FreshwaterandAtlanticwaterinflowstothedeepFreshwaterandAtlanticwaterinflowstothedeep ... (afterLoeng,1991;Loengetal.,1993;Pfirmanetal.,1994; ... Calender ageb (CalyrBP) Sedresearch.bpcrc.osu.edu/geo/publications/Lubinski_et_al_QSR.pdf ·

Quaternary Science Reviews 20 (2001) 1851–1879

Freshwater and Atlantic water inflows to the deepnorthern Barents and Kara seas since ca 13 14Cka:

foraminifera and stable isotopes

D.J. Lubinskia,*, L. Polyakb, S.L. Formanc

a Institute of Arctic and Alpine Research, University of Colorado at Boulder, CB 450, Boulder, CO 80309-0450, USAbByrd Polar Research Center, The Ohio State University, 108 Scott Hall, 1090 Carmack Road, Columbus, OH 43210-1002, USA

cDepartment of Earth and Environmental Sciences, University of Illinois, Chicago, IL 60607-7059, USA

Abstract

Foraminiferal stable isotopes and assemblages from Franz Victoria and St. Anna troughs provide a valuable record of freshwater

and Atlantic Water flows to the northern Barents and Kara seas from deglaciation to present. The d18O and d13C of planktonicNeogloboquadrina pachyderma (s) and benthic Elphidium excavatum were up to 1.4% lower than present at ca 13, 11.5, and10 14Cka (global sea-level corrected), mostly reflecting substantial freshwater inputs coincident with glacial–marine sedimentdeposition. Cassidulina teretis exceeded 40% of benthic foraminifera ca 13 and 10 14Cka, indicating subsurface penetrations of

Atlantic Water. The transition to postglacial marine conditions is marked by a�1% rise in foraminiferal d18O and a sharp fall in %C. teretis soon after 10 14C ka. These changes imply reduced inputs of freshwater and Atlantic Water. Subsequent isotopic andforaminiferal assemblage variations reflect changing Atlantic Water conditions ‘‘upstream’’ in the Nordic Seas and shifts between

the warm Fram Strait and cold Barents Sea branches of Atlantic Water. We hypothesize that glacial-isostatically induced deepeningby up to �150m influenced Atlantic Water inflows to the northern Barents Sea during deglaciation and the Holocene. Thus, effectsof isostatic recovery have to be factored into paleoceanographic reconstructions. # 2001 Elsevier Science Ltd. All rights reserved.

1. Introduction

Sediment cores from the Nordic Seas and the NorthAtlantic Ocean document dramatic shifts in sea-surfacetemperatures (SST) since the Last Glacial Maximumca 20 14C ka; the largest of these SST shifts of �108Coccurred over a period of a few decades accompanied bya ‘‘mode’’ change in ocean thermohaline and Atlanticmeridional circulation (Lehman and Keigwin, 1992;Bond et al., 1993; Koc et al., 1993; Hald et al., 1996;Dokken and Jansen, 1999). The degree to which theseshifts and changes propagated into the adjacentEurasian Arctic (Fig. 1), however, remains poorlyknown. In turn, there is limited knowledge aboutwhether Eurasian Arctic freshwater fluxes (e.g., sea ice,glacier ice, meltwater, and river water) modulatedobserved reductions in thermohaline circulation in theNorth Atlantic and Nordic Seas such as those at ca 14.5,

13.5, 12, and 10.5 14C ka (e.g., Keigwin et al., 1991;Veum et al., 1992; Dokken and Jansen, 1999).Existing Arctic Ocean records show a series of low

d18O meltwater events between 16 and 11.5 14C ka (e.g.,Stein et al., 1994; N�rgaard–Pedersen et al., 1998; Pooreet al., 1999), but poor temporal and spatial resolutionhas hindered determination of their magnitude, number,ages, and sources. Furthermore, these Arctic Ocean sitesprobably integrate freshwater inputs from multipleEurasian and non-Eurasian sources. Records from theadjacent Eurasian shelf are specifically needed to delimitfreshwater fluxes from the northern margins of theBarents and Kara ice sheets and the voluminousnorthward flowing Eurasian rivers.Shelf records from the Eurasian North are also

needed for delimiting changes in meridional AtlanticWater fluxes into the high Arctic (cf. Duplessy et al., inpress). In particular, little is known about the influenceof the Barents and Kara ice sheets on Atlantic Watertransport over the Barents Sea shelf. The ice sheets mayhave physically blocked this flow or indirectly altered itby glacial-isostatically induced bathymetric changes.

*Corresponding author. Tel.: +1-303-735-6619; fax: +1-303-492-

6388.

E-mail address: [email protected] (D.J. Lubinski).

0277-3791/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.

PII: S 0 2 7 7 - 3 7 9 1 ( 0 1 ) 0 0 0 1 6 - 6

Approximately, 2 Sv (106m3/s) of Atlantic Water pre-sently flows through the Barents Sea, about 1/3 to 1/2 ofthe net Atlantic flow into the Arctic Ocean (e.g., Rudelset al., 1994; Schauer et al., 1997a; Zhang et al., 1998).Six sediment cores from 440 to 630m depth in the

Franz Victoria and Saint Anna troughs (Fig. 2) wereused in this study because they are well situated tomonitor variations in meltwater and Atlantic Waterflows to the western Eurasian Arctic immediately upondeglaciation. During the Last Glacial Maximum, theBarents Sea Ice Sheet grounded in the troughs (e.g.,Lubinski et al., 1996; Polyak et al., 1997; Kleiber et al.,2000), which are presently bathed by subsurface AtlanticWater (Figs. 1 and 2). Postglacial sedimentation rates inthese shelf troughs range between �10 and 100 cm/kyr,at least an order of magnitude higher than in the

adjacent Arctic Ocean. Previous studies of sedimentcores from these troughs and the adjacent slope clearlyshowed: (1) Last Glacial Maximum glacier ice groundedout to the shelf edge, (2) initial deglaciation by ca 1514C ka, with initial grounding line retreat as late as ca 1314C ka in some portions of the troughs, (3) discontin-uous retreat, and (4) a glacial–marine to marinetransition ca 10 14C ka (Polyak and Solheim, 1994;Herlihy, 1996; Lubinski et al., 1996; Polyak et al., 1997;Hald et al., 1999; Kleiber et al., 2000). Our new studydiffers by concentrating on reconstruction of environ-mental conditions using higher resolution, mostlycentury to multi-century scale records of foraminiferalstable isotopic and species composition the period sinceca 13 14C ka. It also differs by assembling data on themodern distributions of seawater d18O and certainforaminiferal species to better understand these records.Together, these fossil and modern data provide a basisfor interpreting the history of Atlantic and freshwatervariation for each trough since 13 14C ka; this historygives insight to the controls on environmental change inthis seldom-studied high-latitude region and improvesour understanding of its relationships with the better-studied, climatically connected Nordic seas and NorthAtlantic Ocean to the south.

2. Materials and methods

A piston core and its gravity core trigger werecollected simultaneously from Franz Victoria Troughby the United States Coast Guard icebreaker Polar Starin 1991 (Lubinski et al., 1996). These two cores formcomposite core PG5/JPC5, taken from the eastern flankof Franz Victoria Trough at 460m water depth (Table 1;Fig. 3). Four additional cores were collected from 425 to633m water depths in St. Anna Trough by Russian–American–Norwegian expeditions on board R/V Geo-log Fersman in 1993 and R/V Professor Logachev in1994 (Polyak et al., 1997; Hald et al., 1999; Ivanov et al.,1999; Fig. 3). Although the northernmost cores PL94-29(605m) and PL94-07 (633m) are from areas with lowHolocene sedimentation rates, they are particularlyimportant for our study because they contain rarelycollected deglacial sediments from St. Anna Trough.All cores were sampled at 2–10 cm intervals. Nearly

all samples were 2–2.5 cm thick. The samples were oven-or freeze-dried, weighed, re-hydrated in distilled water,wet-sieved with 2mm and 63 mm screens, dried again,and then re-weighed for particle-size analysis. The>63 mm fraction was dry-sieved to isolate the>125 mm fraction. Foraminifera in this size class werecounted, typically based on >150 calcareous benthicspecimens; respective planktonics ranged from a few to�100 specimens. Agglutinated foraminifera were rareand not studied in detail.

Fig. 1. Overview map of the Nordic Seas and western Eurasian Arctic

showing locations of our cores and those mentioned in the text. Also

shown is the limit of the Eurasian and Fennoscandinavian ice sheets

during the Last Glacial Maximum (LGM), modified from Landvik

et al. (1998), Svendsen et al. (1999), Forman et al. (1999), and Polyak

et al. (2000a); the limits are particularly poorly known in the

northeastern Kara Sea. Isobaths appear at 500 and 1000m depths.

Abbreviations of northern Eurasian land areas are as follows:

SV=Svalbard, FJL=Franz Josef Land, NZ=Novaya Zemlya,

SZ=Severnaya Zemlya, and TP=Taimyr Peninsula. The large

Yenesey and Ob’ rivers are noted.

D.J. Lubinski et al. / Quaternary Science Reviews 20 (2001) 1851–18791852

Paired, simultaneous measurements of d18O and d13Cwere performed on the non-symbiotic planktonicforaminifer Neogloboquadrina pachyderma (sinistral)(n ¼ 134) and the benthic foraminifer Elphidium ex-cavatum forma clavata (n ¼ 83). Additional isotopicsamples of the benthic species Cassidulina teretis (n ¼ 8)provide information where E. excavatum was absent. Allisotopic samples were composed of mature individualsin the 125–250 mm size class. This narrow classminimizes size-dependent effects on isotopic composi-tion (Aksu and Vilks, 1988; Keigwin and Boyle, 1989;Oppo and Fairbanks, 1989; Donner and Wefer, 1994;

Bauch et al., 2000). Furthermore, this size class allowsdirect comparison with many high northern latitudestudies (Spielhagen and Erlenkeuser, 1994; Bauch et al.,1997; Bauch and Weinelt, 1997; N�rgaard–Pedersenet al., 1998). Our N. pachyderma (s) specimens typicallycontain a thick secondary calcite crust, which increasesthe shell density and modifies its chemistry (e.g.,Kohfeld et al., 1996). Most of the specimens had 4 or4.5 chambers; previous studies suggest no isotopicdifference between these morphotypes (Johannessenet al., 1994; Alderman and Lehman, 1995). The majorityof the 225 analyzed samples consisted of about 20

Fig. 2. Map of the Barents and Kara seas showing simplified bathymetry (300 and 1000m isobaths), sediment core sites, study areas, and general

circulation pattern (after Loeng, 1991; Loeng et al., 1993; Pfirman et al., 1994; Schauer et al., 1997b). Mean sea-ice limits (>12% cover) are from

Barry (1989). Three major Atlantic Water current systems are labeled: NAC=Norwegian Atlantic Current, NCC=North Cape Current, and

WSC=West Spitsbergen Current. The position of the winter sea ice limit is similar to that of the Polar Front, which separates cold Polar Water on

the surface from warm Atlantic Water.

Table 1

Core and site information

Core No. Latitude

(8N)Longitude

(8E)Water depth

(m)

Core length

(cm)

Core dia.

(cm)

Priora work

ref.

Number of14C ages

Number of foram

isotopes

Franz Victoria Trough

P191-AR-PG5 81807.10 43825.90 463 117 6 2 2 26

P191-AR-JPC5 81807.10 43825.90 463 421 11 2 9 27

St. Anna Trough

GF93-163 77800.10 70800.30 550 232 11 1 2 49

PL94-67 78819.540 70802.070 443 242 11 3,4 7 75

Pl94-07 80859.690 67832.770 633 161 11.5 3,4 2 6

PL94-29 79859.590 69856.960 605 168 11 3,4 1 2

aReference codes are (1) this study; (2) Lubinski et al. (1996); (3) Herlihy (1996); and (4) Polyak et al. (1997).

D.J. Lubinski et al. / Quaternary Science Reviews 20 (2001) 1851–1879 1853

individuals. Only 14 samples consisted of 510 indivi-duals and all of these were from stratigraphic levels withlow foraminiferal abundance in St. Anna Trough. Themajority of the d18O and d13C values were measured ona Finnigan MAT 252 with a ‘‘Kiel’’ automatedcarbonate analytical device at Woods Hole Oceano-graphic Institution using standard methods (Ostermannand Curry, 2000). The precision of this instrument forthe NSB19 standard, with correction for all knownbiases, is � 0.07% for d18O and � 0.03% for d13C.Small sample (10–30 mg) precision is slightly lower:� 0.12% for d18O and � 0.05% for d13C. Theremaining 20 samples were published in Lubinski et al.(1996). Data are reported in per mil units (%) relative tothe PDB calcite standard.To better understand the foraminiferal isotopic

measurements, we assembled previously publishedmodern seawater isotopic and hydrographic data forthe Barents and Kara seas region (references listed inSection 3). Because no empirically derived isotopic/temperature equations exist for N. pachyderma (s) andE. excavatum (cf., Bemis et al., 1998), we rearrangedShackleton’s (1974) derivation of the equation ofO’Neill et al. (1969) to calculate the isotopic composi-

tion of inorganically precipitated equilibrium calcite:

d18Oe:c:ðPDBÞ ¼ ðd18OwaterðSMOWÞ � 0:27

þ ðð�16:9þ Temperature ð8CÞÞ=� 4:0Þ:

This equation includes a SMOW-to-PDB conversion of�0.27% (Hut, 1987) and yields isotopic values �0.5%higher than an extrapolation of the inorganic equationof Kim and O’Neil (1997). We have chosen to use theShackleton (1974) equation because it is: (1) moredirectly comparable to prior foraminiferal studies usingthe same equation, (2) potentially more accurate thanKim and O’Neil (1997) to the low temperaturesobserved in the troughs, and (3) insufficiently differentthan other equations to alter our main results.The chronologies for the two composite records are

based on accelerator mass spectrometer (AMS) 14Cdeterminations made on 22 samples (Table 2). Datingwas confined to nine collections of calcareous benthic orplanktonic foraminifera selected from abundance peaks,10 small paired pelecypods (51 cm length), and threechitinous polychaete tubes where biogenic carbonatewas absent. All ages were reservoir-corrected bysubtraction of 440 years from the laboratory-reported,

Fig. 3. Study area maps showing bathymetry, sediment cores, and hydrographic stations. Bathymetric maps modified from Cherkis et al. (1991).

Also shown are two cores from recent studies (Polyak and Solheim, 1994; core 45; Duplessy et al., in press; core ASV-880).


d13C-corrected age (Mangerud and Gulliksen, 1975;Olsson, 1980; Forman and Polyak, 1997). Ages wereinferred for all sampled levels by using linear interpola-tion between dated levels. Ages discussed in the text usethe 14C timescale and are given in ‘‘14C ka’’ units.Calibrated calendar ages are shown in the tables and onsome figures to promote comparison with other records;they are derived using the atmospheric calibration dataset after correction for reservoir age (Stuiver andReimer, 1993; Calib version 4.3).

