The depositional environment of the Laptev Sea continental margin: Preliminary results from the R/V POLARSTERN ARK IX-4 cruise D . NURNBERG, D. K. FUTTERER. F. NIESSEN. N. N0RGAARD-PEDERSEN, C. J. SCHUBERT, R . F. SPIELHAGEN and M. WAHSNER Niirnberg, D., Fiitterer, D. K . , Niessen, F.. N0rgaard-Pedersen. N., Schubert, C. J . , Spielhagen, R . F . , Wahsner, M. 1995: The depositional environment of the Laptev Sea continental margin: Preliminary results from the R/V POLARSTERN ARK IX-4 cruise. Polar Research 14(1), 43-53. Marine geological investigations were performed across the Laptev Sea continental shelf and slope, Thirty sampling sites were selected covering a depth range of ca 3500 m. Maximum core recovery was 9 m. PARASOUND sub-bottom profiling was used for site surveying and provided important information on the depositional environment of the continental margin together with sedimentological and stratigraphical investigations. Undisturbed horizontal layering of the sea-floor sediments is a common feature for the Laptev Sea shelf. There is no indication for glaciation of the broad shelf region during the Last Glacial, since moraine deposits are missing. However. a high number of plough marks in places points to recent to sub-recent ice-erosion which has led to an intensive sediment reworking on the shelf. Several broadly incised river channels recorded near the shelf edge are related t o Pleistocene drainage systems of large Siberian rivers which cut into the dry shelves during the Last Glacial Maximum and were subsequently filled during the Holocene. During the Last Glacial we therefore suspect a significant freshwater contribution from the Eurasian continent to the Arctic Ocean. The composition of the normally consolidated core sediments indicates a strong flux of terrigenous material, which is mainly provided by the Siberian rivers. Currents distributing the suspension load and sea ice are supposedly major agents transporting sediments aaoss the shelf to the central arctic deep sea basin. Sediment cores from the upper and middle continental slope exhibit only minor lithological changes. Bioturbated, fine-grained sediments with high organic carbon contents dominate. The presence of free hydrogen sulphide gas within the sediment column indicates that an intense decay of organic matter under reducing conditions is taking place. Sedimentation rates are estimated to be ca. 50 cm/1000 years at the upper slope of the western Laptev Sea, being approximately 10 times higher than at the continental rise. The suboxic t o anoxic environment diminishes at deep sea sites of the western Laptev Sea. where sedimentation rates and influx of organic matter are reduced. D . Nurnberg, D . K . Futterer, F. Niessen. C . J . Schubert, M . Wahsner, Alfred Wegener Imtitute for Polar and Marine Research, P . 0. Box 120161, 0-27568 Bremerhauen, Germany; Nmrgaard-Pedersen. N . , Spielhagen, R . F., GEOMAR, Research Center for Marine Geosciences, Wischhofstrape 1-3, 0-24148 K i d . Germany. Introduction Approximately one third of the Arctic Ocean area is occupied by shallow epicontinental seas (Gierloff-Emden 1982; Carmack 1986), the global importance of which is related to sea ice pro- duction (Colony & Thorndike 1985) and brine formation (Middtun 1985; Aagaard 1988, 1989). The Laptev Sea is supposed to be one of the main sea-ice producing areas in the Arctic where huge amounts of sediments may be incorporated and subsequently transported into the Transpolar drift. The shelf edge lies between 50 and 60m water depths (Holmes & Creager 1974), and as far as 500 km from the continent. Five large sub- marine valleys. which are interpreted as sub- merged extensions of the major rivers in that region (the Lena and Kotuy: Holmes & Creager 1974), cut into the gently dipping shelf plain. An irregular pattern of major shoals, which some- times reach or break the sea surface, characterises the sea-floor topography. In the framework of the joint German-Russian expedition ARCTIC '93 during summer/autumn 44 D . Niirnberg et al. 1993, multidisciplinary research was undertaken for the first time in the Laptev Sea from the coast across the outer shelf and continental slope t o the deep sea. T h e R,’V P O L A R S T E R N worked mainly on the outer shelf. the slope and in the d e e p sea (Futterer 1994). T h e R,WKIREYEV from the Arctic and Antarctic Research Institute ( A A R I , St. Pet- ersburg, Russia) investigated the inner shelf of the Laptev Sea (Kassens &. Karpiy 1994). T h e marine geology programme aboard R/V POLAR- STERN focused on the