this volume). Iron sulphide filaments (Mycelia) and mottles are also a common feature ofhemipelagic sediments. Texturally, hemipelagic sediments are silty clays with 1 to 15% of sand of mainly biogenic origin. They are poorly sorted and show no systematic grading apart from that associated with the compositional cyclicity discussed below. The shape of the grain size cumulative curves (Rivi6re 1977) appears almost uniform or logarithmic, although there may be irregularities due to a typical biogenic input or rare exotic terrigenous material (e.g. ice-rafted debris). Apart from the mixed biogenic-terrigenous aspect of hemipelagites, it is difficult to generalize
Structure massive no primary structures thoroughly bioturbated bedding poor or absent limestone-marlstone cycles
z" '
o_ nn m
Texture
.,=
sandy silty mud very poorly sorted size variation related to biogenic content
I
Composition mixed biogenic-terrigenous mixed planktonics- benthonics biogenic- rich/clay- rich cycles common ice-rafted input at high latitudes
12. Hemipelagite facies model. Rhythmic alternation of biogenicrich and clay-rich intervals. FIG.
o
-
Deep-water fine-grained sediments."facies models further about their composition. The biogenic input may be slight or dominant, calcareous or siliceous depending on surface productivity, carbonate compensation depth, etc. It is commonly of mixed planktonic and in situ (deep-water) benthonic origin. The terrigenous components are mostly uniform in any given area and show very gradual compositional trends towards the source area. Their nature can, of course, be highly varied depending on tectonic, climatic and other factors, and the sediments may also contain a minor to significant fraction of far-travelled (wind-blown, ice rafted, etc.) terrigenous debris. Seismic-reflection profiles of hemipelagic sediments frequently show a draped morphology over bathymetric irregularities suggesting that they were deposited by vertical settling through the water column. Observations of oceanographic processes and of regional patterns of sediment distribution, however, suggest that other rather more complex processes may also have an effect on hemipelagic deposition (e.g. Moore 1969; Damuth & Kumar 1975; Drake et al. 1978; Stanley & Kelling 1978; Karl et al. 1983; Hill & Bowen 1983). In particular, slow lateral displacement of fine suspended sediments in mid-water and bottom water nepheloid layers; sediment dispersion by up and down-canyon normal currents; resuspension and slow diffusion at the shelf break; and sediment creep over gentle slopes all appear to be important processes. These are discussed at more length in papers by Gorsline, McCave, Eittreim, Thornton and Gorsline et al. (all this volume).
Hemipelagite-pelagite cycles Examples of some hemipelagic and pelagic sediments are shown in Fig. 10d-g. An important feature of many, but not all, hemipelagic and pelagic successions is a cyclic alternation of biogenic-rich (pelagite) and clay-rich (hemipelagite or pelagic clay) beds (Fig. 12) (e.g. Dean et al. 1981; Einsele 1982; Einsele & Seilacher 1982; Stow, Wezel et al., this volume). Most of the cyclic sequences that have been described from both modern and ancient successions are limestone-marl cycles with the primary variation being the relative proportions of carbonate and clay, although similar cycles are also known from biogenic siliceous sequences. Other compositional (e.g. organic carbon, trace elements) and textural (grain size) characteristics may mirror the limestone-marl cycle or may vary independently. There is a close correspondence worldwide in the order of magnitude of both Quaternary-Recent and older cycles. They are commonly between 10 and 100 cm thick, with a
629
sedimentation rate of 0.5 to 3 cm/1000 yrs and a periodicity of 20 000 to 100 000 yrs. Possible causes for cyclic variation in the relative abundance of CaCO3 or biogenic silica include: (1) variation in the rate of production of CaCO3 or SiO2 by planktonic organisms; (2) variation in the rate of input of CaCO3 or SiO2 via turbidity currents; (3) variation in the amount of dissolution of CaCO3 or SiO2 either shortly after sedimentation or at depth after burial; and (4) dilution of the biogenic component by variation in the non-carbonate (terrigenous) input. All of these processes have been shown to be important in different cases. The close relationship of the periodicity to the earth's orbital changes and climatic variation suggest that climate is most often the ultimate control, with sea-level changes and fluctuation in the carbonate compensation depth also being important (e.g. Fischer & Arthur 1977). There are rather fewer examples of similar cycles in siliceous biogenic sediments (e.g. Garrison & Fischer 1969; Barrett 1982). Some of these probably have a similar origin to the limestonemarl cycles, but others were more clearly controlled by turbidity current input of the siliceous material (e.g. Nisbet & Price 1974; Folk & McBride 1978).
Environments and facies distribution The various fine-grained facies outlined above are not distributed randomly through the deep sea but occur preferentially in certain environments. They are commonly closely associated and interbedded with each other and with the range of coarser grained facies also found in deep water. Based largely on their occurrence at the present day, we outline here some aspects both of their generalized distribution on a global scale and of their more detailed horizontal distribution within slope, fan and basin plain settings. We then describe their occurrence in characteristic vertical sequences in the ancient record. Our intention here is to highlight only the main aspects of facies distribution and not to attempt rigorous documentation. We believe that more detailed consideration will be a fruitful area for further research.
Global distribution The principal environments of turbidite deposition (both coarse-grained and fine-grained) are the very large areas of continental slopes and rises and oceanic abyssal plains (Fig. 13). Also impor-
630
D.A.V. Stow and D.J.W. Piper e~
.,..a
~A
.=_
t~
O
~,,,~
8 Q
0
O
Deep-water fine-grained sediments: facies models tant are the slopes and floors of smaller marginal seas and land-locked basins. Turbidites are dominant, therefore, around the margins of the continents and particularly those trailing-edge margins that receive most sediment from continental drainage systems and glacial erosion, including much of the Atlantic Ocean, north Indian Ocean, Mediterranean Sea and the circum-Antarctic margin (Inman & Nordstrom 1971). Large amounts of sediment draining from south-east Asia is mostly trapped in back-arc basins or on broad continental shelves of the western Pacific. There is generally less sediment supplied to the coasts along collision margins, although the tectonically unstable and metastable slopes in these areas are important sites of resedimentation via slumping, debris flows and muddy turbidity currents (e.g. Gorsline et al., this volume). Silt turbidites are associated with channels and other conduits across the slope, and are commonly fed to the channel distributary regions on the lower slope (e.g. mid and lower fan lobes) and to more proximal parts of abyssal plains (Piper 1978). Mud turbidites are more widespread, occurring in levee and interchannel areas of the slope and throughout the basin plain. Biogenic turbidites are restricted to regions off carbonate banks and reefs, upwelling zones or seamounts and oceanic ridges, where a biogenic source is available. The distribution of disorganized turbidites is less clearly understood; they occur both in proximal and distal regions of turbidite sedimentation. The chief environments of contourite deposition are closely related to the deep-water thermohaline circulation pattern of the ocean basins and, in particular, to the higher velocity bottom currents that occur principally on the western margins of basins or by the acceleration of flow through restricted passageways. In the North Atlantic, for example (Fig. 14), a number of large sediment drifts can be identified that are made up almost entirely of muddy and silty contourites, and of pelagic or muddy pelagic oozes that have been moulded by bottom currents during deposition. These occur as isolated elongate mounds both in the middle of the ocean and parallel to or projecting from the continental rise, and as irregular dome-shaped mounds near seamounts and other topographic highs. Coarser-grained sand and gravel lag contourites occur in association with muddy contourites in the Straits of Gibraltar and in the western Labrador Sea. On the continental rise of eastern North America contourites are closely interbedded with finegrained turbidite and hemipelagite facies. Pelagic and hemipelagic sediments are inter-
631
bedded to varying degrees with both turbidite and contourite facies wherever they occur. Elsewhere in the deep sea they are the principal facies. Calcareous pelagic oozes are dominant in the Atlantic Ocean and over the shallower mid-ocean ridge parts of the other oceans. Siliceous pelagic oozes occur mainly in two high latitude circumpolar bands, a Pacific equatorial band and under upwelling zones on the western boundaries of the continents. Pelagic clays are confined to the very deepest and central parts of the oceans, in particular the Pacific. Hemipelagites are most abundant near the continents, except where masked by a major input of resedimented facies. At very high latitudes, there is an ice-rafted component within the other facies which, if dominant, gives an ice-rafted hemipelagite facies. Slope aprons, submarine fans and basin plains: horizontal distribution
Slope aprons are morphologically heterogeneous and have a correspondingly complex and irregular distribution of facies (Fig. 15) (Bouma et al. 1978; Doyle & Pilkey 1979; Stow 1984a). On normal terrigenous-supplied slopes there is commonly a mud-line (Stanley & Wear t978) that separates the shallow, higher-energy, sandy shelf facies from the mainly fine-grained slope sediments. Some sand spillover occur along the shelf break as well as the funnelling of coarser sediments down canyons or gullies to isolated depositional lobes. Areas of sediment creep and slumping give rise to resedimented slump and debrite masses and, in some cases, to mud turbidites. On the open slope there is an interbedding of and gradation between fine-grained turbidites, contourites, hemipelagites and pelagites, with the turbidites being more common near channels, the contourites occurring in areas of bottom current activity, and the pelagites increasing in abundance distally. Very rapid progradation occurs off deltas, particularly at times of lowered sea-level. Fluctuations in river discharge may produce graded beds resembling turbidites (see discussion in Stow, Alam & Piper, this volume). Sediment facies include both turbidites and hemipelagites accumulated from suspension fall-out. Both may be little bioturbated because of high rates of deposition. Similar facies occur in high latitudes off active ice margins. At the base of slopes there may be either a smooth gradual transition to the basin plain facies or areas of more positive construction. Isolated channel-fed lobes comprise silt and fine sand turbidites (Piper 1978), slump/debris-flowfed lobes comprise mud turbidites (Stow 1984c),
632
D.A.V.
Stow and D.J.W.
Piper
FIG. 14. Present-day deep-water circulation in the North Atlantic. Areas of bottom-water formation shown in wide stipple. Main contourite drifts shown in close stipple. Contour at 2000 m; mid-ocean ridge in heavy dashes. (After Stow 1982). and contourite mounds or drifts may be constructed parallel to the slope contours. Alongslope trends of sediment characteristics are observed in these drift deposits (Gonthier et al., this volume). Other papers in this volume that discuss facies distribution on modern clastic slopes include those by Hill, McGregor et al., Ballance et al., Krissek, Got, and Gorsline et al (all this volume). Variations on these facies distributions occur on slopes that are controlled by active faulting or diapirism. Resedimented and pelagic biogenic facies are dominant on carbonate slopes (e.g. McIlreath & James 1978; Mullins & Neumann 1979; Heath & Mullins, this volume; Faug6res et al., this volume) and on the flanks of oceanic ridges and seamounts (e.g. Gonthier et al. 1982).
The distribution of facies on submarine fans has received rather more attention in the past than that on slopes (e.g. Mutti & Ricci Lucchi 1972; Walker & Mutti 1973; Walker 1978). Several important papers have addressed specifically the question of distribution of the fine-grained facies (Piper 1978; Nelsen et al. 1978; Stow 1981; Cremer 1981, 1983) so that a relatively clear picture has emerged (Fig. 16). Van Weering & van Iperen (this volume) and Hayward (this volume) also deal specifically with fine-grained facies on deep-sea fans. There is both an elongate and concentric distribution of coarse to fine-grained facies on most fans. Slumps and slides are mostly confined to the slope, upper fan and channel margins. Debrites may be more widespread (Nardin et al.
Deep-water fine-grained sediments: facies models IRREGULAR SLOPE
SMOOTH SLOPE MIXED FACIES
r"
SLUMP MASS
MIXED FACIES
DISORGANIZED TURBIDITES
63 3
DEBRITE MASS
\ ~'9 .'.':..:'. :.. -9 9149 9 9 : ':,, \al__ ~.~. ..-.-......:....-. _ ,~-" _9. . . . . . . . . . . . . . ' . - . .j_ _1_ i~9." ." .'." -.'- . " . . . .a_ .L 9 ..
9
9
,-
.
.
."
-.
|
9
::::. : J
"
...-.... ..-.: : .:. :... ::.. 9 : 9 ;.
~
- ."
9 .'...'.
'
"...,"
.-
".
1
~-~_
"
~
. J .
/ m
9
9 : .... 9149149 9 .
- 9
9 o 9
.
9
tiiii!ii!iiiii 1 MASSIVE SILT
~
TURBIDITE
.Jl_
" . . .
9
9
..-",._,~,, .a.,;~
.,,
.
'
[~=~=~f TURBIDITES
/ I'~('
SILTY CONTOURITES
k,,,~
MUD TURBIDITES 4.- PELAGITES
MUDDY CONTOURITES CHANNEL TO LOBE
k,.,,~~ CONTOURITE DRIFT
FIG. 15. Schematic distribution of fine-grained sediment facies across a typical slope and rise. Note irregular distribution of mixed facies types 9 1979). Coarse-grained turbidites are transported through the channel system and deposited along their length and on the sandy lobes that spread out at their terminations. Fine-grained turbidites are transported either down channels and then laterally by overflow onto the levees and interchannel areas, or as thick unconfined low-density
flows (Bowen et al. 1984). They commonly show both a down-fan and away-from-channel evolution of textural, structural and compositional features. Hemipelagites, pelagites and, in some cases, contourites are interbedded with the resedimented facies in areas of lower energy. Basin plains are the ultimate trap for deep-sea
634
D.A.V.
Stow and D.J.W.
MUD TURBIDITES + PELAGITES ACROSS FAN
Piper SILT TURBIDITES DISTAL CHANNEL + LOBE
iii1
~
I.'.'.'..'.::'3
sea-level
.
5.-:.:. ;..--~" ..."
,,at.
.at.
: . . . . . :-:.:.:
SLUMPS UPPER SLOPE
HEMIPELAGITES q- DISORGANIZED TURBIDITES MID-SLOPE
DOWN-FAN TO ABYSSAL PLAIN
FIG. 16. Schematic distribution of fine-grained sediment facies across a typical large deep-sea fan. Note systematic down-fan and across-fan distribution of mainly turbiditic facies. sediments that have been eroded and transported by currents or that have settled slowly through the water column (Gorsline 1978; Pilkey et al. 1980; Stow, 1984a). They vary widely in their areal extent, depth and shape, including for example, the tiny plains of slope basins and the major oceanic abyssal plains. A single basin plain may be fed from several sources, including channels and fans, the surrounding slopes and surface waters. All sediment facies types are represented in basin plains (Fig. 17), depending very much on
morphological, tectonic, sedimentary and sealevel controls, and the fine-grained facies commonly predominate. Both centripetal and longitudinal facies distributions are observed; turbidites, contourites and hemipelagites/pelagites are often intimately interbedded. Several papers in this volume outline specific examples of basinal sedimentation, including those by Auffret et al., Got, Chough, Thornton, Gorsline et al., Robertson and Crevello et al. (all this volume). Where a basin is fully enclosed and circulation
Deep-water fine-grained sediments."facies models /-..~.