3. Modern environment and proxies

3.1. Modern hydrography of Franz Victoria Trough andSt. Anna Trough

We call the surficial (0 to �60m) water mass in bothtroughs Polar Water (PW). Its temperature rises to>28C at the surface during summer in some regions butis often lower than –18C by 60m depth (Fig. 4). Salinityis often rather low (534.5), largely due to sea-ice melt

Table 2

AMS 14C ages for six sediment cores from the Franz Victoria and St. Anna troughs, northern Barents and Kara seas. arctic Russia

Core

depth

(cm)

Reported

age

(14C yrBP)

Reservoir

corr. agea

(14C yrBP)

Calender

ageb

(Cal yr BP)

Sed

ratec

(cm/cal ky)

Lab. num.d d13Ce

(% vs. PDB)

Materialf Refg

Franz Victoria Trough, P191-AR-PG5

2 440 0� 130 0 32 AA-21397 �19.4 Polychaete tube 1

114 3700 3260� 165 3510 } AA-16408 �19.3 Polychaete tube 1

Franz Victoria Trough, P191-AR-JPC5

12 3745 3305� 60 3530 } AA-16406 �16.8 Polychaaete tube 1

16 3760 3320� 60 3565 14 CAMS-5551 0.0 P-Yoldiella, Thyasira 2

48 5580 5140� 70 5910 14 CAMS-5552 0.0 P-Yoldiella, Cuspidaria 2

88 8360 7920� 70 8710 26 CAMS-5550 0.0 P-Thyasira 2

136 9800 9360� 70 10,545 25 CAMS-5553 0.0 P-Thyasira 2

168 10,640 10,200� 70 11,835 16 CAMS-5549 0.0 BF-C. t. and I. n. 2

200 12,230 11,790� 70 11,820 66 CAMS-5548 0.0 BF-C. t. and E. e. 2

312 13,330 12,890� 80 15,520 } CAMS-5547 0.0 BF-C. t. 2

St. Anna Trough, GF93-163

100 2320 1880� 60 1820 54 AA-14589 0.0 P-Yoldiella (?) 1

195 3775 3335� 50 3570 } AA-14590 0.0 P-Yoldiella (?) 1

St. Anna Trough, PL94-67

17 3420 2980� 70 3170 9 GX-20859 0.0 BF-E.e. and M.b. 3,4

50 6285 5845� 75 6660 46 AA-19029 0.0 BF-M.b. and C.i. 3,4

92 7175 6735� 160 7575 15 AA-20482 0.0 BF-mixed 3,4

115 8610 8170� 115 9150 88 GX-20860 0.0 P-Yoldiella 3,4

171 9225 8785� 120 9790 51 AA-19030 0.0 P-Yoldiella 3,4

190 9390 8950� 120 10,160 36 GX-20861 0.0 P-Yoldiella 3,4

214 10,010 9570� 90 10,820 } AA-19031 0.0 P-Yoldiella 3,4


5 1870 1430� 50 1310 5 AA-16849 0.0 PF-N.p. (s) 3,4

74 13,710 13,270� 130 15,945 } AA-16848 0.0 BF-M.b. 3,4


95 13,730 13,290� 110 15,970 } GX-21067 0.0 BF-M.b., C.i. 3,4

aAll ages have been normalized to a d13C of �25% vs. PDB as per Stuiver and Polach (1977). For all samples, laboratory reported ages have been

reservoir corrected by substracting 440 yr from the laboratory-reported age (Mangerud and Gulliksen, 1975).bCalibration with Calib 4.3 program (cf., Stuiver and Reimer, 1993). In cases with more than one calibrated age, we use the mean of the 1-sigma

range with the largest relative area under the probability distribution.cLinear sedimentation rates calculated between one calendar age and the next older one.dLaboratory codes are (AA) National Science Foundation Arizona AMS Facility, (CAMS) Lawrence Livermore Center for Accelerator Mass

Spectrometry, and (GX) Geochron Laboratories, Krueger Enterprises, Inc.ed13C values have been measured in three polychaete chitin samples and were assumed to be 0.0% for the remaining samples (all carbonates).fMaterial codes are (P) pelecypod, (BF) benthic foraminfera, (PF) Planktic foraminifera, (C.t) Cassidulina, teretis, (I.n.) Islandiella norcrossi, (E.e.)

Elphidium excavatum, (N.I.) Nonion labradoricum, (M.i.) Melonis barleeanus, (C.I.) Cibicides lobatulus, and (N.p.(s)) Neogloboquadrina pachyderma

sinistralgReference codes are (1) this study; (2) Lubinski et al. (1996); (3) Herlihy (1996); and (4) Polyak et al. (1997).


(Pfirman et al., 1994; Ivanov and Korablev, 1997;NODC, 1991). Sea ice typically occurs in the northernhalves of the troughs during summer (e.g., Vinje, 1976;Parkinson et al., 1987; NSIDC, 1953–1990). PolarWater flows northward in much of the St. AnnaTrough, except its northernmost stretches where itmoves more westerly. In the Franz Victoria Trough,Polar Water flows westward to southwestward and isstrongly influenced by the large westward flowingTranspolar Drift current.A strong pycnocline and halocline separates Polar

Water from a distinctly warmer and saltier Atlantic-derived water mass in Franz Victoria Trough and alongthe western flank of St. Anna Trough down to a depthof about 400m (Figs. 4 and 5). We call this Atlanticwater mass Fram Strait Branch Water (FSBW; >08C,�34.9) because it originates from a northward flow ofAtlantic Water through Fram Strait (West SpitsbergenCurrent), as depicted in Fig. 2. A total of �2 Sv passesinto the Arctic Ocean by this route (Aagaard andCarmack, 1989; Gascard et al., 1995; Manley, 1995;Zhang et al., 1998). FSBW sinks while moving eastward,mixing a bit with the overlying Polar Water and deeperwater masses. A portion of this subsurface water mass

moves southward into the troughs as depicted in Fig. 2(Mosby, 1938; NODC, 1991; Pfirman et al., 1994;Ivanov and Korablev, 1997; Schauer et al., 1997a). Itforms a prominent intermediate water mass at core sitesPG5/JPC5 (Franz Victoria Trough), PL94-07 (north-western St. Anna Trough), and PL94-29 (west central St.Anna Trough). FSBW does not quite reach the seafloorat any of the trough core sites (440–630m) today,although mixtures of FBSW and the remaining watermass, Barents Sea Branch Water (BSBW), are particu-larly prevalent near the seafloor in Franz VictoriaTrough.Barents Sea Branch Water (BSBW) is also an

Atlantic-derived water mass, although a much moretransformed one. It exhibits a relatively wide range ofboth temperature and salinity (freezing to 08C, �34.7–35) and, thus, can occupy a depth range from �100mdown to the seafloor. BSBW influences bottom hydro-graphic conditions to some extent at all of our core sites,although it is less common in Franz Victoria Troughthan eastern and southern St. Anna Trough (Fig. 4).BSBW is formed from interactions between the AtlanticWater that flows into the southwestern Barents Sea(�2 Sv) and local Barents and Kara waters (Blindheim,

Fig. 4. Summer temperature and salinity profiles across the Franz Victoria and St. Anna troughs. See Fig. 3 for location. Water masses are as

follows: PW=Polar Water, FSBW=Fram Strait Branch Water, and BSBW=Barents Sea Branch Water. Mixtures of these water masses are

common, particularly in the central St. Anna Trough. Data for both troughs were collected in August 1994 and, thus, are directly comparable. Franz

Victoria Trough data were collected on the R/V Lance (Lothe et al., 1996) while St. Anna Trough data are from the R/V Professor Logachev (Ivanov

et al., 1999). Horizontal scales are different.


1989; Loeng, 1991; Loeng et al., 1993; Zhang et al.,1998). As the Atlantic Water moves northeastwardacross the Barents Sea, it is radiatively cooled, mixedwith fresher, colder shelf waters, and occasionallyentrained by sinking sea-ice produced brines. Theundulating relief of the Barents Sea guides transport,mixing, and entrainment (Pfirman et al., 1994). North-ward flowing BSBW crosses sills as shallow as �150mbut its strongest flows are associated with 200–300m-deep passages that intersect southwestern St. AnnaTrough (Fig. 2). Upon entering southwestern St. AnnaTrough, BSBW moves quickly northward along its

eastern flank and along the bottom (Ivanov andKorablev, 1997; Schauer et al., 1997b). Vigorous mixingoccurs in central St. Anna Trough along a shear zonebetween the northward flowing BSBW and the south-ward flowing FSBW as shown in Fig. 4 (Ivanov andKorablev, 1997).There are two modes of the relatively cold Barents Sea

Branch Water (Schauer et al., 1997b). The first tends tobe a mixture of cold, relatively fresh coastal water andwarm Atlantic Water, the densities of which are similar.This mode is particularly common in eastern St. AnnaTrough at intermediate depths, where it forms several

Fig. 5. Summer water-column properties, including the d18O of seawater, compared to modern d18O values for equilibrium calcite and N.

pachyderma (s), E. excavatum, and C. teretis from selected core-tops (PG5, PL-163, PL-29 box core, and PL-07 box core). Core-top isotopic data are

shown as short thick vertical lines on the rightmost plots. Horizontal lines are drawn at the water depths of each of the six cores used in this study.

(A) Franz Victoria Trough seawater data are from the YMER80 expedition, September 1980 ( .OOstlund and Hut, 1984; .OOstlund and Grall, 1993).

Since data constraining historical variations of temperature and salinity are available and greatly exceed those measured in 1980 in the Franz Victoria

Trough, we show dashed envelopes taken from the historical NODC data set extending back to the 1920s (NODC, 1991). (B) Northern St. Anna

Trough seawater data are from R/V Polarstern ARKXII-1, July 1996 (Mensch et al., submitted). (C) Southern St. Anna Trough data are from the

R/V Ivan Petrov, July/August 1998 (Polyak, in prep). See text for discussion of the calculation of d18O of equilibrium calcite. See Fig. 3 for the loca-

tions of all stations. Water masses are as follows: PW=Polar Water, FSBW=Fram Strait Branch Water, and BSBW=Barents Sea Branch Water.


prominent temperature minima (Fig. 4). The second,denser mode of BSBW is formed from sinking sea-icebrines that entrain relatively warm, northeastwardflowing Atlantic Water (Schauer et al., 1997b). Thebrines are formed during winter on the shallow shelvesof the northern Barents and Kara seas (Midtunn, 1985;Pfirman et al., 1994; Ivanov and Korablev, 1997). Thedenser BSBW mode forms bottom water at all six coresites (440–630m), although it is mixed with FSBW inFVT. Hydrographic data show that the strongestinfluence of dense BSBW on the core sites is in southernSt. Anna Trough (PL94-67 and GF93-163).

3.2. Modern oxygen isotope distributions

Compilations of published d18O, temperature, andsalinity of seawater in the troughs during summer

(Figs. 5 and 6) provide a context for interpreting theforaminiferal d18O record from sediment cores, particu-larly for the Holocene, when the hydrographic systemprobably operated in a similar manner to the present.These compilations also provide a baseline for compar-ing against the clearly different deglacial period. Datafor the summer period are used because foraminiferagenerally calcify at this time at high northern latitudes(e.g., Kohfeld et al., 1996; Wollenburg and Mackensen,1998). Although there is an absence of d18O data forliving foraminifera in the troughs, we were able tocalculate equilibrium calcite values to better understandthe modern isotopic system.At present, the calculated d18Oe.c. values show a

stronger temperature than salinity control in the troughswhen the depths of foraminiferal calcification areconsidered. This result is principally due to a higher

Fig. 6. Comparison of the d18O of seawater and equilibrium calcite with summer salinity and temperature for the Franz Victoria and St. Anna

troughs, using the modern hydrographic data depicted in Fig. 5. Linear relationships with r250:5 are shown with solid lines. Data from the entire

water column at sites 215 and 209 in Franz Victoria Trough have been excluded from calculation of linear relationships because they are�0.2–0.3%lower than surrounding sites, suggesting a systematic measurement error (‘‘out’’=outlier). The upper-left panel shows a linear relationship between

the d18O of seawater and salinity in St. Anna Trough in 1996/1998, with freshwater endmembers of �23% for northern St. Anna Trough (1996) and

�16 to �9% for southern St. Anna Trough (1998); the latter value depends on whether a single very fresh sample is considered an outlier. These

values are generally consistent with riverine d18O compositions and surface current patterns (e.g., Brezgunov et al., 1982; Bauch et al., 1995). There is

no comparable relationship for Franz Victoria Trough in 1980, probably because the lower salinity surface layer is largely derived from melting of sea

ice formed from relatively saline waters. Lower panels are limited to a narrow depth interval, corresponding to the 100–200m calcification depth for

N. pachyderma (s) (see text). The lower-right panel shows particularly strong relationships between the d18O of equilibrium calcite and temperature.