reconstruction of the depo- sitional environment and its change through time. Specific aspects addressed include (1) the spatial and temporal development of the arctic sea-ice cover during glacial/interglacial changes, (2) the varying impact of North Atlantic water masses intruding into the Arctic Ocean via Fram Strait and the Barents Sea. and (3) t h e influence of the large Siberian river systems for the terrigenous sediment supply onto the broad shelf and changes in marine productivity through time which are assumed to be reflected in the sea-floor deposits. Four transects covering the entire continental slope area down to approximately 3500 m were chosen for sedimentological investigations. T h e easternmost Laptev Sea (Transects G and H , Fig. 1) was ice-free during September, allowing the survey to extend far into the d e e p sea basin. T h e western Laptev Sea (Transects E and F) suffered from heavy ice conditions. Transect E was re- stricted to less than 1200m water depth. In this study we present preliminary results from ship- board investigations. Detailed stratigraphical. sedirnentological, geochemical and mineralogical investigations are currently in progress. Methods High resolution subbottom profiling ( P A R A S O U N D ) Bottom and sub-bottom reflection patterns of the uppermost sea-floor sediments received by the P A R A S O U N D sediment echosounder system (Atlas Electronics, Bremen, Germany) document sedimentary features, which can b e used to characterise sedimentary environments and their changes in space and time. T h e aims of the P A R A S O U N D profiling were (1) to select lo- cations for coring, (2) to identify and interpret lateral differences in sedimentary facies along shelf-slope transects a n d between t h e Barents and Laptev seas, and (3) to evaluate major strati- graphic features indicative of past environmental changes. in particular glacial-interglacial tran- sitions. Geological sampling and processing Geological sampling and coring recovered undis- turbed surface and near-surface samples (large box corer) a n d undisturbed, long sediment sequences (kastenlot and gravity core) for strati- graphic, paleoenvironmental and sedimento- Fig. 1. Bathymetric chart of the Laptev Sca indicating coring sites of expedition A R K 1x4 Numbers indicate the Awl number code (71 = PS2471). Inset shows the location of the study area R/V POLAKSTERN The depositional emironwient of the Laptell Sea corrrineriml niargiri 45 logical investigations. Fig. 1 shows the sampling locations. Sediment cores were obtained along four transects from the shelf t o the lower con- tinental slope. Transects F, G and H covered the depth range from approximately 5 0 m t o 3500 m . Coring operations along 'rransect E were terminated at a water depth of approximately 1200 m. T h e large box corer (GKG) recovered 25 undis- turbed sediment surfaces and subsurface sedi- ments (maximum depth 50 cm). T h e gravity corer (SL) was applied at 16 stations and penetrated sediment sequences with a n average length of 430 cm. Maximum recovery was 908 cm. Five kas- tenlots (KAL) recovered sediment sequences with an average sediment thickness of ca. 620 cm. Maximum recovery was 800 cm. T h e sediment cores obtained were routinely photographed, described in detail and graphically displayed. Sediment colours were identified according to the "Munsell Soil Colour Chart" (Kollmorgen Instruments Corp., Newburgh, USA). Samples were coded by an AWI-code (e.g. PS2460-3 SL). For the analysis of the coarse grain fraction, the bulk sediment of selected sediment samples was washed through a 6 3 p mesh, dried, and analysed under a binocular. For X-ray imaging, sediment slabs of ca. 0.5 cm thickness were taken continuously downcore in order to elucidate sedi- mentary and biogenic structures. Grain size distribution and distribution patterns of major sediment components were estimated from smear slide examinations. Abundances of coccoliths could easily be determined under the microscope with crossed nicols, and could be applied as preliminary time constraints. Whole core measurements of magnetic susceptibility in 2 cm intervals were applied for core correlation. The records mainly reflect the content of mag- netite in the sediments investigated as the fer- rimagnetic magnetite exhibits high susceptibilities in contrast to all other rock-forming minerals with weak diamagnetic o r paramagnetic character- istics. A detailed description is given in Nowaczyk (1991) and Futterer (1992). Measurements were performed with a Bartington suscep- tibility control unit M.S. 2 in conjunction with two different sensors. o n e for measurements of entire core segments, the other for paleomagnetic standard samples with volumes u p t o 10 cm3. Dur- ing the analyses, the sample within the sensor was subjected to a weak magnetic field (0.5 mT, f = 565 H z or f = 460 Hz). Results and discussion Sediment echo-sounding P A R A S O U N D profiles reveal that the Laptev Shelf is characterised by a Bat bottom topography (Fig. 2). Moraines, which would give an indication for the glacial history of this area, were not Fig. 2 PARASOUND W E 1" ... 