\ "~
'~
~'~'-~-~-~"
~.~"~.
635
_ see-level
. . .:.... 2 - : . )
..
.... ......-._::.-.:_-..
. ..". "!i:i!i:i!i! ~...
:!vif
\ i \ I,, \~
,,
\
hediiPoerl;aginteSed turbidites
.....
.":::::'".'?.'2":':.':'.'."."_.._
::::::::::::::::::::::::::::
BI.AS!NFRINGE ] ~ ' ~ ~ . . ~ silt-I- mud BASIN PLAIN turbidites thin silt + thick mud turbidites hemipelogites
_V
V t'.!i~
, IRREGULAR SLOPE slump deposits hemipelagites debrites
FIG. 17. Schematic distribution of fine-grained sediment facies within a relatively small enclosed marginal basin. Note irregular to concentric distribution of mixed facies types.
DEEP-WATER SLOPE DEPOSITS "2/" ~
}
Upper slope
/
Slump scars Lower slope
~
Channel-fill
sandstones
DEEP-WATER FAN DEPOSITS Inner fan
Channel fills; including olisthostromes
Middle fan Channel-fill sequences
DEEP-WATER PLAIN DEPOSITS ........... , .... ~ ......... ..........
T ..T ........... ............
,
~
............
Fine/very fine sandstones, siltstones of marked lateral continuity
9~
....... .~....[' . . -'r.........
,:~i?. ~ i o oli~i~o ~
Olisthostromes and slump deposits
T
--rThickening/ coarsening upwards sequences Outer fan Medium-tofine-graded sandstones of good lateral continuity
"'~ ....... ~' ..'....'. ~- ~." :.-:f. .=P.. , , . . . . . . ~ . . . ~ . . . . . . . . . . . ~ ~ I'J
50 m 1
FIG. 18. Typical facies associations for coarse-grained and fine-grained sediments in slope, fan and basin plain environments. (After Mutti & Ricci Lucchi 1972).
636
D.A.V. Stow and D.J.W. Piper
restricted, periodic anoxic bottom conditions can lead to the accumulation of organic-rich sediments (black shale facies) (e.g. Arthur et al., this volume). Facies associations & vertical sequences In ancient record it is often possible to recognize slope, fan and basin plain palaeoenvironments on the basis of the vertical and/or lateral association
I
,
. ,
of the characteristic facies outlined above. Thus, Mutti & Ricci Lucchi (1972) proposed facies associations for each of these three main environments (Fig. 18). Finer-grained facies make up a large part of these facies associations. Within these associations there are commonly smaller scale vertical sequences of beds that appear characteristic of different subenvironments or morphological elements in the deep sea (Fig. 19). For example, fining-upwards sequences
e,l,ee MID-FAN/SLOPE CHANNELS
PROXIMAL CANYON OR C H A N N E L
D E B R I S - SLUMP MASSES
CONTOURITE MOUND
DISTRIBUTARY CHANNELS
SANDY LOBE
SILTY-SANDY DISTAL LOBES
MUD-FILLED CHANNELS
PROXIMAL MUD L O B E
DISTAL SILTMUD L O B E
mud turbldites thin silt lurbidltes
6
send turbldltes
=c::mum
I
p e b b l y sand + gravel "turbidites"
I debrltes
PROXIMAL LEVEE
DISTAL LEVEE
INTERCHANNEL
OPEN S L O P E
slumps contourlte slits bl&ck s h l l e s
pelagites § ~emlpelagltes LEGEND
BASE-OF-SLOPE BASIN WEDGE
OR
OVER-SUPPLIED BASIN
UNDER-SUPPLIED BASIN
RESTRICTED BASIN
FIG. 19. Typical vertical facies sequences for various deep-water sub-environments (or morphological elements). Note variation in scale and in grading characteristics of the different sequences. (From Stow 1984a).
Deep-water fine-grained sedimen ts: facies models of turbidites are often interpreted as representing channel- or canyon-fill deposits, whereas coarsening-upwards sequences are taken as indicative of lobe deposits (Ricci Lucchi 1975; Rupke 1977; Walker 1978). Vertical sequences of predominantly fine-grained facies have been described by Shanmugam (1980); Stow et al. (1982); and Stow (1984a) among others. In this volume, papers by Hill; Picketing; Stow, Alam & Piper; Bourrouilh & Gorsline; Hayward; and Isaacs (all this volume) give further specific examples of sequences of a variety of facies in different environments. Canyons and channels can be filled by muddy sediments rather than coarse-grained facies, although much of the fill comprises slumps and debrites as well as mud turbidites and hemipelagites in a rather chaotic vertical sequence (e.g. Coleman et al. 1983). Stow (1984c) has interpreted thickening-upward followed by thinning upward sequences ofbiogenic-mud turbidites and debrites from the south-east Angola Basin as representing progradational-regradational submarine fan facies. The more distal parts of fan lobes or finer-grained terminal lobes on large muddy fans appear to show symmetrical vertical sequences of silt and mud turbidites (e.g. Pickering 1982b). Irregular or relatively uniform sequences of interbedded turbidites, contourites and hemipelagites/pelagites appear to be more characteristic of levees, interchannel, open slope and some basinal environments. These commonly differ in the relative proportions of the main facies present. More purely contourite sequences occur through contourite drifts (e.g. Gonthier et al., this volume; Stow & Holbrook, this volume), and these apparently show an irregular alternation of sandy, silty and muddy contourite facies, although very little data are yet available. Thick sequences of regularly-spaced pelagite-hemipelagite alternations characterize many low energy slope and basinal environments.
Discussion In this section we highlight briefly some interesting and problematical areas, mostly concerning detailed characteristics of the deep-water finegrained facies outlined above. These are some of the areas of current research and where future advances are likely to be made. Silt-mud lamination Silt-laminated muds are a very common deep-sea facies and occur repeatedly in the geological
637
record. However, their mode of emplacement and the process of lamina formation is not fully understood, and many different mechanisms have been proposed (see reviews by Stow & Bowen 1978, 1980). In many, perhaps most, cases, the lamination is current-induced. Velocity fluctuation within a single current (Lombard 1963), reflection of a turbidity current from the walls of a small basin (Van Andel & Komar 1969), the quasi-cyclic bursting process in boundary layers (Hesse & Chough 1980), and a series of small distinct flows or suspension clouds (Dzulynski & Radomski 1955) have all been suggested as the prime cause of such lamination. Perhaps more widely applicable are the processes of congregational sorting described by Piper (1972b), and of shear sorting of clay flocs from silt grains during final deposition through the base of the bottom boundary layer (Stow & Bowen 1978). This latter mechanism has received some support from experimental work with mud-water suspensions (see Kranck, this volume). Although these kinds of processes would apply equally to bottom currents, the association of abundant laminae with bottom currents as proposed by Hollister & Heezen (1972) has not subsequently been confirmed (see e.g. Hill, this volume; Shor et al., this volume). The laminae that have been described from contourites of sediment drifts are irregular or wispy and discontinuous (Piper & Brisco 1975; Stow 1982; Gonthier et al. this volume). Those of turbidites are more regular, distinct and continuous (e.g. Hill; Stow, Piper & Alam; Pickering; all this volume). In anoxic environments that cannot support a burrowing infauna or epifauna, pelagic and hemipelagic sediments can also preserve a primary lamination. This lamination is not currentinduced but results from periodic changes in the type of sediments supplied. The laminae are thus more truly varves, though not necessarily of seasonal periodicity (e.g. Isaacs, this volume; Robertson, this volume). Arthur et al. (this volume) show that lamination in black shales is in some cases turbiditic, in other cases pelagic and in other cases is better described as a fissility rather than true silt-mud alternation.
Mud fabric & fissility The fabric of sands and gravels has long been used as a diagnostic of depositional process and flow direction and silt fabric in fine-grained sediments can be used in a similar manner (Piper 1972a,b; Stow 1979b). However, there has been very much less work of an equivalent nature on the fabric of muds and shales (see reviews by
63 8
D.A.V.
Stow and D.J.W.
Moon 1972; Bennett et al. 1977; O'Brien 1981; Moon & Hurst, this volume). O'Brien (1981) and Moon & Hurst (this volume) agree that the development of fissility in shales is associated with an original orientation of clay flakes parallel to bedding. However, it remains uncertain whether the parallel orientation results from deposition of clay in the dispersed state as a result of an organic or inorganic deflocculant, or from deposition of oriented floc domains and subsequent mechanical reorientation on burial. Moon & Hurst (this volume) further discuss the importance of shale fissility in the generation and primary migration of hydrocarbons from organic-rich black shales. Geotechnical properties The study of sea floor soil mechanics has lagged behind that of its terrestrial counterpart, but has recently gained great impetus from the offshore oil, gas and minerals industries. In an important synthesis, Bennett et al. (1977) review the work on clay fabric and methodology and begin to relate different clay fabric types to selected geotechnical properties, depth of burial and laboratory consolidation loads. They suggest that for smectite or illite-rich muddy sediments, low void ratios result from high density packing of oriented clay particles, whereas high void ratios develop in sediments having non-oriented chain-like domains of clay particles. The possible relationship of geotechnical properties to particular fine-grained facies deposited by different deep-sea processes has been investigated by Hein & Gorsline (1981) for muddy sediments in various borderland basins off California (see also Gorsline et al., this volume). In a study of various hemipelagic sediments from the Guatemalan margin, including the slope, trench and Cocos oceanic plate environments, Faas (1982 and this volume) identifies marked regional trends in geotechnicai properties. He notes high plasticity in the upper slope mudflow and debris flow deposits, together with high organic-carbon content, and a decrease downslope to lowest values in the trench-fill turbidites. The siliceous hemipelagites of the Cocos plate also have a relatively high plasticity index and are relatively overcompacted compared to oozes. Bioturbation Trace fossil and bioturbation assemblages (ichnofacies) are very common in fine-grained sediments throughout the deep sea and are useful for sedimentological and environmental interpreta-
Piper
tion (Seilacher 1967, 1978; Crimes & Harper 1977; Werner & Wetzel 1982; Wetzel 1982 and this volume). Various attempts have been made to summarize the distribution and character of ichnofacies assemblages in relation to bathymetry, sediment type, ecological stress and biotic diversity and density (e.g. Potter et al., 1980, p. 44). Whereas early work suggested a Zoophycos slope assemblage and a Nereites basinal assemblage, the situation is now seen to be more complex with at least five different assemblages found in the deep sea (Werner & Wetzel 1982). Although there is not a simple relationship between sediment facies and ichnofacies, it appears that in many cases the three main facies groups do have a characteristic suite of burrows and/or degree of bioturbation. Pelagites and hemipelagites are very thoroughly bioturbated and may show a wide range of burrow types with several superimposed tiers being present. Contourites show almost complete bioturbation but insufficient to have destroyed all primary structures. Different burrow types characterize the more muddy and more silty contourite facies. Turbidites tend to be markedly less bioturbated with a concentration of burrows towards the tops of individual beds and a still more restricted diversity of burrow types. Superimposed tiers of burrows may also be present where there has been sufficient time between successive flows. Turbidity currents can apparently introduce an exotic shallow-water burrowing fauna into a deepwater setting (Wetzel, this volume). Ichnofacies assemblages are also proving to be an important characteristic for understanding the depositional environment and origin of black shales. In particular, the nature of the burrows in the interbedded non organic-rich facies and in the transition zone to black shales appear related to bottom water oxygenation (Byers 1977; Ekdale 1980; Stow & Dean 1984). Bioturbation is presumably absent from many black shales because of the anoxic bottom conditions during sedimentation. However, there are many examples of apparently oxic environments and sediments that still lack bioturbation. The other factors that deter burrowing organisms are still poorly known. Textures The analysis of grain size and other textural attributes of fine-grained sediments can provide very useful information on the processes and mechanics of deposition (e.g. Piper 1973; Rivi6re 1977; McCave, this volume; Kranck, this volume). However, textural analyses are more difficult to perform on consolidated material, and
Deep-water fine-grained sediments."facies models hence comparison of ancient and modern sediments is difficult. There is commonly a clear textural distinction between current-deposited and pelagic-settled sediments (e.g. Passega 1964). This can be seen as a sorting parameter, on a plot of coarser one percentile against median diameter, or in terms of the shape of the cumulative grain-size distribution curve. Using Rivi6re's (1977) terminology, the grain-size curves tend towards logarithmic for pelagic/hemipelagic sediments, hyperbolic (coarse tail) for muddy contourites and parabolic (fine tail) for turbidites and silty or fine sandy contourites. The types of grain-size sequence or grading are also diagnostic of the different facies: distinct positive grading and positive grading through grouped silt laminae are characteristic of turbidites; more irregular, alternating positivenegative grading is characteristic of contourites; and negative grading is most pronounced through the turbidite-pelagite transition in biogenic turbidites. Grain size as analysed in the laboratory differs significantly from grain size during deposition in two respects: (1) the clay fraction is analysed in the dispersed state but usually deposited as flocs of various sizes; and (2) the correspondence between grain size and hydraulic equivalence of biogenic and terrigenous material is poorly known. The measurement of grain size of suspended sediment in sea water allows greater insight into the depositional state of fine-grained material (e.g. McCave, this volume; Eittreim, this volume). Kranck (this volume) attempts to distinguish between grain-size populations that have been deposited as flocs and those that have settled in the dispersed state. There has been little serious attempt to determine the hydraulic equivalence of biogenic material (but see Berger & Piper 1972).
Composition and colour There are many different aspects to the composition of fine-grained sediments, some of which have received much attention and others less attention (see, for example, Potter et al. 1980; Tissot & Welte 1978). The biogenic content is clearly important for both biostratigraphic and palaeoecological information. Clay mineralogy has been much used in palaeoenvironmental and provenance studies. Inorganic geochemistry is beginning to be used more for provenance study. Organic geochemistry is vital for our understanding of hydrocarbon source-rock deposition and potential. What seems to be lacking most in our understanding and interpretation of compositional characteristics is better use of a combined
639
approach involving a number of separate compositional studies on the same suite of rocks. Although in individual cases compositional criteria are extremely valuable in distinguishing between turbidite, contourite and pelagite facies (e.g. Piper 1978; Stow & Lovell 1979; Spears & Amin 1981), such criteria are not readily generalized as so much depends on the local source and supply, competition from other components, and so on. The colour of fine-grained sediments is closely related to composition and has been a topic of long-standing interest, but one in which relatively little advance has been made in our understanding of the origin ofcolour and hence in using it as a diagnostic feature (but see McBride 1974; Pettijohn 1975; Potter et al. 1980). Potter et al. (1980) present a useful synthesis of what we know so far and have attempted to quantify the relationship of shale colour to both carbon content and the oxidation state of iron.
