PALEOCEANOGRAPHY, VOL. 12, NO. 4, PAGES 594-603, AUGUST 1997
Barium accumulation in the Atlantic sector of the Southern Ocean: Results from 190,000-year records
C. C. Ntirnberg, G. Bohrmann,and M. Schltiter GEOMAR ResearchCenter, Kiel, Germany
M. Frank Department of Earth Sciences, University of Oxford, Oxford, England
Abstract. Extensive investigationsof sedimentarybarium were performedin the southernSouth Atlantic in order to assessthe reliability of the barium signal in Antarctic sedimentsas a proxy for paleoproductivity. Maximum accumulationrates of excessbarium were calculatedfor the Antarctic zone south of the polar front where silica accumulatesat high rates. The correspondencebetween barium and opal supports the applicabilityof bariumas a proxy for productivity.Within the Antarcticzone north of today'saverage sea ice maximum,interglacial vertical rain ratesof excessbarium are high, with a maximumoccurring during the last deglaciationand early Holoceneand duringoxygen isotopechronozone 5.5. Duringthese periods, the maximumsilica accumulation was supposedlylocated south of the polarfront. Glacialpaleoproductivity, instead, was low withinthe Antarctic zone. Northof the polarfront, significantly higherbarium accumulation occurs during glacial times. The verticalrain rates,however, are as high as in the glacial Antarcticzone. Thereforethere was no evidencefor an increasedproductivity in the glacialSouthern Ocean.
Introduction and alkalinity profiles points to the involvementof barium in the biogeniccycle. The dissimilaritybetween barium profiles During the last 15 years,the searchfor geochemicaltracers and those of nitrate and phosphateindicates the regeneration which allow the reconstruction of the chemical characteristics of barium with the less soluble skeletons of silica and/or of the oceans, past climates, and paleoproductivity was carbonate[Chan et al., 1977]. Apparently,barium is removed intensified [e.g., Berger et al., 1983; Prahl et al., 1989; from surface waters by precipitation within sinking biogenic Boyle, 1988, 1992; Lea and Boyle, 1989, 1990; Elderfield, particles,which dissolvein the deeperwater columnand in the 1991; Rau et al., 1991]. It became apparent that marine sediment. Most particulate barium is composedof barite sediments from upwelling areas show enhanced barite crystals, which act as the main carrier for barium into the concentrations [e.g., Revelle et al., 1955; Goldberg and sediment[Dehairs and Goeyens,1989; Dehairs et al., 1991; Arrhenius, 1958; Church, 1976; Brurnsack, 1989; von Bishop,1988]. A maximumof particulatebarium commonly Breyrnannet al., 1992]. The covariancebetween the barite occurs between 200 and 500 m water depth [Dehairs et al., concentration and an enhanced surface productivity [e.g., 1990; $troobants et al. , 1991]. Goldberg and Arrhenius, 1958; Dehairs et al., 1980; Schmitz, Nevertheless,the processby which barite forms within the 1987; Bishop, 1988] lead us to assumethat bariumis a proxy water column is still not known. According to severalauthors for paleoproductivity [e.g., Jurnars et al., 1989; von [Chow and Goldberg, 1960; Dehairs et al., 1980; Bishop, Breyrnannet al., 1992; Dyrnond et al., 1992; $hirnrnield et 1988], barite formation takes place within micro- al. , 1994]. environmentswithin the water column, where the organic Like other oceanographicparameters, dissolved barium matter of planktic cells or fecal pellets and aggregates showsdrastic horizontal gradients at the oceanographicfronts decomposes.The excesssulfate results from the decayof within the Antarctic CircumpolarCurrent [Chan et al., 1977; labile sulfurin the organicmatter [Chow and Goldberg,.1960; Narnberg, 1995]. Furthermore,in the SouthernOcean, high Dehairs et al., 1980; Bishop, 1988]. Barite can be also concentrationsof dissolvedand particulatebarium occur in the formed during the dissolutionof acantharian-derivedcelestite water column [Chan et al., 1977; Dehairs and Goeyens,1989] [Bernstein et al., 1992]. An active intracellular formation of and within deep-seadeposits [Turekian and Johnson,1966; barite is observedin benthic xenophyophores[Tendal, 1972; $hirnrnieldet al., 1994]. Gooday and Nott, 1982]. Because these large protozoans Vertical profiles of dissolved barium are generally rarely survive the sampling procedure and are difficult to characterized