The true temperature relationship in Franz Victoria Trough may be stronger than appears with this 1980 data set because temperatures in other

years, such as 1994, for the 100–200m depth interval are up to 1.58C higher with no substantial change in salinity (Fig. 4).


temperature than salinity variability below the surfacePolar Water (Figs. 4–6). For our purposes, the surfacelayer should be excluded because even the planktonicforaminifer N. pachyderma (s) appears to be calcifyingbelow this layer, as implied by studies of two regionsadjacent to our trough study areas: the southern NansenBasin and the central Arctic Ocean. The averagecalcification depth of this species in both regions variesbetween 100 and 200m, despite differing hydrographicregimes (Carstens and Wefer, 1992; Bauch et al., 1997).This result, combined with the proximity of our FranzVictoria Trough core site to the southern Nansen Basin,strongly suggests that N. pachyderma (s) in the troughsalso calcify most frequently at 100–200m depth, wheretemperature has a particularly strong influence onisotopic composition (Fig. 6).Coretop N. pachyderma (s) have d18O values consis-

tent with that of equilibrium calcite at 100–200m waterdepth when the foraminiferal vital effect (biotic dis-equilibrium) and temperature variability are considered.The vital effect for N. pachyderma (s) was previouslymeasured in the Arctic at �1% lower than equilibrium(e.g., Kohfeld et al., 1996; Bauch et al., 1997).Correction for this effect yields d18O values ofN. pachyderma (s) that are similar to equilibrium calciteaveraged over the 100–200m depth interval (Fig. 5A).The difference is only �0.2% for Franz VictoriaTrough. The difference in St. Anna Trough is less than0.5% (hydrographic stations A96-7 and IP98-3 wereused for core tops PL-07 and GF-163, respectively).These remaining differences may reflect temperaturevariability at the core sites over time. Temperature ofFSBW can vary by 1–28C in less than a decade in thetroughs (Stanovoy et al., 1998) while the core-topmeasurements are an integration of conditions forapproximately the past century. An additional factor isuncertainty in the exact vital effect for N. pachyderma (s)in the high Arctic, which may differ from that of theSouthern Ocean (e.g., Charles and Fairbanks, 1990;Kohfeld et al., 1996).Modern hydrographic measurements imply that

modern benthic d18O values, like their planktoniccounterparts, are more likely to be influenced bytemperature than salinity. At the depths of the coresites (460–633m; Fig. 5B) today, there is a relativelynarrow range of seawater salinity (50.2) and d18Owater

values (50.4%) compared to the range of temperatures(�38C) and d18Oe.c. (�1%). Core-top benthic forami-niferal d18O values are also broadly consistent withcalculated d18Oe.c. when vital effects are considered. Theisotopic difference between the d18O of E. excavatumand equilibrium calcite of near-bottom waters is �1%for PG5 in Franz Victoria Trough and �0.5% for GF-163 in southern St. Anna Trough (Fig. 5). Thesedifferences are broadly consistent with vital-effectestimates of 0.85% for E. excavatum (Poole, 1994; using

estimated d18O of water) and ��1% for the genusElphidium (McCorkle et al., 1990; Erlenkeuser and vonGrafenstein, 1999). Nevertheless, the vital effect forliving E. excavatum in arctic shelf settings has neverbeen directly measured. In contrast to E. excavatum,little difference was observed between the d18O ofC. teretis and d18Oe.c for core site GF-163, the onlylocation where near-modern C. teretis was measured(Fig. 5C). This pattern is consistent with prior high-latitude studies of C. teretis, which show calcificationclose to equilibrium (Poole, 1994).The observed influence of temperature on foramini-

feral d18O in the Franz Victoria and St. Anna troughsappears to extend over much of the Barents and Karashelves, with the notable exception of the shallower,more river-proximal areas. As shown in Fig. 7, tem-perature more strongly influences isotopic compositionat depths >60m. This figure also shows, however, astronger salinity influence when the data for all depthsare plotted together (upper left scatter plot). This resultis due to the very strong influence of salinity on theisotopic composition of shallow, river proximal waters.A strong salinity influence may have been present in thesubsurface of the Franz Victoria and St. Anna troughsduring deglaciation when nearby ice sheets were meltingand some Eurasian river mouths were displaced north-ward by lowered sea level. In particular, we wouldexpect that the d18O of N. pachyderma (s) would becomeespecially salinity sensitive in the troughs after fresh-water penetrated down to �100–200m depth, thecalcification depth of this species in our study region.Support for this ‘‘depth-threshold’’ theory comes fromthe central Arctic Ocean, where the relatively fresh PolarWater is much thicker than at our core sites and thed18O of modern N. pachyderma (s) is controlled bysalinity rather than temperature (Spielhagen and Erlen-keuser, 1994; Alderman and Lehman, 1995); thecalcification depth is the same in both places (Bauchet al., 1997).

3.3. Modern carbon isotope distributions

Recent studies of the distributions of modernplanktonic d13C in the Arctic (e.g., Spielhagen andErlenkeuser, 1994; Bauch et al., 1997, 2000) provide auseful context for interpreting the N. pachyderma (s)d13C record in our cores. An important result of thesestudies is the understanding that lateral transport ofsurface and pycnocline water from the Eurasian arcticshelves to the central Arctic Ocean results in subsurfacewaters which are better ventilated than inflowing FramStrait Branch Water (FSBW). This ventilation differencecauses N. pachyderma (s) d13C values in central ArcticOcean tow nets and coretops to be �0.5% higher thanin our more southerly region despite much heaviersea-ice cover. Primary productivity may also alter


distributions of planktonic d13C. This factor appears tobe minor to the north of our sites in the southernNansen Basin, as shown by a lack of a d13C increase inN. pachyderma (s) in the upper water column and norelationship between d13C and phosphate concentra-tions (Bauch et al., 2000). Although a strongerproductivity effect might occur at our sites given lesssea-ice cover, it is probably not a dominant control onpresent d13C values because productivity is still lowrelative to more southerly sites (e.g., Rudels et al., 1991).An additional controlling factor on planktonic d13Cdistributions is the low d13CDIC of riverine inflow to theEurasian Arctic, which generally varies between �5 and�10% (Erlenkeuser et al., 1999; Erlenkeuser and VonGrafenstein, 1999). Because this d13CDIC is much lowerthan the �+1% value for Atlantic Water and thesummer discharge of rivers is quite high (Aagaard andCarmack, 1989), mixing between these sources can affectthe d13C of marine foraminifera in freshwater-proximalenvironments. Although this factor probably was notsignificant during the Holocene, it may have beenimportant at times during deglaciation.The environmental controls on the d13C of benthic

foraminifera are more poorly understood than forplanktonics. Nonetheless, they are generally controlled

by the d13C of ambient waters, which in the case of porewaters may be strongly affected by degradation oforganic matter (e.g., Grossman, 1984). The inferredinfaunal habitat of E. excavatum is often associated withorganic-matter decomposition, which shifts the d13Cdisequilibrium to �2 to �3.5% (McCorkle et al., 1990).In contrast, the d13C disequilibrium for the planktonicspecies N. pachyderma (s) is not as pronounced (��0.8to �2%; Labeyrie and Duplessy, 1985; Kohfeld et al.,1996; Bauch et al., 2000). The d13C difference betweenE. excavatum and N. pachyderma (s) in our cores issimilar to other regions, ranging from �1 to �3%.

3.4. Modern distributions of Cassidulina teretis andplanktonic foraminifera

Prevalent benthic foraminifera in our core-tops andcores are the common arctic shelf and/or upper slopespecies Elphidium excavatum, Cassidulina reniforme,Cassidulina teretis, Cibicides lobatulus, Islandiella nor-crossi, Buccella frigida, and Melonis bareleeanus. Forthis study, we concentrate on the percentage of C. teretis(also called C. neoteretis; Seidenkrantz, 1995) relative toall calcareous benthic foraminifera (% C. teretis)because prior core and surficial sediment studies

Fig. 7. Comparison of the d18O of equilibrium calcite versus summer salinity and temperature for four regions of the Barents and Kara seas and

surrounding areas. Lower panel excludes waters above 60m and is more applicable to planktonic and benthic foraminiferal calcification in our

trough study areas. Hydrographic measurements}including the d18O of seawater}are from Ferronskii (1978), Brezgunov et al. (1982), .OOstlund and

Grall (1993), Bauch (1995), Bauch et al. (1995), Mensch et al. (submitted), and Polyak (in prep). Data span the months from June to September and

are irregularly distributed in both space and time (1976, 77, 80, 83, 84, 87, 88, 91, 92, and 98).


demonstrate that this species is common in the Arcticonly where chilled, slightly transformed subsurfaceAtlantic Water is present below �200m depth (Green,

1960; Lagoe, 1979; Mackensen et al., 1985; Mackensenand Hald, 1988; Ishman and Foley, 1996; Jennings andWeiner, 1996; Wollenburg and Mackensen, 1998). In theBarents and Kara seas, C. teretis occurs almostexclusively in Bear Island, Franz Victoria, and St. Annatroughs, all areas with Atlantic Water in the subsurface(Todd and Low, 1980; Polyak, 1985; Mackensen andHald, 1988; Steinsund, 1994; Fig. 8). C. teretis does notoccur in areas where Atlantic Water has been stronglymixed with shelf waters such as in the northeasternBarents Sea. The distribution of C. teretis in the troughsprobably reflects a combination of environmentalfactors associated with Fram Strait Branch Water(FSBW) including its relatively high phytodetritaldeposition, temperature, and salinity.Planktonic foraminifera in the coretops and cores are

almost exclusively of the species N. pachyderma, nearlyall of which are sinistral. The mapped distribution of themodern proportion of planktonic foraminifera relativeto all calcareous foraminifera (% planktonic) resemblesthat discussed above for the proportion of C. teretis(Fig. 8). In general, planktonic percentages are high inthe troughs relative to the shelves and thus, areapproximately coincident with inflow of relativelyunmodified Atlantic-derived waters. Nevertheless, weconsider % planktonic in the troughs to be a less reliableindicator of the presence of FSBW than % C. teretisbecause % planktonic integrates environmental controlsfor both planktonics and benthics, some of which maybe independent.

3.5. Identification of water mass change

Shifts between water masses in Franz Victoria Troughand St. Anna Trough during deglaciation and theHolocene are likely to have been accompanied bydistinctive changes in foraminiferal species and isotopiccompositions. We expect to observe shifts between threemain water masses: Fram Strait Branch Water (FSBW),Barents Sea Branch Water (BSBW), and a freshwater

Fig. 8. Maps of the Barents and Kara seas showing: (A) approximate

shelf area where 200m temperature is >18C, reflecting the occurrenceof relatively unmodified Atlantic Water from both the Barents Sea and

Fram Strait branches, (B) % Cassidulina teretis, and (C) % planktonic

species. C. teretis data are relative to the calcareous benthic

foraminifera, while % planktonic data are relative to all calcareous

foraminifera. The smallest dots show the locations of samples with no

C. teretis (B) and no planktonics (C). Temperature map is from a

recent compilation by Matishov et al. (1998). Most of the foraminiferal

data are from a 10-study compilation for the Barents and Kara seas by

Polyak (1985) and Steinsund (1994). Additional data for the Kara Sea

are from Paulsen (1997) and Polyak (unpublished). Central Arctic

Ocean data are from Wollenburg and Mackensen (1998). Nearly 675

sample sites are shown. Planktonics are dominated by Neogloboqua-

drina pachyderma (s).

3


mass composed of ice-sheet meltwater and/or riverinewater (neither of which occur below the surface watersin the troughs at present). Shifts between the first twowater masses are readily distinguished because therelatively warm FSBW is accompanied by the presenceof C. teretis and relatively low foraminiferal d18O, whilecolder BSBW is accompanied by sparse or absentC. teretis and relatively high d18O. Freshwater flowsvoluminous enough to influence depths down to 100–200m will have low d18O values of N. pachyderma (s) inexcess of those which could be caused by reasonabletemperature increases (assumed to be associated withincreased FSBW); such freshwater flows may also beassociated with low d13C of N. pachyderma (s) due toincreased stratification or proximity to rivers. Theproxies used in this study cannot be used to distinguishice-sheet meltwater from river water, either of whichmay be seasonally stored as sea ice. Nevertheless,independent determinations of ice-sheet margins andriver mouths may provide helpful constraints fordistinguishing these sources.Providing information on the vertical structure of

these three water masses is the d18O difference betweenbenthic and planktonic foraminifera. This differencetends to reflect stratification between the subsurface andbottom portions of the water column (cf. Duplessy et al.,in press). The present d18O difference is mostly due totemperature stratification because of the strong influ-ence of temperature on the d18O composition ofE. excavatum and N. pachyderma (s). Past differencesmay have been more controlled by salinity stratificationor the degree of downward transport of low d18O sur-face water with brines created during sea-ice formation.More problematic to reconstruct are surface-most

waters because none of the proxies used in this studydirectly record surface layer conditions (0–60m). Asdiscussed above, calcification of planktonic foraminiferain Franz Victoria Trough and St. Anna Trough occursin the subsurface at 100–200m depth. Although asubsurface measure, d18O of planktonic foraminiferamay provide indirect information on surface conditions,including the identification of large subsurface fresh-water flows; such flows must have also been present atthe surface.

4. Lithostratigraphy, chronology, and sedimentation rate:

setting a framework for interpretation of foraminiferal

paleoenvironmental proxies

4.1. Lithostratigraphy

Prior detailed lithologic and geophysical studiesdelineated three major seismo- and litho-stratigraphicunits for our cores, which form an upward sequence ofgenerally decreasing proximity to a glacier margin

(Herlihy, 1996; Lubinski et al., 1996; Polyak et al.,1997):The lowest unit, a massive, almost unfossiliferous

gray-to-dark-gray diamicton, was penetrated in onlythree of our six cores: JPC5, PL-07, and PL-29 (Fig. 9).However, seismic data show that the diamicton iswidespread and associated with grounded glacier iceduring the Last Glacial Maximum (Lubinski et al., 1996;Polyak et al., 1997; Kleiber et al., 2000). It is probablycomposed of a spatially varying combination ofsubglacial till, ice-proximal debris flows, and iceberg-turbated till.The middle unit consists of mud, thin sandy layers,

laminated clay, and turbidites. This unit is mostly fossilbarren but contains intermittent abundance peaks. InSt. Anna Trough cores, this unit includes two prominentyellowish- or reddish-brown layers containing iron andmanganese oxides, accompanied by lows in magneticsusceptibility, carbonate, and organic carbon contents.These brown layers provide regional lithostratigraphicmarkers (Polyak et al., 1997), which are particularlyuseful given the paucity of 14C ages in St. Anna Trough.The contact between the diamicton and the middle unitis conformable, consistent with the interpretation of themiddle unit as being deposited in ice-proximal to ice-distal glacial-marine environments.The upper unit is a homogeneous brown-to-olive-gray

mud with generally low sand content and bulk density.It is fossiliferous, bioturbated, and often stained withiron-sulfides. This unit resembles the postglacial intervalin other regions of the Barents Sea, which is associatedwith biologically productive marine conditions, sea-icesedimentation, and re-distribution of sediments bycurrents (e.g., Elverh�i and Solheim, 1983; Elverh�iet al., 1989).

4.2. Chronology and sedimentation rates

The chronostratigraphic framework for the FranzVictoria and St. Anna Trough cores is provided by 10and 12 14C ages, respectively (Table 2). Only fivesamples directly constrain the age of the deglacialperiod, because the middle lithostratigraphic unit isoften fossil barren. Assigning ages to the majorlithostratigraphic units is aided by the occurrence ofprominent unit boundaries and oxidized beds. We inferan age of 513 14Cka for the boundary between thebasal and intermediate unit in the Franz VictoriaTrough and northern St. Anna Trough. The true ageis probably considerably older than 13 14C ka innorthern St. Anna Trough (�15 14C ka?) because�30–50 cm of sediment separate the oldest 14C samples(13.3 14Cka) and the diamicton. This interveningsediment includes a regionally extensive oxidized layer,which probably reflects slow sedimentation (Polyaket al., 1997). In contrast, the oldest 14C samples in


Franz Victoria Trough (12.9 14C ka) lie only a few cmabove a mostly massive diamicton (Lubinski et al.,1996). Thus, the age for the boundary between the basaland intermediate unit in Franz Victoria Trough isprobably younger than in St. Anna Trough. An age ofca 10 14C ka is interpolated from existing 14C ages forthe transition from the intermediate to upper unit inboth troughs.Radiocarbon ages and lithostratigraphy of the degla-

cial sediments (intermediate unit) suggest substantialvariations in sedimentation rate. For example, thesedimentation rate for glacier-proximal clays in FranzVictoria Trough is �100 cm/kyr, about an order ofmagnitude higher than the glacier-distal, coarser-grained sediments.To provide a chronological estimate for undated

deglacial sediments in St. Anna Trough, the 52 cm levelin core PL-07 (a dashed line in Fig. 9) was assigned aca 11 14C ka age based on lithostratigraphic correlationwith the better-dated Franz Victoria Trough deglacialrecord. This level contained sufficient foraminifera for

stable isotopic analysis but not for 14C dating. The52 cm level occurs in the upper portion of the middlelithostratigraphic unit between two distinct sandy layers.Because all deglacial sediments collected in the St. Annaand Franz Victoria trough regions have contained twodistinct sandy layers in a similar stratigraphic sequence(e.g., Lubinski et al., 1996; Polyak et al., 1997; Kleiberet al., 2000), we used the deglacial 14C ages in FranzVictoria Trough core JPC5 and the position of its sandylayers to provide a first-order estimated age of ca 1114C ka for the 52 cm level in core PL-07.Holocene sedimentation rates are better known

and do not show any strong similarities between thetroughs. Early Holocene rates are nearly an order ofmagnitude higher than those of the late Holocene in St.Anna Trough core PL-67 (Table 2), consistent withother cores from southern St. Anna Trough (Hald et al.,1999). In contrast, Holocene sedimentation ratesin Franz Victoria Trough are less variable, withslightly higher rates during the early Holocene andsince ca 3 14Cka.