Diagenesis Of prime importance to all studies of ancient fine-grained sediments is the question of diagenesis. The twin processes of mechanical and chemical diagenesis can so profoundly alter the structural, textural, compositional, colour and fabric characteristics of deep-sea facies, that their use as criteria for distinction and for interpreting depositional history must necessarily change. The clay-rich beds in a pelagite-hemipelagite sequence can be reduced to thin shale partings by carbonate dissolution (e.g. Fischer & Arthur 1977). Primary sedimentary structures can be destroyed and deformational structures introduced during compaction (e.g. Rieke & Chilingarian 1974). Clay minerals undergo progressive alteration with increased burial pressures and temperatures (e.g. Perry & Hower 1970). Organic carbon is subject to progressive maturation into hydrocarbons with increasing temperature (e.g. Tissot & Welte 1978). Fluids released and mobilized during these various processes migrate through the formation and cause further chemical changes (e.g. Neglia 1979). These and numerous other effects of diagenesis we do not discuss here, and were not discussed at any length at the Fine-Grained Sediments Workshop in Halifax. We would like simply to emphasize the importance of considering diagenetic history in any study of fine-grained sediments in the deep sea. One of the key areas of research in the near future must surely be the development of diagenetic models, for both fine-grained and coarse-grained sediments, to complement the facies models we have outlined in this paper.
640
D.A.V. Stow and D.J.W. Piper
Conclusions (1) The wealth of data that exists and the depths of our knowledge on fine-grained sediments has increased dramatically in the past five to ten years. The holding of a workshop on just the deeper water fine-grained sediments and the editing of this volume has enabled us to take stock of where we have reached and where we should be going in this field of research. In this paper we have attempted a synthesis of facies models and have drawn heavily on other papers in the volume. (2) We identify three broad facies groups in the deep sea: turbidites, contourites and pelagites-hemipelagites. These are related primarily to the processes of deposition by, respectively, gravity-driven turbidity currents, thermohaline/wind-driven normal bottom currents, and vertical settling through the water column. Within each of these groups we describe several distinct facies models: turbidites (1) silt turbidites (2) mud turbidites (3) biogenic turbidites (4) disorganized turbidites contourites (1) muddy contourites (2) silty/fine sandy contourites pelagites-hemipelagites (1) pelagic oozes (2) muddy pelagic oozes (3) pelagic clays (4) hemipelagites These facies models, based on a large amount of observational data on both modern and ancient sediments, are intended as a summary of principal sedimentary characteristics and as a basis for interpretation of the depositional processes involved. Neither the complete sets of characteristics nor the full
sequences of sedimentary structures will necessarily be observed in every section examined and we have tried to show, to some extent, the variability to be expected in each case. (3) We have also given a brief summary of the hydrodynamic interpretation for each facies model. However, it should be emphasized that there appears to be a continuum between the various processes that operate in the deep sea and hence there will be a continuum of resultant facies deposited. (4) The different fine-grained facies show certain characteristic patterns of occurrence both horizontally over the surface of the sea floor and vertically in boreholes or in ancient rock successions. They occur interbedded with and adjacent to each other and to coarser-grained facies. These patterns of distribution, which we have briefly described, can be used to help identify depositional palaeoenvironments. (5) Finally, in discussion, we attempt to do little more than highlight some of the interesting and problematical areas of current research in fine-grained deep-water sediments. We mention, in particular, certain aspects of facies characteristics including: silt-mud lamination, fabric and fissility, geotechnical properties, bioturbation, textures, composition and colour, and diagenesis. ACKNOWLEDGEMENTS:In a synthesis paper of this kind we have many people to thank for their helpful and stimulating discussion. In particular, we thank all the participants in the International Workshop on Fine-Grained Sediments held at Halifax, Canada in 1982. DAVS acknowledges financial support from the Natural Environment Research Council, UK, and technical and secretarial support at the Grant Institute of Geology. We are very grateful for the helpful comments of our reviewers, Ali Aksu, Donn Gorsline, Phil Hill and Julian Holbrook.
References ANATRA, S., ACKERMANN,T. & HOMEWOOD,P. 1980. Les facies de l'Ultrahev6tique du Montsalvens (Pr6alpes Externes) et de la r6gion d'Anzeinde (Pr6alpes Internes). Ecol. geol. Heh'., 73/1,283-92. ANKETELL,J.M. & LOVELL,J.P.B. 1976. Upper Llandoverian Grogal Sandstones and Aberystwyth Grits in the New Quay area, Central Wales: a possible upwards transition from contourites to turbidites. Geol. J., 11, 107-8. ARRHENIUS,G. 1963. Pelagic sediments. In: Hill, M.N. (ed.), The Sea. Interscience, New York. 3, 655-727. AUDLEY-CHARLES,M.G. 1965. A geochemical study of
Cretaceous ferromanganiferous sedimentary rocks from Timor. Geochim. cosmochim. Acta., 29, 1153-73. BARRETT, T.J. 1982. Jurassic bedded cherts from the North Apennines, Italy: dyscyclic sedimentation in the deep pelagic realm. In: Einsele, G. & Seilacher, A. (eds), Cyclic and Event Stratification. SpringerVerlag, Berlin. 389-403. BARTOLINI, C., BERLATO,S. & BORTOLOTTI,V. 1975. Upper Miocene shallow-water turbidites from west Tuscany. Sed. Geol., 14, 77-122. BEALL, A.O. & FISCHER, A.G. 1969. Sedimentology.
Deep-water fine-grained sediments."facies models Init. Repts. DSDP, 1, US Govt. Print. Office, Washington, DC. 521-93. BEIN, A. & WEILER, Y. 1976. The Cretaceous Talme Yafe Formation: a contour current shaped sedimentary prism of calcareous detritus at the continental margin of the Arabian Craton. Sedimentology, 23, 511-32. BENNETT, R.H., LAMBERT, D.N. & HULBERT, M.H. 1977. Geotechnical properties of a submarine slide area on the US continental slope northeast of Wilmington Canyon. Mar. Geotech., 2, 245-61. BERGER, W.H. 1970. Biogenic deep sea sediments: fractionation by deep sea circulation. Bull. geol. Soc. Am., 81, 1385-401. 1974. Deep-sea sedimentation. In: Burk, C.A. & Drake, C.D., (eds), The Geology of Continental Margins. Springer-Verlag, New York. 213-4 1. & PIPER, D.J.W. 1972. Planktonic Foraminifera: differential settling, dissolution and redeposition. Limnol. Oceanog., 17, 275-87. BISCAYE, P.E. & EITTREIM, S.L. 1977. Suspended particulate loads and transports in the nepheloid layer of the abyssal Atlantic Ocean. Marine Geol., 23, 155-72. BLANPIED, C. & STANLEY, D.J. 1981. Uniform mud (unifite) deposition in the Hellenic Trench, eastern Mediterranean. Smiths. Contrib. Marine Sci., 13, 40 PP. BOUMA, A.H. 1962. Sedimentology of Some Flysch Deposits. Elsevier, Amsterdam. 168 pp. , MOORE, G.T. & COLEMAN, J.M. (eds). 1978. Framework, facies and oil-trapping characteristics of the upper continental margin. Am. Ass. Petrol. Geol. Studies in Geology, 7. BOWEN, A.J., NORMARK,W.R. & PIPER, D.J.W. 1984. Modelling of turbidity currents on Navy submarine fan, California Continental Borderland. Sedimentology, (in press). BROECKER, W.S. 1974. Chemical Oceanography. Harcourt Brace Jovanovich, Inc., New York. 214 PP. BYERS, C.W. 1977. Biofacies patterns in euxinic basins: a general model. Soc. econ. Palaeo. Min. Spec. Pub., 25, 5-18. CARRASCO-V, B. 1977. Albian sedimentation of submarine authochthonous and allochthonous carbonates, east edge of the Valles-San Luis Potosi platform, Mexico. Soc. econ. Palaeo. Min. Spec. Pub., 25, 263-72. CARTER,L. & SCI-IAFER,C.T. 1983. Interaction of the Western Boundary Undercurrent with the continental margin off Newfoundland. Sedimentology, 30, 751-78. CrIOUGH, S.K. & HESSE, R. 1980. The Northwest Atlantic Mid-Ocean Channel of the Labrador Sea" III head spill vs. body spill deposits from turbidity currents on natural levees. J. sed. Petrol., 50, 227-34. & HESSE,R. (in press). Contourites from the Eirik Ridge, south of Greenland. Sed. Geol. COLEMAN, J.C. 1976. Deltas: Processes of Deposition and Models for Exploration. Continuing Education Publ. Co., Inc., Champaign. 102 pp. --, PRIOR, D.B. & LINDSAY, J.F. 1983. Deltaic -
-
-
-
-
-
641
influences on shelf-edge instability processes. Soc. econ. Palaeo. Min. Spec. Pub., 33, 121-38. COOK, H.E. & ENOS, P. (eds.). 1977. Deep-water carbonate environments. Soc. econ. Palaeo. Min. Spec. Pub., 25. & TAYLOR,M.E. 1977. Comparison of continental slope and shelf environments in the Upper Cambrian and lowest Ordovician of Nevada. Soc. econ. Palaeo. Min. Spec. Pub., 25, 51-82. CREMER, M. 1981. Distribution des turbidites sur l'6ventail du Canyon du Cap-Ferret. Bull. Inst. Geol. Basin d'Aquitaine, Bordeaux., 30, 51-69. -1983. Approches SOdimentologiques et G~ophysique des Accumulations Turbiditiques. Th6se, Doctorat d'Etat, Universit6 Bordeaux I. CREVELLO, P.D. & SCHLAGER, W. 1980. Carbonate debris sheets and turbidites, Exuma Sound, Bahamas. J. sed. Petrol., 50, 1121-48. CRIMES, T.P. & HARPER,J.C. (eds). 1977. Trace fossils. Geol. J. Spec. Issue, 9, 351 pp. CURRAY, J.R. & MOORE, D.G. et al. 1979. Drilling the Gulf of California, DSDP leg 64. Geotimes, 24(7), 1 8 - 2 0 .
DAMUTH, J.E. 1975. Echo-character of the western equatorial Atlantic floor and its relationship to the dispersal and distribution of terrigenous sediments. Marine Geol., 18, 17-45. 1978. Echo character of the Norwegian-Greenland Sea: relationship to Quaternary sedimentation. Marine Geol., 28, 1-36. 1980. Use of high frequency (3.5-12kHz) echograms in the study of near-bottom sedimentation processes in the deep-sea: a review. Marine Geol., 38, 51-75. & KUMAR,N. 1975. Late Quaternary depositional processes on the continental rise of the western equatorial Atlantic: comparison with the western North Atlantic and implications for reservoir rock distribution. Bull. Am. Ass. Petrol. Geol., 59, 2172-81. DAVIES, G.R. 1977. Turbidites, debris sheets and truncation structures in upper Palaeozoic deepwater carbonates of the Sverdrup Basin, Arctic Archipelago. Soc. econ. Palaeo. Min. Spec. Pub., 25, 221-48. DEAN, W.E., GARDNER,J.V. & CEPEK,P. 1981. Tertiary carbonate dissolution cycles on the Sierra Leone Rise, eastern equatorial Atlantic Ocean. Marine Geol., 39, 81-101. - - . , STOW, D.A.V., BARRON,E. & SCHALLREUTER,R. 1984. A revised sediment classification for siliceousbiogenic, calcareous-biogenic and non-biogenic compounds. Init. Repts. DSDP, 75. US Govt. Print. Office, Washington DC. DOYLE, L.J. & PILKEY, O.H. (eds). 1979. Geology of continental slopes. Soc. econ. Palaeo. Min. Spec. Pub., 27. DRAKE, D.E., HATCHER, P.G. & KELLER, G.H. 1978. Suspended particulate matter and mud deposition in Upper Hudson submarine canyon. In: Stanley, D.J. & Kelling, G. (eds), Sedimentation in Submarine Canyons, Fans and Trenches. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 33-41. DZULYNSKI,S. & RADOMSKI,A. 1955. Origin of groove
642
D.A.V. Stow and D.J.W. Piper
casts in the light of turbidity current hypothesis. Acta Geol. Pologne., 5, 47-56. EINSELE, G. 1982. Limestone-marl cycles (peridotites): diagnosis, significance, causes--a review. In: Einsele, G. & Seilacher, A. (eds), Cyclic and Event Stratification. Springer-Verlag, Berlin. 8-53. - & S E I L A C H E R , A. (eds). 1982. Cyclic and Event Stratification. Springer-Verlag, Berlin. 536 pp. EKDALE, A.A. 1980. Trace fossils and stagnation of deep-sea basins. (Abstr.). Bull. Am. Ass. Petrol. Geol., 64, 704. ENOS, P. 1977. Tamabru limestone of the Pozo Rico trend, Cretaceous, Mexico. Soc. econ. Palaeo. Min. Spec. Pub., 25, 273-14. FAAS, R.W. 1982. Gravitational compaction patterns determined from sediment cores on the Guatemalan Transect: continental slope to Middle America Trench to Cocos Plate, on DSDP leg 67. In: Auboin, J., von Huene, R. et al. Initl. Repts. DSDP, 67, US Govt. Print. Office, Washington, DC. 617-38. FAUGI~RES, J-C. et al. 1979. Evolution de la s6dimentation profonde au Quaternaire r6cent dans le Bassin nord-atlantique: corps s6dimentaire et s6dimentation ubiquiste. Bull. Soc. g~ol. France., 21, 585-601. , GAYET, J., GONTHIER, E., POUT1ERS,J. ~; NYANG, I. 1982. La Dorsale M6dio-Atlantique entre 43 ~ et 56~ facies et dynamique s6dimentaire dans plusieurs types d'environements au Quaternaire R6cent. Bull. Inst. Geol. Bassin d'Aquitaine, Bordeaux., 31, 195-215. , STOW, D.A.V. & GONTHIER, E. (1984) Contourite drift moulded by deep Mediterranean outflow. Geology, (in press). FISCHER, A.G. & ARTHUR, M.A. 1977. Secular variations in the pelagic realm. Soc. econ. Palaeo. MiD. Spec. Pub., 25, 19-50. FOLK, R.L. & MCBRIDE, E.F. 1978. Radiolarites and their relation to subjacent 'oceanic crust' in Liguria, Italy. J. sed. Petrol., 48, 1069-102. GARRISON, R.E. & FISCHER, A.G. 1969. Deep-water limestones and radiolarites of the Alpine Jurassic. Soc. econ. Palaeo. Min. Spec. Pub., 14, 20-55. GINSBURG, R.N. 1975. (ed.). Tidal Deposits: ,4 Casebook of Recent Examples and Fossil Counterparts. Springer-Verlag, Berlin. 428 pp. GONTHIER, E., CREMER,M., FAUGERES,J.C. & NIANG, I. 1982. Turbidites et contourites sur la marge sud-arabique (Sukra). Reunion Ann. Sci. de la Terre., Paris. 284 pp. GORSLINE, D.S. 1978. Anatomy of margin basins. J. sed. Petrol., 48, 1055-68. - 1981. Fine sediment transport and deposition in active margin basins. In: Douglas, R.G., Colburn, I.P. & Gorsline, D.S. (eds), Depositional systems of active continental margins. Pacific Section, Soc. econ. Palaeo. Min., Short Course Notes, Notes, Los Angeles, 1981.39-59. GRADSTEIN, F.M., SHERIDAN, R.E. et al. 1981. Leg 76: Western North Atlantic Ocean. Joides Journal, VII. 29-35. GRIFFIN, J.J., WlNDON, H. & GOLDBERG,E.D. 1968. The distribution of clay minerals in the world ocean. Deep Sea Res., 15, 433-59.
