The climate-sensitive Vesterisbanken area (central Sea): Depositional environment and paleoceanography during the past 250,000 years

MARTIN ANTONOW1, PETER MARTIN GOLDSCHMIDT2j3and HELMUT ERLENKEUSER2

1. Freiberg University of Mining and Technology, Institute of Geology, Bernhard-von-Cotta-Str. 2, D-09596 Freiberg, Germany 2. SFB 313, University of Kiel, Heinrich-Hecht-Platz 10, D-24118 Kiel, Germany 3. present address: PCD, Eckernforder Str. 259, D-24119 Kiel, Germany

ABSTRACT Sedimentological, micropaleontological and geochemical studies of sediment cores from the Vesterisbanken region were used to reconstruct the sedimentation pattern, depositional history and paleoceanography for this area over the last 250,000 years. The dating and correlation of the sediments were based on oxygen isotope stratigraphy and absolute ages. The hemipelagic deposits near the Vesteris are characterised by biogenic, ter- restrial and volcanogenic sediment input that varies through time. The area was influenced by sporadic turbidity currents and thermohaline-induced contour currents. Ice-rafted debris occurred nearly throughout the investigated time interval. Primary production was higher during interglacial periods. Filter-feeding epifauna (C. zi~~~t~llt~rst~rji),an indicator of bottom cur- rents, dominated in isotope stages 7 and 5. During deglaciation (stage boundaries 8/7, 6/5 and 2/1, events 3.3 and 3.1), enorinous meltwater inp~~tstabilised the water column, leading to periodic interruptions in deep water renewal. The influence of water masses from the Polar and Atlantic Domains in the were very variable over time. The oceanic fronts in the Vesterisbanken area were always close together, allowing only a narrow Arctic Domain to exist. Although the area is rather unique in relation to other parts of the Greenland Sea, climatic data can be compared favourably to the established Late / sedimentation pattern in the northern North Atlantic. Due to their good correlation to ice core records from Greenland and Antarctica, the Vesterisbanken sediments present a high-resolution picture of global climate change.

INTRODUCTION: THE VESTERIS SEAMOUNT bathylnetry and gave rise to strong basaltic volca- IN THE GREENLAND SEA nism that event~lally generated the Vesteris The Greenland Sea is one of the few links between Seamount (73"301N,9"101W, Fig. 1).Vesterisbanken the Arctic Ocean and the northern North Atlantic. rises from the Greenland abyssal plain up to 133 in This area is reinarkably sensitive to oceanographic water depth (Johnson & Campsie, 1976), thus for- and climatic changes because Polar and Atlantic ming the only intra-plate in the ~~orthern water masses mix here. Morphologically, the North Atlantic. Greenland Sea is framed by major fracture zones The goal of this investigation is to provide a and ridges such as Knipovich and Mohns Ridges to reconstruction of the climatic development and the the east and the Greenland continental slope/shelf sedimentary environment of an area of the to the west (Vogt, 1986). Tectonically, sea-floor Greenland Basin that has been studied relatively spreading since the Paleocene led to a complex little due to a thck ice cover linliting access. The

In: Hass, H.C. & Kaminski, M.A. (eds.) 1997. Contributions to the Micropaleontology and Paleoceanogra- phy of the North Atlantic. Grz!ybozoski Fo~lildatio~lSpccinl Pllblicntion, no. 5, pp. 101-118

a - P6hocny Atlantyk Antonow, Goldschmidt and Erlenkeuser

central reasons and questions to be answered for this study are:

1. The determination of sedimentation in the Vesterisbanken area. - What depositional processes affected the study area? -Was there a temporal variation of different sedimentation patterns? 2. Paleoceanographic determination for the central Greenland Sea area. - Which paleo-surface water masses influenced the area around the Vesterisbanken and how variable is the position of the oceanic fronts? - When did meltwater pulses change the isotopic character of the Greenland Sea? 3. The integration of these results into the pattern of Late Pleistocene-Holocene sedimentation and paleoceanography of the Greenland-Jceland- 1WW 0" 1WE I Norwegian Seas. -+Vlskris Sralnoulll 1.01) Polar "'3nl;lin N41, NorlI~Athtl~lic l)riKl - Does the paleoceanographic development of Pelur Fr,,nl ARD Arclic Lk,mi,in yVC;l Na~~-rrgi;n~~IU,anrla~ll Cx#rrsnl 1Vr.t SpiI.l>wgcn Cnmrrem~l - ~~~~i~ AND ~~l~,,~i~~,,,,,~i~I

- Central Greenland Sea

Influence of

from the East

---

NAW- North Atlantic Water NSDW - Norwegian c;,. 5" \v Deep Water Temperature Salinity Reference Surface Waters Polar Water < 0°C < 34.4 %< Swift (1986) EGC - East Greenland Current (EGC. JMC) -I°C 33 %< Hopkills (1991) JMC - Jan Maycn Current RAC- Return Atlantic Current Intermediate Water Pnquette et al. ( 1985) AIW (IIAC) 0.5 to 3OC 34.9 to 35 Ic(. Berner (1 99 1) ADW- Arctic Deep Water Deep Waters GSDW- Greenland Sea Deep Water AI)W -0.5 to -0.8"C 34.93 to 34.95 %r Hopkins (1991 ) AIW- Atlantic lntermediatc Water GSDW -I. l to -I.3"C 34.88 to 34.90 Tr Swift (1986) PW- I'olar Water

Figure 2. Recent oceanography of the Greenland Sea after several authors (see text).

recovered from the vicinity of the Vesteris Coarse fraction analysis Seamount (Thiede & Hempel, 1991). The present The analysis of the sediment composition of the study investigates sedimentation processes at sites two subfractions 125-250 pm and 250-500 pm was PS1878, PS1882, and PSI892 (Fig. 3, Table 1). To carried out according to the method of Sarnthein answer the questions raised in section 1, these sedi- (1971). When the sample had more than ca. 400 ment cores from the Greenland Sea were analysed particles per subfraction, a microsample splitter in terms of sedimentology, micropaleontology and was used to generate representative subsamples. geochemistry. The sampling interval was 4 cm. The presence and frequency of the coarse fraction analysis is represented in grain percent for each Granulometry subfraction. A Zeiss GSZ microscope was used for A combination of two size-dependent methods the analyses. was used to determine the total grain size. The Ca. 600-700 sediment particles were assigned to the boundaries of the grain size subfractions are based following groups: on the scale of Krumbein (1936). The weighed sediment sample was wet-sieved A terrigenous particles (quartz, rock fragments) to separate the fine (<63pm) and the coarse B volcanogenic particles (>631un) material. The coarse fraction was divided (brown and transparent, porous and amor- into five subfractions (63-125, 125-250, 250-500, phous glass, volcanic fragments) 500-1000, >I000 pn) using standard sieves and an C benthic biogenic particles ATM Sonic SifterTM.The fine fraction was investi- (, spicules) gated using the LumosedTMphotosedimentometer D planktic biogenic particles (measuring range 1-63 pn). The photosedimento- (foraminifera, , diatoms) meter method is based on gravimetric sedimen- E authigenic particles (pyrite, aggregates) tation and is described by Staudinger et 01. (1986). In comparison to conventional sediment decompo- The species and genus of benthic and planktic for- sition (weight %) according to Atterberg (1912), aminifera were differentiated according to Be this optical measuring method (volume %) offers (1977) and Thies (1991). very good reproduction (Syvitski, 1991). Antonow, Goldschmidt and Erlenkeuser