by low concentrationsin the surface water, identify, active intracellular formation was only rarely which increase with water depth [e.g., Wolgernuth and investigated, and its role in the barite budget is still not Broecker, 1970; Li et al., 1973; Chan et al., 1977; Broecker understood. and Peng, 1982]. The similarityof the bariumprofile to silica Additional sources of barite crystals are hydrothermal precipitates [BostrOrnet al., 1973]. It should be noticed, however, that hydrothermal sources are only of local Copyright1997 by the AmericanGeophysical Union. significance [Dyrnondet al., 1992]. In addition to barite PaperNumber 97PA01130. crystals, particulate barium accumulatesin seafloordeposits, 0883-8305/97/97PA-0! 130512.00 by incorporation into or absorption by siliceous or
594 NORNBERGET AL.: BARIUM ACCUMULATION IN THE SOUTHERN OCEAN 595 carbonaceoustests, in organic material, or in aluminosilicates perpendicularto the oceanographicfronts (R/V Polarstern [e.g., Goldberg and Arrhenius, 1958; Chow and Goldberg, cruisesANT-VIII/3 and ANT-IX/4; Figure 1). 1960; Dehairs et al., 1980; Lea and Boyle, 1989, 1990, 1991; The lithology of the cores is dominatedby diatomaceous Lea et al., 1989]. muds (Figure 3). Core PS2082-1 from the Subantarcticzone The diageneticmobilization of barium in sulfate-depleted shows a few intercalations of foraminiferal oozes at the top sediments with often a subsequent reprecipitation in and calcareous muds beneath. The other cores also consist of sulfate-bearingenvironments [Brumsack, 1989; Torreset al., diatomaceousoozes, which predominatein the cores of the 1996] also contributes to the barite budget. However, the Antarctic zone. diagenetic mobilization in anoxic environments, in fact, The age modelsof the surfacesediments are basedon diatom restricts the application of barium as a paleoproductivity and radiolarian biofluctuation stratigraphies and indicator to non-sulfate-depletedsediments [Bruinsack, 1989; biostratigraphies (A. Abelmann et al., unpublished report, von Breymann et al., 1992]. 1992). For core PS2082-1, Mackensen et al. [1994] developedan agemodel by correlatingthe d•80 recordsof the Samples and Procedures benthic foraminiferal speciesCibicidoides spp. and Fontbotia wuellerstorfi, and the planktic speciesNeogloboquadrina The area of investigationcomprises the Atlantic sector of pachyderma sin. and Globigerina hulloides to the SPECMAP the Southern Ocean south of approximately42øS (Figure 1). stack. For core PS1756-5, an age model was constructedby This area can be divided from north to south into the Subantarctic zone south of the Subtropical front, the polar combiningthe d•3C profileof organicmatter (G. Fischer frontal zone situated between the Subantarctic front and the et al., unpublished manuscript, 1997) and the siliceous microfossil biofluctuation measurements (A. Abelmann et al., polar front, and the Antarcticzone, which is coveredby winter sea ice in the southernmostregions [Peterson and Strarnma, unpublishedreport, 1992). The age model of core PS1768-8 1991]. On the investigated transect south of Africa, the is basedon a d•80 recordof theplanktic foraminiferal species location of the polar front varies between 49ø39'S and Neogloboquadrinapachyderma sin. [Niebler, 1995], which is 50ø47'S, and the location of the Subantarctic front varies supported by three accelerator mass spectrometry (AMS) between 45 ø15'S and 47ø25'S [Lutjeharms and Valentine, •4C-dated samplesin the upper section of that core 1984]. The surface waters are rich in nutrients and show (R. Gersonde et al., manuscript in preparation, 1997). For significant gradients in concentration across the core PS1772-8, diatom and radiolarian biofluctuation oceanographicfronts. For the investigationof the barium stratigraphiesand biostratigraphieswere used to determinethe content, undisturbed surface sedimentswere sampled in the stage boundaries (A. Abelmann et al., unpublished report, area from the southernWeddell Sea to the Subtropical front 1992). All the age models of the long sediment cores were during severalR/V Polarstern cruisesfrom December1986 to combined with a 230Thex constant flux model to evaluate January 1994 (Figure 2). In addition, four sediment cores sedimentationrates at high resolution [Frank et al., 1996]. were recovered in the eastern Atlantic Ocean from each of the By normalizationto 230Thex,the sedimentaccumulation rates above mentioned zones on a transect between 43øS and 56øS were corrected for lateral sediment redistribution [Frank et al., 1996] in order to estimate the vertical rain rates of excess barium (VRR Baexcess)from the overlying water column. 