5.0 lan - profiles from the central and eastern part of the Laptev Sea shelf Top between 77% N . 116"58'E and 77"14'N, 116"54'E Bottom between 77"3O'N, 133"37'E and 77"21'N, 133"37'E Horizontal and sub-horizontal layering and Ice gouges (arrows) arc clearly visibte 3 5 h 46 D. Nurnberg et al. observed. PARASOUND penetration was mostly limited to a few metres locally increasing up to 20m. Subsurface reflectors have mostly subparallel geometries. In places, they appear slightly deformed or "folded" (Fig. 2). These subsurface reflectors often toplap against an unconformity and are overlain by a horizontal bed, mostly only a few metres thick. The latter is interpreted as late glacial and Holocene deposits. With exposure of the shelf area, the unconforrnity may relate to the Last Glacial Maximum. The fact that PARASOUND was able to penetrate pre- Holocene Laptev shelf deposits is taken as indi- cation of the lack of an ice load during the Last Glacial. The top of subglacial deposits would be characterised by a strong contrast in acoustic impedance. Therefore, the top of ice-consoli- dated sediments is indicated by sound reflection rather than penetration. For example, during ARK-IX/4. PARASOUND penetration into Pleistocene deposits was not observed on the Barents Shelf, where sediment consolidation due to a late Pleistocene ice cap is proposed by Vorren et al. (1983). Even without consolidation of sedi- ments by an ice sheet, Laptev Shelf sound pen- etration into deeper strata is a surprising observation because relict permafrost from a periglacial tundra environment is expected to be present underneath Holocene deposits. In such a case, permafrost would cause acoustic impedance near the sediment surface probably high enough to prevent further sound penetration. It is too early, however, to speculate on the reasons of the above observation before a detailed survey for perma- frost features in seismic profiles is carried out. Several incised river channels were recorded on profiles near the shelf edge (Fig. 3 ) . The widths of the channels are on the order of a few kilo- metres, the depths are about 10-15 m. All chan- nels are filled with well-stratified sediments. These channels are presumably related to Plei- stocene drainage of the large Siberian rivers, such as the Lena and Kotuy, and were cut into the shelf (e.g. Holmes & Creager 1974). Most likely, this occurred during the Last Glacial when the sea level was proposed to be 120 m below that of today (Chappell & Shackleton 1986), and the coastline was located north of the shelf region near the present shelf/slope boundary. Conse- quently, considerable freshwater drainage most likely occurred during glacial times from the Eura- sian continent into the Arctic Ocean. These find- ings together with the lack of moraine deposits imply that the area of investigation was not cov- ered by an extended ice sheet during the Last Glacial as proposed by Grosswald (1988). Our investigations rather support the predominance of periglacial environmental conditions in the Laptev Sea area as suggested by Dunayev & Pavlidis (1988). East of Severnaja Zemlja and west of the New Siberian Islands, a large number of ice gouges have been observed at the sediment surface (order of magnitude lo3 numbers in total) (Fig. 2). Incisions of 1-101-11 depth are generally flanked by berms on just one or both sides. The total widths are on the order of a few tens of metres. In most cases. it appears that the sediment accumulated in the berms would fill the scars. The relatively small total width of the features compared to their total relief make them clearly distinguishable from drainage channels flanked by levee deposits. Furthermore, the occurrence of ice gouges is strictly limited to a range of water depth between 30 and 90 m water, whereas erosional channels were observed at all water depths on the shelf. Most ice gouges appear to be "fresh" and unfil- led (Fig. 2 bottom). Some incisions show alter- ation as indicated by their