Depositional Models for Fine-Grained Sediment in Western Hellenic Trench, Eastern Mediterranean: ABSTRACT

Sedimentology (1981) 28,273-290

Depositional models for fine-grained sediment in the western Hellenic Trench, Eastern Mediterranean

D A N I E L JEAN STANLEY and A N D R E S M A L D O N A D O *

Division of Sedimentology, Smithsonian Institution, Washing ton, D.C. 20560, U.S.A., and *CSIC, Egipciacas 9, Barcelona-1, Spain

ABSTRACT

Sediment in tectonically active, topographically restricted settings of the western Hellenic Arc, eastern Mediterranean, consists primarily of clayey silt and silty clay. Failure of metastable sediment tempor- arily stored on relatively steep slopes is triggered by earthquake tremors and eustatic oscillations. Re- deposition of these materials by gravitative transport has resulted in markedly different lithofacies from site to site. Most piston cores include three Late Quaternary stratigraphic units that can be correlated with sections in other parts of the eastern Mediterranean; numerous radiocarbon-age determinations enhance the correlation.

Seven fine-grainedsediment types are identified in cores fromeight dist inct depositional environments. Some muds are closely related to specific environments (slump and debrisflow deposits on slope and high-relief environments), or to time (well luminatedmud during the latest Pleistocene-mid-Holocene), or to both (uniform and faintly laminated muds restricted to trench basins), Turbiditic and hemipelagic muds are common throughout the study area. Mud distribution patterns correlate closely with calculated sedimentation rates.

We propose two depositional models for these sediments. The first emphasizes downslope transformations resulting in progressively reduced flow concentration during transport : from slump and debris flow+turbidity current+low density turbidity current or turbid layer mechanisms. The distal end- member deposits settling from low concentration flows are thick, rapidly emplaced, fine-grained uniform muds closely associated with faintly laminated muds. These were ponded in flat trench basin- plains. Planktonic and terrigenous fractions in the turbiditic, finely laminated and uniform muds record mixing of materials of gravitative and suspension origin during redeposition. This sequence prevails under conditions of minimal stratification of water masses, as characterized by the present Mediterranean.

In the second model developed for conditions of well-developed water mass stratification, well laminated rather than uniform mud prevails as the end product of low concentration flows. These very finely laminated and graded muds record particle-by-particle settling from detached turbid layers concentrated along density interfaces; they include material from turbid layers complemented by the normal ‘rain’ of pelagic material. Stratification barriers resulted in region-wide distribution of such deposits, in both slope and trench environments.

INTRODUCTION

Mud, rather than sand, is the dominant Late sediment input during this period and the short Quaternary facies in piston cores recovered from dispersal path between source and depositional basins of the Mediterranean Sea. This finding is site in these small enclosed settings. Fine-grained somewhat surprising in view of the relatively large sediment is not randomly distributed, but occurs

a t distinct times, or in specific regions, or both, oO37-0746/81 /04OO-0273 $02.00 and generally has accumulated rapidly in the 0 1981 International Association of Sedimentologists different environments examined to date (Stanley

274 D. J. Stanley and A . Maldonado

200 210 2 2 O 2 3 O

Fig. 1. Chart showing the position of 24 western Hellenic cores examined in this study (see Table 1); eight distinct depositional environments are recognized. ZB, Zakinthos- Strofadhes system; MB, ' Matapan Deep' trench system; KB, Kythera-Antikythera trench system.

& Maldonado, 1977). Mud distribution patterns vary from basin to basin, within any one basin, and at any one depositional site. Certain mud types are related to the seafloor configuration, depth, distality and palaeoceanographic conditions at the depositional site (Maldonado & Stanley, 1977, 1979).

Preliminary investigations also suggest that fine- grained Mediterranean lithofacies can be differenti- ated on the basis of depositional process, that is, derivation from fluid-driven and suspension mechanisms, or from essentially gravitative transport (Bartolini, Gehin & Stanley, 1972; Huang & Stanley, 1972; Rupke &Stanley, 1974; Maldonado &Stanley, 1977, 1979). As might be expected, distinctions are often ambiguous due to the interplay of several mechanisms involved in the transport of fine-grained sediment in marine continental margin and basin settings (Pierce, 1976). The present study was initi- ated to help clarify some aspects of sedimentation in the western Hellenic Arc, Ionian Sea. It bears on the problem of identifying types and distributions of muds, and on their dispersal in submarine trench settings. Sedimentation of the sand-size terrigenous fraction, which accounts for a relatively small pro-

Table 1. Position, water depth and length of cores recovered in the western Hellenic Arc (Trident cores 32- 37; MarsiIi cores 1-23)

Core station Latitude Longitude Depth Core number (m) length (cm)

R/V Trident (cruise 172, 1975) 32 37" 36.5' 20" 246' 3345 378 33 37" 25.6' 20" 18.2' 3820 339 34 37" 21.5' 20" 19.2' 4060 694 35 37" 22.9' 20" 21-3' 4140 609 36 37" 284' 20" 344' 4147 690 37 37" 38.2' 20" 35.3' 1200 215

R/V Marsili (Hellenic cruise, 1976) 1 36" 34.5' 21" 94' 4789 578 2 36" 32.5' 21" 4.3' 4393 507 3 36" 42.05' 21O33.1' 3030 537 4 36" 16.1' 21" 56.9' 2745 527 5 36" 179' 21'56.9' 2720 279 6 36" 20.1' 21" 52.3' 3050 468 8 36" 48.4' 21" 31.7' 2860 478

10 35" 45.5' 22" 27.3' 4300 565 111 35" 43.9' 22" 24.1' 4362 379 112 35" 44.2' 22" 23.1' 4305 527 12 36" 00.1' 22" 26.7' 3060 567 14 36" 175' 22" 38.9' 1610 506 17 36" 5.4' 2T25.6' 2979 509 18 35" 39.3' 22" 56.6' 2833 518 19 35" 294' 22" 41.5' 4000 528 20 35" 52.4' 22" 19.0' 4510 >lo00 22 36" 20.0' 21" 29.2' 4090 736 23 36" 26.2' 21" 18.1' 4120 570

portion of the Late Quaternary cores, is discussed elsewhere (Feldhausen & Stanley, 1980).

The small, confined sector west and south of the Peloponnesus in the eastern Mediterranean (Fig. 1) is noteworthy because of the remarkable diversity of mud types. Earlier petrologic investigations in this region, involving piston cores (Ryan et a/., 1970; Hieke, Sigl & Fabricius, 1973; Stanley, 1974, 1977; Got, Stanley & Sorel, 1977; Nesteroff, 1977; Vittori, 1978; Stanley et al., 1978a) and Deep-sea Drilling Project cores (Ryan, Hsu et al., 1973; Bartolini et al., 1975), focused extensively on mass- emplaced layers, sandy deposits and sapropels (dark organic-rich layers), and emphasized difficulties in correlating cores from site to site in the Hellenic Trench. Irregular sediment distribution patterns would be expected partly as a response to the complex topography of this area (Hersey, 1965; Pareyn, 1968; Ryane td , 1970; Carter, Flanagan etal., 1972). The many distinct depositional environments within a small geographic sector are directly related to the geologically recent evolution and the continuing

Depositional models for fine-grained sediment 275

intense tectonic activity of this region (many studies summarized in Brunn, Aubouin, Mercier et al., 1976; Biju-Duval & Montadert, 1977; Kallergis, 1979). Frequent triggering of mass sediment failure and almost continuous downslope sediment displacement would be expected in an area characterized by such high seismicity (Papazachos & Comninakis, 1977; Makris, 1978; McKenzie, 1978). The highly irregular configuration of Plio-Quaternary sedimentary series identified on continuous seismic sub-bottom profiles reflects the strong tectonicoverprint on sedimentation (Vittori, 1978; Le Quellec et al., 1978). Other pheno- mena that would have affected Quaternary sedimentation patterns include eustatic oscillations and major basin-wide changes in water-mass circulation of the type recognized throughout the eastern Mediter- ranean (Stanley & Maldonado, 1977).

A combined petrologic and lithostratigraphic approach is used in this study. We believe that interpretation of depositional processes in a setting as complex as the western Hellenic Arc is feasible only after identification of the major mud types, determination of Sedimentation rates, and careful analysis of their specific spatial and temporal distribution patterns.