Fig. 9. Basic lithostratigraphy and correlation of sediment cores from Franz Victoria and St. Anna troughs. Three main units are based on sediment

color, structure, grain-size, bulk density, and magnetic susceptibility. Solid curves=volume magnetic susceptibility (MS); dotted curves=sand

content. Core locations are shown in Fig. 3.


5. Results and interpretation of foraminifera and stable

isotopes: timing of deglacial and postglacial freshwater

and Atlantic Water inflows to Franz Victoria and St.

Anna troughs

5.1. Timing of deglacial inflows

Glacial–marine sediments comprising the deglacialinterval from >13 to ca 10 14C ka are mostly barren offoraminifera, probably due to lowered productivityand/or high dissolution (e.g., Steinsund and Hald,1994). Nevertheless, there are several isolated butimportant abundance peaks, which are roughly centeredon ca 13, 11.5, and 10 14C ka (Fig. 10; Table 2).The oldest abundance peak in our cores, dated ca 13

14C ka, provides evidence for Atlantic Water and fresh-water contributions to Franz Victoria and northernSt. Anna troughs during deglaciation. The dominanceof C. teretis at ca 13 14C ka (Fig. 10) shows a clearpresence of relatively unmodified Atlantic-derived waterin both troughs, probably similar to the present FramStrait Branch Water (FSBW). The planktonic andbenthic foraminiferal d18O records (Figs. 10 and 11A)are dominated by pronounced depletions. To comparethese isotopic values with local values for the present,the influence of global ice-sheet melting on seawaterd18O must be accounted for; we used a d18O decrease of�0.011% per meter rise in global sea level (Fairbanksand Matthews, 1978; Fairbanks, 1989). With thiscorrection, the local d18O values of N. pachyderma (s)and E. excavatum at ca 13 14C ka are 0.8–1.4% lowerthan modern (Fig. 11). These substantially low d18Ovalues are unlikely to solely reflect a temperatureinfluence given that the values equate to temperatures�3–58C higher than present. Such an elevated tempera-ture during deglaciation is inconsistent with environ-mental constraints inferred from benthic foraminiferalassemblages. Nevertheless, a more modest bottomtemperature increase of �1–28C above present isplausible for this time given C. teretis proportions>40%. A smaller difference is possible given uncertain-ties in relating C. teretis and other benthic foraminiferato temperature. Thus, the lower-than-present d18Ovalues may or may not contain a minor temperatureeffect. The values are mostly due to freshwater input; aresult supported by d13C values of N. pachyderma (s)0.6–1.1% lower than present (Figs. 11 and 12). Thisdegree of lowering is in excess of the global effects oncarbon cycling since the Last Glacial Maximum, whichare�0.3–0.4% (Shackleton, 1977; Duplessy et al., 1988;Boyle, 1992, Spero et al., 1997). The low d13C valuesprobably reflect poor ventilation and/or freshwater withlow enough d13C DIC and high enough volume toreduce the overall d13C DIC of seawater. The low d18Ovalues probably also reflect deglacial freshwater end-members with d18O values lower than those of present

riverine inflows, which range from about �14 to �21%(Pechora, Ob, Yenesey, and Lena rivers; e.g., Brezgunovet al., 1982; Bauch et al., 1995).The ca 11.5 14C ka period is distinguished from the

ca 13 14C ka period by having lower % C. teretis and avery low benthic–planktonic d18O difference. TheC. teretis values suggest that FSBW was still present,but may have been less strong than at ca 13 14C ka.Relative to the ca 13 14C ka period, E. excavatum d18Ovalues became �0.4% lighter, while N. pachyderma (s)became �0.3% heavier (Fig. 11). The E. excavatumvalues themselves became 0.4% lower than those for N.pachyderma (s) (see benthic–planktonic d18O differencein Fig. 11). This unusual benthic–planktonic isotopicsignature most likely reflects increased vertical mixing,probably associated with downward transport of waterwith low d18O/high salinity during sea-ice formation(i.e., brine formation). This result does not necessarilyimply that the benthic foraminifera experienced lowerd18O of ambient seawater than their planktonic counter-parts because the vital-effect uncertainties for bothspecies are large enough to explain the �0.4%difference. At ca 11.5 14C ka, freshwater must have beenthe dominant influence because all d18O values remainedwell below the present.By ca 10 14C ka, benthic foraminifera were again

present in trough sediments. The proportions ofC. teretis were >40%, while d18O values were wellbelow present. These conditions were similar to those atca 13 14C ka, with inflows of both Atlantic-derived waterand freshwater.

5.2. Timing of Holocene (postglacial) inflows

The relatively rapid transition to post-glacial condi-tions soon after ca 10 14C ka is one of the strongestsignals in our records and is coincident with alithological change to sulfide-stained, olive-gray mud(Fig. 9). The d18O of N. pachyderma (s) increased by 1%in less than �400 14C years, while that of E. excavatumincreased by 0.7%. The d13C of N. pachyderma (s) alsorose relatively quickly. Combined with a simultaneousdecrease in the proportion of C. teretis, these isotopicshifts reflect a relatively rapid cessation of freshwaterand Fram Strait Branch Water (FSBW) flows at asimilar time in both troughs (Fig. 12). These shifts alsoimply that subsurface and bottom temperatures fellrelatively quickly. These cold, early Holocene conditionspersisted until ca 7.5 14C ka in both troughs and show astrong influence of cold, highly modified water from theBarents Sea branch (BSBW). Carbon isotopic values forN. pachyderma (s) are lower than the Holocene average,suggesting relatively poor ventilation and/or lowerproductivity.The succeeding interval from ca 7.5 to 6 14C ka

is characterized by low d18O of N. pachyderma (s) and


high % C. teretis, due to return of warm FSBW. The�0.3–0.5% decrease in planktonic d18O between ca 8and 7 14C ka suggests that subsurface temperatures may

have risen by �1–28C. The magnitude of this change isconsistent with a replacement of BSBW by FSBW. Arelatively low benthic–planktonic d18O difference in

Fig. 10. Foraminiferal time-series data from: (A) Franz Victoria Trough cores PG5 and JPC5, and (B) St. Anna Trough cores GF-163, PL-67, PL-

07, and PL-29. Average planktonic values are shown with a thick connecting line, benthic values with a thin line. For comparison, the isotopic effect

of global sea-level rise (Fairbanks and Matthews, 1978; Fairbanks, 1989) on the d18O of N. pachyderma (s) is shown. The global sea-level effect on

d18O of E. excavatum is not shown but is a simple offset of the depicted dashed line. Horizontal shading highlights intervals with especially low d18Oin N. pachyderma (s). To place the d13C of N. pachyderma (s), E. excavatum, and C. teretis on the same scale, 1.5% has been added to the

E. excavatum values and 0.5% to the C. teretis values. Similarly 0.5% has been subtracted from the d18O of C. teretis to plot them with

N. pachyderma (s) data. Especially small samples for N. p. (s) (2–4 specimens) in SAT are highlighted (6 for PL-67, 1 for GF-163).


Franz Victoria Trough during the ca 7 14C peak of thisinterval (Fig. 11) suggests that temperature stratificationwas low and that FSBW may have expanded to the seafloor. In contrast, a relatively high benthic minusplanktonic d18O difference in southern St. Anna Troughimplies that FSBW generally overlay BSBW. By ca 6 ka,

temperatures had fallen in both troughs due todecreased FSBW influence and/or increased BSBWinfluence.The period from 6 to 3 14C ka is a period of transition,

reflecting changes in the relative influence of FSBW andBSBW in each trough. Oxygen isotopic values of

Fig. 11. Averaged and corrected foraminiferal time-series data from the Franz Victoria Trough (thick line) and St. Anna Trough (thin line). The

d18O data have been corrected for the influence of global sea-level rise (see text). Horizontal shading highlights intervals with especially low d18O in

N. pachyderma (s).


N. pachyderma (s) and E. excavatum in the two troughsdiverged at ca 6 and 4 14C ka, respectively, leading tosystematically lower d18O values in Franz VictoriaTrough than southern St. Anna Trough during the lateHolocene. This isotopic separation suggests a transitionfrom a shared dominance of mostly FSBW in subsurfacewaters ca 7 14C ka to a stronger influence of FSBW inFranz Victoria Trough than southern St. Anna Trough.After the transitions, oxygen isotopic values ofN. pachyderma (s) and E. excavatum in Franz VictoriaTrough were �0.4% lower than in southern St. AnnaTrough, equivalent to subsurface and bottom watertemperatures �1.58C higher. This interpretation isconsistent with higher % C. teretis in Franz VictoriaTrough than southern St. Anna Trough. The modernhydrographic data in Figs. 4 and 5 show that tempera-tures in Franz Victoria Trough are indeed higher thannear core GF-163 in the southern St. Anna Trough, butthe difference is only �0.758C at the seafloor (Fig. 5).This apparent discrepancy suggests that the hydro-graphic data collected in 1998 for southern St. AnnaTrough may not be representative of recent averageconditions, which are probably colder. The alternativeexplanation}that seawater d18O values in the southernSt. Anna Trough are systematically and substantiallyhigher than in Franz Victoria Trough}is unlikely giventhat available seawater isotopic data suggest the

opposite trend. Temperature stratification was probablylow in southern St. Anna Trough and high in FranzVictoria Trough (benthic minus planktonic d18O differ-ence data), implying that the water columns differedgreatly. BSBW in St. Anna Trough was probablydominant in the subsurface and bottom while it wasmore dominant only at the bottom of Franz VictoriaTrough.Carbon isotopic values for N. pachyderma (s) during

the ca 6–3 14C ka transition period were at theirHolocene maximum, suggesting a relatively well-venti-lated water mass and/or high productivity. The generalincrease in d13C values through the early Holocene up tothis interval (Fig. 11) may reflect ventilation conditionsover a broad area. One possibility is that sea-level rise inunglaciated regions of the Kara and Laptev Sea resultedin a progressive increase in shelf area, increasing thetransport of well-ventilated shelf waters to the ArcticOcean until sea-level stabilization ca 6 14C ka (cf., Bauchet al., 1999).The interval from ca 3 to 2 14C ka is characterized by

relatively low d18O values and high % C. teretis,suggesting an increase in the influence of FSBW in bothtroughs. This result is supported by generally higher %planktonics at this time. d13C values N. pachyderma (s)show the start of a trend toward slightly lower valuesthat extends to the present. This suggests that ventilation

Fig. 12. Summary data and basic interpretation of our Franz Victoria and St. Anna trough records.


and/or productivity might have started to decreaseslightly during the latest Holocene.The interval from ca 2 14C ka to the present is

characterized by generally low d18O values, but higherthan at ca 3–2 14C ka. Combined with a decrease in %C. teretis, the isotopic values suggest a reduction ofFSBW influence in both troughs. However, % plank-tonic generally increased, reaching maximum values forthe entire record at present (Fig. 10). The cause for thismaximum is still not known.

6. Discussion: comparison with other records and causes

for inflow variations

6.1. Deglacial ice margins, stable isotopes, andfreshwater flow sources

During the Last Glacial Maximum ca 20 14C ka, theBarents Sea Ice Sheet covered the shelf out to its edgealong much of the northern and western margins,grounding in the deep shelf troughs while much of theKara Sea probably remained unglaciated, especially inthe south (e.g., Landvik et al., 1998; Forman et al., 1999;Mangerud et al., 1999; M .ooller et al., 1999; Polyak et al.,2000a). If a northern Kara Sea ice sheet existed, it mayhave blocked drainage of the large Ob’ and Yeniseirivers}thus creating the potential for large, river/lakedischarges that may have traveled to Franz Victoria andSt. Anna troughs during deglaciation. Regardless of theice sheet geometry in the Kara Sea, the Barents Sea IceSheet would have blocked any northward AtlanticWater flow via the Barents Sea branch. Re-establish-ment of this important branch required deglaciation ofthe deep troughs and portions of the shallower centralBarents Sea. The re-establishment probably occurredrelatively early, between ca 15 and 13 14C ka}aparticularly data sparse interval for the Barents andKara seas.At ca 15 14C ka, retreat began along the northern and

northwestern Last Glacial Maximum margin of theBarents Sea Ice Sheet (Svendsen et al., 1996; Landviket al., 1998; Andersson et al., 1999; Kneis et al., 1999,2000; Kleiber et al., 2000; see Fig. 13). New isotopicstudies of Arctic Ocean cores PS2138 and PS2446 from995 and 2022m depths adjacent the Franz VictoriaTrough suggest the first major pulse of freshwater alongthe northern margin of the Barents Sea Ice Sheet atroughly ca 14 14C ka (Kneis et al., 2000), generallyconsistent with isotopic records from the central ArcticOcean (Stein et al., 1994; N�rgaard–Pedersen et al.,1998; Fig. 14C and D).By ca 13 14C ka, the ice-sheet had retreated to near-

coastal areas in western and northern Svalbard butprobably still covered shallow areas of the northernBarents Sea (e.g., Forman, 1990; Elverh�i et al., 1995;

Lierdal, 1997; Landvik et al., 1998; see Fig. 13). Ourforaminiferal d18O and d13C data show substantialfreshwater influx to the already deglaciated FranzVictoria and northern St. Anna troughs. This ca 1314C ka freshwater influx was probably derived fromnearby melting glacier ice, although additional influencefrom the Ob’, Yenisei, and other Eurasian rivers cannotbe ruled out. Riverine influences may have includeddamming by a Kara Ice Sheet or progradation of riversto locations closer to the troughs. Progradation mayhave occurred given a global sea level �100m belowpresent at 13 14C ka (Fairbanks, 1989) and unglaciated,shallow shelves in the Laptev Sea (Bauch et al., 1999)and the eastern Kara Sea.By ca 11.5 14C ka, the ice sheet had fragmented into

separate ice caps over the Svalbard, Franz Josef Landand Novaya Zemlya regions (e.g., Elverh�i et al., 1995;Landvik et al., 1998; see Fig. 13). Freshwater waspresent in Franz Victoria and St. Anna troughs ca11.5 14C ka. The apparent proximity of remainingglacier ice to the troughs suggests that it may have beenthe main source of the d18O depletion. A riverine source,however, is also possible given that preliminary analysisof a high-resolution Arctic Ocean record from theunglaciated Laptev Sea margin implies a very largefreshwater discharge into the Arctic Ocean just beforethe start of the Younger Dryas (PS2458, 985m,Spielhagen et al., 1998; see Fig. 1 for location).The ca 10 14C ka period in Franz Victoria and