smoothed topography (e.g. Fig. 2 top left). In places, the well-stratified character of subsurface sediments seems to be locally ploughed despite the presence of a pres- ently flat and undisturbed sediment surface. Fig 3 PARASOUND profile showlng incised river channels (arrows) from the centre of the Lapter Sea near the continental slope between 77"03'N, 131"06'E and - 8.5 km 77"02'N, 130"48'E The depositional environment of the Lapteu Sea continental margin 37 Fig. 4. PARASOUND profile from the lower part of the continental slope along Transect H between 78"41'N, 132"41'E and 78"31'N, 132Yl'E. The uppermost and lowermost sections are magnified in order to elucidate the thinning out of sediment la ye rs . From observations and simulations of for- mation and filling of ice gouges from the Alaskan Beaufort Sea, Barnes et al. (1990) concluded that ice gouges are short-lived and filled within a few years. This interpretation is consistent with our observations in the Laptev Sea. It implies that ice erosion is presently an important shelf process on the Laptev Sea shelf. potentially inducing resus- pension and lateral sediment transport over the shelf. Indeed, PARASOUND profiles indicate that, in places, subsurface reflectors underwent strong deformation (Fig. 2, top). This is inter- preted as a result of strong sediment reworking by ice gouging during the Holocene. Whether ice gouging is related to grounding icebergs. or pressure ridges, or both, is still unclear. The continental slope deposits belong to dif- ferent facies. The central part of the Laptev Sea slope (Transect G , from 77"01,9'N 126"25,4'E to 79"13,7'N 122"51,7"E) is characterised by reflec- tion patterns of slumps down to water depths of ca 2000 rn, followed by channel and levee deposits to the deepest point at 3200m. Channels have distinct or prolonged reflection character. Some 48 D . Niirnberg et al. channels show truncation of older beds. In places. thick undisturbed sections were found. They com- prise subparallel reflectors which drape under- lying topographies. Penetration was down to 50 m sediment thickness. These well-stratified sedi- ments are interpreted as levee facies, which are, in conjunction with channels, typical for submarine fans. Since incised river channels of the Lena drainage system were found o n the continental shelf, t h e fan facies observed below 2000 m water depth may have formed mainly during the Last Glacial. when the riverine transport of continental debris reached to the edge of the present slope/ shelf boundary. Thick, mostly undisturbed, slope deposits were recorded along sections H and F (Fig. 4). Reflec- PS2471-4 S L PS2473-4 KAL PS2474-3 K A L PS2476-4 S L W D: 3.047 rn W D: 1.919 rn WD: 1.494 rn WD: 521 rn EOC = end of core EOC: 7 04 m Fig. 5 . Lithological core description of sediment cores from the Laptev Sea continental slope along Transect F. T h e depositional environment of the Laptev Sea continental margin 49 tions are predominantly numerous down to about 100 m sediment depth. The deposits drape under- lying topographies. Distinct reflectors correlate over long distances and show a pronounced wedg- ing out downslope. Only in the lowest part of section H (deeper than 3000 m) was a major slump identified. Sedirnentological and geophysical investigations All box core tops consist of soft brown mud. The surface deposits from the shelf and upper slope (50-1000 m) contain approximately 10-20% sand- sized material. Terrigenous grains (quartz, feld- spar, rock fragments) predominate. Ostracods, small bivalves, and calcareous and agglutinated benthic foraminifers are common, whereas plank- tonic foraminifers are absent or rare. The sand content decreases significantly downslope and is less than 1% in water depths greater than 3000 m, there consisting almost entirely of planktonic foraminifers. The density of benthonic popu- lations is high on the shelf and uppermost slope, but decreases significantly downslope. The uppermost, brown lithological unit of the sediment cores is thickest on the lower slope (ca. 60cm) and decreases to 3-10cm on the upper slope and shelf (Fig. 5). This strongly bioturbated unit resembles the sediment surface in the box cores and apparently reflects the depth of oxy- genation. On the shelf and upper slope, the under- lying dark grey to black sediments mainly consist of sandy, silty clays. Occasionally, thin sandy layers (turbidites) are intercalated, predomi- nantly in the lower parts of the sediment cores (Fig. 5 ) . The amount of coarse material (> 63 pm) slightly decreases with