METHODS

This study is based largely on 24 piston cores collected in western Hellenic trench slope and trench basin systems on the 1975 Mediterranean cruise of the R/V Trident and 1976 cruise of the R/V Marsili (Fig. 1 and Table 1). High-resolution 3.5 kHz sub- bottom profiles obtained during a survey of this region on cruises of the R/V Bannock (1974) and R/V Marsili (1976) complements the core study; selected records are illustrated by Stanley (1977) and Vittori (1978). Airgun and sparker data collected in this same region (R/V Trident cruise in 1975; R/V Dectru cruise in 1976) supplement topographic information and define the sub-bottom configuration and long-term dispersal patterns of unconsolidated Plio-Quaternary sections underlying the core sites (Vittori, 1978; MEDIBA Group, 1979).

Split cores (3-10 m long) were photographed and X-radiographed. Comprehensive lithologic logs for each core (not presented in this paper) incorporate visual information (colour and texture) of the lithologic sections, sedimentary and biogenic structures (primarily from X-radiographs), and petrography; the latter includes textural and conipo5itional analyses

of about 450 core samples (R. J. Knight, 1977, personal communication). Lithostratigraphic interpretation of core sections is enhanced by the avail- ability of numerous radiocarbon dates (data in Stanley et al., 1978a). For each core, the relative percentage of each mud type and the rates of sedimentation were calculated for three Late Quaternary stratigraphic horizons and for the entire core length. This information is depicted graphically in Figs 2 and 4-1 1 ; the original numerical data are listed in tables available from the authors.

0 BSERVATION S

Depositional environments sampled

The study area occupies the northwest-southeas t trending sector of the western Hellenic Arc, and cores were retrieved between the western-southern Peloponnesus Shelf region west of Crete, and the Mediterranean Ridge (Fig. 1). This fore-arc sector (about 450 km long and 100 km wide) comprises three major structural-physiographic complexes : Zakinthos-Strofidhes (ZB) trench basin-slope complex; ‘Matapan Deep’ (MB) trench basin-slope complex; and Kythera-Antikythera (KB) trench basin-slope complex. Most major physiographic features of the three individual basin-slope systems are parallel or normal to the major north-west-south- east trend of this Hellenic sector, and this configuration is explained in terms of the regional structural framework (Got et al., 1977; Le Quellec, 1979). The ‘Matapan Deep’ (MB), with a maximum depth slightly greater than 5000 m, is the deepest part of the Mediterranean (Hersey, 1965). The other two basins (ZB and KB) are 4148 m and 4614 m deep, respectively (Carter et ul., 1972).

Eight environments (identified in Fig. 1 and shown schematically in Fig. 2) were cored in the three major complexes. These settings are identified on the basis of the close-grid 3.5 kHz surveys made at the core sites, and are grouped in two major categories: trench slope and trench basin. The trench slope environments include: (a) steep, straight, inner trench wall (represented by R/V Trident, or TR, core 37); (b) irregular, broken, high-relief slope, including ‘cobblestone’ (Ryan et al., 1970; Stanley, 1977) topography (TR cores 32, 33 and R/V Marsili, or MA, cores 4, 5 , 6, 17); (c) submarine valley (MA core 8); (d) proximal perched basin (MA core 14); and (e) distal perched basin (MA cores 3, 12, 18). I n this

276 D. J . Stanley and A . Maldonado

Fig. 2. Logs showing the distribution of the upper three Late Quaternary lithostratigraphic units examined in 24 cores from eight depositional environments. Length of each log is proportional to the calculated rate of sedimentation (in cm loo0 yr-l). The theoretical stratigraphic sections in lower left (after Stanley& Maldonado, 1979) showdated key horizons used for regional correlation (C, calcareous ooze; OL, oxidized layer; S, sapropel).

classification slope, amount of relief and distality Maldonado, 1979) is possible. Correlation is further (distance from land) are the most significant attri- refined by numerous radiocarbon dates obtained for butes. these cores (in Stanley e ta / . , 1978a).

The following depositional environments make up Diagnostic sediment types that occur only at the trench basin category: (f) flat trench basin-plain specific stratigraphic levels throughout most of the proper (TR cores 35, 36; MA cores 1 , 22, 11', 20); eastern Mediterranean are also present in the (8) smooth, slightly sloping apron merging with Hellenic Trench. The most useful for regional trench plain (MA core 2, 1 1 2 , 19); and (h) zone in correlation and assignment of approximate age are: flat basin-plain near a high-relief contact (TR core dark sapropel (S), reddish-brown oxidized ooze (OL), 34, MA cores 10, 23). Marsili core 1, collected in the and light yellow calcareous ooze (C). These dated 'Matapan Deep', is the deepest (4789 m) core key horizons (Stanley & Maldonado, 1979) are most studied. easily recognized in slope cores. Sapropels are less

obvious and display markedly different character- istics in trench cores, a phenomenon discussed later

Lithostratigraphy of core sections

Late Quaternary sections in the western Hellenic Arc show more obvious core-to-core variability than do those from topographically less complex regions of the eastern Mediterranean such as the Ionian Basin plain and the Nile Cone. Visual observation reveals a diversity of sediment types and lithostratigraphic sequences, with low comparability between cores, even those collected close together. Nevertheless, examination of the entire core suite enables us to identify key lithostratigraphic criteria in most core sections. Thus, in most cases, correlation between Hellenic cores and complete type-sections defined in more tectonically tranquil regions (Stanley &

in this paper. The position of key horizons with respect to each other is used to identify the section of stratigraphic record recovered in each core (see log in Fig. 2). Three major lithostratigraphic layers, from top towards the base of the Hellenic Arc cores, are defined as follows: Iavpr I , from f o p of core to, and including the upper oxidized layer OL, (about 5700'yr BP to present): luyer 2, from the base of OL, to the top of C, (from about 17,000 to 5700 yr BP); and layer 3, which includes C , and underlying sediment above the second sapropel, S, (from about 23,000 to 17,000 yr BP). Many cores fail to recover sections below layer 3 (at, or below, key horizon S2) because of very high sedimentation rates in this region. Thus, in order to be consistent with the analysis of mud


type distributions, we restrict our observations to the upper three stratigraphic layers.

The lithostratigraphic sequence is moderately to well developed in basin cores in the three trench systems (Fig. 2); in contrast, some cores from the slope (particularly in areas of marked relief) show only parts of, or may display sections a t variance with, the sequence described above. Such anomalies, including disrupted sections, are expected in a tectonically active region of this type where earlier studies have shown that sediment failure (Hieke et a/.,

1973; Got et al., 1977) and downslope redeposition (Stanley, 1977; Vittori, 1978) prevail.

Mud types

Classification of fine-grained sediment types requires consideration of several properties, including gross lithology observed in split cores, structures in X- radiographs, and textural and mineralogical attributes. The Late Quaternary deposits for the most part are clayey silt, silty clay and silt as defined in the

Fig. 3. Selected X-radiographic prints of R/V Trident cores showing the seven major fine-grained sediment types discussed in text. (A) Bioturbated hemipelagic mud (He), and uniform (Un) mud; core 34, 322-356 cm. (B) Well laminated (La) mud; core 35, 240-250cm. (C) Well laminated mud; note pteropod shell (arrow) in finely laminated section. @) Uniform (Un) and faintly laminated (FI) muds; core 36,625-659cm. (E) Turbiditic (Tu) and hemipelagic(He)muds; core 35,202-216 cm. (F) Two turbiditic mud (Tu) layers; core 36, 182-199 cm. ( G ) Muddy debris flow layer (Df); core 32, 325-333 cm. (H) Slump (9); core 33. 240-265 cm.


Shepard textural classification. Seven mud types are distinguished, but it should be noted that gradational varieties between these are also present. Four of the mud types are generally well known in the Mediter- ranean and elsewhere, have been adequately described in the literature, and need no detailed elaboration here. These are: (a) chaotic slumped mud sections, (b) muddy debris flow deposits, (c) turbiditic mud and (d) hemipelagic mud.