St. Anna troughs, like the ca 13 and 11.5 14C ka periods,was also characterized by depleted d18O values reflectingfreshwater inputs. Although little of the Barents Sea iceSheet remained at ca 10 14C ka, melting of its remnantsmay have been especially vigorous from ca 10.2 to 9.514C ka when abrupt arctic warming occurred during theYounger Dryas–Preboreal transition (e.g., Ko

-c et al.,

1993; Cuffey and Clow, 1997; Bj .oorck et al., 1997; Haldand Hagen, 1998). It is also possible that a pronouncedriver discharge from northward displaced Eurasianrivers contributed to the depleted d18O values. Thislatter possibility is supported by low d18O during theearliest Holocene in an apparently more river-proximalcore from 125m water depth in the southwestern KaraSea (Core L0987; Polyak et al., 2000b; see ‘‘R9’’ inFig. 1 for location). Regardless of the source(s), the ca10 ka freshwater pulse appears to be restricted to theBarents and Kara seas region. To our knowledge, it hasnot been observed in the northeastern Nordic Seas, theArctic Ocean, or the Laptev Sea continental margin.The only cores showing relatively low or decreasingd18O at this time, such as core M17732 from the easternNordic Seas (Bauch and Weinelt, 1997; see Fig. 1 forlocation), are accompanied by increased d13C, suggest-ing a lack of freshwater in these regions. It is unlikelythat our ca 10 ka event (dated 10.2 14C ka in JPC5) isrelated to low d18O observed during the short-lived


9.8 14C Preboreal oscillation in the Nordic Seas (e.g.,Bj .oorck et al., 1997; Hald and Hagen, 1998) because theoscillation appears to have been associated with areduction in the strength of northward Atlantic Waterflow; this contrasts with the clear Atlantic influence(FSBW) during our event. Nevertheless, we cannot ruleout the possibility that these events are related given therarity of deglacial samples in our cores, poorly knownreservoir ages, and the prominent 14C plateaus at thistime.

The Franz Victoria and St. Anna troughs andsurrounding marine areas were completely deglaciatedby no later than ca 10 14C ka. Fully marine sedimentappeared in our cores soon after ca 10 14C ka andin others throughout the Barents Sea by no later thanca 9.5 14C ka (e.g., Elverh�i et al., 1990; Polyaket al., 1995). In addition, glacial-isostatically raisedbeaches had already begun to form on Franz Josef Landand eastern Svalbard by ca 10.3 14C ka (Bondevik et al.,1995; Forman et al., 1997).

Fig. 13. Hypothesized freshwater and Atlantic Water flows to the Franz Victoria (FVT) and St. Anna (SAT) troughs for seven time periods since the

last deglaciation. Environmental conditions in the troughs for each time period are listed in the lower right boxes, where FSBW=Fram Strait Branch

Water and BSBW=Barents Sea Branch Water. Potential freshwater sources include nearby ice sheets, whose margins have been modified from

Lambeck (1996); Landvik et al. (1998); Forman et al. (1999); Svendsen et al. (1999), and Polyak et al. (2000a). The margins are particularly poorly

known in the northern Kara Sea. Additional freshwater sources include river outflows from the voluminous Ob’ and Yenisei rivers, whose mouths

probably prograded northward when global sea level was lower. Past river mouth positions in the Kara Sea are approximated using global sea level

from Fairbanks (1989) and bathymetric charts, assuming that the rivers are sufficiently distant from ice loads to have escaped substantial isostatic

depression. We hypothesize that glacial-isostatically induced bathymetric changes may have altered Atlantic Water flow. Past water depths in the

northern Barents Sea are estimated using an Earth rheology-based model (Lambeck, 1996, BK-3 model) and relative sea level data and are shown as

deviations from present depth at 25-m intervals; positive numbers indicate deeper-than-present waters. Depth deviations at 15, 13, and 11.5 ka are

modified from model results at 18, 15, 12, and 10 ka and are only approximate because deglaciation probably proceeds too quickly in this model.

Depth deviations for St. Anna Trough are not shown because BK-3 does not include grounded ice there, a result contradicted by field data collected

after the model run (Polyak et al., 1997). Depth deviations at 10, 9 and 7 ka are based on measured emerged beaches on eastern Svalbard and Franz

Josef Land (e.g., Bondevik et al., 1995; Forman et al., 1997; Lambeck, 1996). Present isobaths at 500 and 1000m are shown, but not labeled.


6.2. Deglacial biotic data and Atlantic Water inflows

Fram Strait Branch Water (FSBW) probably flowedinto the Arctic Ocean intermittently during early phasesof deglaciation between ca 15 and 13 ka. Studies of coresfrom offshore the northern margin of the Barents SeaIce Sheet even suggest that some FSBW flow may haveexisted near the Last Glacial Maximum (Kneis et al.,1999). This result is supported by the presence of someAtlantic Water further south and ‘‘upstream’’ in theNordic Seas ca 19.5–14.5 14C ka (e.g., Hebbeln et al.,1994; Dokken and Hald, 1996; Bauch et al., in press).Evidence for the periodic nature of deglacial Atlanticflow includes prominent decreases at 14.5 (�Heinrichevent H-1) and 13.5 14C ka in the Nordic Seas (e.g.,Keigwin et al., 1991; Veum et al., 1992).The presence of Cassidulina teretis in Franz Victoria

Trough and St. Anna Trough at ca 13, 10 14C ka, and to

a lesser extent at ca 11.5 14C ka in Franz VictoriaTrough, clearly indicates that flow of FSBW to the highArctic occurred during later phases of deglaciation; aresult supported by Atlantic Water inflows to theNordic Seas and independent evidence from the ArcticOcean. A strong deglacial flux of Atlantic Wateroccurred in the southeastern Norwegian Sea ca 1314C ka, linked with the start of deep convection and theappearance of at least seasonally ice-free conditions(e.g., Lehman and Keigwin, 1992; Veum et al., 1992;Koc et al., 1993; Fig. 14K). Northward transport of thisAtlantic Water flux into the Arctic Ocean via at least theFram Strait branch is supported by our results in FranzVictoria and St. Anna troughs as well as ostracod datafrom the central Arctic Ocean (Cronin et al., 1994,1995). FSBW influence in the high Arctic during somedeglacial intervals may have been similar to or greaterthan present given ostracod-based reconstructions of a

Fig. 14. Comparison of foraminiferal records from the Franz Victoria and St. Anna troughs to paleoenvironmental records from the Barents Sea,

central Arctic Ocean, and Nordic Seas. Data sources besides this paper are: (A) PS1295-4, Jones and Keigwin, 1988; (C) PS2170, Stein et al. (1994);

(D) PS2177, N�rgaard–Pedersen et al. (1998); (E) Fairbanks (1989); (F) Berger (1978); (G) Lubinski et al. (1999); (H) modified from Salvigsen et al.

(1992) by Lubinski et al. (1999); (I) PS21842-5, Koc et al., 1993; (J) HM52-43, Koc et al., 1993, (K) Troll 3.1, Lehman and Keigwin (1992); and (L)

Keigwin et al. (1991). For location of all core sites, see Fig. 1.


thicker-than-present FSBW layer in the Arctic Oceanand very high % C. teretis in the troughs. In contrast,the Atlantic Water influence was clearly curtailed duringthe Younger Dryas in the Nordic Seas (e.g., Koc et al.,1993). None of our trough samples date to the YoungerDryas period, suggesting reduced productivity and/orhigh dissolution. The subsequent return of AtlanticWater to the high Arctic is shown by a high proportionof C. teretis (and low d18O values) in Franz Victoria andSt. Anna troughs ca 10 14Cka.

6.3. Comparison of Holocene (postglacial) records inFranz Victoria Trough

Similarity between our JPC5/PG5 composite record(463m water depth) and nearby records of shorterduration (Polyak and Solheim, 1994; Duplessy et al.,in press) shows that JPC5/PG5 is broadly representativeof overall Atlantic Water inflows to the Franz VictoriaTrough during at least the Holocene. In particular,JPC5/PG5 is similar}except for the last three millen-nia}to a new high-resolution trough isotopic recordthat extends from ca 8.7 14C ka to the present (Duplessyet al., in press; core ASV-880; see Fig. 3 for corelocation). Both JPC5/PG5 and ASV-880 contain aprominent low d18O period centered ca 7 14C ka but itsmagnitude is higher in ASV-880 (Fig. 15). This differ-ence is probably at least partly due to a sedimentationrate about twice higher in ASV-880 than in JPC5/PG5 atca 7 14C ka (less bioturbation). The especially strong E.excavatum signal in ASV-880 may also be related to theshallower water depth at ASV-880 (388m); benthicforaminifera at this site lie about 75m higher in thewater column and closer to the warm core of FSBWthan they do in JPC5/PG5. The ASV-880 record alsoshows distinctly lower d13C values during the short-livedlow d18O period centered ca 7 14C ka, which is consistentwith our interpretation of the presence of less-wellventilated FSBW during this period. The higher resolu-tion ASV-880 record appears to show higher frequencyevents such as a 8.2CALka event (�7.6 14C ka) thatbriefly interrupted early Holocene warming in at leastthe Nordic Seas and North Atlantic region (e.g., Alleyet al., 1997; Barber et al., 1999). The main differencebetween the records is for the period fromca 3 14C ka to the present when JPC5/PG5 has generallylower d18O values than ASV-880 (�0.1–0.3% offset).The pattern of isotopic change in each record suggeststhat FSBW had a stronger influence at JPC5/PG5 thanASV-880 over much of this Late Holocene period.

6.4. Holocene (postglacial) Atlantic Water inflows

Very cold subsurface and bottom waters in the FranzVictoria Trough and southern St. Anna Trough fromca 9–7.5 14C ka occurred when surface conditions in the

eastern Nordic Seas and at least the northwesternportion of the Barents Sea were similar to or warmerthan present. The transition to this period is character-ized by an increasing influence of cold BSBW in FranzVictoria and St. Anna troughs while the eastern NordicSeas SST records show a clear warming trend inter-rupted only by the short Preboreal oscillation at 9.814C ka (e.g., Lehman and Keigwin, 1992; Koc et al.,1993; Bj .oorck et al., 1997; Hald and Hagen, 1998;Fig. 13). SSTs in the eastern Nordic Seas had reachedpresent values by ca 9.5 14Cka, followed by the abovepresent values in most areas until at least the middleHolocene (e.g., Koc et al., 1993; Fig. 14J). Sea-surfaceand air temperatures at the Svalbard shore were at theirHolocene maxima from ca 9–7.5 14C ka and remainedabove present until ca 5 ka (Birks, 1991; Salvigsen et al.,1992; Hjort et al., 1995; Wohlfarth et al., 1995;Fig. 14H). Furthermore, glaciers on Svalbard and FranzJosef Land had retreated to positions at or behindpresent limits by ca 10 ka and remained retracted untilthe Late Holocene, probably due to the high summertemperatures (Svendsen and Mangerud, 1997; Lubinskiet al., 1999; Fig. 14G).Relatively warm FSBW returned to Franz Victoria

and St. Anna troughs during the ca 7.5–6 14C kainterval, generally coinciding with maximum AtlanticWater influence in the Nordic Seas. In particular,diatom assemblages throughout the Nordic Seas showmaximum Atlantic Water influence ca 8–5 14Cka (e.g.,Koc et al., 1993). Eastern Nordic Seas SST reconstruc-tions (based on diatoms) show warm, stable conditionsfor a long period from ca 9.5–4 14C ka. In contrast,western Nordic Seas SST reconstructions show a shorterduration peak ca 7.5 14C ka due to westward migrationof Atlantic Water into a region formerly (and subse-quently) dominated by Polar Water (Fig. 14I). Thetiming of this peak is roughly similar to the ca 7 14C kaFSBW maxima in the Franz Victoria and St. Annatroughs. Data from core ASV-880 suggest a duration ofabout 1000 yr for this maxima in Franz Victoria Trough(Duplessy et al., in press).The middle Holocene period from ca 6–3 14C ka is

an interval of transition in Franz Victoria Trough,St. Anna Trough, and adjacent regions. This includes ashift toward a higher influence of FSBW in FranzVictoria Trough and BSBW in southern St. AnnaTrough. This period also encompasses a transition toreduced Atlantic Water influence and lowered SSTs inmuch of the Nordic Seas (e.g., Koc et al., 1993).Although the late Holocene period from ca 4 14C ka

to present is characterized by decreased Atlantic Waterinfluence in surface waters of the Nordic Seas andcoastal Svalbard, the trough records suggest that inflowof FSBW may have increased at times, particularlybetween ca 3 and 2 14C ka. This result is not surprisinggiven that Franz Victoria Trough has a strong FSBW


Fig. 15. Franz Victoria Trough foraminiferal isotopic records for our core JPC5/PG5 and core ASV-880 (Duplessy et al., in press). Unlike the

previous figures, the time-series data are shown against calendar age and are truncated near the Holocene boundary. The d18O data have been

corrected for the influence of global sea-level rise in slightly different ways. An approximate temperature scale from Duplessy et al. (in press) is shown

for the d18O data, which is consistent with modern hydrography data (�+18C at 100–200m depth and�08C at the seafloor) and our interpretation

of strong temperature control on foraminiferal d18O composition during the Holocene. Horizontal shading shows the intervals with especially low

d18O in N. pachyderma (s) noted earlier in our Franz Victoria and St. Anna trough cores. ASV-880 shows no evidence for the youngest interval. It

does show, however, a very well defined event with low d18O and low d13C centered ca 7.5CALka.


signature at present. Our record, however, does appearto conflict with a commonly held view that the warmestsurface conditions in the Nordic Seas ca 8–5 14C ka weretimes of strongest deep-water formation and meridionalAtlantic Water flow. This view has been questionedrecently by Bauch et al. (in press), who suggest that verystrong convection occurs after ca 5 14C ka. In addition,there are a few records ‘‘upstream’’ in the Nordic Seasshowing periods of enhanced Atlantic Water influenceduring the late Holocene (e.g., core T-88-2; Hald andAspeli, 1997; see Fig. 1 for location).