increasing water depth. When exposed to the atmosphere for a sufficient time ( 1 2 4 8 hours), the dark sediment changes to brownish and olive greenish colours. The sedi- ments contain high amounts of organic matter (max. 2%), the main component of which is ter- rigenous in origin (e.g. plant fibres) (Stein & Nurnberg 1995). Formation of free hydrogen sul- phide gas is typical in these sediments, pointing to the decay of organic matter by the activity of sulphate-reducing bacteria under reducing con- ditions. Early diagenetic precipitations of calcium carbonate hexahydrate (ikaite) were found in ca. 235 cm depth of core PS2460-4 from shallow wat- ers (200 m). In the eastern Laptev Sea (transects G , H), the dark subsurface-sediments from the upper slope and outer shelf occur down to ca. 3500m water depth. Sediments from the western Laptev Sea (transects E, F), however, are partly different. The subsurface sediments from water depths greater than 3000 m contain mostly silty clays of brownish to olive greenish colours, being sig- nificantly lighter than sediments from shallower depths. Free hydrogen sulphide gas is no longer present. A few well-preserved shallow marine bivalves ( Y o l d i a amygdalea) were observed in cores PS2458-4 and PS2460-4 (Transect H) at 40G 700 cm core depth. Smear slide investigations of core PS2471-4 revealed the coccolith species E m i - liania huxleyi and Gephyrocapsa sp. to be present down to 400 cm core depth. Maxima in abundance occur in near-surface sediments (ca. 20 cm), between ca 200cm and 300cm, and around ca 380cm (Fig. 6). Mixing of sediment by bioturbation is extremely strong o n the shelf and upper slope. Vertical chitinous worm tubes with a length up to 30 cm indicate a very thick upper mixed sediment layer. Although the degree of benthonic coverage decreases downslope, bioturbation is still an important environmental process, which prevents any stratification. Fig. 6 shows magnetic susceptibility records of sediment cores from Transect F (see Figs. 1 and 5 ) . Continental slope and shelf sediments are typically characterised by magnetic suscepti- bilities of approximately 40-250 x SI. These values are significantly higher than values pre- viously published for central Arctic Ocean sedi- ments. Cores recovered during the ARCTIC '91 expedition of R/V POLARSTERN have magnetic susceptibilities of approximately 25-50 X SI (Amundsen and Nansen Basin) and 15- 30 x 1O-'SI (Gakkel and Lomonosov ridges) (Futterer 1992). The magnetic susceptibilities generally decrease downslope. Shelf and upper slope sedi- ments (<500m water depth) are commonly characterised by magnetic susceptibilities larger than 60 x SI, mainly ranging around 100 x SI with extremes up to 250 x SI. Below ca 500 m water depth, mean magnetic sus- ceptibility values decrease to 2C60 x SI. The decrease of magnetic susceptibilities from shelf to continental slope and into the central Arctic deep-sea and mid-ocean ridge sediments presumably reflects a constant. density-dep- endant loss of heavy, ferrimagnetic minerals of C o re c o rr el at io n a cr o ss t h e L ap te v S ea c o n ti n en ta l sl o p e b y m ag n et ic s u sc ep ti b ili ty L o w er s lo p e I U p p e rs lo p e P S 2 4 7 1 -4 S L P S 2 4 7 2 -4 S L P S 2 4 7 3 -4 K A L P S 2 4 7 4 -3 K A L P S 2 4 7 5 -3 S L 79 " 09 .1 N 11 9" 47 ,6 'E 7 8 °4 0 ,1 N 11 8" 44 ,S E 77 '5 8. 6N 11 8' 34 ,S E 77 '4 02 " ll B 0 3 4 ,5 'E 77 '3 1. 8N 11 8' 27 .8 'E N a n n o - p la n k to n S u s c e p t. ( 1 0 e -5 S I) 0 20 4 0 60 80 80 0 7 0 0 L . - fn M a g n e ti c s u s c e p ti b il it y S u s c e p t. ( 1 0 0 -5 S I) 20 40 60 8 0 S u s c e p t. ( 1 0 8 -5 S I) S U S C eD t. (1 0 e -5 S ib S u s c e ~ t. 11 0 0 -5 S I) 20 4 0 60 80 1 0 0 1 2 0 4 0 SO 12 0 1 W D 2 6 2 0 m 80 12 0 16 0 20 0 P S 2 4 7 6 -4 S L P S 2 4 7 7 -4 S L 77 '2 3. 4N 1 1 8 °1 1 ,6 E 7 7 V 4 .6 N 11 8' 32 ,9 E lo o 15 0 2 si 100 200 400 a 500 ;i 3 800 700 800 D ec re as in g d o w n co re o xi da tio n _ _ _ _ . c - In cr ea si n g a bu nd an ce in p la nk to ni c fo ra m in ife rs I T1 - T3 T u rb ld it e s e q u e n c e s - s e d im e n t e n ri c h e d i n o rg a n ic 0 D a rk g ra y t o b la c k f in e -g ra in e d m a te ri a l B m w n to o li v e f in e -g ra in e d s e d im e n t Z o n e o f o x id a ti o n Fi g. 6 . G eo ph ys ic al , se di m en to lo gi ca l an d st ra ti gr ap hi c re su lt s fr om T ra ns ec t F c ov er in g a de pt h ra ng e of ca . 