The Quaternary record of eastern Svalbard - an overview JON Y. LANDVIK, CHRISTIAN HJOKT. JAN MANGERUD, PER MOLLER and OTTO SALVIGSEN Landvik, J . Y . , Hjort, C., Mangerud. J.. M d l e r , P. & Salvigsen. 0. 1995: The Quaternary record of eastern Svalbard - an overview. P o / a r Research / 4 ( 2 ) , Y5-103. The eastern part of the Svalbard archipelago and the adjacent areas of the Barents Sea were subject to extensive erosion during the Late Weichselian glaciation. Small remnants of older sediment successions have been preserved on Edgeoya, whereas a more complete succession on Kongsoya contains sediments from two different ice-free periods, both probably older than the Early Weichselian. Ice movemeht indicators in the region suggest that the Late Weichselian ice radiated from a centre east of Kong Karls Land. On Bj~rnOya, on the edge of the Barents Shelf, the lack of raised shorelines o r glacial striae from the east indicates that the western parts of the ice sheet were thin during the Late Weichselian. The deglaciation of Edgeoya and Barentsoya occurred ca 10,300 BP as a response to calving of the marine-based portion of the ice sheet. Atlantic water, which does not much influence the coasts of eastern Svalbard today, penetrated the northwestern Barents Sea shortly after the deglaciation. At that time, the coastal environment was characterised by extensive longshore sediment transport and deposition of spits at the mouths of shallow palaeo-fjords. J o n Y . L u n d o i k . The University Courses o n Svalbard ( U N I S ) , P . O . BOX 156. N-9170 Longyearbyen, N o r w a y ; Christian Hjort and Per Moller, Department of Quaternary G e o l o g y . Lund U n i ~ ~ e r s i t y , Solvegatan 13, S-223 62 L u n d , Sweden; Jan Mangerud, Department of Geology, University of Bergen, Alkgr. 4 1 , N - 5007 Bergen, N o r w a y ; O t t o Saloigsen, Norwegian Polar Institute, P . O. B o r 5072 Majorsnra, N-0301 Oslo, N o r w a y . Introduction The Quaternary geology of the Svalbard archi- pelago (Fig. 1) has been the subject of extensive research during the last three decades. T h e inves- tigations have concentrated in particular o n the largest island, Spitsbergen (Fig. 2), and models have been developed for the Late Quaternary glaciations (Boulton 1979; Miller 1982; Landvik e t al. 1992a; Mangerud & Svendsen 1992), glacio- isostatic rebound (Schytt e t al. 1968; Forman 1990; Forman et al. 1995) and the Holocene marine conditions of both Svalbard and the adjac- e n t seas (Salvigsen et al. 1992). However. due to the more difficult accessibility, less progress has been made in understanding the Quaternary geo- logy of the eastern islands of the archipelago such as Edgeoya, Barentsaya and Kong Karls Land (Fig. 2). These islands, however, are critical for the understanding of the geology of the northern Barents Sea, and their geological history is fun- damental for t h e development of regional models. A s a result of both the P O N A M programme (Polar North Atlantic Margins. Late Cenozoic Evolution) and hydrocarbon exploration in the Barents Sea, renewed interest has been expressed in the Quaternary geology of the sea floor and the surrounding land mass. Thus, the eastern islands of the Svalbard archipelago can now con- tribute towards an integrated understanding of the Quaternary development along a profile from the central Barents Sea t o the continental shelf west of Svalbard. In this paper we summarise the main results obtained from the P O N A M expeditions t o Edgeoya. Barentsaya, Kongsoya and Bjomoya, and use the new data t o compare and modify t h e present models of the Quaternary history of Svalbard and the