6.5. Causes for variations in Atlantic Water flows to thetroughs?

Deciphering the causes for changing Atlantic Waterflows to the troughs is not easy given that the core sitehydrographies are strongly controlled by mixtures of thewarm Fram Strait and cold Barents Sea branches ofAtlantic Water and that changes in these mixtures mayreflect changes in either the flux or the hydrographiccharacteristics (i.e., temperature, salinity) of eitherbranch. These variations, in turn, may reflect changesin a number of environmental factors on global toregional scales, including the strength of global deep-water production/thermohaline circulation and regionalpatterns of atmospheric and ocean circulation. Moderndata- and model-based studies show that certain atmo-spheric circulation patterns, such as the North AtlanticOscillation (NAO) and Arctic Oscillation (AO), alter theflow of Barents Sea Branch Water and that feedbackmechanisms with sea ice, temperature, and air pressureact to enhance the flow changes (e.g., Adlandsvik andLoeng, 1991; Loeng, 1991; Zhang et al., 1998). Alsoimportant is the substantial variation in summerinsolation at high latitudes, which peaked at more than10% higher than modern during deglaciation (Berger,1978; Mitchell et al., 1988; TEMPO, 1996; Kerwin et al.,1999; Fig. 14F). This insolation change no doubtcontributed to altered circulation patterns, even duringthe Holocene (c.f., Harrison et al., 1992). The highinsolation seasonality (high summer values, none inwinter) probably also altered sea-ice cover in the Barentsand Kara Seas. Increased seasonality may have alsoincreased production of sea-ice brines, which wouldalter BSBW characteristics and flow. Another importantfactor for understanding past Atlantic Water flow isglacier-isostatically and eustatically induced variationsin bathymetry. Because this factor has seldom beenstudied but is important for paleoceanographic recon-struction, we discuss it in detail in the next section.

6.6. Bathymetric controls on Atlantic Water inflows?

We hypothesize that Barents Sea Branch Water(BSBW) inflows to Franz Victoria and St. Anna troughs

were influenced by glacial-isostatically induced bathy-metric changes in the Barents Sea and, potentially, byeustatically driven bathymetric changes elsewhere in theArctic. The Barents Sea is the only continental shelf innorthern Eurasia that sustained a 2+ km-thick ice sheetduring the last glaciation (Forman et al., 1995;Lambeck, 1996; Svendsen et al., 1999; Howell et al.,2000). Emergence patterns constrain the largest LastGlacial Maximum ice load in the Barents and Kara seasto the north central Barents Sea (Lambeck, 1996;Forman et al., 1997; Zeeberg et al., submitted). Muchof the deglaciation under this load probably took placesometime between ca 13 and 10 14C ka (possibly ca 13–11.5 14C ka), while deglaciation of the thinner ice in thesouthern Barents Sea mostly occurred between ca 16and 13 14C ka (e.g., Vorren et al., 1988; Polyak et al.,1995; Vorren and Laberg, 1996; Forman et al., 1997;Landvik et al., 1998). During the ca 13–10 14C kadeglacial interval in the troughs, glacial-isostatic depres-sion resulted in a much higher-than-present relative sealevel over the northern shelf area; this shelf wasprobably deepened by at least 200m at the start ofdeglaciation (Lambeck, 1996; Howell et al., 2000;Fig. 13). In contrast, the relatively thin ice and earlydeglaciation in the southern Barents Sea resulted in asimilar-to or shallower-than-present bathymetry at ca13–10 14C ka. The Bear Island Trough, however, wouldhave remained sufficiently deep (�400m) to permitAtlantic Water to enter the Barents Sea relativelyunhindered. Even during the early Holocene, thenorthern shelf was up to �100m deeper than present(Bondevik et al., 1995; Forman et al., 1997; Landviket al., 1998; Fig. 14). The northern shelf progressivelyshoaled, with �90% of the rebound complete by ca 614C ka (Forman et al., 1997).At present, sills southwest of Franz Victoria Trough

are too shallow by �50–150m to permit unhinderednorthward flow of BSBW into the Arctic Ocean (Pfir-man et al., 1994; Fig. 2). These sill depths combined withthe scale of glacial-isostatic deepening of the northernBarents Sea shelf suggest that the deepening wasprobably sufficient to permit Atlantic Water travellingthrough the Barents Sea to easily reach the FranzVictoria Trough, a pattern which does not occur today.Such a transit path may help explain why colder-than-present subsurface and bottom waters in Franz VictoriaTrough from ca 9–7.5 14C ka occurred when warmer-than-present conditions were prevalent upstream in theeastern Nordic Seas (e.g., Koc et al., 1993). Continuedshoaling may have acted to progressively restrict theflow of BSBW to Franz Victoria Trough until relativesea levels were stabilized by ca 6 14Cka. Such arestriction may have enhanced the ca 7 14C ka FSBWmaximum and the ca 6–3 ka divergence of the d18Osignals in Franz Victoria and St. Anna troughs.Separating the glacier-isostatic affects from other


forcings}such as insolation and the strength ofthermohaline circulation}will require more data andnew model simulations. For example, simulations forthe period of cold conditions in the troughs ca 9–7.514C ka may shed light on whether the BSBW flows aredue mostly to glacier-isostatically induced circulationchanges, high insolation seasonality with high sea-icebrine production, or other factors.Atlantic Water inflows to the troughs may also have

been affected by bathymetric changes elsewhere in theArctic (i.e., Nares Strait and other eastern CanadianArctic Channels, Bering Strait). Of these, the opening ofBering Strait ca 11 14C ka might be the most importantbecause it introduces relatively fresh Pacific waters intothe Arctic Basin. Preliminary modeling experimentsshow that opening of Bering Strait strengthens mer-idional Atlantic Water circulation through the FramStrait, possibly at the expense of Barents Sea inflow(Forman et al., 2000).

7. Conclusions

Our study of Franz Victoria and St. Anna Troughsediments combines AMS 14C-constrained foraminiferaldata from five cores that delimit inflows of freshwaterand Atlantic Water to the deep northern Barents andKara seas from the last deglaciation to present. The coldBarents Sea Branch of Atlantic Water}which presentlycarries �1/3 to 1/2 of the Atlantic Water into the ArcticOcean}would have been blocked by the Barents Sea IceSheet during the Last Glacial Maximum. Re-establish-ment of this important branch required deglaciation ofthe deep troughs and portions of the shallower centralBarents Sea. Although the re-establishment age remainspoorly known (between 15 and 13 14C ka?), initialdeglaciation of the outermost troughs probably beganca 15 14C ka. The inner Franz Victoria Trough,however, may not have been deglaciated until closer toca 13 14C ka.Foraminifera were common in the troughs during

three deglacial intervals centered at ca 13, 11.5, and 1014C ka. The d18O and d13C of Neogloboquadrinapachyderma (s) and Elphidium excavatum were up to1.4% lower than present at these times (after global sea-level correction); indicating substantial freshwater in-puts that influenced both the subsurface and bottom.Freshwater sources likely included still unknown com-binations of ice-sheet melt and river inflow. Rivers in theKara and Laptev seas may have had particularly stronginfluences when their mouths may have been muchfurther north due to lowered sea level. Cassidulina teretiswas more abundant than present ca 13 and 10 14C ka,showing deglacial penetrations of the trough byAtlantic-derived water, similar to the present FramStrait Branch Water (FSBW). These warm-water

penetrations probably contributed to the low d18Ovalues. Substantial freshwater input ca 10 14C ka (dated10.2 14C ka in core JPC5/PG5) may have been unique tothe northern Barents and Kara seas region; nocorrelative freshwater signal has been identified in theNordic Seas and Arctic Ocean.The transition to postglacial marine conditions in

Franz Victoria and St. Anna Troughs is marked by arapid�1% rise in foraminiferal d18O and a sharp fall in% C. teretis soon after ca 10 14Cka. The isotopicincrease reflects the cessation of the last large freshwaterflow to the troughs and reduced temperatures associatedwith decreased relative influence of warm FSBW. FSBWinfluence remained minor from soon after ca 10 14C kauntil ca 7.5 14C ka, indicated by a near absence ofC. teretis and relatively high d18O values. These coldconditions reflect an enhanced influence of Barents SeaBranch Water (BSBW), possibly due to glacial-isostati-cally induced deepening of the northern Barents Sea by�100m, changes in atmospheric circulation that pre-ferentially steered Atlantic Water into the southwesternBarents Sea, and/or increased generation of sea-icebrines. Subsequent isotopic and foraminiferal assem-blage variations show a return of relatively warm andunmodified FSBW by ca 7.5 14C ka, associated with theAtlantic Water maximum ‘‘upstream’’ in the NordicSeas and possibly with reduced glacial-isostatic shoal-ing. This maximum FSBW period was short lived(�1000 yr) and followed by a middle Holocene shifttoward a higher influence of relatively warm FSBW inFranz Victoria Trough and colder BSBW in southern St.Anna Trough. We hypothesize that this shift may atleast partly reflect continued shoaling, which reducedflow of BSBW to Franz Victoria Trough but did notgreatly affect its path to St. Anna Trough. Thesubsequent Late Holocene period includes a smallFSBW maximum in Franz Victoria Trough, and to alesser extent in St. Anna Trough, ca 3–2 14C ka.

Acknowledgements

This research was supported by National ScienceFoundation awards OPP-9529133, OPP-9529243, andOPP-9725418. Much of the data compilation occurredwhile Lubinski was a Post Doc at Byrd Polar ResearchCenter. Our isotopic samples were measured by RindyOstermann (Woods Hole Oceanographic Institution).Most of the Franz Victoria Trough foraminiferal speciescounts were performed by Sergey Korsun}formerly ofMurmansk Marine Biological Institute, now at theShirshov Institute of Oceanology. Preparation of someof the St. Anna Trough samples was provided byFrancis Snyder (nee Herlihy). Modern seawater datawere kindly provided by Charlene Grall (RosenstielSchool of Marine and Atmospheric Science), Dorothea


Bauch (GEOMAR), Manfred Mensch (Universitt Hei-delberg), and Vladimir Stanovoy (Arctic and AntarcticResearch Institute, AARI). Jean-Claude Duplessyshared his new isotopic data from core ASV-880with us. The following people gave helpful reviews:John Andrews (INSTAAR), Dorothea Bauch, andtwo anonymous reviewers for the journal. To allthese people and institutions we offer our heartfeltthanks.

References

Aagaard, K., Carmack, E.C., 1989. The role of sea ice and other fresh

water in the arctic circulation. Journal of Geophysical Research 94,

14,485–14,498.

Adlandsvik, B., Loeng, H., 1991. A study of the climatic system in the

Barents Sea. Polar Research 10, 45–49.

Aksu, A.E., Vilks, G., 1988. Stable isotopes in planktonic and benthic

foraminifera from Arctic Ocean surface sediments. Canadian

Journal of Earth Sciences 25, 701–709.

Alderman, S.E., Lehman, S.J., 1995. Correlation of coretop delta

(super 18) O and modern salinity gradients in the Arctic Ocean.

AGU 1995 Spring Meeting, Eos, Vol. 76, No. 17 Suppl., American

Geophysical Union, Baltimore, 171pp.

Alley, R.B., Mayewski, P.A., Sowers, T., Stuiver, M., Taylor, K.C.,

Clark, P.U., 1997. Holocene climatic instability: a prominent,

widespread event 8200 yr ago. Geology 25, 483–486.

Andersson, T., Forman, S.L., Inglfsson, O., Manley, W.F., 1999.

Late Quaternary environmental history of central Prins Karls

Forland, western Svalbard. Boreas 26, 292–307.

Barber, D.C., Dyke, A., Hillaire-Marcel, C., Jennings, A.E.,

Andrews, J.T., Kerwin, M.W., Bilodeau, G., McNeely, R., Southon,

J., Morehead, M.D., Gagnon, J.-M., 1999. Forcing of the cold event

of 8,200 years ago by catastrophic drainage of Laurentide lakes.

Nature 400, 344–348.

Barry, R.G., 1989. The present climate of the Arctic Ocean and

possible past and future states. In: Herman, Y. (Ed.), The Arctic

Seas. Van Nostrand Reinhold Co., New York, pp. 1–46.

Bauch, D., 1995. The distribution of d18O in the Arctic Ocean:

implications for the freshwater balance of the halocline and the

sources of deep and bottom waters. Reports on Polar Research 159,

144.

Bauch, D., Carstens, J., Wefer, G., 1997. Oxygen isotope composition

of living Neogloboquadrina pachyderma (sin.) in the Arctic Ocean.

Earth and Planetary Science Letters 146, 47–58.

Bauch, D., Carstens, J., Wefer, G., Theide, J., 2000. The imprint of

anthropogenic CO2 in the Arctic Ocean: evidence from planktic d13Cdata from water column and sediment surfaces. Deep Sea Research

II 9-11, 1791–1808.

Bauch, D., Schlosser, P., Fairbanks, R.G., 1995. Freshwater balance

and the sources of deep and bottom waters in the Arctic Ocean

inferred from the distribution of H218O. Progress in Oceanography

35, 53–80.

Bauch, H.A., Erlenkeuser, H., Speilhagen, R.F., Struck, U., Matthies-

sen, J., Thiede, J., Heinemeier, J., 2001. A multiproxy reconstruction

of the evolution of deep and surface waters in the Nordic Seas over

the last 30,000 years. Quaternary Science Reviews 20, 659–678.

Bauch, H.A., Kassens, H., Erlenkeuser, H., Grootes, P.M., Thiede, J.,

1999. Depositional environment of the Laptev Sea (Arctic Siberia)

during the Holocene. Boreas 28, 194–204.

Bauch, H.A., Weinelt, M.S., 1997. Surface water changes in the

Norwegian Sea during the last deglacial and Holocene times.

Quaternary Science Reviews 16, 1115–1124.

Bemis, B.E., Spero, H.J., Bijma, J., Lea, D.W., 1998. Reevaluation of

the oxygen isotopic composition of planktonic foraminifera:

Experimental results and revised paleotemperaturae equations.

Paleoceanography 30, 150–160.

Berger, A., 1978. Long-term variations of daily insolation and

Quaternary climatic changes. Journal of Atmospheric Science 35,

2362–2367.

Birks, H.H., 1991. Holocene vegetational history and climatic change

in West Spitsbergen}plant macrofossils from Skardtj�rna. The

Holocene 1, 209–218.

Bjorck, S., Rundgren, M., Inglfsson, O., Funder, S., 1997. The

Preboreal oscillation around the Nordic Seas: terrestrial and

lacustrine responses. Journal of Quaternary Science 12, 455–466.

Blindheim, J., 1989. Cascading of Barents Sea bottom water into the

Norwegian Sea. Rapports et Proces-Verbaux des Reunions, Conseil

International Pour L’ Exploration scientifique de la Mer 188, 49–58.

Bond, G., Broecker, W., Johnsen, S., McManus, J., Labeyrie, L.,

Jouzel, J., Bonani, G., 1993. Correlations between climate records

from North Atlantic sediments and Greenland ice. Nature 365,

143–147.

Bondevik, S., Mangerud, J., Ronnert, L., Salvigsen, O., 1995.

Postglacial sea-level history of Edge�ya and Barents�ya, easternSvalbard. Polar Research 14, 153–180.

Boyle, E.A., 1992. Cadmium and d13C paleochemical ocean distribu-

tions during the stage 2 glacial maximum. Annual Review of Earth

and Planetary Sciences 20, 245–287.