GRIGGS, G.B. & KULM, C.D. 1970. Sedimentation in Cascadia Deep Sea Channel. Bull. geol. Soc. Am., 81, 1361-84. HALL, B.A. & STANLEY, D.J. 1973. Levee-bounded submarine base-of-slope channels in the Lower Devonian Seboomook Formation, northern Maine. Bull. geol. Soc. Am., 84, 2101-10. HARMS, J.C. & FAHNESTOCK,R.K. 1965. Stratification, bed forms and flow phenomena (with an example from the Rio Grande). Soc. econ. Palaeo. MiD. Spec. Pub., 12, 84-115. HEEZEN, B.C., HOLLISTER, C.D. & RUDDIMAN, W.F. 1966. Shaping of the continental rise by deep geostrophic contour currents. Science., 152, 502-8. HEIN, F.J. & GORSLINE,D.S. 1981. Geotechnical aspects of fine-grained mass flow deposits: California continental borderland. Geo-Marine Letters, 1, 1-5. HESSE, R. 1975. Turbiditic and non-turbiditic mudstone of Cretaceous flysch sections of the East Alps and other basins. Sedimentology, 22, 387-416. -• CHOUGH, S.K. 1980. The Northwest Atlantic Mid-Ocean Channel of the Labrador Sea: II Deposition of parallel-laminated levee muds from the viscous sub-layer of low density turbidity currents. Sedimentology, 27, 697-71 I. HICKS, D.M. 1981. Deep-sea fan sediments in the Torlesse zone, Lake Chau, South Canterbury, New Zealand. N Z J. Geol. Geophys., 24, 209-30. HILL, P.R. 1981. Detailed Morphology and Late Quaternary Sedimentation on the Nova Scotia Slope South of Halifax. Unpublished Ph.D. Thesis, Dalhousie Univ. Halifax, NS, Canada. 331 pp. - & BOWEN, A.J. 1983. Modern sediment dynamics at the shelf-slope boundary off Nova Scotia. Soc. econ. Palaeo. MiD. Spec. Pub., 33, 265-78. HOFFERT, M. 1980. Les 'Argiles Rouges des Grands Fonds' clans le Pacifique Centre-est: Authigenkse, Transport, Diagen~se. Thesis, Universit6 Louis Pasteur de Strasbourg, M6moire No. 61,231 pp. HOLLISTER,C.D. & HEEZEN, B.C. 1972. Geologic effects of ocean bottom currents: western North Atlantic. In: Gordon, A.L. (ed.), Studies in Physical Oceanography. Gordon Breach Science Pubs. 2, 37-66. HOMEWOOD, P. & WINKLER, W. 1977. Les calcaires d6tritiques et noduleux du Malm des M6dianes Plastiques dans le Pr~alpes fribourgeoises. Bull. Soc. Frib. Sc. Nat., 66, 116-40. HORN, D.R., EWING, M., HORN, B.M. & DELACH,M.N. 1971. Turbidites of the Hatteras and Sohm Abyssal Plains, Western North Atlantic. Marine Geol., 11, 287-323. Hsu, K.J. & JENKYNS, H.C. (eds.). 1974. Pelagic Sediments: on land and under the sea. Spec. Publ. Int. Ass. Sediment., 1,447 pp. IMOTO, N. & FUKUTOMI, M. 1975. Genesis of bedded cherts in the Tamba Belt, southwest Japan. Assoc. Geol. Collabor. Japan Monogr., 19, 3542. INGERSOLL, R.V. 1977. Submarine fan facies of the upper Cretaceous great valley sequence, north and central California. Sed. Geol., 21, 205-30. INMAN, D.L. & NORDSTROM,C.E. 1971. On the tectonic and morphologic classification of coasts, d. Geol., 79, 1-21.
Deep-water fine-grained sediments."facies models ISAACS, C.M. 1981. Lithostratigraphy of the Monterey Formation, Goleta to Point Conception, Santa Barbara Coast, California. Am. Ass. Petrol. Geol. Field Guide, 4, 9-24. JACOBI, R.D. 1982. Microphysiography of the SE North Atlantic and its implication for the distribution of near bottom processes and related sedimentary facies. Bull. Inst. Geol. Bassin d'Aquitaine., 31, 31-46. JENKYNS, H.C. 1978. Pelagic environments. In: Reading, H.G. (ed.) Sedimentary Environments and Facies. Blackwell Scientific Publications, Oxford. 314-71. 1984. Pelagic environments. In: Reading, H.G. (ed.), Sedimentary Environments and Facies. 2nd edition. Blackwell Scientific Publications. Oxford. (In press). JIPA, D. & KIDD, R.B. 1974. Sedimentation of coarser grained interbeds in the Arabian Sea and sedimentation processes of the Indus Cone. In: Whitmarsh, R.E., Weser, O.E., Ross, D.A. et al., Initl. Repts. DSDP, 23, US Govt. Print. Office, Washington, DC. 471-92. JOHNSON, T.C., CARLSON, T.W. & EVANS, J.E. 1980. Contourites in Lake Superior. Geology, 8, 437-41. KAGAMI, H. 1979. Transformation of opaline silica in sediments from Bay of Biscay and Rockall Bank. In: Montadert, L., Roberts, D.G. et al. (eds), Init. Repts. DSDP, 48, 757-64. US Govt. Print. Office, Washington, DC. KALIN, O., PATACA, E. & RENE, O. 1979. Jurassic pelagic deposits from southeastern Tuscany: aspects of sedimentation and new biostratigraphic data. Ecol. Geol. HelL,., 72/3, 715-62. KARL,H.A., CARLSON,P.R. & CACCHIONE,D.A. 1983. Factors that influence sediment transport at the shelf break. Soc. econ. Palaeo. Min. Spec. Pub., 33, 219-32. KELTS, K. & ARTHUR, M.A. 1981. Turbidites after ten years of deep-sea drilling--wringing out the mop? Soc. econ. Palaeo. Min. Spec. Pub., 32, 91-127. KENNEDY,W.J. 1980. Aspects of chalk sedimentation in the southern Norwegian offshore. In: The Sedimentation of North Sea Reservoir Rocks, Norwegian Petroleum Society, Geilo. KOLLA, V., EITTREIM,S., SULLIVAN,L., KOSTECKI,J.A. & BURCKLE,L.H. 1980. Current-controlled abyssal microtopography and sedimentation in the Mozambique Basin, Southwest Indian Ocean. Marine Geol., 34, 171-206. KUENEN, PH.H. 1964. Deep-sea sands and ancient turbidites. In: Bouma, A.H. & Brouwer, A. (eds). Turbidites--Developments in Sedimentology, 3. Elsevier. 3-33. LAUGHTON, A.S., BERGGREN, W.A. et al. 1972. Initl. Repts. DSDP, 12. US Govt. Print. Office, Washington, DC. LISlTZIN, A.P. 1972. Sedimentation in the world ocean. Soc. econ. Palaeo. Min. Spec. Pub., 17, 218 pp. LOMBARD, A. 1963. Laminites--a structure of flyschtype sediments. J. sed. Petrol., 33, 14-22. LOVELL, J.B.P. & STOW, D.A.V. 1981. Identification of ancient sandy contourites. Geology, 9, 347-9. LUNDEGARD, P.D., SAMUELS, N.D. & PRYOR, W.A. -
-
643
1980. Sedimentology, petrology and gas potential of the Brallier Formation--Upper Devonian turbidite facies of the central and southern Apalachians. US Dept. Energy Report DOE/METC/5201-5,220 pp. MALDONADO,A. & STANLEY,D.J. 1976. The Nile Cone: submarine fan development by cyclic sedimentation. Mar. Geol., 20, 27-40. MCBRIDE, E.F. 1974. Significance of colour in red, green, purple, olive, brown and grey beds of Difunta Group, NW Mexico. J. sed. Petrol., 44, 760-8. MCCAVE, I.N. 1972. Transport and escape of finegrained sediment from shelf areas. In: Swift, D.J.P., Duane, D.B. & Pilkey, O.H. (eds), Shelf Sediment Transport: Process and Pattern., 10, 225-48. 1979. Diagnosis of turbidites at Sites 386 and 387 by particle-counter size analysis of the silt (2-40/~m) fraction. In: Tucholke, B.E., Vogt, P.R. et al. Initl. Repts. DSDP., 43, 395-405. --, LONSDALE,P.F., HOLLISTER, C.D. & GARDNER, W.D. 1980. Sediment transport over the Hatton and Gardar contourite drifts. J. sed. Petrol., 50, 1049-62. MCILREATH,I.A. & JAMES,N.P. 1978. Facies models 13. Carbonate slopes. Geoscience Canada., 5, 189-99. MOON, C.F. 1972. The microstructure of clay sediments. Earth Sci. Rev., 8, 303-321. MOORE, C. 1974. Turbidites and terrigenous muds: DSDP leg 25. In: Simpson, E.S.W., Schlich, R. et al., Initl. Repts. DSDP., 25, 441-79. MOORE, D.G. 1969. Reflection profiling studies of the Californian continental borderland: structure and Quaternary turbidite basins. Geol. Soc. Am. Spec. Pap., 107, 1-142. MULUNS, H.T. & NEUMANN, A.C. 1979. Deep carbonate bank margin structure and sedimentation in the northern Bahamas. In: Doyle, L.J. & Pilkey, O.H. (eds). Geology of Continental Slopes., Soc. econ. Palaeo. Min. Spec. Publ., 27, 165-92. MUTTI, E. 1977. Distinctive thin-bedded turbidite facies and related depositional environments in the Eocene Hecho Group (South-central Pyrenees, Spain). Sedimentology., 24, 107-31. -& R i c o Luccm, F. 1972. Le torbiditi dell' Appenino settentrionale: introduzione all'analisi di facies. Mere. Soc. geol. ltal., 11, 161-99. - - , NILSEN,T.H. & RIccl LUCCHI,F. 1978. Outer fan depositional lobes of the Laga formation (Upper Miocene and Lower Pliocene), East-Central Italy. In: Stanley, D.J. & Kelling, G. (eds), Sedimentation in Submarine Canyons, Fans and Trenches. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 210-23. NARDIN, T.R., HE1N, F.J., GORSLINE,D.S. & EDWARDS, B.D. 1979. A review of mass movement processes, sediment and acoustic characteristics, and contrasts in slope and base-of-slope systems versus canyonfan-basin floor system. Soc. econ. Palaeo. Min. Spec. Pub., 27, 61-73. NAYLOR, M.A. 1980. Origin of inverse grading in muddy debris flow deposits--a review. J. sed. Petrol., 50, I I I 1-6. 1981. Debris flow (olistostromes) and slumping on a distal passive continental margin: the Palombini limestone-shale sequence of the northern Apennines. Sedimentology, 28, 837-52.