Slump (Fig. 3H)

This type, the most obvious of the group, displays deformed stratification and, in some cases, a chaotic mix of several mud types and sharp truncation surfaces, or disrupted internal bedding, or both. Slumps include en mane displacement of most of the sediment types described below; some original structures of these allochthonous muds are generally preserved. Slump layers range in thickness from several decimetres to over three metres. Similar deposits from the western Hellenic region have been illustrated by Hieke et al. (1973, fig. 9).

Muddy debrisflow deposit (Fig. 3G)

This type displays a coarse mix of mud clasts of vary- ing size in a matrix of muddy sand, silt or mud and, unlike slumps, does not display any original stratification features. Layer thickness ranges from several decimetres to over one metre. Comparable units recovered in Mediterranean cores are illustrated in Ryan etal . (1970, fig. 21).

Turbiditic mud (Fig. 3E, F)

This type is identified in X-radiographs on the basis of graded bedding. The lower part of such units is often laminated (parallel or low-angle cross- lamination), and the upper part may appear uniform. Colour of turbiditic mud is closely related to the lithostratigraphic unit in which it occurs: dark olive- grey (5Y2/2) to greyish-orange (lOY4/2) in sapropel- related Layer 2; very pale orange, greyish-orange, or pale yellowish-brown (1 OYR8/2,10YR7/4,1 OYR6/2) in layers 1 and 3 associated with calcareous ooze. Layer thickness is highly variable, ranging from a few centimetres to over one metre. Compositional textural parameters usually display progressive vertical changes associated with grading. The sand and silt Fractions commonly include variable proportions of terrigenous and bioclastic material.

Similar turbiditic mud in the Mediterranean has been illustrated by Rupke & Stanley (1974, fig. 4) and Maldonado & Stanley (1977, fig. 2).

Hemipelagic mud (Fig. 3A, E)

Somewhat less obvious than turbiditic mud, this type is recognized primarily on the basis of structures seen in X-radiographs and on petrography. Core sections are commonly bioturbated, or show speck- ling produced by a high content of planktonic tests dispersed in the mud matrix (Rupke & Stanley, 1974, fig. 4). Hemipelagic mud, as used in this paper, includes all sediment varieties inferred to have been deposited primarily by settling through the water column, including calcareous and organic oozes and all sapropel-related deposits as defined in Maldonado & Stanley (1976a). Colour is not a definitive criterion and, as in the case of turbiditic mud, is most closely related to the lithostratigraphic layer in which it occurs, i.e. dark in sapropel-related layer 2, and pale yellowish-orange or brown in calcareous ooze- related layers 1 and 3. Individual layers generally range from a few to not more than 20 cm.

Three other mud types are distinguished in the Hellenic region: (e) well laminated, (f) faintly laminated and (g) uniform. These probably have been confused in some previous studies with turbiditic or hemipelagic deposits, remain poorly defined in the literature, and warrant more detailed description.

Well laminated mud (Fig. 3B, C)

This type is characterized by relatively thick sections of distinct, thin ( < 1 mm to a few mm), parallel laminae observable in both split cores and X-radiographs. Some varve-like mud layers show grading, and this is reflected by fining-up texture and vertical compositional changes. The absence of bioturbation, stratification and structures typically produced by bottom currents is noteworthy. Within a single section carbonate content, of largely planktonic origin, shows large vertical fluctuations. Occasional large, asymmetric planktonic tests are observed. These are always flat lying and buried by thin individual laminae (arrow in Fig. 3C). Sand content accounts for less than 2 % of the total sediment, and the clay size fraction ( < 4 pm) may range from 40 to 70 O/( within each varve-like layer. Carbonate per- centages range from less than 30 yo to more than 40 %; the sand fraction consists of more than 50% plank-


tonic foraminifera. Total organic matter content is rarely more than 5%. Sediment colour generally ranges from light greenish-grey (58G/1) to dark grey (N3). Sections of superposed, well-defined laminae are variable in thickness, and range from a few centimetres to more than one metre.

Dark, finely laminated muds are interbedded, and thus can be confused, with sapropels (see hemipelagic mud). Sapropels sensu stricto, however, have a considerably higher organic matter content (generally > loo/) and do not show grading. Gradations between well laminated muds and sapropels occur, and distinction in each case requires detailed petrologic analysis.

Uniform mud (Fig. 3A, D)

These greyish-yellow (5Y8/4) or greyish-orange (10YR7/4) muds are characterized by an almost total lack of structures, as noted in split core and X- radiographic analysis. Locally, very faint lamination, or grading or bioturbation may be present. Colour and petrographic variations are, at best, subtle. Generally such units are more than 1 metre thick, but may be as thick as the largest core section recovered ( 1 0 m). Grain size analyses from several of these cores indicate very low (usually < 0.5%) or absent sand fraction, whereas the proportion of clay ranges from about 35 to 60%. The uniform mud layer within any one core, however, shows little vertical change in the clay content (variations of 5 ?A). Carbonate content shows almost no vertical variation (between 34 and 36 yo). In comparison with other mud types, the sand fraction of uniform mud constitutes a moderate to large proportion of terrigenous components (as in the case of turbiditic mud), and a somewhat lower content ( < 40 yo) of planktonic foraminifera.

Faintly laminated mud (F ig. 3D)

This mud type can be identified only in X-radiographs, where it is generally recognized by a vague to poorly defined basal contact and some vertical gradation reflecting textural and subtle compositional variations. Faint lamination at the base of such units is sometimes observed, and thicknesses are generally of the order of several centimetres. Colour and petrography are most often comparable to uniform muds described above, with which this type is generally associated. Faintly laminated mud may also be gradational with the turbiditic and well laminated mud types, and genetically is probablyrelated to these.

IS

TEMP 0 R A L-SPAT 1 A L DE P 0 SIT I 0 N A L TRENDS

The first step in the semi-quantitative analysis of the spatial and temporal distribution of the fine-grained sediments entailed measurement in each core of the total length of sections formed by each of the seven mud types. From this we calculated for each core the relative proportion of mud types in each of the three lithostratigraphic layers. Coarse-grained layers (mostly sand) are discarded in the calculations.

All data from the 24 cores were combined to determine the overall relative proportion of mud types for all core sections without regard to stratigraphic or depositional considerations (total core section in Fig. 4). For the entire region examined, the mud types occur in the following order of diminishing importance: turbiditic and hemipelagic mud clearly predominate; well laminated mud is common; uniform mud and slumps are present;

E N T I R F C O R E P O P U L A T I O N

TOTAL CORE

> 23

SECTION

CM/lOOO YRS

>

LAYER 2

' I0 CM/I 000 YRS

LAYER 3

Fig. 4. Lithologic logs showing relative proportions of major mud types in the entire core population, including the total core section and those comprising stratigraphic layers I , 2 and 3. Log lengths are proportional to calculated sedimentation rates in Figs 4-10.

$1. I > ? X


faintly laminated mud accounts for low proportions; muddy debris flow deposits are rare. The total core section serves as a generalized standard that obscures the effects of regional differences, o r specific depositional environment, or time.

Spatial distribution of mud types

All core data were treated to show the relative proportions of mud types for the three regions studied with regard to time, and indicate differences from north-west to south-east along the Hellenic Arc (Fig. 5). Regionally, it is apparent that cores from the Kythera-Antikythera complex in the south-east sector are characterized by the highest relative proportion of turbidites, while those in the Zakinthos- Strofadhes complex to the north-west comprise an important turbiditic-hemipelagic-laminated mud suite. Overall proportions of mud types in cores of the central 'Matapan Deep' complex are inter- mediate. Slumps, uniform mud and faintly laminated mud are present in low amounts in the three complexes and show no marked regional trend.

Treatment of the total core section data (Fig. 5) shows that slope cores are dominated by large proportions of hemipelagic and turbiditic mud types, but low proportions of well laminated and slump units. In contrast, the trench basin core data show that turbiditic and uniform muds are dominant, and that well laminated and hemipelagic muds are im-

R I G I O N I N V I R O N M I N I

"MA TAPAN DEEP"

I V S I E M

22 CM/iOm YRS

U I H E M I P E L n C I C MUD

WELL-LAMINATED M U D

TRENCH- ' RELAIED

J V I R O N M I N I

=UNIFORM MUD IURBIDII IC MUD

F A I N I L Y L a M I N A I E D M U D > ~ ~ ~ $ c ~ ~ ~ ! ~ ~

Fig. 5. Total core log section for the entire core population, three trench systems examined in the western Hellenic Arc, and slope- and trench-related environments.

portant, but subsidiary types. Slumps occur in both the slope and trench settings, whereas faintly laminated mud is primarily present in trench cores.