30øW 20ø 10ø 0ø 10ø 20øE j_{ii I I I I I I I I I I I I Ill I 111I I IIII I II I I I IIII I I11I I Illl I I '130ø For the determination of the barium content, the decarbonated sediment samples were dissolved by acid • I Meansummer digestion. The sample material (100 mg) was weighed 30øH• !•$eaiceedge Meanwinter Cap•e••• directly into Savillex Teflon cups. The digestion was performed by applying 3 mL HF (40%), 3 mL HNO3 (65%), • :""::•Maximumwinter Basi and 4 mL HC104 (60%) at 180øC for 10 hours. SUBANTARCTICZ• su•;o½C• PS2082-1 ß Subsequently,the acid was evaporated to dryness, and the residue was dissolved by 1 mL HC1 (30%). Finally, the •) •a• POLARFRONTALZONE Teflon cups were filled to 50 mL with deionized water and • PS1756-5ß stored in precleanedpolyethylene bottles. The sampleswere 50ø --• 50ø analyzed by inductively coupled plasma atomic emission ANTARCTICZONE spectrometry (ICP AES). Accuracy was controlled by ...... ,..•..,.:•...... ,.•...... PS1768-8 ß ß .::..- •$:'•4•:_..::-.'--'-'•5i:•'.'•:'.-•:.--,•...... Bouv.e t.I..s....L.-::.:•:--.:•...-.•.:.-:-.• ...... analyzing international reference material (MAG-1). Deduced from replicates, the precision of the method is about 2% for • -'.i•:•:..-:½d•i::..'-:•'.::..!'•:----:.-i:•:•:•;•.?:::•:•:::•::::•g?•::i:;•:i:•:.•.:.•.•`.•::•i•:•:.•.•.•::::•*•:?•?•;:•:•::•.:•:?.•:•::•;;:[:;•:•:*•.:-:....•:Ai+g::.:•-:.•?-.':•i•ii:::-:'"'""'-'•-'•:,--•ß :.:.:::.• '"'•::•:•ii::..:-•:.-'•i:.½:.-•ß .•...... i•:.'•i•i...... •---'---:-:::::::::::?--'-::•ji ...... •g;7•::•::•:•?:•:•:•:•[i.•.•.`•d•?•f:[i[i::•::•:•:•i::•!ii[::•i?•:d• :"'-::'•-:'.-•::•::•:::::: ß I'""'"'"•'"'•'•'""•"'•''""'•'•'••••'i':i• ! barium and 3% for aluminum. A detailed descriptionof the ...... methodwas reportedby Nfirnberg [1995]. • "..:.'":::,•:: ::•::•::•::• ::•::•::• ':"%'"L...... •...... I1.::.•.•:•:•5•::•i:S.Sandwich •'iiii:•:•i•::::::::!ii::i::::,:•,• •ii:::i::::..'"..i?:::::::::i •:•:•:•:•:::::::::::::::::::::::::::....• Before interpreting the barium signal as a J'•111•mm'mmmm m•m.m. "immm'm mmm'm mmmm m •I::•: :•:'::i 'l 'm:'im i'mm N mm' mmmmm mmmmmmmm mmmmmmmm mmmm' paleoproductivityindicator, diagenetic redistributionhas to 30øW 20ø 10ø 0ø 10ø 20øE be considered.In areaswith high sedimentaccumulation and a Figure 1. Shown are locations of sediment cores high input of organicmaterial, sulfate reduction could develop studied. Surface water circulation and oceanographic near the sedimentsurface due to the microbialdegradation of fronts are according to Hellruer et al. [1985], organic carbon. A diageneticmobilization of barite occursin Whitworth and Nowlin [1987], and Peterson and anoxic sediments, often exhibiting a subsequent Stramma [ 1991 ]. reprecipitationin the sulfate-bearingenvironment [Bruinsack, 596 NORNBERGET AL.: BARIUM ACCUMULATION IN THE SOUTHERN OCEAN
70oW 60ø 50ø 40ø 30ø 20ø 10ø 0 ø 10ø 20øE 30ø ;"...... • 30ø : ....-......
U • - :•"•...... 4('."'.'e:::•F • 40ø I ."•' •;•'• ...... • ' • : '...... • ...... -.-•,-----.- •...... •...... ß I Eton :•:, •'•.•:']•.•.-."• • .-..': '•: ß • •:•'•:DD• -.•:•:•;•...... •?•:•?"'•';;'•:.-- -•• • ArgentineBasin :•::•:•e•:::•:;,. ' ...... •'...... -...... , '•;• •:e.•;...... '.•;.:' , e• ß
.0 ß / ß :' • • ::e...
'e ß d Rise I
I/ % Weddell Sea 70 ø 70 ø / /%l' / / ß .. I :•,- ...... ,.? / I .... .:•...... - •:?•? r' •'•'::::"•'•:'•' '.•:.•.••' • '""-e.;.",M • (.•.::•/ß •:•;' %/ I I I
ExcessDetr. correctionbarium = 0.0067 (ppm) I •...... _<500 I
' ?:•::::' Antarctica • >1000-2000>soo-•ooo I ...... •.•:.: •.- %%/% Shelf•'/%/%/% /%/%/%/% • >2000-3000 %/% % % % % % % % % % %%%% __ ß
I / / / / / / / / / / /Filchnerlce / ! %%%%%%%%%%...... 0 Surface sample ! 80ø5i%•-•--•-••"•"•"•••-•••"•-•••••••• % % % % % % % % % .,.,.,•.. . % % % 80ø5 70oW 60ø 50ø 40ø 30ø 20ø 10ø 0ø 10ø 20øE
Figure 2. Concentrations of excess barium in surface samples from the southern South Atlantic. Surface water circulation and oceanographic fronts are according to Hellruer et al. [1985], Whitworth and Nowlin [1987], and Peterson and Stramma [1991].