4O OO m . Sc he m at ic l it ho lo gy l og s, t he a bu nd an ce o f pl an kt on ic fo ra m in if er s a nd c oc co li th s an d th e de pt hs o f th e ox id at io n zo ne a re i nd ic at ed . C or e co rr el at io n (s ha de d ar ea s) is b as ed o n m ag ne ti c su sc ep ti bi li ty r ec or ds . W D = w at er d ep th The depositional environinent of the Laptev Sea continental margin 51 terrigenous origin (e.g. magnetite) during long- range sediment transport. Depositional environment The geophysical and geological investigations allow a preliminary characterisation of the outer Laptev Sea depositional environment. In general, the sediment composition indicates a strong supply of terrigenous material from the continent to the Laptev Sea shelf edge. The relatively high content of plant fragments and the occurrence of freshwater and shallow marine bivalves in cores from the middle and lower continental slope underline the importance of a large-scale downslope transport of sediments. Most of the fine-grained material is probably transported in suspension by currents. The Lena River, with an outflow of approximately 520 km’/ year (Aagaard & Carmack 1989), is one of the largest river systems flowing into the Arctic Ocean. The Kotuy River adds another 105 km3/ year (Aagaard & Carmack 1989). The Lena sus- pension load is calculated t o amount to 17.6 mil- lion tons/year (Gordeev & Sidorov 1993), which is relatively low compared to the 100 million tons/ year of the McKenzie River (Milliman & Meade 1983). Significant amounts of sediment may also be contributed by sea ice. Dethleff et al. (1993) and Nurnberg et al. (1994) already demonstrated the importance of the Laptev Sea for Arctic sea- ice formation and resuspended sediment entrainment by sea ice. Dethleff (1995) estimates the amount of sea-ice sediments exported from the Laptev Sea shelf to approximately 10 million tons/year. Whether the entrained portion of shal- low marine sediments in sea ice is subsequently released and adds to the slope sedimentation will remain unclear as long as the lack of strati- graphical information prevents accurate estimates of sediment mass balances in this area. It is, however, probable that major portions of these sediments leave the Laptev Sea area within the transpolar ice drift (Nurnberg et al. 1994). Sedi- ment-rafting by glacier ice seems to be of minor importance in the Laptev Sea. Core-top sedi- ments and sediment cores from the continental slope contain only minor portions of coarse sand and gravel, which may be indicative of iceberg transport. The typical black spots colouring the subsurface sediments dark grey to black are related to iron and manganese precipitations, which must be attributed to the decay of marine organic matter by the activity of sulphate-reducing bacteria (e.g. Emelyanov et al. 1982). This process is active o n l y under reducing conditions. The strong anaerobic bacterial activity (in Futterer 1994) indicates a rapid burial of high amounts of marine organic matter before a sufficient oxidation from sea- water could take place. This process apparently is less important for the western Laptev Sea deep sites, where brownish and olive greenish colours indicate a “normal” deep-sea oxidation and/or a lower supply of marine organic matter to this area. Here, the sediment colour is presumably determined by iron and manganese oxides, clay minerals, etc. Though most Laptev Sea continental slope cores only exhibit insignificant lithological changes, thus preventing visual core correlation, the magnetic susceptibility provides an excellent tool for core correlation and for determination of relative sedimentation rates across the continental slope. Transect F (Fig. 6) comprising 7 magnetic susceptibility logs elucidates the depositional pat- tern along the continental slope. The sus- ceptibility record from core PS2471-4 (3047 m water depth) unambiguously correlates with all cores located on shallower slope positions. A distinct susceptibility peak at ca. 50 cm core depth in core PS2471-4 is present in all other cores from Transect F , but becomes increasingly pronounced when reaching the upper slope at 187m water depth (indicated by light shading in Fig. 6). Here. this peak appears at approximately 560 cm core depth in core PS2477-4, indicating that sedi- mentation rates apparently are approximately 10 times higher near the shelf break than in the deep sea. The continuous downslope wedging-out of the approximately 6 m thick homogenous sedi- ment package from the shelf break is supported by the PARASOUND echo-sounding profile along Transect H (Fig. 4). The strongly decreasing sediment accumulation with greater water depths is also suggested from the increasing amount of planktonic foraminifers in the coarse fraction of the surface sediments (Fig. 6). Although conditions for planktonic life become unfavourable to the north due to per- ennial sea ice and colder waters, the downslope decreasing dilution of terrigeneous material makes planktonic foraminifers become the dom- inant coarse fraction component in surface sedi- ments from the lower slope, whereas they are rare in upper slope deposits. 