Observed regional differences probably reflect an uneven core distribution, and this can be tested by comparing all core sections from slope settings with all cores from trench basin settings, without regard to specific region or time (Fig. 5) . The proportion of mud types in the KB complex is most similar to that of the trench system, and this in part reflects the larger number of cores recovered from trench basins and perched basins in this region. In contrast, the proportion of mud types in the MB complex is most similar to that of slope deposits, also indicating the larger proportion of cores recovered from slope environments in this area.

The total core section analysis indicates that the differences between mud-type suites in slope and trench basin systems are of a larger magnitude than differences between the three regions sampled. A more precise determination of the correlation between depositional site and mud type prompted an analysis of the proportion of each fine-grained deposit present in each of the eight environments (Figs 6, 7 and 8). Correlations are discussed in order of decreasing relative overall importance of mud types in the entire core population.

Turbiditic mud is present to abundant throughout the study area, except on steep, straight, inner trench slopes ; this type dominates core sequences recovered in trench plains and their margins (particularly aprons) and in proximal perched basins. Hemipelagic mud is found in all environments, but is relatively more important on the straight, steep, inner trench wall and other slope environments, and occurs in lower proportions in, and on the margins of, trench basin-plains. Well laminated mud is recovered in most environments, but does not account for a large proportion of the total core section in any of these. Un$orm mud is an important constituent only in trench basins and immediately adjacent margins; it also occurs in distal perched basin sections, but in minor proportions. Slumps prevail in slope areas of broken relief, and also as an important constituent in perched slope basins and on trench basin-aprons. Faintly laminated mud is recovered primarily in trench basins, but also occurs in low amounts on distal perched basins and steep slope sequences. Muddy debrisflow deposits recovered in slope areas of broken relief and perched basin account for the least commonly recovered sections (important in only two of the cores).

Depositional models for fine-grained sediment 28 1

PROXIMAL DISTAL BROKEN SUBMARINE PERCHED PERCHED TRENCH

RATE OF SEDIMENTATION STEEP

r' w 2- -10

TURBlDlTlC M U D

SLUMP & DEBRIS FLOW DEPOSITS

TRENCH PLAIN

FLAT TRLNCH M A R G I N

. .'.,'.'.'.

I Fig. 6. Logs showing distribution of mud types in the eight depositional environments, during stratigraphic layer 1 time.

SUBMARINE

I 3

TURBlDlTlC MUD

PROXIMAL PERCHED

BASIN

DISTAL PERCHED

BASIN

. ... ....

TRENCH is TRENCH PLAIN B .. .. ...

FLAT TRENCH MARGIN

I

Fig. 7. Logs showing distribution of mud types in the eight depositional environments during stratigraphic layer 2 time. IS 2

282 W . J. Stanley and A . Maldonado

PROXIMAL DISTAL FLAT BROKEN SUBMARINE PERCHED PERCHED TRENCH TRENCH TRE N C H

VALLEY BASIN BASIN APRON PLAIN M A R G I N

RECOVERED)

LAYER 3 S L W P 6 DEBRIS FLOW DEPOSITS

Fig. 8. Logs showing distribution of mud types in the eight depositional environments during stratigraphic layer 3 time.

Temporal distribution of mud types

A general comparison is made of mud types for the entire core population and those deposited during each of the three periods considered (layers 1, 2 and 3), disregarding depositional environment (Fig. 4). This synthesis shows that: (a) two of the mud types dominate the fine-grained suite during each of the three periods, and (b) these mud-type assemblages change with time. The generalized changes with time shown in Fig. 4 are most closely comparable to those observed in the ‘Matapan Deep’ region (Fig. 9). In the Zakinthos-Strofadhes region there is a higher proportion of uniform mud in layer 3 ; in the Kythera- Antikythera system there is a somewhat lower proportion of well laminated mud in layer 2 and a much lower relative amount of uniform mud in layer 3. These regional variations probably reflect the uneven core distribution in these complexes.

Significant differences of mud-type assemblages with time become apparent when trench and slope core sections are compared without regard to geographic setting (Fig. 10). The time-related effects are

even more apparent if comparisons are made between mud types and specific depositional environments in layers 1 ,2 and 3 (see Figs 6, 7 and 8) : well laminated mud is widespread, independent of specific environment, but essentially restricted to sections deposited during layer 2 time; uniform mud is restricted to trench sections of layers 1 and 3, and is absent in layer 2 sections. Hemipelagic mud, turbiditic mud and slumps are independent of time-related factors.

SED IMEN TAT 1 0 N RATES : VARIATIONS IN TIME AND SPACE

The relative amounts of mud types are plotted on core-section logs whose lengths are proportional to sedimentation rates calculated using the dated lithostratigraphic boundaries of layers 1 , 2 and 3 (Fig. 2). Some cores did not recover complete lithostratigraphic sections and in these cases an estimated mini- mum accumulation rate was calculated on the basis of the core length recovered. The computed sedimentation rate for theentirecore population exceeds 23 cm/

Depositional models for fine-grained sediment 28 3

Fig. 9. Logs showing distribution of mud types in three western Hellenic Arc systems, from north-west to south- east. The changes in each region with time (for stratigraphic layers 1, 2 and 3) are also shown.

1000 yr (Fig. 4). Overall core section accumulation rates are highest in layer 1 ( > 34 cm/1000 yr) and lowest in layer 2 ( < 10 cm/1000 yr). The somewhat lower value in layer 3 ( > 25 cm/1000 yr) is an arti- fact introduced by the failure of some cores to recover completely the older sections.

A regional increase from north-west to south-east is observed, i.e. from about 16cm/1000 yr in the Zakinthos-Strofadhes region, to > 22 cm/1000 yr in the 'Matapan Deep' region, to > 30 cm/1000 yr in the Kythera-Antikythera region (Fig. 5) . Differences in sedimentation rates are more reliably depicted when specific depositional environment, rather than region, is considered: averaged rates of > 13 cm/ 1000 yr for all slope core sections, and > 34cm/ 1000 yr for all trench basin core sections (Fig. 5). The decrease in sedimentation rate during layer 2 time is most pronounced in trench basin core sections (Fig. 10).

The averaged accumulation rate, in centimetres per 1000 years for each specific environment (Fig. 2), reveals a general downslope increase as follows: steep inner trench slope (7 cm), submarine valley (12 cm), broken high-relief slope (14 cm), proximal perched basin ( > 22 cm), distal perched basin (> 18 cm), trench basin-apron (30 cm), flat trench margin ( > 30 cm), and trench basin proper ( > 35 cm). This pattern of basinward-increased sedimentation rates is well shown by core sections in layers 1 and 3 (Figs 6 and 8). In contrast, sections deposited

Fig. 10. Logs showing contrasting distribution of mud types in slope- and trench-related environments: the changes in each environment with time (for stratigraphic layers 1 ,2 and 3) are also shown.

during layer 2 time record an irregular accumulation pattern between proximal slope and distal basin environments (Fig. 7). Rates calculated for sections that include the uniform mud type are the highest recorded in the Mediterranean (locally > 200 cm/1000 yr).

DEPOSITIONAL MODELS FOR FINE- GRAINED S E D I MEN T

The data presented indicate that some fine-grained deposits accumulated only in specific depositional environments (uniform mud), or are only found at certain times (well laminated mud). The distribution of other mud types, in particular hemipelagic and turbiditic muds, and to a lesser extent slumps, appears independent of location and age. Some types typically occur together and form distinct assemblages: turbiditic and well laminated muds; faintly laminated and uniform muds; and slump and debris flow deposits. Moreover, mud-type assemblages and rates of sedimentation are most similar in stratigraphic layers 1 and 3 ; layer 2 core suites are clearly distinct from these two.