Barents Sea. The Late Quaternary record The pre-Late Weichselircn Sediments which record the Late Quaternary ice growth and decay have been preserved in several sections o n Spitsbergen (Boulton 1979; Troitsky et al. 1979; Miller 1982; Miller et al. 1989; Landvik et al. 1992a; Mangerud & Svendsen 1992; Man- gerud e t al. 1992) and an updated glaciation curve for the last intergIacial/glacial cycle of Svalbard and the northern Barents Sea was recently sug- gested by Mangerud e t al. (1996), based o n cor- relation of t h e different sediment successions. 96 J . Y . Landuik et al. These studies inferred from the Spitsbergen sites that the northern Barents Sea was covered by ice sheets during most of the Weichselian. except for two intervals at around 80 ka and 30-50 k a . T h e Barents Sea part of this reconstruction is based on the assumption that high relative sea levels during ice free periods on Spitsbergen were caused by glacial loading in the northern Barents Sea. During the PONAM expeditions. pre-Late Weichselian sediments were found on Edgesya and on Kong Karls Land (Fig. 2 ) . O n northeastern Edgesya. ice-free conditions are reflected by shell fragments embedded in beach sediments (Bon- devik et al. 1995). T h e fragments yield infinite radiocarbon ages. and mean amino acid ratios (aIle/Ile total) of 0.082 t- 0.019 (n = 6). These ratios are similar to those from the oldest ice- free period on Kongsoya (Ingolfsson et al. 1995). which are postulated to b e of pre-Eemian age. In Visdalen on western Edgeoya (Fig. 2 ) . coarse-grained fan delta sediments were Fig. I . Regional map showing the main areas discussed in the text. The present oceanic surface circulation is indicated. deposited along the northern valley slope at a time when the seal level was > 5 7 m above the present, and these were later covered by sub- glacial till (Moller e t al. 1995). Sub-till sediments have also been reported from the adjacent Rosen- bergdalen by Boulton (1990). Unfortunately. neither of these latter units could be dated. T h e islands of Kong Karls Land (Fig. 2 ) hold a key position with regard to understanding the last interglacial/glacial cycle of the northern Barents Sea. Shells from Svenskoya have been dated to 4 0 , 0 0 0 ~ ~ and older, and four drift- wood logs from Kongsoya were more than 40,000 years old (Salvigsen 1981). O n e whalebone from eastern Kongseya was dated to 33,600 ? 970 which was considered to be a minimum age by Salvigsen (1981). From northwestern Kongsoya, Ingolfsson et al. (1995) report sediment suc- cessions from two ice-free periods older than the Late Weichselian. Biostratigraphical studies sug- gest that the climate then was similar to or slightly warmer than the present (Ingolfsson et al. 1995). Quaternary record of eastern Soalbard 80 I 1.5. 21' 4 97 Fig. 2. Deglaciation dates from Edgeoya and Barentscdya. The two oldest dates have been listed from most of the study areas. Details are given in Table 1 . The insert map also shows the location of Kongsfjorden ( K ) , Isfjorden ( I ) , Van Mijenfjorden (VM) and Agardhbukta ( A ) . The glaciers on Edgeoya and Barentscdya are shaded. 98 J. I'. Lardilik et al. T h e older period is characterised by amino acid ratios (aIle/Ile total) of 0 . W ? 0.009. and coin- parison with the amino acid ratios in the Eeniian deposits a t Kapp Ekholm (Mangerud & Svendsen 1992) and luminescence dates. both support a pre- Eemian age. The vounger period. with amino acid ratios of 0.043 ? 