-
-
-
-
644
-
-
-
D.A.V. Stow and D.J.W. Piper
NEGLIA, S. 1979. Migration of fluids in sedimentary basins. Bull. Ass. Am. Petrol. Geol., 63, 573-97. NELSON, C.H. 1976. Late Pleistocene and Holocene depositional trends, processes, and history of Astoria deep sea fan, North East Pacific. Marine Geol., 20, 129-73. , NORMARK,W.R., BOUMA,A.H. & CARLSON,P.R. 1978. Thin-bedded turbidites in modern submarine canyons and fans. In: Stanley, D.J. & Kelling, G. (eds.). Sedimentation in submarine canyons,fans and trenches. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 177-89. NISBET, E.G. & PRICE, I. 1974. Siliceous turbidites: bedded cherts as redeposited, ocean ridge-derived sediments. Spec. Publ. Int. Ass. Sediment., 1, 351-66. NORMARK, W.R. & PIPER, D.J.W. 1972. Sediments and growth pattern of Navy deep sea fan, San Clemente Basin, California Borderlands. J. Geol., 80, 198-223. O'BRIEN, N.R. 1981. SEM study of shale fabric--a review. In: O'Hare, A.M.F. (ed.) Scanning Electron Microscopy 1981 I. SEM Inc., Chicago. 569-75. , NAKAZAWA, K. & TOKUHASHI, S. 1980. Use of clay fabric to distinguish turbiditic and hemipelagic siltstones and silts. Sedimentology, 27, 47-61. PASSEGA, A. 1964. Grain size representation by CM patterns as a geological tool. J. sed. Petrol., 34, 830-47. PERRY, E.A. & HOWER, J. 1970. Burial diagenesis in Gulf Coast pelitic sediments. Clays Clay Minerals., 18, 165-77. PETERSON, M.N.A. & EDGAR, N.T. et al. 1970. lnitl. Repts. DSDP, 2. US Govt. Print. Office, Washington, DC. PETT1JOHN, F.J. 1975. Sedimentary Rocks. 3rd edn. Harper & Row, New York. 628 pp. PICKERING, K.T. 1982a. The Kongstjord Formation--a late PreCambrian submarine fan in NE Finnmark, N Norway. Norges geol. Unders, 367, 77-104. 1982b. Middle-fan deposits from the Late PreCambrian Kangsfjord Formation submarine fan, NE Finnmark, N Norway. Sed. Geol., 33, 79-110. PILKEY, O.H., LOCKER, S.D. & CLEARY, W.J. 1980. Comparison of sand-layer geometry on flat floors of 10 modern depositional basins. Bull. Am. Ass. Petrol. Geol., 64, 841-56. PXPER, D.J.W. 1972a. Sediments of the Middle Cambrian Burgess Shale, Canada. Lethaia., 5, 169-75. 1972b. Turbidite origin of some laminated mudstones. Geol. Mag., 109, 115-26. 1973. The sedimentology of silt turbidites from the Gulf of Alaska. In: Kulm, L.D., von Huene, R. et al., Initl. Repts. DSDP, 18. US Govt. Print. Office, Washington, DC. 847-67. 1978. Turbidites muds and silts on deep-sea fans and abyssal plains. In: Stanley, D.J. & Kelling, G. (eds). Sedimentation in Submarine Canyons, Fans and Trenches. Dowden, Hutchinson & Ross, Stroudsburg, Pa. 163-76. & BRISCO, C.D. 1975. Deep water continental margin sedimentation, DSDP leg 28, Antarctica. In: Hayes, D.E., Frakes, L.A. et al. Initl. Repts. DSDP, 28, US Govt. Print. Off., Washington, DC. 727-55. -
-
-
- - - , NORMARK,W.R. & INGLE,J.C. 1976. The Rio Dell Formation: a Plio-Pleistocene basin slope deposit in Northern California. Sedimentology, 23, 309-28. POTTER, P.E., MAYNARD, J.B. & PRYOR, W.A. 1980. Sedimentology of shale. Springer-Verlag, New York. 303 pp. REINHARD'r, J. 1977. Cambrian off-shelf sedimentation, central Appalachians. Soc. econ. Palaeo. Min. Spec. Pub., 25, 83-112. R i c o LUCCHI, F. 1975. Miocene paleogeography and basin analysis in the Periadriatic Apennines. Reprinted from Geology of Italy, C. Squyres, (ed.). Pet. Explo. Soc. Libya., 111. 1978. Turbidite dispersal in a Miocene deep-sea plain: the Marroso-Arenacea of the Northern Apennines. Geol. Mijnbouw., 57, 559-76. -1981. The Miocene Marroso-Arenacea turbidites, Romagna and Umbria Apennines. IAS 2nd E U R MJG., Excursion Guidebook. RIEKE, H.H. & CHILINGARIAN,G.V. 1974. Compaction of Argillaceous Sediments. Elsevier, Amsterdam. 424 pp. RIVIERE, A. 1977. MOthodes Granulom~triques: Techniques et Interpretation. Masson, Paris. 170 pp. RUPKE, N.A. 1975. Deposition of fine-grained sediments in the abyssal environment of the AlgeroBalearic Basin, western Mediterranean Sea. Sedimentology, 22, 95-109. -1977. Growth of an ancient deep-sea fan. J. Geol., 85, 725-44. & STANLEY, D.J. 1974. Distinctive properties of turbiditic and hemipelagic mud layers in the AlgeroBalearic Basin, western Mediterranean Sea. Smiths. Contrib. Earth. Sci., 13, 40 pp. SCRU'rTON, R.A. & STow, D.A.V. (1984). Seismic evidence for early Tertiary bottom current controlled deposition on the Charlie Gibbs Fracture Zone. Marine Geol., in press. SEILACHER,A. 1967. Bathymetry of trace fossils. Marine Geol., 5, 413-29. 1978. Use of trace fossil assemblages for recognising depositional environments. In: Basan, P.B. (ed.), Trace Fossil Concepts. Soc. econ. Palaeo. Min. Short Course No. 5, Oklahoma. SHANMUGAM, G. 1980. Rhythms in deep sea, finegrained turbidite and debris-flow sequences, Middle Ordovician, eastern Tennessee. Sedimentology, 27, 419-32. SHEPARD, F.C., PHLEGER, F.B. & ANDEL, TJ. H. VAN. (eds) 1960. Recent Sediments, Northwest Gulf of Mexico. Am. Ass. Petrol. Geol. Mere., Tulsa, Oklahoma. 394 pp. --, MARSHALL, N.F. McLOUGHL1N, P.A. & SULLIVAN, G.G. 1979. Currents in submarine canyons and other sea valleys. Am. Ass. Petrol. Geol. Studies in Geology. 8, 179. SHOR, A.N. & POORE, R.Z. 1978. Bottom currents in ice rafting in the North Atlantic: interpretation of Neogene depositional environments of Leg 49 cores. In: Luyendyk, B.P., Cann, J.R. et aL Initl. Repts. DSDP, 49. US Govt. Print. Off., Washington, DC. 859-72. --, LONSDALE, P., HOLLISTER, C.D. & SPENCER, D. 1980. Charlie-Gibbs fracture zone: bottom-water -
-
-
-
-
-
Deep-water fine-grained sediments." facies models transport and its geologic effects. Deep Sea Res., 27A, 325-45. SPEARS, D.A. & AMIN, M.A. 1981. A mineralogical and geochemical study ofturbidite sandstones and interbedded shales, Man Tor, Derbyshire, UK. Clays Clay Minerals., 16, 333-45. STANLEY, D.J. 1981. Unifites: structureless muds of gravity-flow origin in Mediterranean basins. GeoMarine Letters., 1, 77-83. 1983. Parallel-laminated deep-sea muds and coupled gravity flow--hemipelagic settling in the Mediterranean. Smiths. Contrib. Marine Sci., 19, 19 pp. -& KEELING, G. (eds). 1978. Sedimentation in submarine canyons, fans and trenches. Dowden, Hutchinson & Ross, Stroudsborg, Pa. & MALDONAOO, A. 1979. Levantine Sea-Nile Cone lithostratigraphic evolution: quantitative analyses and correlation with paleoclimatic and eustatic oscillations in the late Quaternary. Sed. Geol., 23, 37-65. -- & -1981. Depositional models for fine-grained sediment in the western Hellevic Trench, eastern Mediterranean. Sedimentology, 28, 273-90. & WEAR, C.M. 1978. The 'mud-line': an erosiondeposition boundary on the upper continental slope. Marine Geol., 28, M 19-M29. --, SWIFT, D.J.P., SILVERBERG,N., JAMES, N.P. & SUTTON, R.G. 1972. Late Quaternary progradation and sand 'spillover' on the outer continental margin offNova Scotia, southeast Canada. Smiths. Contrib. Earth Sci., 8, 88 pp. STOW, D.A.V. 1977. Late Quaternary Stratigraphy and Sedimentation on the Nova Scotian Outer Continental Margin. Unpub. Ph.D. Thesis, Dalhousie University. 1979a. Distinguishing between fine-grained turbidites and contourites on the Nova Scotian deep water margin. Sedimentology, 26, 371-87. 1979b. Method for silt and sand fabric analysis in deep-sea cores. J. sed. Petrol., 39, 627-30. 1981. Laurentian Fan: morphology, sediments, processes and growth pattern. Bull. Am. Ass. Petrol. Geol., 65, 375-93. 1982. Bottom currents and contourites in the North Atlantic. Bull. Inst. Geol. Bassin d'Aquitaine., 31, 151-66. 1984a. Deep-sea clastics: where are we and where are we going? In: Brenchley, P.J. & Williams, B.P.J. (eds), Sedimentology: Recent Developments and Applied Aspects. Geol. Soc. Lond. Spec. Pub. (in press). 1984b. Anatomy of debris flow deposits. In: Hay, W.W., Sibuet, J.C. et al., Initl. Repts.,DSDP, 75. US Govt. Print. Office, Washington, DC. 1984c. Turbidite facies, associations and sequences in the south-eastern Angola Basin. In: Hay, W.W., Sibuet, J.C. et al., Initl. Repts. DSDP, 75. US Govt. Print. Office, Washington, DC. (in press). Deep-sea clastics. In: Reading, H.G. (ed.), Sedimentary Environments and Facies. 2nd edition. Blackwell Scientific Publications, Oxford. & BOWEN, A.J. 1978. Origin of lamination in -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
645
deep-sea fine-grained sediments. Nature, 274, 324-28. & -1980. A physical model for the transport and sorting of fine-grained sediments by turbidity currents. Sedimentology, 27, 31-46. & DEAN, W.E. 1984. Middle Cretaceous black shales at Site 530 in the southeastern Angola Basin. In: Hay, W.W., Sibuet, J.C. et al., Initl. Repts. DSDP, 75. US Govt. Print. Office, Washington, DC. & LOVELL,J.P.B. 1979. Contourites: their recognition in modern and ancient sediments. Earth Sci. Reviews, 14, 251-91. & SHANMUGAM,G. 1980. Sequence of structures in fine-grained turbidites: comparison of recent deepsea and ancient flysch sediments. Sed. Geol., 25, 23-42. --, BisHoP, C.D. & MILES, S.J. 1982. Sedimentology of the Brae Oil Field, North Sea: fan models and controls. J. Petrol. Geol., 5, 129-48. TAYLOR, P.T., STANLEY, B.J., SIMKIN, T. & JAHN, W. 1975. Gillis seamount: detailed bathymetry and modification by bottom currents. Marine Geol., 19, 139-57. THOMSON, A.F. & THOMASSON, M.R. 1969. Shallow to deep water facies development in the Dimple Limestone (Lower Pennsylvanian), Marathon Region, Texas. Soc. econ. Palaeo. Min. Spec. Pub., 14, 57-78. TISSOT, B.P. & WELTE, D.H. 1978. Petroleum Formation and Occurrence. Springer-Verlag, Berlin. VAN ANDEL, T.H. & KOMAR, P.D. 1969. Ponded sediments of the mid-Atlantic ridge between 22 ~+23 ~ North latitude. Bull. geol. Soc. Am., 80, 1163-90. VAN DER LINGEN, G.J. 1969. The turbidite problem. N Z J. Geol. Geophys., 12, 7-50. VON STACKELBER6,U., VON RAD, U. & ZOBEL, B. 1979. Asymmetric sedimentation around Great Meteor Seamount (North Atlantic). Marine Geol., 33, 117-32. WALKER, R.G. 1975. Upper Cretaceous resedimented conglomerates at Wheeler Gorge, California: description and field guide. J. sed. Petrol., 45, 105-12. 1978. Deep water sandstone facies and ancient submarine fans: Models for exploration for stratigraphic traps. Bull. Am. Ass. Petrol. Geol., 62, 932-66. & MUTrl, E. 1973. Turbidite facies and facies associations. In: Middleton, G.V. & Bouma, A.H. Turbidites and deep water sedimentation. Pacific Section, Soc. econ. Palaeo. Min., Short Course Notes, Pacific Section, Anaheim. -
-
-
-
-
1
1
9
-
5
8
.
WERNER, F. & WEa'ZEL, A. 1982. Interpretation of biogenic structures in oceanic sediments. Bull. Inst. Geol. Bassin d'Aquitaine., 31,275-88. WE'rZEL, A. 1982. Biogenic sedimentary structures in a modern upwelling area: the NW African continental margin. In: Suess, E. & Thiede, J. (eds), Coastal Upwelling: Its Sediment Record. Plenum Press, New York. 123-44. WEZEL, F.C. 1973. Diacronismo degli eventi geologicio-
646
D.A.V. Stow and D.J.W. Piper
ligo-miocenici nelle Maghrebidi. Revista Mineraria Siciliana, Anno XXIV. N 142-144, 219-32. WILSON, J.L. 1969. Microfacies and sedimentary struc-
tures in 'deeper water' lime mudstones. Soc. econ. Palaeo. Min. Spec. Pub., 14, 4-19.
D.A.V. STow, Grant Institute of Geology, University of Edinburgh, West Mains Road, Edinburgh EH9 3JW, Scotland. D.J.W. PIr'ER,Atlantic Geoscience Center, Geological Survey of Canada, Bedford Institute of Oceanography, PO Box 1006, Dartmouth, Nova Scotia, Canada B2Y 4A2.
Index Abyssal basins 437-50 hills, GLORIA survey 147-51 see also Basins Accretion, Tertiary, Guatemalan Transect 563 see also Compaction Accumulation rates biogenous 487-94 Cretaceous period 417-31 Santa Barbara 400, 485 sediments 483-94 terrigenous, Miocene 486-7 versus tectonic rates 431 Acoustic facies, interpretation of GLORIA 145-51 methods, scattering 72 stratigraphy 187 see also Echo-sounders Acritarchs, fossil 344 Actinolite, turbidite, Ulleung Basin 194 Active rift basins, Santa Barbara, Santa Cruz 409 Advection, in augmented transport 401 Aeolian dust sediment input 601,606 transport 17-19 turbidites 161-2 Aegean Sea see Mediterranean, East Africa, SW, shelf, organic carbon 534 see also Cape Verde Basin Aggregates, sediment, properties 48 Aggregation and transport in suspension 46-55 Airgun profiles 1 9 3 - 5 Albian black shales 545 see also Black shales strata, Atlantic-Tethyan 552-3 Algae, calcareous 199 Allochthonous sequences 512, 515 Alpine piston core see Core analysis Alum shales, Swedish, sedimentology 511-25 biostratigraphy 514 Cambrian deposits 512 depositional environment 523 discussion 521-4 lithofacies 517-21 stratigraphy 513, 515-17 Aluminium, in surface sediments 364, 369 Amino acids, proteins, recent sediments 589-90 Amphiboles 127, 178 Anacapa Current, California 381 Anaerobic conditions see Oxygen content Analytical methods see Methods, analytical Angola basin Cretaceous 550 Zaire sediments 95-112 Anisotropy, magnetic susceptibility see Magnetic susceptibility, anisotropy Anoxicity see Oxygen content Apatite, content of calcareous siltstone 333, 484 Aptian strata, Atlantic-Tethyan 552
Aragonite and green mud, Bahamian 475 Little Bahama bank 200-5 Argillaceous sediments, microstructure 579-90 see also names of components: Clays; Marls; Mudrock; Shales; Siltstones Ash falls, periodicity of volcanic events 18 Ash layers lower and upper Tephra 172, 175 Ulleung basin 187, 191 Asterosoma 338-9 Atlantic, central reconstruction, Lower Cretaceous 450 see also North Atlantic; South Atlantic Atomic absorption spectrophotometry 364 Attapulgite, calcareous sediments 210, 212 see also Illite Atterberg limits, sediments 565-9 Autochthonous sequences, Swedish Caledonides 512, 515-16 Bahamas, fine-grained carbonates 199-207 conclusions 207 global CO2 budget 207 methods 200-2 mineralogy 202--4 processes 204-7 setting 202 Bahamian Trough carbonates, source rock potential 469-80 basin sediments 471-4 carbon content, analysis 474-8 depositional setting 471-4 discussionconclusions 478-9 methods 470-1 Baltic Sea, Cambrian deposits 511-12 Barents Sea Group outcrop 343 Barium in slope sediments 364, 369 Barremian turbidites, organic carbon 547, 552 Basin(s) deep marine, translational plate boundary 481-94 marginal, schematic distribution 635 plains, definition 633 schematic distribution, sediments 6 3 4 - 7 and sills, sediment distribution 405-12 slopes, prodeltas, Norway 343-6 Basnaering formation, Norway 344-5, 356, 3 5 7 Bathymetric mapping (SEABEAM) 150, 156 Bathymetry-controlled trace fossils 595 B/itsl]ord formation, Norway 344 Bedforms, wavy contourites, Lake Superior 293 Benthic boundary layer 71 deep water indicators 337 food content, deep sea 600-6 Biodeformational structures 598 Biofiltration see Zooplankton Bioflocculation, in aggregation 46-52