Sediinentation and accr~rnt~lationrates According to the stratigraphy determined by the isotope dates (see following section), the samples between the isotope fixpoints can be dated using linear interpolation. Note, that several sediment sequences (turbidites) are not taken into account due to the sporadic sedimentation processes (Antonow, 1995). However, the assumptions of constant sedimentation and no compaction in the sediment core can lead to errors in the determined linear sedimentation rates (LSR). In order to take these factors into account, accum~~lationrates were calculated. The dry bulk density (DBD) is calculated from the wet bulk density (WBD), the porosity (POR) and the density of seawater (1.025 gcm3 at O°C) Figure 3. Bathymetry of the Greenland Basin and loca- according to Elvmann & Thiede (1985): tions of the stations around Vesterisbanken. (1) DBD = WBD - (1.025~.cm~.POR.~O-*) Geochemistnj (CaCOy TOC, S) The total organic carbon (TOC) was determined The DBD and the LSR are used to calculate the bulk photometrically using a Leco CS-125~'' analysis accumulation rate (ARBULK;van Andel et nl., 1975): system. An aliquot of each sample was decarbona- ted with 0.25n HCI and was also analysed. The dif- (2) ARB,,, [gem-* . k Y 1 ference of the measured values results in the anor- = DBD [gem"] . LSR [cm . ky-'] ganic carbon percentage, which leads to the carbon proportion when it is multiplied with a stochiome- The accumulation rates take sediment colnpaction tric factor (8.333). This assumes that the anorganic into account with the assumptions that the pore carbon percentage is actually present as CaC03. A space is totally filled wit11 water and that no dia- "by-product" of the C/S analysis is the sulphur genetic effects occur (van Andel ef nl., 1975; Tl~iede content of the sample. The preparation and ana- et al., 1986). lysis of samples using this method, as well as a dis- To calculate the accumulation rate of indivi- cussion of arising errors, was described in detail by dual parameters, their proportion to the total sedi- Wolf (1991). The Leco samples were made every 10 ment is multiplied with ARBULK: cm in the kastenlot cores and every 4-5 cm in the box cores. All geochemical measurements were (3) ARparameter[g.cm-* . ky-'I carried out on both the total sample and the fine = Parameter [weight-%]. ARBULK[gcrn-* . ky-'1 fraction (<63 ~un).This allows estimates to be made regarding the input of both coccoliths (fine frac- Accumulation rates of components in the coarse tion) and planktic and benthic foraminifera (coarse fraction uses the equation: fraction). (4) ARco,,,p [g'cm-' . ky-'] = Component [%] . AR,h3F,m[gcm-2. ky-'1 Table 1. Core locations (PS: Polarstern, GKG: boxcore, Wolf (1991) presented a detailed error analysis of KAL: kastencore). results obtained with these equations.

Core Lat~tude Longitude Water Depth Recovery STRATIGRAPHY Number (North) (West) (m) (cm) Absolute dating ("T) Four samples were dated using the AMS 14C PSl878-2 (GKG) 73" 15.10' 09' 00.94' 3038 40 method. The required 1200-1300 specimens of PSl878-3 (KAL) 73" 15.33' 09' 00.73' .W48 469 N. pach!yderina sin. were picked out of the 125-250 pm fraction and measured at the ETH Ziirich FS1882-1 (GKG) 73' 35.52' 08" 23.80' 3169 12 (Institut fiir Mittelenergiephysik) according to the PS1882-2 (KAL) 73" 35.96' 08" 19.29' 3175 650 standard methods (see Kromer et a/., 1987). The measured ages were corrected according to Bard PS1892-1 (GKG) 7.1' 44.05' 09" 37.52' 3125 18 (1988), Bard ef 01. (1990) and Winn et 01. (1991). The 151892-3 (KAL) 73" 44.06' 09" 41.17' 3002 436 dates are listed in Table 2. Vesterisbanken (central Greenland Sea) during the past 250,000 years

Analysis of stable isotopes (6180, 613c) Duplessy (1978) and Ganssen (1983). The sampling Stable isotope analyses were carried out using 20- interval was 4 cm. The isotope ratios 61s0 and 613c 30 tests of N. yacl~!lderr?la sin. (125-250 ~un)per were measured by means of a Finnigan MAT 251 sample according to the method described by mass spectrometer at the C-14 Laboratory of the Universitv of Kiel, Germany. The results were cor- related to' the PDB standarh (Craig, 1957; Craig & Table 2. AMS ages of sediment samples. Gordon, 1965). Statistical errors make LID to 0.09%11, for 6180 values and ~tpto 0.04%o for the 613C Scdiment Corc depth, I4C agc 14C agc Stdnd'ird measurements (H. Erlenkeuser, pers. comm., 1994). CI'I'L' orlginnl (ka) corrected deviation An oxygen isotope stratigraphy correlating the (cm) (k4 (16) investigated- sediment cores was established by Antonow (1995) by comparing the isotope data to 28.530 28.130 330 the stratigraphy of Shackleton & Opdyke (1973, PS1882-2 (KAI.) 18.630 18.230 1976), to the SPECMAPscale (Imbrie ct nl., 1984; 40.610 40.210 1.020 Martinson ct nl., 1987) and to more recent results PSl892-3 (KA1.) 32.230 31.830 from sites in the Norwegian-Greenlaid Sea

Table 3. Oxygen isotope stratigraphy and age/depth correlation of the sediment cores PS1878, PS1882, and PSl893 (note * markes 14C ages).

Core depth Core depth Core depth Termination Age (ka) Reference PSI878 PSI882 PS 1892 (cm) , (c1n) (cm) End 65 I 85 Beginning

85

End TI Beginning

End I11

V Vogelsang (1990); S: Sarnthein et 01. (1992);W: Weinelt (1993);M: Marti~~sonct 01. (1987); I: Imbrie et al. (1984) 106 Antonow, Goldschmidt and Erlenkeuser

Cores PS1878-21-3 Cores PS1882-11-2 Cores PS1892-11-3

3180 (%c) vS. PDU 3I3c (Rc) vs. PDlI 8'0 (Zc) vs. PDU 2'3~(Zc) VS.PDU di80 (Ir)vs. PDll 313c (%<) vs. PDIS 5 1 3 2 -01.5 0.0 0.5 1.11 5 4 3 I 11.0 0.5 1.11 5 4 3 2 4.5 ll.11 0.5 1.11

. .

0

50

iln, < 1.50

21s

250

300

350

4l*

.Is0

SIMI

550

h(n1

651 kplh laf fcml

Figure 4. Stable isotope data (6180,613C) of N.pncllyderrim sin. versus depth of the cores PS1878, PSI882 and PSI892 (boxcore and kastencore). Oxygen isotope stages are marked.

(Vogelsang, 1990; Sarnthein et al., 1992; Weinelt, LSR for the Vesterisbanken area have increased sig- 1993; Jiinger, 1994). nificantly (Table 4). The maximum amount is 10 Generally, the ratios of the stable isotopes 6180 cm/ky at the end of Stage 3. The Vesterisbanken and 613c in the sediment cores presented in Fig. 4 LSR agree well with the valiies from the central are typical for rniddle/late Pleistocene to Holocene Greenland Sea (Jiinger, 1994) and with data from values in the Norwegian-Greenland Sea (Vogel- the east Greenland continental margin (Nam et nl., sang, 1990; Weinelt, 1993; Jiinger, 1994; Antonow, 1995; Nam, 1996) since isotope Stage 7. However, 1995, 1996; Nam el al., 1995). The sedimentary record from the sites near Vesterisbanken repre- sents the paleoceanographic history back to oxy- Table 4. Mean linear sedimentation rates (LSR) since gen isotope Stage 8 (PS1892). Two of the cores Stage 8, (PSI878 and PS1882) contain high resolution records of Stages 2 and 3 (Fig. 4). Oxygen Duration LSR LSR LSR The oxygen isotope stratigraphy and isotope stages (ky) PSI878 PSI882 PSI892 age/depth correlation of sediment cores PS1878, (un/ky) (crn/ky) (crn/ky) PSI882 and PSI892 are presented in Table 3. The 1 12.050 4.W 2.24 5.39 F180 stratigraphy was complemented by the four 2 12.060 5.72 3.65 0.00' 14c-~~Sages and a magnetostratigraphic frame- 3 34.890 3.17 3.75 0.72' work (Antonow, 1995; Nowaczyk & Antonow in 4 14.910 2.21 2.21 press). 5 55.9.30 1.64 1.05 6 59.770 2.05) 1.62 Sedimentation and accumulation rates 7 54.570 1.71 In general, the linear sedimentation rates (LSR) of 8 58.820 3.76 " the Vesterisbanken cores vary between 0.4 and ca. 6 cm/ky. Since ca. 59 ka (Stage boundary 4/3), the * Note: Low LSR are caused by erosion. " Data available only for the last 4,000 years of Stage 8. Vesterisbanken (central Greenland Sea) during the past 250,000 years

the Vesterisbanken LSR vary significantly lilha,lngy la~luprSlrali~ntphy 'IX0 M'pDI' r)IJc[%,I vs. PDII since Stage 3. Akh I*u?rn Ul'hicknes In cml Iq" 9 6 .3 I1 S 4 3 2 -0.5 11.11 0.5 1.I1 Accumulation rates were calculated I1 from the LSR (see section 3). These results are discussed in the following sections.