0.008. is interpreted to be of either Earlv Weichselian or Eemian age (Ingolfs- son e t al. 1995). Several scenarios for the progression of the last interglacial/glacial cycle of Svalbard and the northern Barents Sea have been suggested (Lindner et al. 1984; Larsen et al. 1991; Landvik et al. 1997a: Mangerud X r Svendsen 1992; Man- gerud e t al. 1996). Through time, these models have evolved a better foundation for recon- structing the area of the Barents Sea proper. Previous models (Larsen e t al. 1991; Mangerud & Svendsen 1992) suggested almost contem- poraneous glacials and interstadials over Spits- bergen and the Barents Sea. whereas the later ones (Landvik et al. 1992a; Mangerud e t al. 1996) suggest a long period of ice cover over the sea to the east of the Svalbard islands during the Early Weichselian. If the younger ice-free period o n K o n g s ~ y a (Ingolfsson e t al. 1995) is of an Early Weichselian age. we assumed that the whole Bar- ents Sea must have been deglaciated. However, the time control so far is t o o rough to justify any revision of the latest models (Mangerud et al. 1996) for the last interglacial/glacial cycle of the Svalbard/Barents Sea region. The d w r r m i c s of rlw L o r p Weichselian ire sheet The extent of the ice sheet over Svalbard and the Barents Sea during the Late Weichselian has been the subject for debate for three decades ( e . g . Schvtt et al. 1Y68; Boulton 1979: Salvigsen 1981; Mangerud et al. 1992; Elverhai et al. 1993). Based on the post glacial uplift pattern (Salvigsen 1981; Forman 1990). and the distribution of sea floor sediments. moraine ridges. and subglacially formed flutes ( E h e r h s i ct al. 1993). there is a consensus today that the ice sheet covered at least the northern Barents S e a . For a short period. outlet glaciers a d w n c e d to the continental shelf \vest o f Spitsbergen (Manperud et al. 1992; Svend- sen rt :it. 1992) and to the western Barents Sea shelf (Elverhmi et al. 1993). Recent models of ice sheet extent based on uplift data also suggest that the ice sheet extended to the shelf edge to the west and north o f Svalbard (Lambeck 1995,1996). However, less is known about the exact timing and dynamics of t h e ice sheet. T h e pattern of glacioisostatic rebound shows a centre of maximum uplift in the northern part of the Barents Sea (Schytt e t al. 1968; Forman 1990). Bondevik et al. (1995) present detailed relative sealevel curves from E d g e ~ y a and B a r e n t s ~ y a , and their revised isobase m a p for 10,000 "'C years BP suggest a centre of uplift southeast of Kong Karls Land (Fig. 2). This compares with the iso- static models of Lambeck (1995, 1996) who also suggested that t h e ice sheet had its maximum ice thickness southeast of Kongsaya. Glaciological modelling of the ice sheet (Siegert & Dowdeswell 1995) suggests a similar location for the maximum ice thickness. Based on reconstructions of ice flow directions, it has been suggested that the Svalbard and Barents Sea Ice Sheet consisted o f several confluent ice domes and not only o n e single ice d o m e centered over the Barents Sea (Landvik & Salvigsen 1985; Mangerud et al. 1992). However, this ice flow pattern may represent only phases of growth. or decay, of t h e ice sheet. In the peripheral region of the former ice sheet, ice streams have been shown to drain along the B j s r n a y a and Storfjorden troughs o n the Barents Shelf (Elverhoi et al. 1993), and along Van Mijenfjorden and Isfjorden o n Spitsbergen (Man- gerud e t al. 1992). Different sets of ice movement indicators (see review by Salvigsen e t al. 1995) show phases when ice flow radiated from a centre east of Svalbard. Evidence for an ice sheet centred east of Kong Karls Land is also found on Kongs- s v a . where the sediments from the younger ice- free period were deformed by a