647
648
Index
Biogenic detritus 77 indicators 117-18 see also Copper; Manganese mudrock, terminology 10 sediments 19 traces, use for bathymetric interpretation 598-606 see also Bioturbation turbidites, summary 617-20 Bioturbation and dissolved oxygen 532-3, 548-9 fine-grained sediments 595-607 biogenic traces 599, 600 conclusions 606-7 environment and ichnofauna 601-5 methods and study areas 595-8 model 598-601 North Atlantic contourites 249-50 Nova Scotia muds 313, 316 in rise canyon sediments 325-7 summary and discussion 638 and trace fossils 338-9 Birnessite (iron sulphide) in sapropels 501 Bituminous limestone 511,515-23 mudstone, Swedish Alum Shales 520 shales see Black shale Bivalves, Nuculid 337 Black laminated mudstone, Swedish Alum Shales 517-19 Black Sea, organic carbon 530, 534 Black shales 588-90 Albian 545 bacterial sulphate reduction 446 and claystones 437-50 model for deposition 527-40 Neogene see Sapropels organic carbon content 527-54 Pacific Mid-Cretaceous 551 sequences, bioturbation, lamination 540-8 South Atlantic 535, 540-9 Blake-Bahama basin, W. North Atlantic 437-50 'black shales' 444-6 'couplets' 446-7 depositional model 447-50 diagenesis 447 lithologies 438-40 marly chalks 444-6 regional setting 437-8 varve-type lamination 440-4 fo~xnation, lithology 437-50 sediments, diagenesis 448 Blake plateau, Bahamas, carbonate sediments 200 Bottom currents activity, lake 293-306 North Atlantic 245-6 Bottom-water oxygenation, inhibitory levels 530 see also Oxygen content Bouma sequence and episapropels 504 Bahamian trough 473 Blake-Bahama basin 441 E and F divisions 209
modified silt facies model 613 Obock Trough 218-19 partial, Halifax Formation 127 Salir Formation 456 sandy turbidites, standard 4 silt turbidites 613 turbidite 87,112,347-9 by X-radiography 328 Boundary-layer, benthic 71 Box cores, sampling method 295 see also Core analysis; DSDP sites Breccias, intraformational 350, 354 Burrowing see also Biogenic traces: Bioturbation indicator of oxidizing sediment, varve-type lamination 447 Calcarenites 227-39 Calcareous sediments New Zealand 331-41 Obock Trough 209-21,331 Calcilutites, Scaglia Rossa 223-39 bedding 228-30 bioturbation 234 burrowing 234 colour 232 composition 231-3 discussion 234-9 structure 234 texture 234 Calcirudite, Scaglia Rossa 231-2, 236 Calcite abundance in benthic foraminifers 488-9 calcareous sediments, NZ. 336-7 calcilutites, depositional model 231-6 classification 9 lithifaction and coalescence 522-3 magnesian, Bahama Bank 199, 202-7 Calcium analysis 364, 366, 369, 370 compensation depth 437, 445 see also Carbonate compensation California Coast Range 363-70 California Continental Borderland, fine-grained sediment transport 395-415 fine sediment transport, canyons 402-4 offshore sewage transport 402 regional setting 396-7 subsurface shelf turbidity patterns 400-2 surface water turbid plumes 397-400 transport scheme 412 turbidity, patterns, basins and sills 405-12 see also Santa Barbara Basin: Santa Paula Creek Cambrian stratigraphy, Alum shale 511-23 Canyons, submarine thalwegs 325-6, 328 transport and sedimentation 400, 402-4, 419 petit-Rhine canyon 121 US mid Atlantic region 319-29 Zaire canyon 95-8 Cape Verde Basin, Southern long-range sidescan sonar 145-51 acoustic facies interpretation 149-51 sonograph coverage 147-9 seismic and sediment facies 153-67
Index Cape Verde Basin, Southern (cont.) Aeolian turbidites 161 correlation between pattern and morphology 160-1 hemipelagic sediments 161 hiatuses 163-5 lithostratigraphy 156-9 processes 156-9 regional setting 155 Carbon-14 analysis 595 Nova Scotian muds 314 Carbon dioxide, global budget 205 Carbon rich strata, organic see Organic carbon Carbonates 9, 199-207, 209-21 Bahamian trough 469-80 muds, organic carbon 469-79 radiolarian nanofossils 437, 440-50 sedimentation and upwelling 488 see also Calcilutites: Calcite Carbonate compensation depth (CCD) 453,463, 540 see also Calcium compensation Casagrande Plasticity chart 569 Cascade-feeding process 171-2, 181 see also Mudflows Cascading, and particle concentration 395, 401 Chalk 10, 28 hemipelagic, Salir Formation 453-66 marly nanofossil 438-50 Challenger, Glomar
3-4
Chemical composition see named elements Cherts 227-39, 454, 484 Chlorite (clay) 19, 337, 365-6, 444, 456 see also Illite; Kaolinite; Montmorillonite Chloritoids 178 Chondrites 501 Chromatograms, gas 471 Chronology, turbidites, Santa Barbara Basin 391 Circulation deep water 245 and sediment movement 395-412 Classification, of fine-grained sediments 6-12 Clastic(s) bioclastic and terrigenous, Cretaceous 438-50 rocks, terminology 6-12 turbidites see Turbidites Clay 5, 19 clayballs 326-7 claystones and black shale 437-50 compacted, microfabric 582-5 freshly sedimenting 579-82 sediments, major microstructural changes 588 see also Chlorite; Illite; Kaolinite; Laterite; Montmorillonite; Mudrock Coacervates, in flocculation 580 Coarse-grained sediments, formation of fans 343 Coccolith(s) 206, 210, 331,338 and diatom zonation 483 ooze 3 Cocos oceanic plate 563-77 Cohesive material, erosion, transport, deposition 40-1 Colorimetry 364 Columbia river, Oregon-Washington 363-70
Columbus basin see Bahamian Trough Compaction artificial 583 characteristics, Quaternary sediments 563-71 and erosion 41-2 Conglomerates, Salir Formation 454--66 Ventura basin 421-30 Conglomeratic limestones, Cambrian 516 Continental margin, tectonically active, sedimentation 377-92 Continental shelf, effect of waves, currents 340 Continental slope Norway 343 Nova Scotia 311-16 sedimentation pattern and sea-level 325 US mid-Atlantic 319-20 Continental source area, hemipelagic sediment formation 363 Contour currents 293-306 Nova Scotian mud turbidites 317 Contourite(s) 122, 245-54, 257-72, 293-306 Blake-Bahama Formation 441,443 and chalks 463 facies, comparison with turbidite and hemipelagite 291,316-17 comparison with turbidites 349 criteria 620-4, 631 destruction by bioturbation 290 history 5 lamination 290 mottled silts, facies 282-4 muds, homogeneous, facies 284-7 muddy 5, 12 North Atlantic, comparison with sandy contourites 253, 275-91 North Atlantic, generalized plot 252 sandy 5, 12 North Atlantic, comparison with muddy contourites 253, 275-91 summary 620-4 muddy 621-2 'sequence' 623-4 silty-sandy 622-3 and turbidites, criteria 293 turbidite, hemipelagite, discussion 31 6-17 see also Turbidites Convoluted bedding 385-7 Copper atomic absorption spectrometry 117-18 pelitic sediments 115-25 in surface sediments 364, 369 Core analysis sites Blake-Bahama Formation 438 Cape Verde Basin 153 Central Pacific Ocean 597 Eastern Mediterranean 499 Faro Drift, Gulf of Cadiz 276 Great Bahama Bank 470 Guatemalan Transect 564 Lake Superior 294 Little Bahama Bank 200-1 North-west Africa, continental margin 596 Nova Scotia continental margin 258 Nova Scotian Slope 312
649
650
Index
Core analysis sites (cont.) Oregon-Washington Slope 364 Santa Barbara Basin 378 Sulu Sea, Borneo 597 Ulleung Basin, Japan 187 United States mid Atlantic continental rise 321 West Hellenic Arc margin 170 Zaire river deep sea fan 99 Cores, box, gravity 595 Coriolis effects, deep-water circulation 317, 409-10, 443 Coulometrics, TOC system 471,474-9 Coulter Counter, in deep sea suspensions 73, 75 'Couplets', radiolarian carbonates 437, 440-4 see also Varve-type lamination Cretaceous, Lower, Central Atlantic reconstruction 450 middle, climate 550 Cucullea, calcareous sediments 331,335 Current transport see Transport Currents bottom 245-7, 252-4 lake 293-306 pelagic depositional 275 turbidity 275, 317 Cyclicity in sediments, test for 446 Debris-flows 565 see also Mudflow Debrites 247-8 Decompaction analysis 571-3 Deep Sea Drilling Project see DSDP Deep-water circulation 409-10 fine-grained sediments, historical outline 3-6 terminology 12 morphology, terminology 11 Deformation see Slides; Slumps; Soft sediments Delta and pro-delta deposits, N. Norway 343,359-60 Density discontinuities, and thermoclines 400-1 Density profiles, wet bulk 565 Deposition fine-grained sediments 59-61 field measurements 61 laboratory measurements 59-61 theory 59-61 related to erosion and transport 36 Depositional sorting, in laminations 347, 349 Destabilization of sediment by biological material 45-6 Detritus biogenic 77 terrigenous 77 Diagenesis Blake-Bahama sediments 441,448-50 clays and muds 582 deep sea sediments 528 summary 639-40 Diapirs 112, 121 Diatom(s) benthic effect on sedimentation stabilization 44-5 productivity, Atlantic Miocene 554 concentration patterns 399 and volcanism 482
Diatomaceous muds, pelagic 563 Diatomaceous ooze 10 Diatomite 10, 28, 427, 484 Dictyonema shales 523 Discontinuous lamination, 490-4 Discordance, erosional-depositional 343, 355 Disorganized turbidites, summary 620 Distortion, in shear orientation of clay 585-7 Dolerite 454, 517 Dolomite 336-7, 493 Domains, microstructural studies, definition 581 Down-core analysis 303-6 Downslope mass movement 574-7 see also Mass flows; Slides; Slumps Drift card trajectories, surface currents, Santa Barbara 406 DSDP cores Atlantic sites 527, 535-9, 544-54, 551 Blake-Bahama basin 437-50 Japan basin 185 North Atlantic 246-8 Pacific sites 527, 542-3, 548-54 slides, criteria 343 West Hellenic Arc 170-7 East Cape, North Island, New Zealand 331-41 Echo-sounders, profiles 150, 156, 258 see also seismic profiles Elements, in surface sediments 364, 369 see also names of individual elements Elzone, particle analyser, electronic 295-306 en echelon tectonic zones 453 Enteromorpha, effect on sedimentation stabilization 44-5 Entrapment, in deposition patterns 399 Epidote 178, 194 Episapropels 502-9 Erosion-deposition model, lake sediment 301-3 Erosion, fine-grained sediments 36-46 biological influences 44-6 and compaction 41-2 critical condition 36-41 rate, physico-chemical parameters 42-4 equation 42-4 related to shear stress 42-4 Erosion, stoss-side 348-9 Erosional-depositional discordant surfaces, Norway 343, 351,355 Exuma Sound see Bahamian Trough 'Fabric rolls', in sheared clay 587 Facies models: deep water fine-grained sediments 611-40 contourites 620-4 discussion--conclusion 637-40 environment and facies distribution 629-37 fine-grained turbidites 612-20 pelagites and hemipelagites 624-9 processes--facies 611-40 Facies analysis, turbidite 106 Faecal pellets, zooplankton 395, 399-400, 412, 431 ocean deposits 443-4 sinking rates 528 Fan, deep sea, schematic distribution 634
651
Index Fan deposits clastic submarine 343-60, 453-66 Laurentian 85-91 Lower Nfieringselva member, Norway 359 Zaire 95-112 Faro Drift, Gulf of Cadiz contourites 275-91 comparison, turbidites, hemipelagites 291 discussion 290 oceanography and bathymetry 275-8 sediment facies 278-87 vertical sequence 287-9 Fault, palaeotransform 344 Fecal pellets see Faecal pellets Feldspar 231,336, 454 Ferrohastingsite 194 Fine-grained sediments, depositional processes deposition rates, biological productivity 22, 28 bioturbation 23, 28 dilution 23 oceanic dispersal systems 22 pelletization 22 sea-level position 22 tectonics 21 depositional sites, oceanic, deep sea and basin floors 27 shelves 26 slopes 27 marine, accumulation rate 61 aggregation 46-54 cohesion 35-6 critical deposition stress 61 distribution with depth 57 erosion 36-44, 62-3 historical outline 3-6 nepheloid layers 57-8 particle settling velocity 62 particle size distribution 49, 50-4, 62 radionuclides accumulation 61 settling velocity 54-6 shear stress 60 stabilization and destabilization 44-6 stratification 57 terminology 12 transport 46, 71,412 vertical flux 56-7 origins, bulk properties 20 cycles 20, 21 discharges, human influence 21 transfer to deep water, debris flows 25 mass movement 26 nepheloids 23 plumes 23 resuspension 24 turbidity currents 25 Fissility, in shales and muds 582-5, 587-90, 637-9 Flaser bedding, as current indicator 459-63 Floc-deposited sediments see Turbidites Floc size distributions 52 Flocculation clay 580-90 coacervates 580 Floods amplification of mass movements 397, 400, 408-9
suspended sediment 380-3, 390 Flow initiation, and long distance transport 612 Fluid shear 47-8, 60-2 Fluidal flows, Atterberg limits 565 Fluorite 200 Fluorometric titration 98 Fluvial discharge, terrigenous matter 528, 549 Fluvio-deltaic deposits Lower Nfieringselva model 359 mineral signatures 364, 372-3 and nepheloid layers 343, 349 post glacial varves 379 Flux, vertical and horizontal 56-7 Flysch deposits 136-42, 339-40 Foraminifera 260, 331 benthic 453 Foraminiferal deposits 199, 209-10, 214, 219 in calcilutite 231,237 in contourites 282, 283, 287 evidence of sea depth 421,425 Forward light scattering, suspended matter 72-3 Fossil communities, trace 595 Freeze drying, surface sediment samples 364 Garnet 178, 194 Gas chromatograms 471 Glaciation, Pleistocene-Holocene 115, 119 Glauconitic sandstones 511,520--3, 547 Glomar Challenger drill ship 1968-1983 5 GLORIA (long range sidescan sonar system) 145-51,319-24 Goldenville Formation, Nova Scotia shales 127--42 Grading 9-11 indicator of turbidity current flow 349 positive and negative 279, 287 see also Depositional sorting; Grain size; Stratification Grain size analysis, factor analysis, quartz sand 325-6 frequency curves, N. Atlantic contourite 281 pelitic sediments 115-19 Santa Barbara basin 378-9 summary 639 X-ray diffraction 364-5 see also Coulter Counter discussion 10, 11 distribution curves, Faro Drift 281 N. Atlantic contourites 251 GRAPE, wet bulk density profiles 565, 571 Graphoglyptid traces 603, 606 see also Nereites