Brezgunov, V.S., Debol’skii, V.K., Nachaev, V.V., Ferronskii, V.I.,

Yakimova, T.V., 1982. Characteristics of the formation of the

oxygen isotope composition and salinity upon mixing of sea and

river waters in the Barents and Kara seas. Water Resources 9,

335–345.

Carstens, J., Wefer, G., 1992. Recent distribution of planktonic

foraminifera in the Nansen Basin, Arctic Ocean. Deep-Sea Research

39, S507–S524.

Charles, C.D., Fairbanks, R.G., 1990. Glacial to interglacial changes

in the isotopic gradients of the Southern Ocean surface waters. In:

Bleil, U., Thiede, J. (Eds.), Geologic History of the Polar Oceans.

Kluwer Academics Publishers, Dordrecht, pp. 519–538.

Cherkis, N.Z., Fleming, H.S., Vogt, P.R., Czarnecki, M.F., 1991.

Bathymetry of the Barents and Kara Seas. (Map and Chart Series

MCH047). Geological Society of America, Boulder.

Cronin, T.M., Holtz, T.R.J., Whatley, R.C., 1994. Quaternary

paleoceanography of the deep Arctic Ocean based on quantitative

analysis of Ostracoda. Marine Geology 119, 305–332.

Cronin, T.M., Holtz, T.R., Stein, R., Spielhagen, R., Ftterer, D.,

Wollenburg, J., 1995. Late Quaternary paleoceanography of the

Eurasian Basin, Arctic Ocean. Paleoceanography 10, 259–281.

Cuffey, K.M., Clow, G.D., 1997. Temperature, accumulation, and ice

sheet elevation change in central Greenland through the last

deglacial transition. Journal of Geophysical Research C102,

26,383+.

Dokken, T.M., Hald, M., 1996. Rapid climatic shifts during isotope

stages 2-4 in the polar North Atlantic. Geology 24, 599–602.

Dokken, T.M., Jansen, E., 1999. Rapid changes in the mechanism

of ocean convection during the last glacial period. Nature 401,

458–461.

Donner, B., Wefer, G., 1994. Flux and stable isotope composition of

Neogloboquadrina pachyderma and other planktonic foraminifers

in the Southern Ocean (Atlantic sector). Deep-Sea Research I 41,

1733–1743.

Duplessy, J.-C., Ivanova, E., Murdmaa, I., Paterne, M., Labeyrie, L.

(in press). Holocene paleoceanography of the northern Barents Sea

and variations of the northward heat transport by the Atlantic

Ocean. Boreas.

Duplessy, J.C., Shackleton, N.J., Fairbanks, R.G., Labeyrie, L., Oppo,

D., Kallel, N., 1988. Deepwater source variations during the last


climatic cycle and their impact on the global deepwater circulation.


Elverh�i, A., Andersen, E.S., Dokken, T., Hebbeln, D., Spielhagen,R., Svendsen, J.-I., S�rflaten, M., R�rnes, A., Hald, M., Forsberge,

C.-F., 1995. The growth and decay of the Late Weichselian Ice Sheet

in Western Svalbard and adjacent areas based on provenance studies

of marine sediments. Quaternary Research 44, 303–316.

Elverh�i, A., Nyland-Berg, M., Russwurm, L., Solheim, A., 1990. Late

Weichselian ice recession in the central Barents Sea. In: Bleil, U.,

Thiede, J. (Eds.), Geological History of the Polar Oceans: Arctic

Versus Antarctic. Dordrecht, Kluwer, pp. 289–307.

Elverh�i, A., Pfirman, S.L., Solheim, A., Larssen, B.B., 1989.

Glaciomarine sedimentation in epicontinental seas exemplified by

the northern Barents Sea. Marine Geology 85, 225–250.

Elverh�i, A., Solheim, A., 1983. The Barents Sea ice-sheet: a

sedimentological discussion. Polar Research 1, 23–32.

Erlenkeuser, H., Speilhagen, R.F., Taldenkova, E., 1999. Stable

isotopes in modern water and bivalve samples from the southern

Kara Sea. Berichte zur Polarforschung 300, 80–90.

Erlenkeuser, H., von Grafenstein, U., 1999. Stable oxygen isotope

ratios in benthic carbonate shells of ostracoda, foraminifera, and

bivalvia from surface sediments of the Laptev Sea, summer 1993 and

1994. In: Kassens, H., Bauch, H.A., Dmitrenko, I., Eicken, H.,

Hubberten, H.-W., Melles, M., Theide, J., Timokhov, L. (Eds.),

Land-ocean Systems in the Siberian Arctic: Dynamics and History.

Springer, Berlin, pp. 503–514.

Fairbanks, R.G., 1989. A 17,000 year glacio-eustatic sea level record:

influence of glacial melting rates on the Younger Dryas event and

deep-ocean circulation. Nature 342, 637–642.

Fairbanks, R.G., Matthews, R.K., 1978. The marine oxygen isotope

record in Pleistocene coral, Barbados, West Indies. Quaternary

Research 10, 181–196.

Ferronskii, V.I.-Principal Investigator, 1978. Investigation of the

distribution of oxygen-18 and tritium in waters of the Barents and

Kara seas for the prognosis of hydrologic changes in the estuarine

areas of the Ob’ River and Kara and Pechora seas. Institute of

Water Studies, USSR Academy of Sciences, 101p.

Forman, S.L., 1990. Post-glacial relative sea level history of north-

western Spitsbergen, Svalbard. Bulletin of Geological Society of

America 102, 1580–1590.

Forman, S.L., Inglfsson, O., Gataullin, V., Manley, W.F., Lokrantz,

H., 1999. Late Quaternary stratigraphy of western Yamal Peninsula,

Russia; new constraints on the configuration of the Eurasian ice

sheet. Geology 27, 807–810.

Forman, S.L., Lubinski, D., Miller, G.H., Snyder, J., Matishov, G.,

Korsun, S., Myslivets, V., 1995. Postglacial emergence and

distribution of late Weichselian ice-sheet loads in the northern

Barents and Kara seas, Russia. Geology 23, 113–116.

Forman, S.L., Maslowski, W., Andrews, J.T., Lubinski, D.J., Steele,

M., Zhang, Y., Lammers, R., 2000. Researchers explore Arctic

freshwater’s role in ocean circulation. EOS, Transactions, American

Geophysical Union, 81, 169–174.

Forman, S.L., Polyak, L., 1997. Radiocarbon content of pre-bomb

marine mollusks and variations in the 14C reservoir age for coastal

areas of the Barents and Kara seas, Russia. Geophysical Research

Letters 24, 885–888.

Forman, S.L., Weihe, R., Lubinski, D., Tarasov, G., Korsun, S.,

Matishov, G., 1997. Holocene relative sea-level history of Franz

Josef Land, Russia. Geological Society of America Bulletin 109,

1116–1133.

Gascard, J.-C., Richez, C., Rouault, C., 1995. New insights on large-

scale oceanography in Fram Strait: The West Spitsbergen Current.

In: Smith, W.O.J., Grebmeier, J.M. (Eds.), Arctic Oceanography:

Marginal Ice Zones and Continental Shelves. American Geophysical

Union, Washington, DC, pp. 131–182.

Green, K.E., 1960. Ecology of some arctic foraminifera. Micropa-

leontology 6, 57–78.

Grossman, E.L., 1984. Stable isotope fractionation in live benthic

foraminifera from the southern California borderland. Palaeogeo-

graphy, Palaeoclimatology, Palaeoecology 47, 301–327.

Hald, M., Aspeli, R., 1997. Rapid climatic shifts of the northern

Norwegian Sea during the last deglaciation and the Holocene.

Boreas 26, 15–28.

Hald, M., Dokken, T., Hagen, S., 1996. Palaeoceanography on the

European arctic margin during the last deglaciation. In: Andrews,

J.T., Austin, W.E.N., Bergsten, H., Jennings, A.E. (Eds.), Late

Quaternary Palaeoceanography of the North Atlantic Margins.

Geological Society, pp. 275–287.

Hald, M., Hagen, S., 1998. Early Preboreal cooling in the Nordic Seas

region triggered by meltwater. Geology 26, 615–618.

Hald, M., Kolstad, V., Polyak, L., Forman, S.L., Herlihy, F.A.,

Ivanov, G., Nescheretov, A., 1999. Late-glacial and Holocene

paleoceanography and sedimentary environments in the St. Anna

Trough, Eurasian Arctic Ocean margin. Palaeogeography, Palaeo-

climatology, Palaeoecology 146, 229–249.

Harrison, S.P., Prentice, I.C., Bartlein, P.J., 1992. Influence of

insolation and glaciation on atmospheric circulation in the North

Atlantic sector: implications of general circulation models experi-

ments for the Late Quaternary climatology of Europe. Quaternary

Science Reviews 11, 283–299.

Hebbeln, D., Dokken, T., Anderson, E.S., Hald, M., Elverh�i, A.,1994. Moisture supply for northern ice-sheet growth during the Last

Glacial Maximum. Nature 370, 357–360.

Herlihy, F.A., 1996. A marine sedimentary record of the Last Glacial

cycle in the Franz Victoria and Saint Anna Troughs, northern

Barents and Kara Seas, Russia. Master’s thesis, Ohio State

University.

Hjort, C., Mangerud, J., Andrielsson, L., Bondevik, S., Landvik, J.Y.,

Salvigsen, O., 1995. Radiocarbon dated common mussel Mytilus

edulis from eastern Svalbard and the Holocene marine climatic

optimum. Polar Research 14, 239–243.

Howell, D., Siegert, M.J., Dowdeswell, J.A., 2000. Modelling the

influence of glacial isostasy on Late Weichselian ice-sheet growth in

the Barents Sea. Journal of Quaternary Science 15, 475–486.

Hut, G., 1987. Consultants group meeting on stable isotope reference

samples for geochemical and hydrological investigations. Report to

the Director General. International Atomic Energy Agency.

Ishman, S.E., Foley, K.M., 1996. Modern benthic foraminifer

distribution in the Amerasian Basin, Arctic Ocean. Micropaleontol-

ogy 42, 321–345.

Ivanov, V., Korablev, A., 1997. Water masses structure and dynamics

in the North Kara Sea in summer 1996. In: Aagaard, K., Hartmann,

D., Kattsov, V., Martinson, D., Stewart, R., Weaver, A. (Eds.),

Conference on Polar Processes and Global Climate. Arctic

Climate System Study (ACSYS), Orcas Island, Washington, USA,

pp. 103–105.

Ivanov, G.I., Nechsheretov, A.V., Ponomarenko, T.V., Krylov, A.A.,

Stein, R., 1999. Sedimentary Environment in the St. Anna Trough.

Berichte zur Polarforschung 306, 95–109.

Jennings, A.E., Weiner, N.J., 1996. Environmental change in eastern

Greenland during the last 1300 years; evidence from foraminifera

and lithofacies in Nansen Fjord, 68 degrees N. The Holocene 6,

179–191.

Johannessen, T., Jansen, E., Flat�y, A., Ravelo, A.C., 1994. Therelationship between surface water masses, oceanographic fronts

and paleoclimatic proxies in surface sediments of the Greenland,

Iceland, and Norwegian Seas. In: Zahn, R.E.A. (Ed.), Carbon

Cycling in the Glacial Ocean: Constraints on the Ocean’s Role in

Global Change. Berlin, Springer, pp. 61–85.

Jones, G.A., Keigwin, L.D., 1988. Evidence from Fram Strait (788N)for early deglaciation. Nature 336, 56–59.

Keigwin, L.D., Boyle, E.A., 1989. Late Quaternary paleochemistry of

high-latitude surface waters. Palaeogeography, Palaeoclimatology,

Palaeoecology 73, 85–106.


Keigwin, L.D., Jones, G.A., Lehman, S.J., 1991. Deglacial meltwater

discharge, North Atlantic deep circulation, and abrupt climate

change. Journal of Geophysical Research 96, 16,811–16,826.

Kerwin, M., Overpeck, J.T., Webb, R.S., DeVernal, A., Rind, D.H.,

Healy, R.J., 1999. The role of oceanic forcing in mid-Holocene

Northern Hemisphere climatic change. Paleoceanography 14,

200–210.

Kim, S.T., O’Neil, J.R., 1997. Equilibrium and non-equilibrium

oxygen isotope effects in synthetic carbonates. Geochemica Cosmo-

chemica Acta, 3461–3475.

Kleiber, H.-P., Knies, J., Niessen, F., 2000. The late Weichselian

glaciation of the Franz Victoria Trough, northern Barents Sea; ice

sheet extent and timing. Marine Geology 168, 25–44.

Kneis, J., Mller, C., Nowaczyk, N., Stein, R., 2000. A multiproxy

approach to reconstruct the environmental changes along the

Eurasian continental margin over the last 160,000 years. Marine

Geology 163, 317–344.

Kneis, J., Vogt, C., Stein, R., 1999. Late Quaternary growth and decay

of the Svalbard/Barents Sea ice sheet and paleoceanographic

evolution in the adjacent Arctic Ocean. Geo-Marine Letters 18,

195–202.

Koc, N., Jansen, E., Haflidason, H., 1993. Paleoceanographic

reconstructions of surface ocean conditions in the Greenland,

Iceland and Norwegian Seas through the last 14 ka based on

diatoms. Quaternary Science Reviews 12, 115–140.

Kohfeld, K.E., Fairbanks, R.G., Smith, S.L., Walsh, I.D., 1996.

Neogloboquadrina pachyderma (sinistral coiling) as paleoceano-

graphic tracers in polar oceans: Evidence from Northeast Water

Polyna plankton tows, sediment traps, and surface sediments.


Labeyrie, L.D., Duplessy, J.C., 1985. Changes in the oceanic 13C/12C

ratio during the last 140,000 years: high latitude surface water

records. Palaeogeography, Palaeoclimatology, Palaeoecology 50,

217–240.

Lagoe, M., 1979. Recent benthonic foraminiferal biofacies in the

Arctic Ocean. Micropaleontology 25, 214–224.

Lambeck, K., 1996. Limits on the areal extent of the Barents Sea ice

sheet in Late Weichselian time. Global and Planetary Change 12,

41–51.

Landvik, J.Y., Bondevik, S., Elverh�i, A., Fjeldskaar, W., Mangerud,

J., Siegert, M.J., Salvigsen, O., Svendsen, J.-I., Vorren, T.O., 1998.

The Last Glacial Maximum of Svalbard and the Barents Sea area:

ice sheet extent and configuration. Quaternary Science Reviews 17,

43–75.

Lehman, S.J., Keigwin, L.D., 1992. Sudden changes in North Atlantic

circulation. Nature 356, 757–762.

Loeng, H., 1991. Features of the physical oceanographic conditions of

the Barents Sea. Polar Research 10, 5–18.

Loeng, H., Ozhigin, V., Adlandsvik, B., Sagen, H., 1993. Current

measurements in the northeastern Barents Sea. ICES Statutory

Meeting.