topographically independent glacier moving from the northeast (Ingolfsson et al. 1995). Even if these sediments are of a n Early Weichselian age, we assume that such an ice movement must be attributed to the large Late Weichselian ice sheet. On a larger scale, there is a systematic variation in the sediment distribution on the Barents Sea floor that fits the theoretical zonation of glacial erosion and deposition under an ice sheet as pro- posed by Sugden (1977,1978). T h e shallow banks north of 74". where bedrock is exposed over ca 5 0 % of the area and sediment thickness is <25 ni (Solheim & Kristoffersen 1984), can be ascribed to a zone of erosion. This contrasts with the zone of deposition t o the southwest, where large sedimentary wedges increase in thickness towards the shelf margin, e.g. in the B j ~ r n - 0yrenna and Storfjordrenna troughs (Solheim & Quuternary record of eastern Siialb~itd 99 Kristoffersen 1984; Vorren e t al. 1990). Also the distribution of pre-Late Weichselian sediments on the Svalbard islands fits such a model. As reported here. Edgeoya and Barentsoya were subjected t o extensive glacial erosion during the last glaciation, whereas most sediment accumu- lations of pre-Late Weichselian age are found o n western Spitsbergen (Landvik & Salvigsen 1985; Miller et al. 1989; Lonne & Mangerud 1991; Landvik et al. 1992a; Mangerud & Svendsen 1992). Several of these accumulations are in glac- ially-shaped fjord and valley basins or on the strandflat, and show that the Late Weichselian glacial erosion was probably less important in forming the glacial topography of western Spits- bergen, The Late Weicliselian deglaciation The western margin of the Barents Ice Sheet reached the shelf break to the west during the Late Weichselian (Elverhoi e t al. 1993; Mangerud et al. 1992), and the glacier must also have sur- rounded or covered BjQrnQya o n the south- western margin of Spitsbergenbanken. As indicated by many radiocarbon dates, the ice sheet started to recede from the outer coast of Spitsbergen at 13,000 BP (Forman 1990; Man- gerud e t al. 1992), whereas the central fjord region first became ice free ca 10,000 BP (Man- gerud e t al. 1992). From the eastern part of the archipelago, the timing of the deglaciation had before the P O N A M expeditions been based on only a few dates from Edgeoya and Barentsoya (Knape 1971: Nagy 1984; Forman 1990) and from Kongsoya (Salvigsen 1981). There were n o satis- factory date$ for the deglaciation of B j o r n ~ y a until the studies of Wohlfarth et al. (1995). Dating of the local deglaciation has been an aim of the P O N A M studies, and a regionally well- distributed set of new dates from Edgeoya and B a r e n t s ~ y a is now available (Table 1, Fig. 2). These indicate a contemporaneous deglaciation at around 1 0 , 3 0 0 ~ ~ along both the eastern and western coasts of the islands. However, dating of the oldest part of the sea level curves from the area (Bondevik e t al. 1995), indicates that the deglaciation occurred ca 500 radiocarbon years later. T h e deglaciation probably progressed by rapid calving in both Storfjorden and Olgastretet (Fig. 2), and remnants of the glacier remained in the interior of Spitsbergen (Mangerud et al. 1992) and Edgeoya (Landvik e t al. 1992b; Moller e t al. 1995). W e assume that the rapid emergence between 10,000 and 9000 BP. (Salvigsen 1981; Bondevik e t al. 1995) is as a response to a major unloading just prior t o 1 0 , 0 0 0 ~ ~ . The cor- respondence in time between the deglaciation dates of ca 10,000 B P for both Agardhbukta (Fig. 2) o n the western side of Storfjorden (Salvigsen & Mangerud 1991) and o n K o n g s ~ y