Vesterisbanken standard profile 511 The raw data sets for several segments of the sediment column had to be synthesised to characterise the overall sedimentation pattern and paleoceanography of the IIWI Vesterisbanken region (Fig. 5). The follow- ing core segments were used in the core synthesis: PS1878-2 (GKG), 040 cm: undi- sturbed surface and near-surface sedi- 1541 ments; PS1878-3 (KAL), 45-435 cm: highest resolution of meltwater events in Stages 2 and 3; PS1882-2 (KAL), 310-535 cm: higher resolution of Stage 5 than in PS1892-3 ZIWI (KAL); and PS1892-3 (KAL), >220 cm: con-

L3l 12 X 4 II Figure 5. Standard profile of the Vesterisbanken Inuninlrd Smiam. I'l'hirknr- In cnt I Oiy~unla,l#>pc SI;~gcs ,\~vI hx 1 region: Lithology and stable oxygen and carbon isotope records of N. pacl~ydcrnmsin. during the lilh#++!lr I.%md 0Ilva~~*.l;g~e l';r~a Vulr,~~licl;~r~~c!:.tin.\ 0I..~IMIII:IIU~~ SCIIJIIICIII I';ICIC\ past 250,000 years.

Figure 6. The variability of climatic and sedirnentological records demonstrated by the example of a PSI878 sediment core section. 108 Antonow, Goldschmidt and Erlenkeuser

tinuous sedimentation and the maximum stratigra- PI), which consists mainly of planktic foramini- phic time period. Using this standard profile, it is fera and terrigenous material (quartz and rock possible to determine general trends of the chan- fragments), forms a relatively small part of the total ging environments in this area. sediment. Following the 61X0maximum is a sec- tion (85-80 cm, 14.9-13.5 ka) with decreasing 813c VARIATIONS OF CLIMATIC AND SEDI- conditions where the very small coarse fraction MENTOLOGIC SIGNALS: AN EXAMPLE OF A amount is primarily composed of sediment aggre- GLACIALIINTERGLACIAL TRANSITION gates and terrigenous material. The variations in the isotopic, geochemical and Starting at ca. 80 cm (13.5 ka), the TOC content sedimentological signals indicate climatic trends rises steadily. Drastically reduced 61X0conditions that are important for correlating the cores and for correlate with increased terrigenous input, as is determining sedimentological and paleoceano- typical for warnzirz,y phnses, i.e. glacier melting graphical trends (for the complete data set and dia- events (Fig. 6). A short period of light oxygen iso- grams of all investigated parameters, see Antonow, tope values is acconlpanied by increased terrige- 1995). An example of these variations and their nous input (IRD), decreased biogeiiic input and a implications is shown in the topmost metre of the decrease in the CaC03 content. The rest of the core sediment profile from Core PSI878 (Fig. 6). Due to consists of large proportions of biogenic compn- the very similar trends for the isotopic and sedi- nents (including benthic species). The increased mentological signals in the short box core (GKG) TOC content, the high 813c conditions and the and the long kastenlot core (KAL), the two can be moderate 6180 values are typical for irzteylncinl combined to provide a single normalised core. conditions. Benthic filter-feeders (C. zo~rellerstofi) The lower portion of the section (100-85 cm, also appear. A layer of tephra (volcanic glass) 16.3-14.9 ka) shows "heavy" 6l80 conditions (Fig. appears at 56-45 cm (9 ka); its coarse grains lead to 6). Very low TOC and low carbonate contents indi- losses in determining the water content (Fig. 6). cate dncial conditions. The coarse fraction (>63

.- .....

,

E Pelagic sedinwl~ts

D Upper la~ilinnludunit (Clay -sill)

,111 C Rippled unit, w:ivy 01' convolulr: I~III~II:~~(sill - s:llld)

IVI I3 Lower laniinatrd 11ni1(sancl)

A G~.:ldcdunit (lining apw;lrd. 1111 gravel -sand)

2x1

Volcanic Aslirs

1VI

.I,#, Turbidilc Fncics

Figure 7. Lithology of the long sediment cores (kastencores). cycles exhibit an influence of turbidity cur- rents (turbidite units after Bouma 1962). Vesterisbanken (central Greenland Sea) during the past 250,000 years 109

SEDIMENTARY PALEOENVIRONMENT successions of coarse-grained (>63 pn) black to The Transpolar Drift transports large amounts of fine-grained (<63 pm) olive green volcanic (basalt- sediment-laden sea ice mainly from the Eurasian ic) ash layers. These ash layers are composed of continental shelves across the North Pole to the glass and rock fragments together with some Fram Strait, supplying mainly fine-grained ice-raf- crystalline components and can be classified as ted detrit~~s(IRD) to the northern North Atlantic volcanoclastic turbidites (Fig. 7). They indicate an Ocean floor (e.g. Wollenburg, 1993). During glacial intensive Late Quaternary volcanic eruption phase eras, another source of IRD was iceberg calving of since about 105 ka (Fig. 5). More detailed core the Greenland, Scandinavian and Barents Sea Ice descriptions of the cores are given in Thiede & Sheets. Within the Greenland Sea, IRD is currently Hempel (1991) and Antonow (1995). distributed by the southward-flowing East Sedimentation in the area of Vesterisbanken Greenland C~~rrentand the cyclonic Greenland Sea was influenced by near-bottom currents (Michels, Gyre, the Jan Mayen Current and the Return 1994; Antonow, 1995). There is a basic difference Atlantic Current. IRD abundance in Pleistocene between gravitative and thermohaline current sediments documents the persistent climatic con- mechanisms in the marine environment. trasts of severe polar conditions and iceberg mel- Gravitative current mechanisms create turbidites ting in parts of the Greenland Sea and in the Fram (Bouma, 1962) while thermohaline mechanisms Strait (Bischof, 1990; Spielhagen, 1991; Baumann et induce bottom currents that follow bathymetric nl., 1994; Goldschmidt, 1994, fliis voliinie; Nam et nl., contours, leading to the deposition of contourites 1995). The input of IRD is discussed in detail in the (Hollister & Heezen, 1972). following section. Deposits of both current mechanisms are evi- All sediment cores are made up of grey, green dent in the study area (Antonow, 1995). The lami- and brown lithogenic hemipelagic muds in the nated sections of the investigated cores were usual- grain size range of clay to sandy silty clay. The ly interpreted as distal turbidites. But some of the cores closest to Vesterisbanken are interrupted by investigated sediment characteristics (sorting, bio-

Meall Grain Sixc Mean Grain Sizc Mean Grail1 Size (~111) Cm) @111) O 50 IM) 150 0 50 100 150 0 50 IUO I50 O 0 0 0 0

5 5 20 20 10 10 50 15 15 40 40

20 20 60 60 Ion 25 25

30 30 80 80

35 35 150 100 100 40 40

45 45 I20 120 200 50 50 140 140 55 55

60 60 160 160 250 I 2 3 I 2 3 I 2 3 Age Agc AB 0ka) Sorti~ig (ka) Sorting Sorting (k;~)