A model is developed for layers 1 and 3 mud dispersal patterns, and a modified scheme is discussed separately for layer 2 sections. The two schemes integrate the nature and proportions of the mud-type suites, their distribution patterns and their accumulation rates, and both call into play deposition from

I S - 3

284

OCCURRENCE PRIMARILY AREAS OF RELIEF D‘STR’BUTloN IN STUDY AREA

SUSPENSION RELATED PROCESS

GRAVITATIVE PROCESS

DEPOSIT

DISTAL BASIN PLAIN

D. J. Stanley and A. Maldonado

r-1 t t I . ” / I

I

both gravitative and suspension processes. Factors considered in each case are topographic attributes of the depositional site, dispersal paths and distance from source, effects of water mass stratification, and changes in flow mechanisms (transformation) that occur during the course of transport events.

Depositional model 1 : minimally stratified conditions

A summary of data for layers 1 and 3 core sections, schematically depicted in Fig. 11, indicated that fine- grained sediment at these specific times was emplaced by a sequence of transport mechanisms. The model emphasizes : (a) a downslope-directed, gravity- induced, progressively less dense continuum of processes resulting from alteration of grain- support mechanisms during flow, and (b) transport through the water column displaying minimally stratified conditions such as that displayed by the present Mediterranean.

Slump deposits, oneend-member of thecontinuum, occur widely but are most common in cores recovered on or near slopes in areas of relief, including ‘cobblestone’ topography, and on the margins of perched and trench basins. Evidence of widespread slumping is recorded by the appearance of slump scars and displaced sediment series on seismic profiles (Got eta]., 1977; Vittori, 1978) and 3.5 kHz records (Stanley, 1977).

Middleton & Hampton (1973) suggested on theoretical grounds that en rnasse displacement of mud- rich slumps gives rise to less dense sediment gravity flows such as debris flows (relatively thick mixtures of granular solids, mud and water). Such transformation (Stanley, Palmer & Dill, 1978b) involves

akeration of grain-to-grain fabric as water is in- corporated during movement. Sequences recovered on the ‘cobblestone’ type terrain (TR core 32 for example) typically display a slump -muddy debris flow assemblage. Experimental work (Dangeard, Larsonneur & Migniot, 1965 ; Morgenstern, 1967) indicates that slumps also give rise to high-density turbulent flows. Indirect evidence that some fine- grained turbidites were emplaced by flows which evolved from either slumping or debris flows origin- ating on the slope, is provided by thecommonoccur- rence of slump-turbidite associations and ubiquitous distribution of turbidites in the study area.

Dispersal path and distality are factors that should be considered in interpreting the distribution of turbiditic mud. The low proportion of turbidites on inner trench walls does not provide a realistic record of processes that occur on these steep slopes but rather suggests that such proximal environments are bypassed or serve only as temporarydepositionalsites. This is consistent with observations made in many sectors of the Mediterranean that show that gravita- tively emplaced materials do not remain long on steep slopes; such metastable deposits tend to fail, move downslope and accumulate preferentially on lower gradient surfaces in more distal basins (Stanley, Rehault & Stuckenrath, 1980). Episodic displacement of unstable sediment is probably triggered by earthquake tremors. Major lowering of sea-level during the Wiirm also influenced downslope redeposi tion, and the enhanced proportion of turbidites in Layer 3 (about 23,000-1 7,000 yr BP) is almost certainly related to this factor. The depositional scheme in Fig. 11 emphasizes that core sections of uniform mud, unlike turbiditic mud, are localized primarily in deep trench basin-


plains and their flat margins (not aprons); minor proportions of this apparently structureless fine- grained material also are present insomedistal perched basins. This distribution pattern thus implicates the importance of topographic restriction and distality of the depositional site. Moreover, the uniform mud- type is time-related : accumulation was negligible during deposition of layer 2 (17,000-5700 yr BP) sections. High-resolution (3.5 kHz) sub-bottom profiles in the three areas show that the thickness of uniform mud in some cases exceeds 10m, and that these deposits display typical attributes of ponded sediment, i.e. stratification is laterally continuous, and of almost uniform thickness throughout the trench basin-plains. Four such units recovered from the major trench basin-plains were radiocarbon dated (TR core 35; MA cores 1, 20, 23). Two to three radiocarbon dates from each of these units show the same age for the entire uniform mud layer. This and the lack of structures imply rapid emplacement, probably from single transport events (Stanley & Knight, 1979). The radiocarbon dates probably represent the age of the material originally deposited in proximal slope settings rather than the time of failure and emplacement in trench basin-plains.

Uniform layers are interpreted as the products of sediment gravity flows which carried large volumes of mud primarily in suspension from the slopes to deep distal basin-plains. The uniform muds are somewhat finer-grained (lower proportions of sand and silt fractions) than graded turbiditic muds with which they are associated, and both types are interbedded with faintly laminated mud. Their concentration, even distribution and continuity in distal trench basins, high rates of accumulation, somewhat finer texture than turbiditic mud, and composition (typically a mixture of bioclastic and terrigenous components) indicate that uniform muds were released from low density or lower velocity turbidity current flows. Some uniform muds probably represent end-members of slump +turbidity current, o r debris flow -+turbidity current transformation events. Not all dilute end-products, however, were necessarily released from a turbidity current (usually implying grain support by turbulent, moderate to high concentration flows). We believe some uniform muds were deposited from turbid layers of low concentration (cf. Moore, 1969) capable of transport- ing over long periods suspended material from fluvial sources (Drake, Kolpak & Fischer, 1972), coastal or shelf-edge erosion, or erosion of the slope (Pierce, 1976).

This interpretation invoking both turbidity currents and turbid layer flows accounts for the association of uniform and vaguely laminated turbiditic muds. Much higher rates of sedimentation calculated for trench basins tend to support this contention. More- over, the strong correlation for core sections deposited during lower stands of sea-level (layer 3) records the effects of increased rates of off-shelf sedimentation. The absence of uniform mud in layer 2 core sequences is attributed to density interfaces associated with water mass stratification that developed during this period: these density interfaces hampered the movement of low-concentration flows along the bottom.

The fine-grained sediments in the Hellenic Arc are not all of gravitative origin. Hemipelagic mud, rich in material settling from suspension (‘rain’) through the minimally stratified water column(Fig. 1 l), occurs in all depositional environments, but is relatively more important in the slope than in the trench basin settings. This uneven distribution is mostly the result of dilution, i.e. a masking effect by almost continuous reworking by gravity flows and deposition further downslope of large volumes of sediment in deep distal settings. This displacement of hemipelagic muds from topographic highs and seaward transport to deeper, more distal settings by gravitative flows would account for the high planktonic content in some turbiditic, faintly laminated and uniform muds (Fig. 1 I), and for the high sedimentation rates in perched basins and trench basins.

The higher proportion of hemipelagic mud in layer 1 than in layers 2 and 3 sections does not necessarily imply a substantial increase in the total amount of hemipelagic material deposited during the past 5700 years, but rather, less masking by decreased downslope deposition from gravitative processes during this short time span. The relative increase of hemipelagic relative to gravitative deposits in layer 1 sections probably records the diminished and less direct input from shelf and fluvial sources during this period of slow rise and stabilization of sea-level (Morner, 1971). This phenomenon is well shown by cores recovered on steep slopes, topographic highs and areas of broken, or ‘cobblestone’, relief, which are least likely to retain mud types emplaced by gravitative processes (Got e t a / . , 1977; Stanley, 1977).

Depositional model 2: stratified water mass conditions

A modification of model 1 is needed to explain: (a) the much-increased proportion of well laminated

286 D. J. Stanley and A. Maldonado

IS US PENSION^ I

I H P l in ri I H P I i t . PI i n ~ i

PROCESS

ROLE OF STRATIFICATION DETACHED TURBID LAYER FLOW] 111.m.mD INTERFACE 1111111111111111 A s BARfiER 1

I 7-yr- 5- 7 I DEPOSIT FROM GRAVITATIVE a SUSPENSION-RELATED PROCESS I+ ------WELL LAMINATED MUD-,)

I DISTAL BASIN OCCURRENCE PRIMARILY AREA OF RELIEF WIDESPREAD DISTRIBUTION IN STUDY AREA PLAIN

Fig. 12. Fine-grained depositional model applicable to conditions of well-developed water-mass stratification, emphasiz- ing process, resulting mud type, and its area of occurrence.

mud, (b) rather low overall sedimentation rates, and ( c ) general absence of uniform mud in layer 2 sequences (Fig. 12). This latest Pleistocene to early- mid Holocene time is characterized by rapid rise of sea-level (Morner, 1971), and marked palaeoclimatic changes in the eastern Mediterranean region resulting in water circulation in this almost completely enclosed ocean (Ryan, 1972; Stanley, Maldonado & Stuckenrath, 1975).