a (Salvigsen 1981) also suggests that the marine part of the ice sheet vanished d u e to rapid calving. A t this time, the nature of the ice flow across high areas thus changed from being regionally governed by the configuration of the Barents Ice Sheet, to being more locally affected by the topography beneath the last ice remnants on the islands. O n B j ~ r n o y a (Fig. l), glacial striae reflect only a local ice cap centred on the southern and middle part of the island (Salvigsen & Slettemark 1995). The ice cap may also have been a remnant of the ice sheet, formed by rapid withdrawal of the western margin of the Barents Ice Sheet. Basal dates from lake sediments on the island show that most of the local glaciers had vanished before 9800 BP (Wohlfarth e t al. 1995). The Holocene The sediment records of Edgeoya and B a r e n t s ~ y a show that the physical environment at the Weichselian/Holocene transition was different from today. T h e large transverse ridges found at the mouths of several valleys were interpreted by Biidel (1968) t o have been formed by longshore currents. They were later reinterpreted by Nagy (1984) as ice contact deltas, and taken as evidence for a halt in the deglaciation, called the “Visdalen event”. New investigations of the sediment suc- cessions in Visdalen o n Edgecdya (Fig. 2 ) (Moller e t al. 1995) support Biidel’s view, and show that thick piles of sediments were deposited as fan deltas which also fed longshore currents, depositing large ridge-formed spits at the mouths of inundated valleys between 10,000 and 9000 BP. T h e elevations of other similar ridges at the mouths of shallow paleo-fjords on E d g e ~ y a and Barentsoya, suggest that these are of the same age as the o n e in Visdalen. The formation of large spits (Miiller et al. 1995) during a period of rapid regression (Bondevik et al. 1995), suggests that the longshore sediment transport during the Preboreal was considerably greater than today. In the situation immediately following t h e deglaciation, we assume that uncon- lY 7 5 ) F ra n kc n h a lv a va K dp p Z ic h cn lla d a lc n H u m h B li fj o rd d a lc n A lh rc ch th rc cn D ia n a d a le n D yr d a le n G u ld a le n S rn el le da lc n R a d d e d a le n V is d a le n T a la ve ra 7X 03 4' N 2 1" 20 'F . 78 '33 " ?1 "5 X 'E 7X "l l' N 2 I" SS 'E 7H "I: 'N 23 "( W I'E 77 "S Y )'N 2 2" S Y 'E 7 7 "5 7 'N 2 3Y 15 'E 77 % " 23 "l h' E 77 04 5" 22 "4 5' E 7 7 "5 3 'N 2 1" 3Y 'E 7 7 5 7 " 22 "Z X 'E 78 "0 0' N 21 "3 6' E 7 8 "0 3 'N 2 1 "I S 'E 7 8 "1 5 'N 2 1" 07 'E IO .2 65 2 Y S 94 50 t I I 0 Y 61 5 5 11 11 YS O 5 2 7( 1 X 77 (1 i 5 0 YX XS 5 13 (1 96 20 -c 13 11 l( 1. 33 0 -c I l (1 YS YO f Y II Y 40 5 2 71 1 03 60 f l (1 5 10 .2 (M l % Y 5 95 9. 5 t 1 05 9 x4 0 t 1 30 Y 70 5 ? 1 35 96 75 2 50 YY YO ? 80 10 .0 15 l r 75 95 65 - c 8 0 l( l. lY 0 f 3 20 98 20 f 1 30 98 10 2 2 00 10 .3 70 lr 60 X 7- 77 6 X 7- 77 11 x7 -x 1 5 X X -2 M I (1 B I Ih X X -S ?S A X X -5 27 xx -5 x3 XX -5 6X X 7 -X l( l X 7- X O 6 X 6- 70 6 X 6- 70 5 xx -7 37 86 -4 18 X 6 - 4 2 ( ) G B X S 8 6 -5 7 2 A 86 .5 72 8 - 88 -6 75 87 -7 02 G B 12 3 A A R -9 0 8 A A R -8 38 S R R -2 2 0 4 I. a n d vr k C I a l (1 99 2h ) La nd vr k ct a l. ( IY 9 2 h ) U m d c v ik C I ;II (l Y Y 5) B on dc vi k c t ;iI (l Y Y 5 ) F or rn un ( 1 0 9 0 ) H o n d e vi k c t at . (1 99 5) H o n d cv ik c t al . (1 9 9 5 ) L a n d vi k ct a l. (I Y Y Z h ) L a n d vi k ct a l. ( 1 9 9 2 h ) R o n n cr t d L .;i nd vi k ( IY Y 3) K o n n cr i d l .; in dv ik (1 99 3) Ih in d e vi k c t a l. ( lY Y 5 ) D o n d cv ik e t al . (1 99 5) A d ri cl ss o n e t a l. ( IY Y Z h) H a n se n & K n u d se n ( IY Y S ) H