- Mean Grain Sizc (pm) - Sorting 1 Intervals of hotto111currents I

Figure 8. Sorting of sediments of the Vesterisbanken area as an indicator of bottom current influence during the past 250,000 years. Antonow, Goldschmidt and Erlenkeuser

Cibicides Cibicidoides Onilnnniis turbation, thickness, normal and inverse grading, ln~f~~i~~s wueilers~nrji ~~rnbormrrrs lamination) are typical for contour currents ~~mln-*l ~p"ln%l 1zrrln.l 1 (Bouma & Hollister, 1973; Hesse & Chough, 1980; 0246810I205 1015m~(O12.74 5 Chough & Hesse, 1985). Most episodic current events (turbidites) occur during cooler phases (Stage boundary 8/7, Events 6.4,6.2,4.2 and 2.2) or they occur during meltwater pulses during the entire Stage 3 (Figs. 5 and 8). There are intervals where the sediments are very well sorted (sorting values <2; see Friedman and Sanders, 1978). These well-sorted layers generally correlate with smaller grain sizes of the silt fraction (>lo ~un,Fig. 8). During warmer times, thermoha- line circulation is stable, resulting in an intense cur- rent influence by contour currents (Event 5.5, Figs. 5, 8). Planktic and/or benthic fossils are rare in Greenland Sea sediments during glacial maxima (Baumann, 1990; Birgisdottir, 1991; Struck, 1992; Bauch, 1993; Jiinger, 1994). But the local abundance of coccolithophorids documents that biogenic sedi- A a I\"I#,IW ment input was never completely interrupted (lull s1itgw during glacial eras (Gard, 1987, 1988; Gard & Backman, 1990). Figure 9. Occurrence of C. lobntulz~s,C. zvrrellerstorfi and 0. umbonatils in the sediments of the Vesterisbanken area The carbonate production is related to the pro- (standard profile) during the past 250,000 years. perties of water masses and sea-ice conditions (Kellogg, 1976,1980). The Vesterisbanken area was very often controlled by the cold East Greenland Current and the sea-ice coverage witlun the last continental margin (see Nam et al., 1995; Stein ct al., 250 ky. This resulted in a dominant occurrence of 1996). Thus, the variation in calcium carbonate monospecific N. pach!jderma sin. throughout time content reflects strong climate-induced changes in as also recorded at sites along the East Greenland the inflow of warm Atlantic surface water masses

AN AH AN ,\ R AH 'l'rrri~e#~ou~ AH V~~lc;ln~vgcnlc All lil~~g~~~lc IlaIk Sallrm~nl Scdinwnl> 63 pnl Ilulk-GIGO 'l'otal Or~~nieCarlnn Carn~~"mtm#\ C~mqx>ttv~~l\ C~ntl~nne~~lx lg*cn~~.li:.l I ~g.s,wl.k? .I I ~g*at+ky.l I.1 1g*cnr2-~?.81 ~g*nn.l.ky.It IE*C~I.~.~Y.~I 1x*cmw2 -kv-ll

II 2 4 L X 10120 2 4 6 X (I 02 OAO.6 0.U I 0 0.113 0.06 0.090 0.4 0s 12 0 2 4 6 8 0 11.3 ILL ID!)

50

IM

150

250 hgeIL;~) l~~t~qw s1;,gw

Figure 10. Accum~ilationrates (AR) of bulk sediment, the coarse fraction, the bulk carbonate and TOC contents, and AR of the terrigenous, volcanogenic and biogenic components of the Vesterisbanken region during the past 250,000 years. Vesterisbanken (central Greenland Sea) during the past 250,000 years

and in paleoproductivity near the sea-ice edge (see (1990) in Fig. 11) were caused by strong melting Baumann et al., 1993). events of the Greenland Ice Sheet (Vogelsang, 1990; Of course, primary production is much higher Weinelt, 1993; Junger, 1994; Nam et al., 1995; Stein during interglacial periods (Fig. 9). Suspension-fee- et al., 1996). ding epifauna (C, wriellerstor/i , 0. umbonatus), an Within the last 250,000 years, the paleoceano- indicator of bottom currents (Struck, 1992), domi- graphic development of the central Greenland Sea nate in parts of Stages 7, 5 and 1 (Fig. 9). The pro- was driven by the contrasting water masses of the portion of total organic carbon (0.2 to 0.3 weight % Polar and the Arctic (influenced by Atlantic water of bulk sediment) indicates "normal" (hemi)pela- masses) Domains as well as the resulting surface gic sedimentation in large parts of the abyssal water circulation pattern (cyclonic Greenland Sea Greenland Sea (Junger, 1994; Antonow, 1995). The Gyre). A nearly continuous IRD signal in the sedi- accumulation rates indicate varying depositional ments, even during moderate climate phases (such patterns in the Vesterisbanken area over the last as Events 7.5, 7.3 and 7.1), is a sign of uninterrup- 250 ky (Fig. 10). ted iceberg melting, underlining the relative proxi- mity (and thus the climatic contrast) to the Polar PALEOCEANOGRAPHY OF THE VESTERIS- Domain within the Greenland Sea. Phases of BANKEN AREA IN COMPARISION TO strong ice cover in the Greenland Sea during Event OTHER REGIONS OF THE GREENLAND- 7.4 are also seen in the . Weakened ICELAND-NORWEGIAN SEAS circulation is indicated by low carbonate values in Oxygen isotope signals in the Greenland-Iceland- Norwegian Sea sediments (Henrich et a/., 1989). Norwegian (GIN) Seas primarily display the inter- In Stage 6 (except for Event 6.5), broad expan- change of temperate Atlantic waters with interme- ses of the Norwegian Sea were under a permanent diate water masses of the Arctic Domain. Isotope sea ice cover; isotope values (N. pnch?/derrnn sin.) signals in the Greenland Sea are less affected by indicate a significant shift to "heavy" values (Fig. temperature because they are mainly influenced by 11). In the Greenland and Iceland Seas, on the other Polar water masses. Thus, the 6180 variations in hand, the times of permanent sea ice cover were the Greenland Sea mainly reflect variations in the shorter. Large amounts of meltwater flowed into global ice volume (ice effect). Extremely low 6180 these seas from the instable Greenland Ice Sheet values which strongly diverge from the ice effect during Events 6.5 and 6.3 (Jiinger, 1994; Antonow, ("mean seawater" signal according to Vogelsang 1995,1996; Nam et al., 1995).

hlen~~scnwalcr Earl Cree~~land Crcenln~ldScn Creenlnnd Scil Crce~~l~ntlSen Nor\rcgin~~.%I lccla~~dSC:I l",,,*p conlincnlnl nlargin slnndard prnlile L1U3 s,,,p. PSI73l Vcslcrishn~iliet~ PS 1736 PSI%Wl Z3MZ atno [%,I ,,s. pnll alXolv,lVS. PDU i)lxo[%,I v~. PI^ also [%,IVS. PDU ;)lXo[%,I Vs. I~DII alXo [%,IVS.. PDIS alXo [%,IVS. IIDIS 8.5 1.11116 #).41.415 1 I .I 2 6 4 J 2 5 I J 2 5 4 3 2 5 I J 2 5 1 .I 2 t,

511

It*)

131

IN1

29) (k;,, \'t%<.l%:,nxII'J'M)

Figure 11. Oxygen isotope records of Vesterisbanken sediments in comparison to isotope records (N.pnclryderrrin sirr.) of cores from the Norwegian-GreenlandSea. Mean oxygen isotope ratios of seawater (ice effect) in the NGS related to Present after Vogelsang (1990). 112 Antonow, Goldschmidt and Erlenkeuser