Gravity cores 295 Gravity faulting, submarine slopes 384, 387 Grey mudstone, Swedish Alum Shales 520 Guatemalan Transect, Quaternary sediments 563-77 data analysis 570-3 decompaction values 573-5 discussion and summary 573-7 location DSDP sites 564 methods 563-8 plasticity and compaction 569-70 Gulf of Cadiz see Faro Drift Gullies, on basin slope 343, 356, 359-60 see also Canyons
652 Gully system, dendritic
Index 319, 322
Halifax Formation, Nova Scotia shales 127--42 alga 473 Hallam's model, Liassic concretions 522 Halloysite 336-7 Harmattan winds 595 'HEBBLE' definition 25 method, measurement ofs.p.m. 79 Hellenic Arc margin, west 169-96 composition 175-9 distribution 172-5 general lithology 172-5 morphology 169-70 processes 179-80 stratigraphy 170-2 structure 169-70 texture 175-9 thickness 179 Hemipelagic chalks, fan sequence, Miocene SW Turkey 453-66 depositional process, chalks 459-63 geological setting and sedimentary facies 453-9 lateral and vertical distribution, chalks 463-6 Salir Formation 453 Hemipelagic cores, in palaeoclimate studies 383 Hemipelagic sediments basin model 377-92 Cape Verde Basin 153-66 dispersal 370-2 formation on continental margin 363 Miocene 527-54 Monterey Formation, California 481-95 from multiple sources 373 recent, Bahamas 200 reworked 209 Santa Barbara Basin 377-9 Ulleung basin 191 Wiirmian grey muds 121 Hemipelagic settling definition 311 turbidity currents 431 and turbiditic muds 453 Hemipelagites 12, 20, 115-18, 121 comparison with contourite 291,316-17 criteria 624-9 cycles 629 gradation with contourites 254 muddy pelagic ooze 627 Obock Trough 209 and pelagites, summary 624-9 turbidite, contourite discussion 316-17 see also Pelagites Heyes, North and South Canyons see United States mid-Atlantic region Hiatuses, Southern Cape Verde Basin 163 Hjulstr6m's curve 59-60, 289 Holocene features California Borderland 380 Nova Scotian muds 311 Sulu Sea 605 Horizontal flux, particles 56-7 Hornblende 178, 194, 365-8 Halimeda,
Humic acids 589-90 Hydrocarbons, sources of 469-79 see also Organic carbon Hydrographic features see Basins and sills Hydrography Angola Basin 9 6 - 7 ocean currents 155, 165 see also Circulation; Turbidity currents Ice-rafting 295 Ichnofacies (ichnocoenoses) trace fossil communities 595, 600, 603-7 Ichnofauna, changes 601-7 Illite in calcilutites, Scaglia Rossa 23 California Coast Range 366 definition 19 East Cape, New Zealand 336-7 Guatemalan Transect 563, 574 Hellenic Arc margin 176-8 NW Mediterranean 116 Obock Trough 210-12 and plasticity index 574 ratios, other clay minerals 116-17 Zaire 95 see also Chlorite; Kaolinite; Montmorillonite; Smectite Imbricate structure 587 Imbrication, tertiary 563 Inter canyon slopes see Slopes Intermediate water masses see Oxygen content, oxygen-minimum Internal waves, and resuspension of sediments 395, 397, 400~ Intra-slide beds, Norway 351 Ionian Basin, characteristic clay minerals 176-9 Isopleths, sand and silt, from box core sampling 378-83 Iron 122, 364-6, 369-70 measurement, atomic absorption spectrometry 117-18 pelitic sediments 115 red/green shales 590 Japan
see
Ulleung Basin
Kaolinite artificial compaction 583 in calcilutites, Scaglia Rossa 231-2 definition 19 East Cape, New Zealand 336-7 Guatemalan Transect 563, 574 Hellenic Arc margin 176-8 Obock Trough 212 ratios, other clay minerals 116 Zaire 95 Kernels, coagulation aggregation 46-51 breakup 51-4 Kerogens in Bahamian carbonates 469 type II, Tissot 541 see also Hydrocarbons: Organic carbon Keweenaw Peninsula see Lake Superior
Index Kongsfjord Formation 344-60 Kithira basin 173-9 Klamath mountains, Oregon 363-70 Korea see Ulleung Basin Laconia system, West Hellenic Arc basin 169-80 Lacustrine contourites, Lake Superior 620 Lake Superior, contourite texture 293-307 core-top analysis 295-306 down-core analysis 303-6 methods 295 Sonar records 298 Laminated black marlstone, North Atlantic 540-54 Lamination, Ingrams system 9 bioturbated 282, 408-9 see also Varve-type lamination Laminites 9 terminology 12 Lamont measurement, forward light scatter 73, 76-7 LANDSAT see Satellite imagery Lateral fining, contourites 293 Laterite (clay) 19 Laurentian fan, turbidite sampling 85-91 Laurentian Ice Sheet, retreat 295, 303 Lavas, pillow 517 Lead- 210 measurement 327-8 profiles 293 Leco Corp WR-12 carbon determinator 471 Lensoid deformations, mudflows 385 Light-scattering, forward 73 Limestone bituminous 511, 515-23 nanofossil 438-50 Linear programming, conversion of XRD results 363-71 Lithifaction, Great Stinkstone, Sweden 522 Lithostatic load, Bishop 569 Lithostratigraphy see Stratigraphy Long-range sidescan sonar system see GLORIA Lophocterium and N. Atlantic contourites 279 Lutites Angola Basin 112 definition 9 suspension cascading, fine laminations 441,443 see also Clay; Mudrock Lyngbya, effect on sedimentation stabilization 44-5 Magnesian calcite 199, 202-7 Magnesite, West Hellenic Arc 178 Magnesium, Oregon-Washington sediments
364,
369-70
Magnetic signature 161,164 Magnetic susceptibility, anisotropy (AMS) Cape Verde Basin 161, 163-4 North Atlantic contourites 250 Nova Scotia continental rise contourites core samples, results 264-71 discussion 271-2 Holocene sedimentation 259-61 methods 261-4 Magnetite 261
257-73
653
Manganese, measurement, atomic absorption spectrometry 117-18, 122 in surface sediments 364, 369 Marine sediments, chemical and mineral relationships 371 Markov sequence analysis, non-turbidite association 316-17 turbidite association 315 Marl, definition 9, l0 Marlstone laminated black, N. Atlantic 540-54 Scaglia Rossa 225-7 Mass-flow processes, in seismically active basins 185, 188 Mass gravity transport, definition 12 Mass movements fine-grained sediments 395-412, 417-31 processes 377 Ulleung basin 195 Santa Barbara Basin 384-7 scars, volume calculations 385, 391-2 tectonically active margin 384-7 Mass wasting, slope, events leading to 360 Massive bedding, vague lamination 490-4 Matapan Trench 169-80 Mediterranean, Eastern 497-508 Mediterranean, North-west margin, during late Quaternary 115-25 basin plain 121-2 composition of < 40 #m fraction 115 continental rise 121 continental slope 119-21 dynamic sequence 122 geochemical criteria 117 granulometry of < 40 ~m fraction 115-16 mineralogical criteria 116-17 palaeoclimatic sequence 122 ungraded sequences 122-4 Mediterranean outflow, general effects 276 Mediterranean, Western 169-80 Mesozoic-Cenozoic carbon-rich sediment, model for deposition 527 bottom-water oxygenation 530-2 discussion 550-4 DSDP analysis 527-8, 535-9 iithology, black shales 540-9 organic matter, preservation 528-9 productivity indicators 532-3 sedimentation rate 529-30 Metamorphism, extent in continuous coastal exposure 344 Methodology, principal text list 8 Methods, analytical atomic absorption spectrophotometry 364 calorimetry 364 coulometrics, Total Organic Carbon 471,474 Coulter Counter 73, 75 electronic particle analysis (Elzone) 295-306 freeze drying 364 GRAPE, wet bulk density 565, 571 Lamont measurement, forward light 72 lead-210 measurement 293, 327 Leco Corp WR-12 carbon determinator 471 linear programming 363
654
Index
Methods, analytical (cont.) Markov sequence 315-17 mineral abundance data 365-70 pipette analysis 295 Q mode factor analysis 363 radiocarbon dating (C14) 325-6, 595 Rock Eval, GEOCHEM 471,475 scanning electron microscopy (SEM) 204, 206 sediment trap measurements 75-7 thermovaporization gas chromatography 471,475 transmissometry 405, 407 viscometer measurement 40 wet bulk density profiles 565, 571 X-ray diffractometry 202, 335, 364 Methods, statistical end member composition 364-5 linear programming 364-5 see also specific headings Mica, enrichment 440 Micrite, in chalk beds 456 Microfabric of muds and shales see Muds; Shales Middle America Trench analysis 570-1,576 sediments 563 Mineral abundance data, from XRD results by linear programming 363, 369-71 Mineral composition, East Cape sediments, New Zealand 336-7 Miocene-Pleistocene, sediment accumulation, Ventura Basin 419-20 Miocene sequence, Santa Barbara 484 Mississippi Delta, deposits 581 Monterey Formation, California 481-94 Montmorillonite (clay) 19, 336-7, 456, 563, 574 Mud(s) beds, under Newtonian fluid flows, behaviour 35-69 calcareous 247-54 and fissility summary 637-8 and shales, fabric 579-93 major structural changes, clay sediments 588 microfabric, compacted clay 582-5 sedimenting clay 579-82 shear, effects of 585-7 synthesis and conclusion 587-90 turbidites, ideal facies model 615-17 Mudflow, deposition 384-92 Mudrocks and anoxic conditions 523 biogenic 9, 10 black, Cambrian 516, 517-20 calcareous sandy 331-41 classification 9 parallel-laminated 343-8 scanning electron microscopy (SEM) 335 structureless 343-9 see also Clay; Silt; Siltstone; Slate
Nepheloid flow 441 layer, benthic 124 circulation 247, 252, 340-1 deep ocean 57-8 Pacific and Atlantic 77-9 plumes 395, 400-12 Nephelometer, Lamont, SPM sensing 72-3, 76-9 Nereites 339 ichnofacies 595, 603, 606-7 New Zealand, calcareous sediments, upper slope to shelf 331-41 debris flow deposits 339 flora and fauna 337-8 interpretation and discussion 339-40 lithological descriptions 332-5 palaeocurrents 339 stratigraphy and facies 331-2 trace fossils 338-9 X-ray diffraction mineralogy 335-7 Normal bottom currents 12 North Atlantic contourites (overview) 245-54 bioturbation 249 bottom circulation 245-6 continental rise contourites 248 facies 254 grain size distribution 251 processes 252--4 sediment drifts 246-8 sedimentary characteristics 248-9 see also Faro Drift; Nova Scotia Slope Eastern and Northern DSDP sites 538-9 sedimentation rate, cretaceous 551-3 Western, DSDP sites 536--7 North-west Africa margin 595-6, 600-7 Norway, sediment transport deposition, late Cambrian 343-60 basin slopes and prodeltas 343 discussion 360 erosional-depositional discordant surfaces 355-6 LNM depositional model 359 Lower N/ieringselva Member (LNM) Finnmark, facies 344-50 soft-sediment deformation 350-2 synthesis, palaeocurrent data 356 Nova Scotia, continental rise, contourite, anisotropy of magnetic susceptibility 257-73 see also Magnetic susceptibility, anisotropy Halifax Formation, shales and slates 127-42 slope muds 311-17 discussion 31 6-17 physiography 311-12 sediments 312-16 stratigraphy 312 Nucleation kernels, in concretions 522 Nuculid bivalves, calcareous siltstones 337
Nfieringselva Member, Lower, (LNM) Norway 343-60 Namibian Shelf, organic carbon 534-5 Nanoplankton, calcareous 338 National Oceanic and Atmospheric Administration 319-20
Obock Trough (Gulf of Aden) late Quaternary calcareous clayey-silty muds 209-22 comparison of sequences 219-22 general characteristics 209-14 sequence interpretation 219 turbidites, characteristics 214-19
Index Obock Trough (cont.) frequency and origin 221 Oil, Alum Shale Formation 511 see also Hydrocarbons; Organic matter; Total organic carbon Oil seeps 417, 430 Olistrosome, rose diagrams 339 Olivine 194 Ooze, classification 10 pelagic 12 periplatform 199-207, 239 Optical methods, deep sea suspensions 73 Ophiolite, derived sediments 453 Oregon-Washington continental slope, fine-grained sediments 363-75 continental source area contributions 363-4 data and results 365-70 discussion--summary 370-4 sampling and analytical methods 364-5 Organic carbon vs carbonate carbon, Aegean seas 500 NW Africa sediments 598 -rich strata, Atlantic Ocean distribution 540-54 model for deposition 527-40 Sulu Sea sediments 597-8 Organic matter carbonates, generated hydrocarbons 469-71 and detrital minerals, Monterey formation 489-94 marine sediments, classification 589-90 terrigenous 528 'Orsten' see Bituminous limestone Orthoclase 231 Overburden pressure and compaction trends 569-70 see also Atterberg limits Oxygen content and decay 405-8 deep sea, bottom water 527-54 general criteria 533 and laminations 489-94 mud sediments 377-92 oxygen-minimum zones, black shale deposition 530, 549-51 varve-type lamination 444-6 and organic carbon 521-3 and pelagic settling 459 reducing environment 588 and sapropels 497-509 Oxygen- 18 determinations 595 Oxygenation of carbonate rocks 469-78 Pacific North-west, US continental slope see Oregon-Washington Pacific oceanographic circulation, Eastern 399 Palaeoclimate and anoxic conditions 497, 505 Palaeocurrents, data, LNM, Norway 356 East Cape, New Zealand 339 Palaeozoic, Lower, stratigraphy 513, 516, 519 Palygorskite 444 Parallel-laminated mud, Ulleung Basin 191 Particle analyser, electronic 295-306 Particle analysis, statistical treatment 295-305 Particles, aggregation and size 46-51 Particulate matter see Suspended particulate matter
655
Pelagic carbonates, Blake-Bahama Basin 437-50 clay, definition 12 deposition, Blake-Bahama Basin 437-50 diatomaceous muds 563, 569 ooze 12, 626 sedimentation, 'black shale' 527-54 history 3 Pelagites criteria 624-5 cumulative frequency curves 115-17 gradation with contourites 254 and organic-rich sediments, Western North Atlantic 437 Scaglia Rossa limestones 237 see also Calcilutites; Hemipelagite Pelites definition 9 NW Mediterranean margin, late Quaternary 115-25 palaeoclimatic sequences 122 West Mediterranean Basin 121-2 see also Mudrocks; Pelagites Pelleting, in sedimentation 380 Peloponnesus see Hellenic Arc Periplatform oozes, sources of hydrocarbons 202-7, 471-9 Peruvian margin, organic carbon 534 Petroleum production, diatomaceous deposits 481 see also Hydrocarbons; Organic carbon Phosphorus accumulation rate, organic carbon 534 Phyllites, black 517, 523 Piper's sequence, turbidites 109, 316, 613 biogenic turbidites 618 Pipette analysis, grain size 295 Piston core analysis see Core analysis Plagioclase 231,336-7, 365-8 Plane lamination, Obock Trough 214 Planktic analysis, East Cape sediments, New Zealand 337 Plankton productivity 444 Planolites 501 Plasticity Casagrande chart 569 indices, Guatemalan Transect 574-7 Plate tectonics, history and discussion 5 Plumes, surface water 396-400 Pohutu formation, calcareous mudstones 331-2 Ponding, in synsedimentary reworking of deposits 122-4 Porcelanite 484 Pore size, volume, and plasticity 569 Post glacial clays, lake 293-306 Potassium 364, 366, 369-70 Primary production, and benthic food content 600--1 surface waters 528 Pro-delta, and delta deposits 343, 359 lobes 119-23 Providence Channel sediments, North-west 199 Pteropods, source ofaragonite 199, 210 Pycnoclines, and sediment movement 400 Pyrite, black laminated mudstone, Swedish Alum 517, 519-23