a n sc n & K n u d rc n ( 1 9 9 5 ) F vr rn a n ( IY Y O ) B o n d cv ik c t al . (l Y 9 5 ) B o n d e vi k c t a l. (1 99 5) N ag y (1 9 8 4 ) M ti ll e r et a l. ( 19 95 ) M o ll e r e t al . ( 19 95 ) F o rr n a n ( 19 90 ) Quaternary record of eastern Svalbard 101 solidated glacial sediments deposited on unstable slopes were the most important source for the sediments deposited i n the alluvial fans and fan deltas. The latter process may have been enhanced by an early Holocene hydrographic regime which was different from the present. The Svalbard waters experienced a stronger influx of Atlantic water during the early and middle Holocene as reflected by the influx of thermo- philous molluscs already around 9500 BP (Sal- vigsen et al. 1992). However, a detailed Preboreal climatic record based on foraminifera from raised glacimarine deposits at eastern Edgeoya suggests that the steady amelioration of the marine con- ditions was interrupted by a short cooling between 9600 and 9 4 0 0 ~ ~ (Hansen & Knudsen 1995). From findings of subfossil Mytilus edulis, a mini- mum age of 8 8 0 0 ~ ~ was obtained for the first major influx of Atlantic water reaching Edgecbya (Hjort et al. 1995). This shows that the circulation in the northwestern part of the Barents Sea was different during the early Holocene, and it poss- ibly may also have influenced the strength and pattern of the near-shore currents in the area. At the same time, during the Preboral, the glaciers on Edgeoya receded beyond their present limits (Ronnert & Landvik 1993) and the modern flora of the area was established (Bennike & Hedenas 1995). Most glacier margins i n the area are today characterised by the retreat from young moraines formed by stacking of sheets of sedi- ments during the Little Ice Age (Adrielsson et al. 1992; Ronnert & Landvik 1993; Dowdeswell & Bamber 1995). Conclusions The collaborative efforts during the PONAM expeditions to eastern Svalbard and Bjorncbya resulted in a new understanding of the geological development of this part of the Svalbard archi- pelago and the Barents Sea. The main results are summarised below. 1. Glacial erosion during the Late Weichselian removed most of the older sediments from the eastern parts of the Svalbard archipelago and adjacent parts of the Barents Sea. The erosion i n this region was more extensive than on the western coast of Spitsbergen. where several older sediment successions have been preserved. 2. Deposits from two pre-Late Weichselian periods, with climates similar to or slightly warmer than the present on Kong Karls Land, show periods when the Barents Sea was corn- pletely deglaciated. 3. Both the improved isobase reconstructions and isostatic modelling suggest that the Late Weichselian ice sheet had its maximum thicknes, southeast of Kong Karls Land. 4. Detailed relative sea level curves from E d g e ~ y a and Barentsaya show that the glacio- isostatic uplift 10,000-9000 BP was rapid, responding to a major unloading just prior to 5. During t h e last deglaciation, the marine- based part of the ice sheet over eastern Svalbard underwent rapid calving, and most of the glacier fronts receded beyond the present day coasts of E d g e ~ y a and Barentscbya before 10,300 BP. 6. The coastal environment immediately after the deglaciation was characterised by extensive longshore sediment transport. Large spit ridges were formed at the mouths of several valleys. 