Lothe, T., Stanovoy, V.V., Fomitchev, N.I., 1996. Physical Oceano-

graphy. In: Solheim, A., Forsberg, C.F., (Eds.), Norwegian Polar

Institute’s cruise to the nothern margin of Svalbard and the Barents

Sea 25/7-2/9, 1994: Marine Geology/Geophysics and Physical

Oceanography (Cruise Report) Norsk Polarinstitutt, Oslo,

pp. 56–66.

Lubinski, D.J., Korsun, S., Polyak, L., Forman, S.L., Lehman, S.J.,

Herlihy, F.A., Miller, G.H., 1996. The last deglaciation of the Franz

Victoria Trough, northern Barents Sea. Boreas 25, 89–100.

Lubinski, D.J., Forman, S.L., Miller, G.H., 1999. Holocene glacier

and climate fluctuations on Franz Josef Land, Arctic Russia, 808N.Quaternary Science Reviews 18, 85–108.

Mackensen, A., Hald, M., 1988. Cassidulina teretis Tappan and C.

laevigata d’Orbigny; their modern and late Quaternary distribution

in the northern seas. Journal of Foraminiferal Research 18, 16–24.

Mackensen, A., Sejrup, H.P., Jansen, E., 1985. The distribution of

living benthic foraminifera on the continental slope and rise of

southwest Norway. Marine Micropaleontology 9, 275–306.

Mangerud, J., Gulliksen, S., 1975. Apparent radiocarbon ages of

recent marine shells from Norway, Spitsbergen, and Arctic Canada.

Quaternary Research 5, 263–273.

Mangerud, J., Svendsen, J.-I., Astakhov, V.I., 1999. The age and

extent of the Barents and Kara Sea ice sheets in Northern Russia.

Boreas 28, 46–80.

Manley, T.O., 1995. Branching of Atlantic Water within the Green-

land-Spitsbergen Passage: an estimate of recirculation. Journal of

Geophysical Research 100, 20,627–20,634.

Matishov, G., Levitus, S., Smolyar, I., Zyev, A., Golubev, V., Androv,

N., Slobodin, V., 1998). Climatic atlas of the Barents Sea 1998:

temperature, salinity, oxygen (CD-ROM) Murmansk Marine

Biological Institute/US National Oceanographic Data Center.

McCorkle, D.C., Keigwin, L.D., Corliss, B.H., Emerson, S.R.,

1990. The influence of microhabitats on the carbon isotopic

composition of deep-sea benthic foraminifera. Paleoceanography

5, 161–185.

Mensch, M., Frank, M., Stein, W., Bayer, R., 2000. Water mass

distribution and circulation in the Eurasian Basin of the Arctic

Ocean. Journal of Geophysical Research-Oceans. Personal commu-

nication.

Midtunn, L., 1985. Formation of dense bottom water in the Barents

Sea. Deep-Sea Research 32, 1233–1241.

Mitchell, J.F.B., Grahame, N.S., Needham, K.H., 1988. Climate

simulations for 9000 years before present: seasonal variations and

effect of the Laurentide Ice Sheet. Journal of Geophysical Research

93, 8283–8303.

Moller, P., Bolshiyanov, D.Y., Bergsten, H., 1999. Weichselian

geology and palaeoenvironmental history of the central Taymyr

Peninsula, Siberia, indicating no glaciation during the last global

glacial maximum. Boreas 28, 92–114.

Mosby, H., 1938. Svalbard waters. Geofysiske Publikasjoner 12, 85.

NODC (1991). Global ocean temperature and salinity profiles

(NODC-02), Atlantic Ocean CD-ROM. National Oceanographic

Data Center, Washington, D.C.

N�rgaard–Pedersen, N., Speilhagen, R.F., Thiede, J., Kassens, H.,1998. Central Arctic surface ocean environment during the past

80,000 years. Paleoceanography 13, 193–204.

NSIDC (1953–1990). AARI 10-Day Arctic Ocean EASE-Grid Sea Ice

Observations. National Snow and Ice Data Center. NSIDC

Distributed Active Archive Center, Boulder, Colorado.

Olsson, I., 1980. Content of 14C in marine mammals from northern

Europe. Radiocarbon 22, 662–675.

O’Neill, J.R., Clayton, R.N., Mayeda, T.K., 1969. Oxygen isotope

fractionation in divalent metal carbonates. Journal of Chemical

Physics 51, 5547–5558.

Oppo, D.W., Fairbanks, R.G., 1989. Carbon isotope composition of

tropical surface water during the past 22,000 years. Paleoceano-

graphy 4, 333–351.

Ostermann, D.R., Curry, W.B., 2000. Calibration of stable isotopic

data: an enriched d18O standard used for source gas mixing

detection and correction. Paleoceanography 15, 353–360.

Ostlund, H.G., Grall, C., 1993. Arctic Tritium, 1973–1991. University

of Miami.

Ostlund, H.G., Hut, G., 1984. Arctic Ocean water mass balance

from isotope data. Journal of Geophysical Research 89,

6373–6381.

Parkinson, C.L., Comiso, J.C., Zwally, H.J., Cavalieri, D.J., Gloersen,

P., Cambell, W.J., 1987. Arctic Sea Ice, 1973–1976: Satellite Passive-

microwave Observations. National Aeronautics and Space Admin-

istration, Washington.

Paulsen, B.E., 1997. Utbredelse av levende og d�de bentiske

foraminiferer i Karahavet. Thesis, University of Troms�.


Pfirman, S.L., Bauch, D., Gammelsr�d, T., 1994. The northern

Barents Sea: water mass distribution and modification. In:

Johannessen, O.M., Meunch, R.D., Overland, J.E. (Eds.), The

Polar Oceans and their Role in Shaping the Global Environment.

American Geophysical Union, Washington, DC, pp. 77–94.

Polyak, L.V., 1985. Foraminifers of bottom sediments of the Barents

and Kara Sea and their stratigraphic significance. Ph.D. Thesis,

University of Leningrad.

Polyak, L., Forman, S.L., Herlihy, F.A., Ivanov, G., Krinitsky, P.,

1997. Late Weichselian deglacial history of the Svyataya (Saint)

Anna Trough, northern Kara Sea, Arctic Russia. Marine Geology

143, 169–188.

Polyak, L., Gataullin, V., Epshtein, O., Okuneva, O., Stelle, V., 2000a.

New constraints on the limits of the Barents-Kara Ice Sheet during

the Last Glacial Maximum based on borehole stratigraphy from the

Pechora Sea. Geology 28, 611–614.

Polyak, L.V., Lehman, S.J., Gataullin, V., Jull, A.J., 1995. Two-

step deglaciation of the southeastern Barents Sea. Geology 23,

567–571.

Polyak, L., Levitan, M., Gataullin, V., Khusid, T., Mikhailov, V.,

Mukhina, V., 2000b. The impact of glaciation, river-discharge,

and sea-level change on Late Quaternary environments in the

southwestern Kara Sea. International Journal of Earth Sciences 89,

550–562.

Polyak, L.V., Solheim, A., 1994. Late- and post-glacial environments

in the northern Barents Sea west of Franz Josef Land. Polar

Research 13, 197–207.

Poole, D.A.R., 1994. Neogene and Quaternary paleoenvironments on

the northern Norwegian shelf. Ph.D. Thesis, University of Troms�.Poore, P.Z., Osterman, L., Curry, W.B., Phillips, R.L., 1999. Late

Pleistocene and Holocene meltwater events in the western Arctic

Ocean. Geology 27, 759–762.

Rudels, B., Jones, E.P., Anderson, L.G., Kattner, G., 1994. On the

intermediate depth waters of the Arctic Ocean. In: Johannessen,

O.M., Meunch, R.D., Overland, J.E. (Eds.), The Polar Oceans and

their Role in Shaping the Global Environment. American Geophy-

sical Union, Washington, DC, pp. 33–46.

Rudels, B., Larsson, A.-M., Sehlstedt, P.-I., 1991. Stratification and

water mass formation in the Arctic Ocean: some implications for the

nutrient distribution. Polar Research 10, 19–32.

Salvigsen, O., Forman, S.L., Miller, G.H., 1992. Thermophilous

molluscs on Svalbard during the Holocene and their paleoclimatic

implications. Polar Research 11, 1–10.

Schauer, U., Muench, R.D., Rudels, B., Timokhov, L., 1997a.

Impact of eastern Arctic shelf waters on the Nansen Basin

intermediate layers. Journal of Geophysical Research 102,

3371–3382.

Schauer, U., Rudels, B., Loeng, H., Jones, P., Muench, R., Swift, J.,

Bjoerk, G., 1997b. Barents Sea Water input to the Eurasian Basin

through St. Anna Trough. In: Aagaard, K., Hartmann, D., Kattsov,

V., Martinson, D., Stewart, R., Weaver, A. (Eds.), Conference on

Polar Processes and Global Climate. Arctic Climate System Study

(ACSYS), Orcas Island, Washington, USA, pp. 236–238.

Seidenkrantz, M.-S., 1995. Cassidulina teretis Tappan and Cassidulina

neoteretis new species, 1994: stratigraphical marker species for deep

sea and outer shelf areas. Journal of Micropalaeontology 14,

145–157.

Shackleton, N.J., 1974. Attainment of isotopic equilibrium between

ocean water and the benthonic foraminifera Uvigerina: isotopic

changes in the ocean during the last glacial. Centre National

de la Recherche Scientifique Paris. Colloque CNRS No. 219, pp.

203–210.

Shackleton, N.J., 1977. Carbon-13 in Uvigerina: Tropical rainforest

history and the equatorial Pacific carbonate dissolution cycle. In:

Anderson, N.R., Malahoff, A. (Eds.), The Fate of Fossil Fuel CO2

in the Oceans. Plenum, New York, pp. 401–428.

Spero, H.J., Bijma, J., Lea, D.W., Bemis, B.E., 1997. Effect of seawater

carbonate concentration on foraminiferal carbon and oxygen

isotopes. Nature 390, 497–500.

Spielhagen, R.F., Erlenkeuser, H., 1994. Stable oxygen and carbon

isotopes in plaktic foraminifers from Arctic Ocean surface sedi-

ments: reflections of the low salinity surface water layer. Marine

Geology 119, 227–250.

Spielhagen, R.F., Erlenkeuser, H., Heinemeier, J., 1998. Deglacial

freshwater outflow from Lena River (Siberia). Quaternary Environ-

ment of the Eurasian North (QUEEN), Second Workshop,

St. Petersburg, p. 48.

Stanovoy, V.V., Pivovarov, S.V., Sokolov, V.T., Karpiy, V.Y.,

Lebedev, N.V., 1998. Climatic changes on the sea boundaries of

the Barents Sea. In: Proceedings of the First International

Conference, The Barents Sea Impact Study (BASIS), Arctic and

Antarctic Research Institute (AARI), St. Petersburg.

Stein, R., Nam, S.-I., Schubert, C., Vogt, C., Futterer, D.,

Heinemaeier, J., 1994. The last deglaciation event in the eastern

central Arctic Ocean. Science 264, 692–696.

Steinsund, P.-I., 1994. Benthic foraminifera in surface sediments of the

Barents and Kara Seas: modern and late Quaternary applications.

Dr. Scient. Thesis, University of Troms�.Steinsund, P.I., Hald, M., 1994. Recent calcium carbonate dissolution

in the Barents Sea: paleoceanographic applications. Marine Geology

117, 303–316.

Stuiver, M., Polach, H.A., 1977. Discussion, reporting of 14C data.

Radiocarbon 19, 355–363.

Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised

CALIB 3.0 14C age calibration program. Radiocarbon 35,

215–230.

Svendsen, J.I., Elverh�i, A., Mangerud, J., 1996. The retreat of the

Barents Sea Ice Sheet on the western Svalbard margin. Boreas 25,

244–256.

Svendsen, J.I., Mangerud, J., 1997. Holocene glacial and climatic

variations on Spitsbergen, Svalbard. The Holocene 7, 45–57.

Svendsen, J.I., Astakhov, V.I., Bolshiyanov, D.Y., Demidov, I.,

Dowdeswell, J.A., Gataullin, V., Hjort, C., Hubberton, H.W.,

Larsen, E., Mangerud, J., Melles, M., Moller, P., Saarnisto, M.,

Siegert, M.J., 1999. Maximum extent of the Eurasian ice sheets in

the Barents and Kara Sea region during the Weichselian. Boreas 28,

234–242.

TEMPO (1996). Potential role of vegetation feedback in the climate

sensitivity of high-latitude regions: a case study at 6000 years B.P.

Global Biogeochemical Cycles 10, 727–736.

Todd, R., Low, D., 1980. Foraminifera from the Kara and Greenland

seas, and review of Arctic studies. US Geological Survey Profes-

sional Paper 1070, 30 p.

Veum, T., Jansen, E., Arnold, M., Beyer, I., Duplessy, J.-C., 1992.

Water mass exchange between the North Atlantic and the

Norwegian Sea during the past 28,000 years. Nature 356,

783–785.

Vinje, T.E., 1976. Sea ice conditions in the European sector of the

marginal seas of the Arctic, 1966–1975. Norsk Polarinstitutt rbok,

163–174.

Vorren, T.O., Hald, M., Lebesbye, E., 1988. Late Cenozoic environ-

ments in the Barents Sea. Paleoceanography 3, 601–612.

Vorren, T.O., Laberg, J.S., 1996. Late glacial air temperature,

oceanographic and ice sheet interactions in the southern Barents

Sea region. In: Andrews, J.T., Austin, W.E.N., Bergsten, H.,

Jennings, A.E. (Eds.), Late Quaternary Palaeoceanography of the

North Atlantic Margins. Geological Society Special Publication No.

111, pp. 303–321.

Wohlfarth, B., Lemdahl, G., Olsson, S., Persson, T., Snowball, I.,

Ising, J., Jones, V., 1995. Early Holocene environment on Bj�rn�ya(Svalbard) inferred from multidisciplinary lake sediment studies.

Polar Research 14, 253–275.


Wollenburg, J.E., Mackensen, A., 1998. Living benthic foraminifers

from the central Arctic Ocean: faunal composition, standing stock

and diversity. Marine Micropaleontology 34, 153–185.

Zeeberg, J.J., Lubinski, D.J., Forman, S.L., in press. Holocene relative

sea-level history of northern Novaya Zemlya}implications for Late

Glacial ice-sheet loading in the Barents and Kara seas. Quaternary

Research.

Zhang, J., Rothrock, D., Steele, M., 1998. Warming of the Arctic

Ocean by a strengthened Atlantic inflow: model results. Geophysical

Research Letters 25, 1745–1748.


Documents

FreshwaterandAtlanticwaterinflowstothedeepFreshwaterandAtlanticwaterinflowstothedeep ... (afterLoeng,1991;Loengetal.,1993;Pfirmanetal.,1994; ... Calender ageb (CalyrBP) Sedresearch.bpcrc.osu.edu/geo/publications/Lubinski_et_al_QSR.pdf ·