52 D. Nurnberg et al. Fig 7. Map of the Siberian Arctic showing glaciation models from Grosswald (1988) and Dunayev & Pavlidis (1988). Assuming a Holocene age for the upper ca. 50 cm of core PS2471-4 based o n the abundance maximum of Gephyrocapsa sp., which has proven to be a useful indicator for arctic interglacial sediments ( G a r d 1988). we speculate from the core correlation presented in Fig. 6 that the upper approximately 6 m of sediments in cores PS2476- 4 and PS2477-4 were entirely deposited during Holocene times. Corresponding sedimentation rates of ca. 50cm/1000 years a r e in accordance with rates of 40 cm/1000 years calculated from a radiometrically dated sediment core from the Laptev Sea shelf (Kassens, pers. corn. 1994). The homogeneity of the sediments suggest that environmental conditions did not change sig- nificantly during this time interval. T h e cores from the lower slope, which often contain sandy layers, presumably span over a longer time interval. possibly several glacial-inter- glacial cycles. T h e occurrence of the coccolith Gephyrocapsa sp. at 2 0 G 3 0 0 c m in the lower slope core 2471-4 (Fig. 6) might be related to the Last Interglacial (oxygen isotope stage 5). Conclusions The sedimentary environment of the Laptev Sea continental margin is characterised by com- paratively high bulk sedimentation rates. T h e deposition of relatively high amounts of ter- rigenous and marine organic matter a n d , in many areas, bacterial activity and insufficient oxidation lead to t h e presence of hydrogen sulphide gas within the dark grey t o black sediments. Since there is n o major change either in lith- ology or in reflection geometry, it is speculated that these deposits are of Holocene age. A pre- liminary coccolith stratigraphy, in combination with the magnetic susceptibility records allowing the correlation of cores across the entire slope, support this assumption. High sedimentation rates near the shelf edge, which are presumably caused by a strong lateral transport of debris on the shelf in a n off-shore direction, decrease with increasing water depth. Downslope wedging-out of undisturbed sediment layers, slumps, and fan deposits o n the lower slope a r e typical for pro- grading slopes. Whether progradation is still active today is questionable since (1) t h e large Lena Delta shoreline is generally retreating ( Z e n - kovitch 1985) and (2) the present suspension load of the L e n a River is relatively low (Gordeev & Sidorov 1993). Important changes affecting t h e relationship between continental supply and destructive marine forces may have occurred dur- ing postglacial time. T h e lack of moraines and diamictons and of The depositional environment of the Laptev Sea continental inargin 53 considerable amounts of dropstones on the outer shelf do not support the assumption of a glaciated Laptev Sea shelf during the Last Glacial. An unglaciated Laptev Sea shelf is also supported by the incised river channels observed on the outer shelf, which were cut into the exposed shelf in late Pleistocene times. A continuous freshwater supply into the Arctic Ocean during the Last Glacial under presumably periglacial conditions as suggested by Dunayev & Pavlidis (1988) (Fig. 7) is more in agreement with our observations. The post-glacial environment is generally charac- terised by ice-erosion and sediment reworking as indicated by significant numbers of ice-gouges. Acknowledgements. - For data discussion we thank E . Reimnitz. R . Stein, H. Eicken, and H. Kassens. Thanks are also due to two reviewers, who improved the manuscript. N. N.-P. and R . F. S . were financially supported by the BMFT- “Palloklimaprojekt”. This is contribution No. 910 of the Alfred Wegener Institute for Polar and Marine Research. References Aagaard, K. 1988: Some thoughts on the large-scale circulation of the Arctic Ocean. Second Conf. Polar Meteorol. Oceanogr. March29-31,1988,1-3. Am. Meteorol. SOC., Madison, Wisc. Aagaard, K. 1989: A synthesis of the Arctic Ocean circulation. Rapp. P. u. Reun. Cons. Int. Explor. Mer. 188, 11-22. Aagaard, K. & Carmack. E. C. 1989: The role of sea ice and other fresh water in the Arctic circulation. J . Geophys. Res. 94 (CIO), 14485-14498. Barnes, P. W., Hunter, R . E . , Lee. A , , Reimnitz. E . & Weber. W. S . 