At the start of Termination 11, isotopically F180, see Fig. 11) during Event 4.2 indicates contra- "light" meltwater again covered large areas of the sting oceanographic conditions within the central GIN Seas (Fig. 11). The meltwater pulses were tre- Greenland Sea (cores PSI900 and PS1736) and the mendous throughout these seas, leading to inter- Vesterisbanken area (Junger, 1994; Antonow, 1995, ruptions in convection (see Haake & Pflaumann, 1996). 1989). Isotopically "light" Atlantic Surface Water A reduced range of species (N. padz!/der~~lais masses can be easily traced, especially in the the only planktic species present) indicates severe Norwegian and Iceland Seas (Fig. 11). The inten- paleo-ecologic conditions, as was the case through- sive thermohaline surface water circulation in- out the GIN Seas during Stage 3 (see Kellogg, 1976, creased the influence of temperate water masses in 1977, 1980; Bauch, 1993). The benthic fauna was the Vesterisbanken area. The cores in the centre of also reduced: only 0. lrinborzatrrs was present in sig- the Greenland Sea Basin indicate very high rates of nificant amounts at the beginning of Stage 3. The deep water renewal starting at Event 5.51; this is isotope signals (Fig. 11) give evidence of early mel- demonstrated by highly benthic 6I3c conditions ting events that have been triggered by the weak (0. rrmbo~~atlrs;Jiinger, 1994).Based on bentluc oxy- inflow of warmer Atlantic water masses, as is gen isotope data, Birgisdottir (1991) and Bauch recorded at the East Greenland continental margin (1993) postulated a very low influence of Polar (PS1730) as well as in the central Greenland Sea water masses in the Iceland Sea. As Stage 5 con- (cores PS1736, PS1900). Thus Stage 3, with its com- tinued, the Norwegian Sea water masses showed mon, strong meltwater pulses and the constant relatively stable Atlantic Water influence, while in alternation of ice-free surface waters and nearly the central Greenland Sea alternating influences of complete ice cover, is an extremely variable, clima- water masses separated by the Arctic Front have tically sensitive period in the Greenland Sea (Fig. been reported (Henrich et nl., 1989, 1992; Junger, 11). During Stages 2 and 3 the sedimentation rates 1994; Antonow, 1995, 1996). are much higher than those within other time inter- In Stage 4, high rates of terrigenous IRD input vals (Table 4). Winnowing and redeposition of fine are seen not only in the Vesterisbanken area, but sediment particles along the Greenland continental also in the Norwegian Sea (Henrich et al., 1989), the shelf, the slope and the adjacent Greenland basin Iceland Sea (Birgisdottir, 1991), the Greenland Sea by the EGC are likely to have occured (see also (Hamich, 1991; Jiinger, 1994), and the Fram Strait Mienert et a]., 1992; Nam et al., 1995). (Spielhagen, 1991; Hebbeln, 1992). This fact is in A drastic increase in the carbonate content contrast to the situation near the east Greenland (spread of biogenic production) at the beginning of continental margin, where an IRD signal is Termination I can be seen in sediments througho~~t distinctly reduced (Nam et al., 1995; Stein et al., the GIN Seas (see Blaume, 1992; Henrich ct al., 1996).The significant isotopic shift (by up to 0.25%0 1989; Koq Karpuz & Jansen, 1992). As in earlier

Figure 12. Ice-rafted debris (>I25 pm) of the Vesterisbanken sediments in comparison with the extent of the Barents Sea and Scandinavian ice sheets. Vesterisbanken (central Greenland Sea) during the past 250,000 years

episodes, the melting of the continental ice sheets Spielhagen (1991) interpreted the flow and ice resulted in meltwater input. As in the Norwegian cover of the EGC as stable over (at least) the last Sea (Blaume, 1992), turbidity currents occurred at 200,000 years. great depths in the Greenland Sea (Mienert et al., Mangerud & Svendsen (1992) used a continen- 1993; Antonow, 1995). tal litho- and biostratigraphy for the last 140,000 years to reconstruct glacier oscillations in the MARINE VESTERISBANKEN RECORDS VS. Svalbard/Barents Sea area. Baumann et al. (1994) TERRESTRIAL DATA: LAND-SEA CORRELA- compared these terrestrial data with the marine TION IRD signal of cores from the northern Greenland IRD records vs. oscillations of glaciers neighbo- Sea and found that the terrestrial data agreed well ring the northern North Atlantic with the marine records. Variations in the input of terrigenous material (20- The IRD signals of the Vesterisbanken sedi- 80 grain % IRD) can be seen in the Vesterisbanken ments correlate well with both the data from sediment cores (Fig. 12). The terrigenous particles Mangerud & Svendsen (1992) and with the recon- (>I25 pm) consist mainly of monocrystalline struction of fluctuations of the Scandinavian Ice (quartz, feldspar) and polycrystalline rock frag- Sheet of Mangerud (1991). The recent surface cir- ments (slate, volcanic fragments, sandstones and culation pattern (Norwegian Current - Atlantic siltstones). Although their source areas cannot be Return Current - EGC/Jan Mayen Current; Fig. 1) precisely determined, it is likely that most of the and the results of Spielhagen (1991) regarding the rock fragments (excepting the volcanic fragments) longevity of the EGC show that icebergs from the come from , Svalbard and the Barents Sea Norway/Svalbard area could have reached the area (see Bischof, 1990; Baumann et al., 1994; central Greenland Sea in the last 250 ky. Nearly Goldscl~midt,1994, this vol~rnle).IRD originating in every time that the corztinerztal glaciers had reached Greenland is generally carried southwards by the beyond the continental shelf or were retreating East Greenland Current (EGC) and would thus from it, the irlnrine IRD record shows a peak (Fig. probably not be found in the study area. 12). Glaciers become instable when passing the

Mid-.Iuly Insolation GRIP Ice Core Vcsterisbanke~~ VOSTOK Ice Corc Isotope 6S0N [\v/m 3'80 [%o] VS.PDB alXO[%a] VS.PDI3 aD [%,o] Stages 370 420 470 520 45 -40 -35 -30 5 4 3 2-500 -475 -450 -425 400

(Mnpnfter Mnyewski el nl. I')OJ)

A gc Ucrgcr IY: I)a~isgaartl ct al. (1993) Jo~lzclel al. (Im) Lnulrc II1)')I) (IYX7,19891

Figure 13. The oxygen isotope signal (N. pncl~~crnmsin.) of the Vesterisbanken standard profile correlates with iso- tope data from Greenland (GRIP)and Antarctica (Vostok). Antonow, Goldschmidt and Erlenkeuser

continental shelf, resulting in iceberg calving (see In general, the oxygen isotope data of the n-ta- Heinrich, 1988), while glacier retreats with intense rine Vesterisbanken area are in good agreement iceberg calving indicate the beginning of warming with the terrestrial Greenland Ice Sheet isotopic phases (see Bauma~et al., 1994; Goldschmidt, record (Fig. 13). Extreme meltwater input reflects 1995; Nam et al., 1995). In both cases, melting ice- climatic fluctuations near Vesterisbanken during bergs releasing their sediment load indicate tl-te isotope Stages 6 and 3. Deviations of the climate extension of glaciers beyond the continental shelf records of the different sites are caused by regional or tl-te retreat therefrom. Although the marine sig- influences within the Norwegian-Greenland Sea. nal varies from core to core, it is much more Orbital forcing of the documented Greenlai-td detailed and continous than the terrestrial signal. Sea climate records is evident. "Light" isotope However, IRD peaks in the study area do not values result from more intensive insolation have to indicate the existence of glaciers (Fig. 12). (Berger & Loutre, 1991) and coincide wit11 the At certain times during Stages 5 and 3, the IRD sig- warm substages 7.5, 7.3, 7.1, 5.5, 5.1 and nal must have been modified by iceberg-covered Termination I (Fig. 13). water masses of the Jan Mayen Current branching The isotope record of tl-te Vesterisbanken sedi- off of the EGC. Locnl melting processes in the ments corresponds better with the data froin the Vesterisbanken area may also play a role. Vostok ice core (Antarctica) than those derived from the Greenland ice core (GRIP), wl-tich is com- Isotope signals (N. pachyderma) vs. ice core data plicated by internal deformation of the ice older Data from the Greenland Ice Core Project (GRIP) than ca. 100 ka towards the glacier's basal zone. document higl-t-resolution series of temperature The sediments around Vesterisbanken thus arcluve oscillations during the last 250,000 years the globnl climatic record. The thermohaline circu- (Dansgaard et nl., 1993). These so-called lation is strongly col-tnected to climatic variations, "Dansgaard-Oeschger cycles" are caused by cl~an- which affects a rapid translation of the Polar Front ges in the 6180 isotope ratios of the air masses and is well documented by the changing influence above the Greenland ice sheet and correlate well 01 different water masses. The sites around wit11 paleosurface water mass temperatures deter- Vesterisbanken are situated in the narrow "battle- mined with N. pnchyderiiza of North Atlantic sedi- field" of oceanic frontiers and this makes it pos- ment cores for the last 90,000 years (Bond et al., sible to correlate their isotope signals and the l-tigl-t- 1993). resolution ice core records to oceanic circ~~lation