Most noteworthy are the well laminated muds which during layer 2 time account for a larger proportion of core sections than all other types (Fig. 4) and are recovered in most depositional environments (Fig. 7). Well laminated mud is only slightly more abundant in trench basins than in slope settings (Figs 5 and 7). The excellent preservation of thin laminae, dark colour of some layers, total absence of bioturbation and bottom current structures, and high bioclastic fraction content of the well laminated facies suggest a close relation to water-mass properties. Depositional features of this type would most likely prevail at times of marked water stratification and euxinic conditions. It is recalled that the youngest sapropel layer (s, = about 9000-7000 yr BP) was deposited during part of the time represented by Layer 2 (Stanley & Maldonado, 1979). This sapropel development is closely associated with basin-wide stratification of eastern Mediterranean waters (Stanley, 1978) and oxygen-depleted bottom conditions as recognized by Ryan (1972), van Straaten (l972), Stanley et a[. (1975), Vergnaud-Grazzini, Ryan & Cita (1977), and others.

We believe that the well laminated mud type represents, at least in part, the distal end-product of

bottom-following gravitative flows such as turbidity currents (Piper, 1978; Stow & Bowen, 1978) and turbid layer flows (Moore, 1969). However, the downslope movement of some low concentration flows was probably modified by pycnocline-density interfaces of the stratified water column. It appears that the thin, often graded layers of this mud type settled out from periodically introduced flows that spread over broad regions as continuous or detached turbid layers (Pierce, 1976), rather than as channel- ized flows moving downslope along the bottom. The resulting dispersion of material over wide areas would account for the lower sedimentation rates calculated for layer 2 core sections (Figs 4, 7, 9 and 10). Some higher-concentration or more rapidly moving turbidity currents were able to penetrate stratification interfaces and flow downslope along the bottom to more distal sectors, thus accounting for classic silt turbidites and turbiditic mud in perched and trench basin-plains. This is recorded by the somewhat higher sedimentation rates in these environments than on steep slope settings during layer 2 time.

The low proportion of hemipelagic mud layers in trench basins (Fig. 7) is explained in two ways: (a) hemipelagic material settling through the water column accumulated with sediment from low-density gravity flows (turbid layers and others) concentrated along density interfaces (thus, hemipelagic components form an integral part of well laminated mud facies), and (b) the masking of largely suspension- derived deposits by denser, bottom-following sediment gravitative flows that reached distal trench basins.


In previous studies most dark, well laminated sections of Hellenic Arc cores have been called sapropels on the basis of split core examination (Nesteroff, 1973; Hieke et a/., 1973; Stanley et al., 1978a). How- ever, not all dark, well laminated muds are sapropels, and more than split core examination is required to distinguish these two facies. Both types are dark due to deposition in reduced to completely anoxic settings. We restrict the use of the term sapropel, however, to dark grey to black layers consisting largely of pelagic material, a high (generally > 10 %) total organic matter content, and commonly a basin- wide distribution (Ryan & Cita, 1977). In the study area, sapropels sensu strict0 are most obvious in slope cores. We have shown earlier that laminated muds in the Hellenic Arc are derived in part from gravitative flows (they comprise a terrigenous as well as a biogenic fraction of silt and sand size), and not only from pelagic suspensions. As a result of dilution by gravity flow input, the organic matter content in well- laminated mud is much lower (3 to 5 % ) than in true sapropels, and colour is grey rather than black,

Some well laminated mud core sections may be confused with turbiditic mud since both types show grading and stratification of silt-size material. The former usually shows thicker core sections formed by the fine, well-defined horizontal laminae and a total absence of cross-lamination (Fig. 3B, C) in contrast to the more commonly recorded laminated turbiditic muds (Fig. 3 F, and also illustrated in Rupke & Stanley, 1974; Maldonado & Stanley, 1976b; Piper, 1978; Stow & Bowen, 1978). Moreover, the texture and composition of the well laminated facies favours accumulation by slow particle-by-particle settling through the water mass rather than by more direct emplacement by bottom-following gravity flows. As shown in Fig. 12, the interplay between suspension and gravitational mechanisms requires that detailed petrography as well as X-radiography be used to distinguish these two facies.

Comparing the models

In the two depositional schemes presented here, the fine-grained facies are viewed as end-members deposited from dense mass flows, many of which were triggered as slumps, by the intense seismic activity affecting this mobile fore-arc region. The tectonic imprint results in almost continuous reworking basinward of metastable sediment and sediment gravity flows. The specific regional distribu-

tion of gravitative deposits largely reflects downslope dispersal affected by the structurally controlled complex topography. The long-term consequence of remobilization of sediments results in displacement from slope to perched basins and subsequently to trench aprons and basin-plains.

The major difference between the two models is in the fine-grained end-members deposited from lower- concentration or slower gravity-influenced mechanisms including turbidity currents and turbid layer flows: uniform mud prevails in distal basins under conditions of minimally stratified water masses, and well laminated mud dominates the fine-grained suite over large areas at times of well-developed stratification. I t is remarkable that the attributes of the water masses can so directly affect overall sedimentation patterns even in a region as tectonically active as the Hellenic Arc, where gravitative and redeposition processes prevail.

CONCLUSIONS

Seven principal fine-grained mud types are identified in dated Late Quaternary cores recovered in eight depositional environments of the western Hellenic Arc. Five mud types (slump, muddy debris flow deposits, mud turbidite, faintly laminated mud and uniform mud) are of gravitative origin, and collec- tively account for the largest proportion of fine- grained material in the study area. Hemipelagic mud of suspension origin is only slightly less important volumetrically than the turbiditic mud which clearly predominates in this region. Well laminated muds involve deposition from both gravitative and suspension mechanisms. Differences in the proportion of mud types between trench slope- and trench basin-related environments are more pronounced than the regional variations observed between the Zakinthos-Strofadhes, ‘Matapan Deep’, and Kythera-Antikythera trench complexes. Certain fine-grained facies are closely related to specific environments (slumps, debris flow units on slope and high-relief environments), or to time (well laminated mud in stratigraphic layer 2) , or to both (uniform mud). Turbiditic and hemipelagic muds, commonly found throughout the region, are generally more independent of either spatial or temporal factors.

The formulation of two depositional models for fine-grained sediments is based on a quantitative analysis of the distribution patterns of the different types in time and space, complemented by analysis of

288 D . J . Stanley and A . Maldonado

sediment accumulation rates in the three trench slope and basin complexes. Both emphasize downslope changes of grain-support mechanisms, or transformation, during gravitative flow events which would explain proximal to distal differences in the mud- type assemblages recorded in the three slope-trench systems. The first model involves progressively reduced concentration during the basinward transport, i.e. dense slump + debris flow + turbidity current + low-density or low-velocity turbidity current and/or turbid layer mechanisms. The distal end-member type is thick, fine-grained uniform mud deposited primarily in the flat trench basin-plains. This rapidly emplaced end-member forms sections deposited at times of minimal water-mass stratification such as presently exists in the Mediterranean. Increased proportions of uniform-turbiditic mud assemblages were emplaced during eustatic low sea- level stands, and progressive downslope redeposition would account for the much higher sediment accumulation rates recorded in the more distal environments.

The second model for fine-grained sedimentation, applicable at times of well-developed water-mass stratification, is at variance with the first primarily with respect to the distribution of clayey silt end- products emplaced by low-concentration flows. At times of stratification, very finely laminated mud, rather than uniform mud, prevails as a result of particle-by-particle settling from turbid and detached turbid layers concentrated along density interfaces. These sediments, retained along pycnoclines, were probably complemented by the normal ‘rain’ of pelagic and hemipelagic material. The density interfaces not only retard settling but also ensure that sediments are spread and eventually deposited over broad areas of the slope and trench basins.