7 . Subfossil Mytilus edulis indicate a major influx of Atlantic water into the northwestern Barents Sea between ca 8800 and SO00 BP. 10,000 BP. Acknowkdgemeni.s. - This paper is mainly the result of the expeditions within the PONAM (Polar North Atlantic Margins. Late Cenozoic Evolution) programme in I Y Y I and 1993. Fund- ing for the fieldwork was provided by the research councils in Denmark, Norway, Sweden and the U . K . , as well as by the participants own institutions. The Norwegian group also received financial support from Norsk Hydro. Saga Petroicurn. Statoil and the Norwegian Petroleum Directorate. and the Swedish group from the Polar Research Secretariat Helicopter transport in the field was made possible by a grant from the European Commission to the European Science Foundation The Norwegian Polar Institute provided log~stic support, and the use of R / V LANCE during initial reconnaissance work and for transport to and from eastern Svalbard. Norwegian Polar Institute expedition leaders T . Siggerud (1991) and J. Hdugland (1993) were most helpful to the two PONAM expeditions. I n 1991, B. Torgerhagen (pilof) and K. Nilsen (engineer) skillfully operated the helicopter from a base camp at Kapp Lee a t Edgeoya. H . Eggenfellnet assisted in base operations. The Norwegian Coast Guard transported two field parties to Bjornoya in 1993. All operattons in eastern Svalbard werc made possible through kind permissions from the Governor on Svalbard, who also assisted with helicopter transport from Kongsoya in 1993. To all these persons and institutions we offer our sincere thanks Anally. all parttctpants in the PONAM project are thanked for the open communication. and all the lively discussions that have contributed to the results we sec today. References Adnelsson, L , Johansson. K & Hjort, C 1992 Deglaciation and Holocene stratigraphy tn a Llttle Ice Age comporite 102 J . Y . Lnndiiik er al nioriiinc i n tront nt Seidbrecn. Edgeoyd. edstern Stalbard. Liindqii(i R e p . 3.7. 15.7-1MI Bennikc. 0 B H e d e n ~ s . L 1YY5: Earl! Holocene land floras m d faun;i, troni Edeco!cl. eaztcrn S\alhard. Pulur R n . l J f 2 J . 205-2 I4 B o n d e \ i k . 5 . . Mangerud. J . Ronncrt. L 8: Salvigscn. 0. 1995. Postgld~iiil rea le\cl histor! of Edgeo!a and Barentsoya. enatern S\,ilhdrd P d i i r R r ~ e i r r ~ I i I J f 2 j . 153-180. Boulton. C; S . l Y 7 Y : Glucial history o f the Spit5hergen archi- pclagi, and the problem n f d Bdrent\ Shelf ice sheet. Borefir 8. 31-i: Boulton (I S. IYYO: Sedimentar! dnd ,ca le\el change5 during glacial c\cle\ and their control on rlacimarine facie, archi- tecture. Pp 15-52 i n Dowdest\ell. J . A 8: Scourse. J D. 1 e d \ 1. ~ ; / ~ i ~ i i i i i i r i i i r c'iii'iruiimriits P r o i r s s r i o i i r l sediineiio GCOI Snc Spec Puhl 53 Budel. J I Y h S . D i e luiigc Landhehung Spitrhergens im L'mkreis dc\ Frecni,in-Sundc\ und der Olga-Sirasre W r r r o h . Grogr. Arh. 1-7 1 . 1-21 D w d e s \ r i . l l . J ,A 6 Banihcr. J L l Y Y 5 : On thc glaciology ol t . d g e d \ a ,tiid B.ircntw!a. S\ulhard Polor Res. 14f21. lil5- 1 2 2 F~eld\ka,ir. L\' . Solheini. A . S F l m d Berg. 51 h. Ku\.uurm. 1. lYY3. 1 - h ~ Barent\ Sea Ice Sheet - ,A model ( 1 1 i t \ g i o \ \ t h dnd dec'i! during the l a \ t ice maxiniuni Q i u r r \ t i Kt.1. 1 2 . Xh.LX7i. F{,rni'in. \ I Y Y i I . P o ~ t - g l ; i c i a l rcIati\c w a - l c \ e l hictor! ot north- nestern Spihbergen. S\albard Grol. So<. Am. Brill. / 0 2 , I . ~ ~ I ~ - I ? Y I ~ t i i r i n d n . \ . L u h i n k i . D . M i l l e r . G . H . . Sn!der. I . 5ldtisho\. ( i . K i i r w n . S. & 5l!\li\ct5. \' I Y Y j . Po,tglacial emergencc dnci di\trihutitm ( i f l a i c H'eich\c'lian ice-theet loads in the northcrn B a r e n h and K a r . ~ s u \ . Kursia. Geol. 23. 112-1 I6 H a m c i i . A ti K n u d w n . K . L . IY95. Kccent foraminifera1 di\trihutiun i n Freemansundct and earl! lfolocene