0-lsotnlr Slreligr;~yli! Scasolrul Mcllwstcr Icc-lal'liltg Uuttn~tiCurrcnt SCBOIB~IB~I~ Sea Levcl CIt;~s~ge S,.,LC. 3180 [?,I bs. I'l)U Opn \\'alcr 1)ischsrp lnle~~sity Vnlcso~ic,\cti~,iI> ,1111

tiia~t~lybascd on bascd on ha%d on based on bawl ott c'I1:~ppcll& cnrbonarr conrrnr isotope rrcol.ds IRD-signal current indicators reph~.aOL‘CLI~~CIICE: Shi~ckletoti( 108hl

Figure 14. Synoptic overview of the environmental conditions of sedimentation in the Vesterisbanken area tor the past 250,000 years. Sea level changes after Chappel & Shackleton (1986). Vesterisbanken (central Greenland Sea) during the past 250,000 years 115

(see Zahn, 1994). Research Center for Marine Geosciences and at the Special Research Project SFB 313 at Kiel University CONCLUSIONS encouraged and helped us: Britta Jiing-er, A general synthesis of the important sedimentation Heidemarie Kassens, Henning Bauch, Helmut events and paleoceanographic determinations for Beese, Frank Blaume, Andre Freiwald, Christian the time since the end of Stage 8 is shown in Figure Hass, Jens Holemann, Riidiger Henrich, Harald 14. The studies carried out on Vesterisbanken sedi- Hommers, Klas Lackschewitz, Klaus Michels, ments indicate the following: Jiirgen Mienert, Stefan Nees, Jan Rumohr, Robert 1 The paleo-envirom~entwithin the studied time Spielhagen, Jorn Thiede, Thomas Wagner, Hans- interval is very variable. The sediments result from Joachim Wallrabe-Adanls, and Rainer Zahn. varying formation and transport processes: sedi- This paper benefitted from the review of Riidiger mentation patterns dominated by terrigenous, bio- Stein and the critical comments of Christiai~Hass. genic and volcanogenic input alternate. Changes in Thanks to Bernhard Fiirst for technical assistance the lithofacies usually occur due to climatic varia- in the preparation of the manuscript. Financial tions; these changes are related to transitions bet- support was given by the German Research ween the oxygen isotope stages and substages. Foundation sponsoring the Special Research Large-scale meltwater events occurred during Project SFB 313. This is SFB 313 publication num- Terminations (111, I1 and I) and during fluctuations ber 304. during the entire Stages 3 and 2. 2. Hen~ipelagicsedimentation in the Vesteris- REFERENCES banken area was mainly current-controlled and Aagaard, K., Swift, J. & Carmack, E. 1985. Thermohaline indicates variable ice cover. In addition to biogeni- c~rculationin the Arctic Mediterranean Seas. \or11 1101 of cally dominated sedimentation patterns that were Geopllysical Resenrcll, 90,4833-4846. relatively constant over time, there were also time Antonow, M. 1995. Sedimentationsmuster Lun den Vesteris Seamount (zentrale Griinlandsee) in den letz- intervals in which episodic current-induced near- ten 250.000 Jahren. GEOMAR Report, 44, 1-127. bottom particle transport was important. Particle Antonow, M. 1996. S ;it uartsre Sedimentation im deposition by suspension c~~rrentsdominate over Gebiet roil Vesteris%ar%m: Paldozearrugri~phi,rhr contour current processes. und klimatische Ableitungen fiir die zentrale Cronlandsee. Zcr~tralblatt fiil- Geologic ILII~ 3. Turbiditic volcanic sediments indicate episodes l'aliiontolo~ie,Teil 1 (1/2), 7-19. of volcanic activity at different times. Ca. 105 ka, Atterberg, A. 1912. Die mechanische Bodenanalyse und intense volcai~icactivity occurred at Vesteris- die Klassifikation der Biiden Mittelschwedens. banken. Eruptions occurred mainly when the sea O~terl~ntiol~oleMitteil~~n~tv~ fiir Bodenkunde, 1-314. level was high (Fig. 14). No volcanic activity took Bard, E. 1988. Correction of accelerator mass spectrome- try 14c ages measured in planktonic foraminifera: place since the beginning of the Holocene. It Paleoceanographic implications. lJ~nl~oc~~~r~o~ra~~l~y,3, nppenrs that the Vesterisbanken volcanic episodes 635-645. were coupled to warming phases (crustal rebound Bard, E., Hamelin, B., Fairbanks, R.C. & Zindler, A. lC)C)O. due to ice sheets disappearing from neighboring Calibration of the 14c timescale over the past 30,001) years using mass spectrometric U-Th ages from land masses?). Barbados corals. Nat~~w,345, 241-213. 4. Despite the modification of the depositional Bauch, H. 1993. Planktische Foraminiteren in1 environment caused by the existence of Vesteris- Europaischen Nordmeer - ihre Bede~~tungfiir die banken, general paleoclimatic information can be alao-ozeanographisclie Interpretation wahrend der gleaned from the sediments. The sedimentological Ltzten 600.000 JaIire Bericlrt~ .,IS dcra So,id~~:fill~ sclll~l?gsbt~reicl313, 40, 1-108. and paleoceanographical data from tlus part of the Baumann, K.-H. 1990. Veranderlichkeiten dcr central Greenland Sea can be integrated into the Coccolithenflora des Euro aischen Nordmeeres im pattern of Late Quaternary sedimentation of the ,~ingquartdr.Berichh~ ails $ni ~~~~~~~IEC~IIII~~S~I~I~L~~C~I northern North Atlantic and present a picture of 373,22,1-146. globally controlled climate changes. Baumann, K.-H., Lackschewitz, K.S., Spielha Ten, R.F. & Henrich, R. 1993. Reflection of continentnkice sheets 5. The correlation between terrestrial glacier oscil- in late Quaternary sediments from the Nordic Seas. lations and marine IRD signals is very good. The Zcl~tralblattfiir ~iologie~lnd Paliiontolo~yic~, Teil 1, 718, climate record in the marine deposits using oxygen 897-9 12. isotope data (N. pnclzyderrizn sin.) has a very good Baumann, K.-H., Lackschewitz, K.S., Mangerud, I., Spielhagen, R.F., Wolf-Welling, T.C.W., Henrich, R. & resolution in the Vesterisbanken area and agrees Kassens, H. 1994. Reflection of Scandinavian ice sheet well with data from high-resolution ice cores from fluctuations in Norwegian Sea sediments during the polar landmasses. last 150,000 years. Q~lnter~~nnjResearch, 43, 185-197. Be, A.W. 1977. An ecolo ical zoographic and taxonomic ACKNOWLEDGEMENTS review of recent pla&tonk foraminifera. in: Ramsay, A.T.S. (ed.), Ocerlrlir 11licropalec7ntology, 1-100, Many colleages and friends from the Geomar Academic Press, London. 116 Antonow, Goldschmidt and Erlenkeuser