High sediment accumulation rates calculated for all trench basin cores (> 35 cm/1000 yr) attest to an intensive regime involving almost continuous erosion, slope bypassing, redeposition and eventual downslope transport to distal depressions (Stanley et al., 1980). In contrast, well laminated mud (model 2 end-member) records substantially lower accumulation rates as a result of more widespread transport over the study area at times of water mass stratification. The depositional types most closely affected by stratification are uniform, well laminated and faintly laminated muds. These three fine-grained types, in some instances, have been confused with better- defined facies such as mud turbidites, hemipelagic muds and sapropels with which they are sometimes

associated and may be genetically related. The more obvious compositional and textural differences between these mud types are obscured as a result of homogenization from the continued redeposition to more distal sites and incorporation of pelagic components during transport (Vittori, 1978; Feld- hausen & Stanley, 1980). Only with a detailed petrographic and X-radiographic methodology is it possible to identify the sequence of transport episodes responsible for displacing sediment between shelf-inner trench slope settings and deep distal sectors.

This investigation shows that fine-grained, lithostratigraphic sections in the Hellenic Arc are generally different from those examined elsewhere in the eastern Mediterranean Sea. The difference, however, is primarily a function of the proportion of mud types forming the Late Quaternary sequences rather than the presence of any specific facies or transport process unique to this fore-arc region. The almost continuous remobilization and basinward transport of sediment in a variety of diverse, often steep and tectonically mobile environments within a small region close to the source area, and deposition in topographically restricted environments, account for the observed marked core-to-core differences. The variable distribution patterns of sediment types are also attributable to the climatic overprint that influenced eustatic oscillations and water mass stratification.

All the major Hellenic mud types were also deposited in other, more open, settings of the eastern Mediterranean, but in different proportions and thicknesses. The fine-grained depositional model we present here can be tested in other restricted and structurally mobile Mediterranean settings and may be applicable to some deep-water analogues preserved in the rock record.

ACKNOWLEDGMENTS

We thank Drs A. Brambati, University of Trieste, N. D. Watkins (deceased), University of Rhode Island, and H. Got, University of Perpignan, for organizing cruises on the R/V Bannock (1974) and Marsili (1976), R/V Trident (1975), and R/V Dectra (1976), respectively. We are indebted to them for helping ensure the success of these various expedi- tions, and for generously providing us with the cores and 3.5 kHz records analysed as part of this study. Special appreciation is expressed to Dr R. J. Knight, Petro-Canada Ltd, for his assistance in collecting the


Marsili cores in 1976 and conducting petrographic analyses during the first stage of this investigation. Valuable assistance in the petrologic study also was provided by Mr H. Sheng, Smithsonian Institution. Drs M. A. Hampton, R. J. Knight, J. K. Leggett, D. J. W. Piper and J. R. Southard reviewed the manuscript. This phase of the Hellenic Trench Study, part of the Mediterranean Basin (MEDIBA) Project, was funded by Smithsonian Research Award 89191028.

REFERENCES

BARTOLINI, C., GEHIN, C . & STANLEY, D.J. (1972) Morphology and recent sediments of the Western Alboran Basin in the Mediterranean Sea. Mar. Geol.

BARTOLINI, C., MALESANI, P.G., MANETTI, P. & WEZEL, F.C. (1975) Sedimentology and petrology of Quater- nary sediments from the Hellenic Trench, Mediter- ranean Ridge and Nile Cone, DSDP Leg 13 cores. Sedimentology, 22, 205-236.

BIJU-DUVAL, B. & MONTADERT, L. (Eds.) (1977) Struc- tural Hislory of the Mediterranean Basins. Editions Technip, Paris, 448 pp.

BRUNN, J.H., AUBOUIN, J., MERCIER, J. et al. (1976) La geologie des regions Egeennes. Bull. Soc. gPo1. Fr. 2, 567 pp.

CARTER, G.T., FLANAGAN, J.P. et al. (1972) A new bathy- metric chart and physiography of the Mediterranean Sea. In: The Mediterranean Sea: a Natural Sedimentation Laboratory (Ed. by D.J. Stanley), pp. 1-23. Dowden, Hutchinson & Ross, Stroudsburg, Pennsylvania.

DANGEARD, L., LARSONNEUR, C., & MIGNIOT, C. (1965) Les courants de turbidite, les coulees boueuses et Ies glissements: resultats d'experiences. C.r. hebd. SPanc. Acad. Sci., Paris, 261, 2123-2126.

DRAKE, D.E., KOLPAK, R.L. & FISCHER, P.J. (1972) Sediment transport on the Santa Barbara-Oxnard Shelf, Santa Barbara Channel, California. In: Shelf Sediment Transport (Ed. by D.J.P. Swift, D.B. Duane and O.H. Pilkey), pp. 307-331. Dowden, Hutchinson & Ross, Stroudsburg, Pennsylvania.

FELDHAUSEN, P.H. & STANLEY, D.J. (1980) Hellenic Trench sedimentation: an approach using terrigenous distributions. Mar. Geol. 38, m21-m30.

GOT, H., STANLEY, D.J. & SOREL, D. (1977) North- western Hellenic Arc: concurrent sedimentation and deformation in a compressive setting. Mar. Geol. 24,

HERSEY, J.B. (1965) Sediment ponding in the deep sea. Bull. geol. Soe. Am. 76, 1251-1260.

HIEKE, W., SIGL, W. & FABRICIUS, F. (1973) Morpho- logical and structural aspects of the Mediterranean Ridge SW of the Peloponnesus (Ionian Sea). Bull. geol. Soe. Greece, 10, 109-126.

HUANG, T.-C. &STANLEY, D.J. (1972) Western Alboran Sea: sediment dispersal, ponding and reversal of currents. Tn: The Mediterranean Sea: a Nalrrral

13,159-224.

21-36.

Sedimentation Laboratory (Ed. by D.J. Stanley), pp. 521-550. Dowden, Hutchinson & Ross, Strouds- burg, Pennsylvania.

KALLERGIS, G. (Ed.) (1979) Proceedings of the VI Colloquium on the Geology of the Aegean Region. Inst. Geol. Mining Res. Athens, Greece, 1405 pp.

LE QUELLEC, P. (1 979) La Marge Continentale Ionienne du PPloponnPse. Ge'ologie et Structure. Unpublished Thesis, University of Paris VI, 210 pp.

LE QUELLEC, P., MASCLE, J., VITTORI, J., GOT, H. & MIRABILE, L. (1978) La fosse de Matapan (Mer Ionienne): nouvelles donnees sur sa structure. C.r. hebd. Sianc. Acad. Sci., Paris, 287 D, 431-434.

MCKENZIE, D.P. (1978) Active tectonics of the Alpine- Himalayan belt : the Aegean Sea and surrounding regions. Geophys. J . R . astr. SOC. 55, 217-254.

MAKRIS, J. (1978) Some geophysical considerations on the geodynamic situation in Greece. Tectonophys. 46, 251-268.

MALDONADO, A. & STANLEY, D.J. (1976a) Nile Cone lithofacies and definition of sediment sequences. Proc. 9th Int. Congr. Sediment., Nice, 6, 185-191.

MALDONADO, A. & STANLEY, D.J. (1976b) The Nile cone: submarine fan devlopment by cyclic sedimentation. Mnr. Geol. 20, 27-40.

MALDONADO, A. & STANLEY, D.J. (1977) Lithofacies as a function of depth in the Strait of Sicily. Geology, 5 , 11 1-1 17.

MALDONADO, A. & STANLEY, D.J. (1979) Depositional patterns and Late Quaternary evolution of two Medi- terranean submarine fans: a comparison. Mar. Geol.

MEDIBA Group (1979) La marge occidentale du PClopon- nkse: DonnCes nouvelles sur la structure et le recouvre- ment sedimentaire. In: VI Colloquium on the Geology of the Aegean Regions, Athens 1977 (Ed. by G. Kallergis), pp. 1043-1051. Inst. Geol. Mining Res. Athens.