656 Pyrolysate index in TOC Pyroxene 194
Index 475-9
Q-mode factor analysis 363 Quartz East Cape, New Zealand 336-7 laminae 517, 520-3 Oregon continental slope 365-6 sands, US mid-Atlantic region 327 Quartzose silt 440, 442 Quaternary sedimentation, NW Mediterranean 115-24 Question Set Approach, Potter's 6, 7 Radioactive waste disposal, sea bed, preliminary study 145-51 Radiocarbon dating indicator of sediment accumulation 325-6 sapropels 498-509 Radiolarian carbonates, nanofossil 437, 440-50 clay, black and green 548 ooze 12 Rainfall, and varve lamination 380 Reducing sediments see Sapropels Resedimentation processes 12 Ripple-markings 343, 348-50, 3 5 6 - 7 Ripples asymmetric 380 interference 349 River discharge 17, 18, 21 Rivers, sediment load 95-112 see also Fluvio-deltaic Rock-Eval analysis, GEOCHEM 471,475 Rose diagrams 339 Salinity, dependence of erosion rate 45 Salir Formation, SW Turkey 453-5 Salt alum 511-23 Sampling methods anisotropy, magnetic susceptibility 261 box cores 295 gravity cores 295 piston cores see Core analysis seismic reflection profiles see Seismic profiles Shipek grab sampler 200 sidescan sonar image, GLORIA 319-24 San Andreas fault 482 Sand(s) content and compaction 4I-2 fractions, Faro Drift 283 quartz 326-7 shelf, as tracers 327-9 -silt-clay, terminology 9 spillover 341 waves, Hueneme Sill 405 Sandstones 129-42 calcareous, muddy 336-7 siltstones, Swedish Alum Shales 520 wave rippled fine-grained 343-50 Santa Barbara Basin, California Continental Borderland 377-94 current transport, fine-grained sediment 380-3 hemipelagic sedimentation 377-9
mass movement 384-7 setting and previous studies 379-80 stratigraphy 388-92 hemipelagic deposits 481-95 geology 481-3 layering 489-93 lithologic sequences 492-3 palaeogeography 482 sediment composition 483-9 stratigraphy 483 synthesis 493-4 Santa Cruz Basin Fault 482 Santa Monica Basin 396 Santa Paula Creek, California fan-lobes and fine-grained sediments 417-33 conclusions 430-1 sediments, zones 1-4, 421-30 structural setting, stratigraphy, Ventura Basin 419-21 tectonic and sedimentary controls 417-19 Santa Ynez Fault 481-2 Sapropels 172-8 marine sediments, Eastern Mediterranean 497-508 definition 497 episapropels 502 lithofacies 498-501 methods 498 organic rich deposits 501-4 as palaeoceanographic indicators 504-8 petrological classification 508 X-radiographs 505, 507 Mediterranean 531 Black Sea 549 Eastern N. Atlantic 548 Sarl 9, 10 Satellite imagery (LANDSAT), turbid plumes 397-400 Scaglia Bianca, central Italy 225 Variegata 225 Scaglia Rossa limestones, calcilutites 223-41 broader implication 239 characteristics 228 bedding 228-30 bioturbation and burrowing 234 colour 232-3 composition 231-2 structure and texture 234 depositional model 237-9 facies 227-8 interpretation 234-6 stratigraphy 223-7 turbidite and pelagite characteristics 237 Scattering measurements, forward suspended matter 72-3 Schuppen structure see Imbricate Sea-level changes and tectonic effects 359 effect on sedimentation patterns 325 Sea MARC 1 (mid range sidescan sonar system) 319 SEABEAM, bathymetric mapping 150, 156 SEABED cruise see Seismic survey Seabed Working Group, radioactive waste disposal 145
Index Sediment(s) accumulation, by lead-210 dating 327-8 measurement, by radiocarbon dating 325-6 budgets 371-3 draping 351,354-5 drifts, North Atlantic 246-54 mass gravity processes 169 slides, slumps see Slides; Slumps transport, down-canyon 329 implications of plasticity 574 processes, East Cape, New Zealand 340 US mid-Atlantic, canyons 319 mass wasting 329 trap measurements, Western Atlantic basins 75-7 /water interface, oxygen levels 533 see also Contourites; Fine-grained sediments; Turbidites Sedimentary sequences, reconstruction 115-24 structures, graphic symbols 11 Sedimentation rates Alum shales 517-23 and accumulation rate 532, 548-9 carbon-14 dating 252 deep sea 598-607 equation 372 glacial and inter-glacial 474, 477 and mass wasting 360 Mesozoic-Cenozoic strata 529-30 organic carbon content, North Atlantic 551 Swedish Alum Shales 517, 518 Seismic activity, sediments 417, 430 airgun profiles 170 reflection profiles 175, 185, 193, 209, 212, 319-20 in bathymetric maps 293, 294 Faro Drift 277 Ulleung basin 190 and sediment coring survey 153-66 SEM, calcareous sediments 332-5, 337 Serpentinite 454, 456 Settling velocity, particles 55-6 Sewage plume 402, 403 Shales Alum see Alum shales black see Black shales definition, Potter Question Set 7 Halifax Formation 129-42 and pressure 582 successions 6-7 terminology 8-10 see also Argillite; Clay; Claystone; Silt; Siltstone; Slate Shear-compressional stress 563 'Shear sorting' in laminated turbidites 587 Shear stress, related to erosion rate 42-4 Shearing forces, clays, effect on microstructure 585-7 Shelf sediments, dispersal by turbidity flows 327 Shipek grab sampler 200 Sidescan Sonar images 293-303 GLORIA 319-24 mid-range (Sea MARC 1) 319 Silica (silicon) 231-39, 364, 369 rich sediment, and high productivity 481-94
657
see also Chlorite; Illite; Pyrite; Quartz Siliceous turbidites 618-19 Siliciclastic sediments 9 Silicoflagellates 97 Sillimanite 194 Silt, classification 20 -sand-clay, terminology 9 turbidites, summary 613-15 Siltite see Siltstone Siltstone 129-42 calcareous sandy 331-41 current rippled 343-9 SEM 333 Sisquoc Formation, Miocene layering 490 Slide(s) diagnostic features in DSDP cores 343 folds, overturning, Lower N~ieringselva Member 351 sheets 352, 353 from soft sediment deformation 344-60 Slope(s) aprons 343 definition 631 devoid of canyons, gullies 377 failure, triggering mechanisms 378 mass wasting, events leading to 360 sediment, redeposition 528-9 -to-shelf sequences see New Zealand, calcareous sediments Slump scars 174, 187, 191-3 Slump packets 331 Slump-slides, Lower Nfieringselva Member 351-2 Ulleung basin 191--4 Slumping, in canyon erosion 328 sea floor 112 Smarl 9, 10 Smectite 116-17, 122, 176-8, 210, 212, 231,365, 444 see also Illite Sodium 364, 369 Soft-sediment deformation breccias, intraformational 350, 352 fluidization structures, Norway 350, 352 slides 343, 350, 358-60 synsedimentary faults 351,355 Sohm Abyssal Plain, turbidite sampling 85-91 Soil Classification, Unified 569 Sonar, images, sidescan 319-24 studies, sidescan 258 Sonograph, long range sidescan (GLORIA) 145-51 South Atlantic, Western, DSDP sites 535 South East Atlantic Ocean, surface current patterns 97 Spectrophotometry, atomic absorption 364 Spinel 178 Stabilization of sediment by biological material 44-6 Stagnation, and sapropels 497-509 Statistical analysis, particle analysis 295-305 Statistical methods, computer contouring, SYMAP 379-80 Staurolite 194 Stinkstone see Bituminous limestone Stokes'-deposited sediments 83-91 Storms, effect of carbonate transport 203
658
Index
Storms (cont.) frequency and intensity, Lake Superior evidence 305-6 Stow sequence, turbidites 619 Stratification terminology 10-11 Stratigraphic correlation, Southern Cape Verde 158 Stratigraphy Cape Verde Basin southern 156-9 Scaglia Rossa limestones 223-7 West Hellenic Arc basin 170-2 Strontium 364, 366, 3 6 9 - 7 0 Subarkoses 346 Sugars, in marine sediments 589 Sulphate reduction, organic carbon sediments 530, 532 Sulu Sea, bioturbation features 597-603, 604-5 Surf, internal 400-1 Surface currents, Santa Barbara Basin 381 Suspended particulate matter measurement, forward light scattering 71, 72 gravimetric 73 particle counting 75 particle size, spectra 77 variations, regional and temporal 72 Suspended sediment, continental slope, annual input 363 Swedish Alum Shales, sedimentology see Alum shales Swedish Caledonides succession 515 Swedish Deep Sea Expedition 71 Symbols, graphic 11 Systemic sedimentology, history 3 Tectonic(s) activity, Santa Barbara Basin 377-92 controls, and volcanic centres 482 factors, carbon-rich strata, Pacific 548 Pacific, Miocene 554 stresses, rift basin 409 transform-margin 417 zones, en echelon 453 Telegraph cables, corrosion studies 276 Tephra upper and lower, ash layers 172-5 volcanic 331-8 Terebellina, New Zealand and California 339 Terminology 6-12 descriptive 6-10 interpretative 10-12 Terrestrial organic matter, periodicity 446-7 sedimentation, Miocene 527-54 Terrigenous detritus 77 nepheloid settling 549 sedimentation rates 390-2 silt as indicator 381 Tethyan sequence, organic carbon 540, 550 Tethys sea, and Atlantic 444 Textural analysis see Grain size Thalweg, canyon 325-6, 328 Thermovaporization gas chromatograms 471,475-8 Titanium, California Borderland 364, 366, 369-70 TOC (total organic carbon) carbonate muds 469-79 profile and Plasticity Index, Guatemalan
Transect 576 Tongue of the Ocean see Bahamian Trough Trace fossils, calcareous slope sediments 338-9 Trace metals, Alum shales 511, 515 Transform motion see Tectonics Transition probability, in chalk/turbidite deposition 463-6 Transmissometer casts, deep water and surface water 405, 407 Transport California Coast Range 363-70 current velocity 288-91,319 down slope continental rise 377 long distance, and flow initiation 612 processes, fine-grained sediments 395-412 related to erosion 35-63 aggregate strengths and sizes 48-51 and aggregation 46-8 biological influences 44 breakup kernels 51 compaction 41 conclusion 61-3 deposition 59-61 erosion 36 rate 42 Newtonian fluid flows 35 settling velocity 55-6 vertical and horizontal flux 56-8 Trilobites 511,522 Trough sediments, Lake Superior 295, 301-6 Turbidite(s) analysis, Angola Basin 98-112 NW Mediterranean 116-24 Bouma standard sequence 4, 613 chronology, Santa Barbara Basin 390-91 and contourites, distinguishing criteria 293, 316-17 criteria 612-17 deposition, and aeolian dust 603 model 83, 84 facies 12 comparison with contourite 316-17 floc-deposited 83-91 global distribution 630 grain-size characteristics 83-92 analytical methods 87 discussion 90-1 material 85 muddy, sedimentary core structures 86, 87-8, 90-1 results 87 settling model 84-5 ichnocoenoses 595-607 interbedded 293 muddy, carbon rich 527-54, 563 see also Black shales pelagite, end-members 223, 237 revolution, history 4-5 and sapropels 504-9 SE Atlantic Ocean, climatic zonation 100 sedimentation, principal areas of world 630 sequences, Nova Scotian muds 312-16 shear sorting 83-91 summary, biogenic, 617-20
Index Turbidite(s) (cont.) disorganized 620 mud 615-17 silt 613-15 Stow sequence 619 type sequence 218, 220 X-radiography 98, 102, 104-5, 107-8 see also Contourites Turbidity currents cause of graded bedding 4 deposition processes, California Borderland 405-12 Obock Trough 214-21 Scaglia Rossa 234-9 effects of diapirs, Cape Verde 163 frequency and chalk horizons 453 transport, Santa Barbara Basin 377-9 see also Mass flow Turbidity patterns, basins and sills 405-12 subsurface shelf 400-2 Turkey, SW, Salir Formation 453-66 'Type sequence' see Turbidite sequence Tyvofjell Formation, Norway 344 Ulleung Basin, Sea of Japan, turbidites and associated deposits 185-94 acoustic stratigraphy 187-8 geological setting 185-6 hemipelagic settlements 191-4 physiography 186-7 processes 195-6 provenance 194-5 turbidites 188-91 Umbro-Marchean Apennines see Scaglia Rossa limestones Ungraded deposits see Contourites Unified Soil Classification see Soil classification Unifites, Angola Basin muds 109-11 terminology 12 United States mid-Atlantic region, canyons, late Quaternary 319-29 see also Wilmington Canyon Upwelling, classic mechanism 399-400 intensity, in planktic production 488-94 of nutrients 447 and productivity, deep sea 532-41,548-9 Uranium 511,515 Vague lamination (massive bedding) 490-2 Varve-type lamination Blake-Bahama Basin 437, 440-4 criteria 637 Monterey Formation 491 post glacial, Santa Barbara 379-80 Santa Barbara deposits 485, 4 9 0 - 2 Vegetation changes, in interpreting sediments 303 Ventura basin, California see California Continental Borderland tertiary deposits 420 VERTEX program, sediment trap arrays 76-7 Vertical flux, particles 56-7
659
Viscometer measurements, cohesive material, critical erosion stress, 40-1 Volcanic activity, carbon rich strata 548 events, active margins, periodicity 18 sources 17,18 tephras 172-5, 331-8 Water chemistry, varves and cyclical alternations 446 content of sediment 37-40 depth, and wave height, sandy shelves 340 trace fossil indicators 595-607 sampling, objectives and methods 73-5 Wave height and water depth, sandy shelves 340 Waves, effect on sedimentation 331,340-1 see also Internal waves; Sand waves Western North Atlantic see Blake Bahama basin Wet bulk density profiles 565, 571 Wilmington Canyon, canyons, late Quaternary 319-29 continental slope 319-20 discussion 329 physiography 320-4 processes 328 sediments 325-8 stratigraphy 325 see also Canyons Wind transport see Aeolian transport Winnowing, current 380 mechanisms, shelf to slope 340-1 X-radiography core analysis 325, 326, 379, 385-92 piston cores 379, 388, 389 turbidite 98, 102, 104-5, 107-8 sapropels 498-509 X-ray diffraction end member compositions, using discharge-weighting 364-70 mineralogy 335-7 techniques 200-1 Zaire deep-sea fan 95-112 conclusions 112 geological setting 95 grain size analysis 101, 109-11 hydrography 96 methods 97 study area 96 X-radiography 102, 104-8 Zaire River, suspended matter load 98, 99 Zephiran chloride, sterilisant 471 Zonation, bioturbational levels 600 Zoophycos
in biogenic traces, fine grained sediment 599-604 in Mediterranean sapropels 501 stratigraphy, Nova Scotia 312 Zooplankton, biofiltration of faecal pellets 395, 399-400, 412, 431