Beryer, A & Loutre, M.F. 1991. Insolation values for the c mate of the last 10 million years. @llntcri7nr!y Science Gard, G. 1987. Late Quaternary calcareous naru~ofossil Rcz?iezc),10, 297-317. biostratigraphy and sedimentation patterns: Frnni Birgisdottir, L. 1991. Die palao-ozeanographische Ent- Strait, Arctica. Pnleocennogrnpl~~y,5, 761-787. wicklung der Islandsee in den letzten 550.000 Iahren. Gard, G. 1988. Late Quaternary calcareous nannofossil Brricl~te011s ~PIIISo~~derfor~~I~~~i~gsbereicI~ 313, 34, 1-112. biocl~ronologyand paleo-oceanography of Arctic and Bischof, J. 1991. Dropstones im Europiiischen Nordmeer- Subarctic Seas. A4cdd. Stockholrirs LTI~Iv.Grol. Irrst., 275, Indikatoren fur Meeresstromunge~~in den letzten 1-45. 300.000 Iahren. Bericl~te 011s denr Sondl~~forscl~r~ngs-Gard, C. & Backman, J. 1990. Synthesis of Arctic and b~:miclr313, 30, 1-127. Sub-Arctic coccolith biochronology and history of Blaume, F. 1992. Hochakkumulationsgebiete am norwe- North Atlantic drift water influx during the last ~iscl~enKontinentalhang: Sedimentologische Ab- 500,000 years. 117: Bleil, U. & Tl~iede,J. (eds.); Gcoloyictil kilder Tc)pugraphie-gefulirtsl Strbmungsmuster. Ilistory of tlre Pol~7rOCL~I~S: Arctic vrrsrls Ailttlrctic, 417- Bcricl~tc011s rltnr Sor~tie~:forsclii~i~gsI~er~~iclr333, 36, 1-150. 436, Kluwer Academic Publishers, Dordrecht. Bond, G.C., Broecker, W., johnsen, S. & McManus, J. Goldschmidt, P.M. 1994. The ice-rafting history in tlie 1993. Correlations between climate records from Norwe~rian-Greenland Sea for the last two North Atlantic sediments and Greenland ice. Nntirrc, glacialfinter ~lac~alcycles. Berrcl~tc,nrrs ticrir Sorldrlfor- 365 (6442), 143-147. scl~~in~sli~~r~ic~113, 50, 1-103. Bo~~ma,A.H. 1962. Serliii~entolo,y!yoJsoir~r,fl!ysclr deposits. A Goldschmidt, P.M. 1995. Accum~~latic>nrates of coarse- ~rnplricn ~pranclito fncies inl~r/~retntion,168 pp, Elsevier, rained terrigenous sediment in the Norwegian- ~msterdam. ereenland Sea: Signals of continental glaciation. Bouma, A.H. & Hollister, C.D. 1973. Deep ocean basin Mnrir~eGeolog!y, 128, 137-151. sedimentation. In: Middleton, G.V. & Bounia, A.H. Goldschmidt, P.M. tllis i~olrrrrrt.The ice-r'ifting history of (eds.), Zlrbidites nnd deep i~lntt~.seiiir~~rr7tntioi7. SEPM coarse-grained sediments in the Norwe~ian- short course, Anaheim, 79-118. Greenland Sea for the last two glacial/intelekcid Bourke, R.H., Newton, J.L., Paquette, R.G. & Tuimicliffe, cycles. M.D. 1987. Circulation and water masses of the East Haake, F.W. & Pflaun~ann,U. 1989. Late Pleistocene tola- Greenland shelf. ]o1~ri7nlof Gec7pl1ysicnl Rescnrcll., 92, miniferal stratigraphy on the Voring Plateau, 6729-6740. Norwegian Sea. Borens, 18, 343-356. Chappel, J. & Shackleton, N.I. 1986. Oxygen isotopes and Hamicl~,A. 1991. Sedimentologische Untersuchungen sea level. Nntr~rr,324, 137-140. eines Kernes (GIK 21906-2) der Gr(i1ilandsee. unpubl. Chough, S.K. & Hesse, R. 1985. Contourites from Eirik diploma thesis, Univ. Kiel, 78 pp. ridge, south of Greenland. Sedinzentnry Geology, 41, Hebbeln, D. 1992. Weiclwelian glacial history of the 185-199. Svalbard area: correlating the marine and terrestrial Ciais, P., Petit, I.R., Jouzel, J., Lorius, C., Barkov, N.I., records. Borens, 21, 295-304. Lipenkov, V. & Nikolajev, V. 1992. Evidence for an Henrich, R., Kassens, H., Vogelsang, E. & Tliiede, J. 1989. early Holocene climatic optimum in the Antarctic Sedimentary facies of glacial-interglacial cycles in tlie deep ice-core record. Clinrnte Dyrmriiics, 6, 169-177. Norwegian Sea during the last 350 ka. Mnrinc GI-ology, Craig, H. 1957. Isotopic standards for carbon and oxygen 86, 283-319. and correction factors for mass-spectrometric anal sis Henrich, R. 1992. Beckenanalyse des Europiiischen of carbon dioxide. Gmclrir~iicnrind Cosrr~ocl~irnicnLtn, Nordmeeres: Pelagische und glaziomarine Sedime~it- 12, 1-133. einfliisse im Zeitraum 2.6 Ma bis Rezent. ~~npubl. Craig, H. & Gordon, L.I. 1965. Deuterium and ox gen-18 habil. thesis, Uruv. Kiel. variations in the ocean and the marine atmospkere. In: Hesse, R. & Chough, S.K. 1980. The Northwest Atlantic Tongiorgi, E. (ed.), Stnble isotopes in ocranogrnpllic strr- Mid-Ocean Channel of the Labrador Sea: 11. dics nrld paleoterrzpernt~~res.Third SPOLETO conference Deposition of parallel laminated levee-muds from the on Nuclear Geology, Consilio Nazionale della viscous sublayer of low density turbidity currents. Ricerche, Laboratorio di Geologia Nucleare, Pisa, 9- Srdin~oltology,27, 697-711. 130. FIollister, C.D. & Heezen, B.C. 1972. Geolo yic effects of Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, ocean bottom currents. In: Gordon, A.L. hd.), Stridi(zs D., Gundestrup, N.S., Hammer, C.U., Hvidber , C.S., in ph~sicnl oceni~o~yrayli~y- A tribute to Gcor