1990: Sediment transport by ice gouging: application of model experiments to the arctic continental shelf. Pp. 10% 111 in Ackley, S. F. & Weeks, W. F. (eds.): Sea ice Properties and Processes. CRREL Monograph 9&1, Cold Regions Research and Engineering Laboratory, Hanover, New Hampshire. Carmack, E . C. 1986: Circulation and mixing in ice-covered waters. Pp. 641-712 in Untersteiner N. (ed.): The Geophysics of sea ice. NATO AS1 Series B 146. Chappell, J . & Shackleton, N. J. 1986: Oxygen isotopes and sea level. Nature 324, 137-140. Colony. R. & Thorndike, S . 1985: Sea ice movement as a drunkard’s walk. J . Geophys. Res. 90 ( C l ) . 965-974. Dethleff, D . 1995: Die Laptevsee - eine Schlusselregion fur den Fremdstoffeintrag in das arktische Meereis. Unpubl. Ph.D. thesis. Univ. Kiel. 111 pp. Dethleff, D . , Numberg. D . , Reimnitz, E . , Saarso, M. & Sav- chenko, Y. P. 1993: East Siberian Arctic Region Expedition ’92: The Laptev Sea - Its significance for Arctic sea ice formation and Transpolar sediment flux. Ber Polarforsch 1201 93. H E . Dunayev. N. N. & Pavlidis, J. A . 1988: A Model of the Late Pleistocene Glaciation of Eurasiatic Arctic Shelf. Pp. 7&72 in Kotlyakov, V. M. & Sokolov. V. E . (eds.): Arctic Research - Advances and prospects, Vol. 2. Proceedings of the C O I I - ference of Arctic and Nordic Countries on coordination of research in the Arctic, Academy of Sciences of the USSR. Leningrad, December 1988. Emelyanov. E. M.. Pilipchuk, M. F., Volostnykh. B. V . , Khandros. G . S. & Shaidurov, Yu. 0 . 1 9 8 2 : Fe and Mn forms in sediments in the geochemical profile of the Baltic Sea. Baltica 7, 15S171. Fiitterer, D. K. (ed.) 1992: ARCTIC‘91: The expedition ARK- VIII/3 of RV “Polarstern” in 1991. Ber. Polarforsch. 107/92. 267 pp. Fiitterer, D. K. (ed.) 1994: The Expedition ARCTIC93 Leg ARK 1x14 of RV “Polarstern” 1993. Ber. Polarforsch. I # / 94. 244 pp. Gard, G. 1988: Late Quaternary calcareous nannofossil bio- chronology and paleo-oceanography of Arctic and Subarctic seas. Medd. Stockholm Uniu. Geol. Inst. 275. W 5 . Gierloff-Emden, H . G . 1982: Das Eis des Meeres. Walter de Gruyter, Berlin, 767-940 pp. Gordeev, V. V. & Sidorov, I. S . 1993: Concentrations of major elements and their outflow into the Laptev Sea by the Lena River. Mar. Chem. 43. 3S45. Grosswald. M. G . 1988: Late Pleistocene ice sheet in the Soviet Arctic. Pp. 18-23 in Kotlyakov, V . M. & Sokolov, V. E . (eds.): Arctic Research - Advances and prospects Vol. 2. - Proceedings of the conference of Arctic and Nordic Countries on coordination of research in the Arctic. Academy of Sciences of the USSR. Leningrad, December 1988. Holmes, M. L. & Creager, J. S. 1974: Holocene history of the Laptev Sea continental shelf. Pp. 211-229 in Herman, Y. (ed.): Marine Geology and Oceanography of the Arctic Seas. Springer Verlag, Berlin. Kassens, H . & Karpiy. V. Y . 1994: Russian-German Coop- eration: The Transdrift I Expedition to the Laptev Sea. Ber. Polarforsch. 151194. 168 pp. Middtun, L. 1985: Formation of dense bottom water in the Barents Sea. Deep-Sea Res. 32 ( l o ) , 123S1241. Milliman, J . D. & Meade, R. H . 1983: World-wide delivery of river sediment to the oceans. J . Geol. 91. 1-21. Nurnberg, D.. Wollenburg, I . , Dethleff, D . , Eicken, H . . Kassens, H.. Letzig, T., Reimnitz, E. & Thiede. J . 1994: Sediments in Arctic sea ice - Implications for entrainment. transport and release. Mar. Geol. 119, 185-214. Nowaczyk, N. 1991: High-resolution magnetostratigraphy of Late-Quaternary Arctic marine sediments. Ber. Polarforsch. 78/91. 187 pp. Stein, R. & Nurnberg, D. 1995: Organic carbon and biogenic opal in surface sediments from the Laptev sea and the adjac- ent continentalslope. In Kassens, H . . Thiede, J., Hubherten. H . W., Piepenburg, D . , Priamikov, S . & Timokhov. L. (eds.): Second Workshop on “Russian-German Cooperation in and around the Laptev Sea”, St. Petersburg. November 1994. Ber. Polarforsch. (First version. February 1995). Vorren, T. 0.. Hald, M., Edvardsen, M. & Lind-Hansen, 0.- W. 1983: Glacigenic sediments and sedimentary environ- ments on continental shelves: General principles with a case study from the Norwegian shelf. Pp. 61-73 in Ehlers, J . (ed.): Glacial deposits in north-west Europe. Balkema, Rotterdam. Zenkovitch, V. P. 1985: Arctic USSR. Pp. 863-871 in Bird, E. C. F. & Schwarz, M. L. (eds.): The World’s Coastline. Van Nostrand Reinhold Company, New York.