MIDDLETON, G.V. & HAMPTON, M.A. (1973) Sediment gravity flows: mechanics of flow and deposition. In: Turbidites and Deep- Wafer Sedimentation (Ed. by G.V. Middleton and A.H. Bouma), pp. 1-38. Pacific SOC. econ. Paleont. Miner. Section Short Course, Anaheim, California.

MOORE, D.G. (1969) Reflection profiling studies of the California Continental Borderland : structure and Quaternary turbidite basins. Spec. Pop. geol. Soc. Am. 107,142 pp.

MORGENSTERN, N. (1967) Submarine slumping and the initiation ofturbidity currents. In: Marine Geotechniqlte, (Ed. by A.F. Richards), pp. 189-220. Urbana University, Illinois Press.

MORNER, N.A. (1971) Eustatic and climatic changes during the last 20,000 years, and a method of separating isostatic and eustatic factors in an uplifted area. Palaeogeogr. Palaeoclirn. Palaeoecol. 9, 155-1 81.

NESTEROFF, W.D. (1973) Petrography and mineralogy of sapropels. In: Initial Reports of the Deep Sea Drilling Project, Vol. 13 (Ed. by W.B.F. Ryan, K.J. Hsii et al.), pp. 713-720. US. Government Printing Office, Washington.

NEsTERofF, W.D. (1977) The recent sedimentation in eastern Mediterranean along the Hellenic Arc (Medor- 75 cruise). VI CON. Geol. Aegean Region, Sept. 1977, Ahstrocts, p. 65. Athens, Greece.

21, 215-250.


PAPAZACHOS, B.C. & COMNINAKIS, P.E. (1977) Modes of lithospheric interaction in the Aegean area. In : Struc- furaf History of the Mediterranean Basins (Ed. by B. Biju-Duval and L. Montadert), pp. 319-322. Editions Technip, Paris.

PAREYN, C. (1968) Observations geologiques et sedi- mentologiques dans la fosse Ouest de la mer Ionienne. Annls Inst. ocianogr., Paris, 46, 53-69.

PIERCE, J.W. (1 976) Suspended sediment transport at the shelf break and over the outer margin. In: Marine Sediment Transport and Environmental Managemen t (Ed. by D.J. Stanley and J.P.W. Swift), pp. 437-458. Wiley (Interscience), New York.

PIPER, D.J.W. (1978) Turbid.ite muds and silts on deep- sea fans and abyssal plains. In: Sedimentation in Sub- marine Canyons, Fans, and Trenches (Ed. by D.J. Stanley and G. Kelling), pp. 163-176. Dowden- Hutchinson & Ross, Stroudsburg, Pennsylvania.

RUPKE, N.A. & STANLEY, D.J. (1974) Distinctive properties of turbiditic and hemipelagic mud layers in the Algkro-Balearic Basin, western Mediterranean Sea. Smithson. Contr. Earth Sci., Washington, 13, 40 pp.

RYAN, W.B.F. (1972) Stratigraphy of Late Quaternary sediments in the eastern Mediterranean. In: The Mediterranean Sea : a Natural Sedimentation Labora- tory (Ed. by D.J. Stanley), pp. 149-169. Dowden, Hutchinson & Ross, Stroudsburg, Pennsylvania.

RYAN, W.B.F. & CITA, M.B. (1977) Ignorance con- cerning episodes of oceanwide stagnation. Mar. Geol. 23, 97-215.

RYAN, W.B.F. & HsU, K.J. et al. (1973) Initial Reports of the Deep Sea Drilling Project, Vol. 13, US . Govern- ment Printing Office, Washington, 1447 pp.

RYAN, W.B.F., STANLEY, D.J., HERSEY, J.B., FAHL- QUIST, D.A. & ALLAN, D.T. (1970). The tectonics and geology of the Mediterranian Sea. In: The Sea, Vol. 4 (Ed. by A. Maxwell), pp. 387-492. Wiley (Interscience), New York.

STANLEY, D.J. (1974) Modern flysch sedimentation in a Mediterranean island arc setting. In: Modern and Ancient Geosynclinal Sedimentation (Ed. by R. H. Dott, Jr and R. H. Shaver). Spec. Publs Soc. econ. Paleont. Miner., Tulsa, 19, 240-259.

STANLEY, D.J. (1977) Recent tectonic overprint on cobblestone deposition in the Northwestern Hellenic Arc. In : Structural History of the Mediterranean Basins (Ed. by B. Biju-Duval and L. Montadert), pp. 433-445. Editions Technip, Paris.

STANLEY, D.J. (1978). Ionian sea sapropel distribution and Late Quaternary paleoceanography in the eastern Mediterranean. Nature, 274, 149-152.

STANLEY, D.J. & KNIGHT, R.J. (1979) ‘Giant’ mudflow deposits in submarine trenches: Hellenic basins and slopes in eastern Mediterranean. Bull, Am. Ass. Petrol. Geol. 63, 532-533.

STANLEY, D.J., KNIGHT, R.J., STUCKENRATH, R. & CATANI, G. (1978a) High sedimentation rates and variable dispersal patterns in the western Hellenic Trench. Nature, 273, 110-113.

STANLEY, D.J. & MALDONADO, A. (1977) Mud-facies definition for paleoenvironmental interpretation. In: Program and Abstracts, AAPG-SEPM Meeting, June 1977, Washington, pp. 106-107.

STANLEY, D. J. & MALDONADO, A. (1979) Levantine Sea-Nile Cone lithostratigraphic evolution: quantitative analyses and correlation with paleoclimatic and eustatic oscillations in the Late Quaternary. Sedim. Geol. 23 37-65.

STANLEY, D.J., MALDONADO, A. & STUCKENRATH, R. (1975) Strait of Sicily depositional rates and patterns and possible reversal of currents in the Late Quaternary. Pala2ogeogr. Palaeoclim. Palaeocol. 78,279-291.

STANLEY, D.J., PALMER, H.D. & DILL, R.F. (1978b) Coarse sediment transport by mass flow and turbidity current processes and downslope transformations in Annot Sandstone Canyon-Fan valley systems. In: Sedimentation in Submarine Canyons, Fans and Trenches (Ed. by D.J. Stanley and G. Kelling), pp. 85-115. Hutchinson & Ross, Stroudsburg, Pennsylvania.

STANLEY, D.J.,REHAuLT, J.-P. &STUCKENRATH, R. (1980) Turbid layer-bypassing model : the Corsican Trough, Northwestern Mediterranean. Mar. Geol. 37, 19-40.

STOW, D.A.V. & BOWEN, A.J. (1978) Origin of lamination in deep sea, fine-grained sediments. Nature, 274,

STRAATEN, L.M.J.U., VAN (1972) Holocene stages of oxygen depletion in the deep waters of the Adriatic Sea. In : The Mediterranean Sea: a Natural Sedimentation Laboratory (Ed. byD.J. Stanley), pp. 63 1-643. Dowden, Hutchinson & Ross, Stroudsburg, Pennsylvania.

VERGNAUD-GRAZZINI, C., RYAN, W.B.F. & CITA, M.B. (1977) Stable isotopic fractionation, climate change and episodic stagnation in the eastern Mediterranean during the Late Quaternary. Mar. Micropaleont. 2, 353-370.

VITTORI, J. (1 978) CaractPres structuru-sPdimentaires de la couverture Plio-Quaternaire au niveau des pentes et des fosses HellPniques de PPloponnBse (CrPcc). Un- published Thesis, UniversitC Paul Sabatier, Toulouse, 194 pp.

324-328.

NOTE ADDED INPROOF

The name unifte is applied t o the nearly structureless, often thick, layers of clayey silt and silty clay that are formed by the uniform and faintly laminated mud facies described in this study. Such deposits generally appear but are not truly compositionally homogeneous and show a subtle fining-upward trend, and are deposited by gravity flow transport events. The term unifite was formally introduced by D. J. Stanley at the University of Urbino, Italy, in October 1980 a t the Conference on ‘Sedimentary Basins of Mediterranean Margins ’.

(Manuscript received 1 November 1979; revision received 27 February 1980)

Depositional Models for Fine-Grained Sediment in Western Hellenic Trench, Eastern Mediterranean: ABSTRACT

Documents