Raynaud, D., Petit, J.R., Barkov, N.I., Korotkevitch, Y.S. & Kotlyakov, V.M. 1989. The Antarctic climate Paquette, R.G., Bourke, R.H., Newton, J.F. & Perdue, W. over the late Glacial period. Qlmtertlnry Resernrch, 31, 1985. The East Greenland polar front in autumn. 135-150. Jo~~rnalof Geoplrysicnl Resenrch, 90,4866-4882. Junzer, B 1994. Tiefenwasseremeuerung in der Rudels, B. 1990. Haline convection in the Greenland Sea. ronlandsee wahrend der letzten 340.000 Jahre. GEO- Deep Sen Research, 37,1491-1551. MAR Report, 35,l-103. Samthein, M. 1971. Oberflachensedimente im Persischen Kellog , T.B. 1976. Late Quatemary climatic changes: Golf und Golf von Oman TI. Quantitative evignce from deep-sea cores of Norwegian and Kom onentenanalyse der Crobfraktion. ,,Meteor" Greenland Seas. In: Cline, R.M. & Hays, J.D. (eds.), ~orscf.-~r~ebi1.,C5, 1-113. Investigation of Late Quaternar aleoceano raphy and paleoclimatology. ~eologica~Liety of jinericn Samthein, M., Jansen, E., Amold, M., Duplessy, J.C., Meinoir, 145, 77-110. Erlenkeuser, H., Flat@ A, Veum, T., Vogelsang, E. & Weinelt, M.S. 1992. 833 0 ..time-slice reconstruction of Kellogg, T.B. 1977. Paleoclimatolog and paleo-oceano- meltwater anomalies at Termination I in the Nortl~ yphy of the Norwegian and Ereenland Seas: The Atlantic between 50 and 80°N. In: Bard, E. & Broecker, ast 450,000 years. Marine Micropnleoi~tol~~~y,2, 235- W.S. (eds.), TIie lost iieglncintion: nbsollite nrzd rndiocnrb- 249. oiz chronologies, 183-200, Springer, Berlin. Kellogg, T.B. 1980. Paleoclimatology and paleoceanogra- Shackleton, N.J. & Opdyke, N.D. 1973. Oxygen isotope phy of the Norwegian and Greenland seas: glacial- and paleomagnetic stratigraphy of Equatorial Pacific interglacial contrasts. Borens, 9,115-137. core V28-238: Oxygen isotope temperatures and ice Koc Karpuz, N. & Jansen, E. 1992. A high-resolution dia- volumes on a lo5 year and lo6 year scale. Ql~nterr~nry tom record of the last deglaciation from the SE Resenrclz, 3,39-55. Norwegian Sea: documentation of rapid climatic Shackleton, N.J. & Opdyke, N.D. 1976. Oxygen isotope changes. Poleocc.nr~ogmphy,7,499-520. and paleomabmetic stratigraphy of Pacific core V28- Koltermann, K.P. 1987. Die Tiefenzirkulation der 238 Late Pliocene to Latest Pleistocene. Grolo,yical Griinland-See als Folge des thermohalinen Systems Societ!/ of Anrericn Mmroir, 145, 449-464. des Europaischen Nordmeeres. unpubl. PhD thesis, Spielhagen, R.F. 1991. Die Eisdrift iii der FramstraBe Univ. Hamburg, 287 pp. wahrend der letzten 200.000 Jahre. GEOMAR Rt~port, Krumbein, W.C. 1936. Application of logarithmic 4,l-133. moments to size frequency distributions of sediments. Spielhagen, R.F. & Thiede, J. 1994. Late Quatemar than- ]oilrnnl of Sedinlentnry Petrology, 6,35-47. es in the Arctic ocean ice cover. Beric&e zrlr Mangerud, 1.1991. The last ice age in Scandinavia. Strine, ~o~n~rsdirng.144, 101-105. 34,15-30. Staudinger, G., Hangl, M. & Pechtl, P. 1986. Quick optical Mangen~d,J. & Svendsen, JI. 1992. The last interglacial- measurement of particle distribution in a sedimentati- lacial period on Spitsbergen, Svalbard. Q~mternnry on apparatus. Pnrticle Chnmcteristics, 3, 158-162. tCieIfce RFD~~'ZUS,11, 633-664. Stein, R., Grobe, H., Hubberten, H., Marienfeld, P. & Martinson, D.G., Pisias, N.G., Hays, J.D., Imbrie, J., Nam, S. 1993. Latest Pleistocene to Holocene changes Moore, T.C. & Shackleton, N.J. 1987. Age dating and in glaciomarine sedimentation in Scoresby Sund and the orbital theory of the Ice Ages: Development of a along the adjacent East Greenland Continental Idgh-resolution 0 to 300,000 years chronostratigraphy. Margin: Preliminary results. Geo-Mnrii~eLetters, 13, 9- Qrinternnry Resenrch, 27, 1-29. 16. Mayewski, P.A., Wumkes, M., Klinck, J., Twickler, M.S., Stein, R., Nam, S., Grobe, H. & Hubberten, H. 1996. Late Putscher, J.S., Ta lor, K.C., Cow, A.J., Waddington, Quaternary glacial history and short-term ice-rafted E.D., Alley, R.B., 2~bb, J.E., Grootes, P.M., Meese, D.A., debris fluctuations 1 the East Greenland conti- Ram, M., Whalen, M. & Wilson, A.T. 1994. Record nental margin. In: azt8rews, J.T., Austin, W.E.N., Drilling Depth Struck in Greenland. EOS, 75, 113-124. Bergsten, H. & Jennin s, A E (eds.), Late Quatemar); paleoceanography of i%e norih Atlantic margins, 135- Michels, K. 1994. Striimunessortierune<, auartarer ~rdimentedes ~ur0~6ischc;Kordrneeres: ' Analvse 151, Geologicnl Society Specin1 Plrblicntion, 111, London. Sinkcrscl~\vindi~keitb-Vertriluncen.fl,~rirlr/~~ nrrs Stein, R., Nam, S., Schubert, C., Vogt, C., Fiitterer, D. & dern ~onde"rforsch~rl1gs~ereic/l313, 55, ly127. Heinemeier, 1. 1994. The last deglaciation event in the Mienert, J., Kenyon, N.H., Thiede, J. & Hollender, F.-J. eastern central Arctic Ocean. Sc~eizce,264, 692-696. 1993. Polar continental margins: studies off East Struck, U. 1992. Zur Palao-Okologie benthischer Greenland. EOS, 74, 225. Foraminiferen im Europaischen Nordmeer wahrend Mienert, J., Andrews, J.T. & Milliman, J.D. 1992. The East der letzten 600.000 Jahre. Bericllte nr~s derir Greenland continental margin (65"N) since the last Sonderforschlrizgsl~ereic/l313, 38, 1-89. deglaciation: Changes in seafloor properties and Syvitski, J.P.M. 1991. Principles, niphlods, nnd opplicntiorz cq' ocean circulation. Mnriize G~ology,106, 217-238. pnrticle size ni~nlysis.368 pp, Cambridge University Nam, S.I. 1996. Late Quaternary glacial/interglacial Press, Cambridge. changes in paleoclimate and paleoceanographic circu- Thiede, J. & Hempel, G. (eds.) 1991. Die Expedition ARK- lation along the East Green and continental margin. TIS-VII/l mit FS ,,Polarstern" 1990. Berichte ziir unpubl. PhD thesis, Univ. Bremen. Polnrforsclz~~rzg,80, 1-137. Nam, S.T., Stein, R., Grobe, H. & Hubberten, H. 1995. Thiede, J., Suess, E. & Muller, P.J. 1982. Late Quatemary Late Quaternary glacial-interglacial changes in sedi- fluxes of major sediment components to the sea floor ment composition at the East Greenland continental of the northwest African continental slope. 111: \I. Rad, margin and their paleoceanographic implications. U. et nl. (eds.), Geolo~'~jof the Nortlzruesf Afi.icni1 coilti- Mniire Geology, 122, 243-262. neiztnl nlnrgin, 605-631; Springer, Berlin. Nowaczyk, N.R. & Antonow, M. in press. Magneto- Thies, A. 1991. Die Benthos-Foraminiferen im Euro- stratigraphy of Greenland Sea Sediments I - pischen Nordmeer. B~riclite nus dein Sondtv:for- Identification of the Mono Lake, Laschamp and Biwa sdilrngsbereidr313, 31, 1-73. I geomapetic polarity events. Geopllyslcnl Jolrrnnl Van Andel, T.H., Heath, G.R. & Moore, T.C. 1975. 118 Antonow, Goldschmidt and Erlenkeuser

Cenozoic history and aleoceanography of the central letzten 60.000 Jahre - Hinweise aus stabile11 Isotopen. Eqllatoriai Pacific. ~ca~o~icnlSocirty ofA~rriricnMnnoii, Bciiclite nlis den1 So~~dcifi~rschrr~~~sb~~r~?icI~313, 41, 1-106. 143, 134. Wi~m,K., Samthein, M. & Erlenke~~sel;H. 1901. 6180 Vinje, T. & Finnekasa, 0.1986. The ice transport throu ~h stratigra hy and chronology of Kiel sediment cores the Fram Strait. Norsk Wlniinstit~ittSkriftrr, 186, 1-3b. from tEe East Atlantic i