1989; Torres et al., 1996]. Within the Soctia Sea, Niirnberg the water column. Such scavenging is not expected in the [1995] measuredpore water in a sedimentcore, which is built Weddell Sea sediments, because the A1/Ti ratios in sediment up of oxic to suboxic sedimentsas is indicated by pore water cores show no downcore variation and therefore are indeed of sulfate concentrations of 28-29 mmol/kg. Mobilization and crustal origin [Bonn, 1995]. South of South Orkney Island, redistribution of barium were not expected, becausethe pore the A1/Ti ratios vary considerably (W. Bonn, personal water barium concentrations resemble those within the barite communication, 1996), thus pointing to a significant saturationzone publishedby Church and Wolgemuth[1972]. scavenging of aluminum. Nevertheless, the differences Although we have no pore water data for the investigated between the barium data corrected by Ti and A1 are minor sediment cores, we assume that mobilization and (108_+33 ppm; n = 123). There are no titanium data for redistribution of barium did not take place, becausethe cores the investigated sediment cores to get the scavengedportion have similar lithology, Corgaccumulation rates, and linear of aluminum. Niirnberg [1995] comparedthe Ba/A1 ratios in sedimentation rates to the one in the Scotia Sea. sediment cores with the accumulation rates of excess barium, In order to apply the barium signal as a proxy for which were both used to derive information about the paleoproductivity,the total barium had to be correctedfor the productivity. Both parameterscorrelate well in core sections nonbiogenic portion. Aluminum, the most characteristic with high terrigenousinput. In core sectionswith aluminum element of aluminosilicates, is assumed to represent the concentrations less than 2%, however, no correlation is terrigeneous portion of clay minerals [Shimmield et al., observed. We therefore conclude that 2% of the aluminum is 1990]. Murray et al. [1993] proposed that a portion of presumably scavenged by biogenic particles. Assuming a dissolvedaluminum is scavengedby biogenic particleswithin consistently lowered aluminum content by 2%, the excess NORNBERGET AL.:BARIUM ACCUMULATION IN THESOUTHERN OCEAN 597
PS2082-1 PS1756-5
Ba (ppm) Baexcess(ppm) AI (%) Isotope Ba (ppm) Baexces,(ppm) AI (%) Isotope Stages 0 10002000 3000 1000 2000 3000 2 3 4 5 S6tages 0 1000 2000 3000 1000 2000 3000 2 3 4 5 o.
100'
2O0
3O0
4O0
500
6O0 7O0 800 500 9OO 6OO
1000
1100 70O
1200 8O0
1300
1400 9O0 - Terrigenous barium ß, Terrigenous barium ß Total barium ß Total barium
PS1768-8 PS1772-8
Ba (ppm) Baexcess(ppm) AI (%) Isotope Ba (ppm) Baexc.ss(ppm) AI (%) Isotope 0 1000 2000 1000 20002 4 6 8 Stages 0 500 10001500 500 10001500 2 4 6 Stages o cole lOO
200
' 2 300
400
300 500
600 4OO
700 .½: 500 4 •oo
600
lOOO
1 lOO
1200
13oo
1400 ß, Tarrigenous barium Tarriganous barium ß Total barium Total barium
Legend Foraminifera-bearingmud, ' .[• Diatomaceousmud •.... Diatomaceousooze Foraminifera ooze, and Nannofossil mud Figure 3. Barium and aluminumconcentrations in sedimentcores. Stratigraphyis according to Mackensen et al. [1994],Niebler [1995],Frank et al. [1996], G. Fischeret al. (unpublished manuscript, 1997), and A. Abelmann et al. (unpublishedreport, 1992). barium would be enhancedby approximately134 ppm. Such For the Atlantic sector of the Southern Ocean, a regional error is, to our mind, negligible in view of the up to (Ba/A1)crustratio is developed from three core-top samples, 4700 ppm high excess barium concentrations. which were situated beneath the Filchner Ice Shelf and thus Owing to lack of titanium data, we normalized the total were characterizedby a negligible biogenic input for a long barium by aluminum and applied the following calculation: time period [Schliiter, 1991]. The barium contentin these samplesis thereforeof terrigenousorigin. The corresponding (1) Baterr= Alsamplex (Ba/A1)crust (Ba/A1)crustratio is 0.0067, which is applied to calculate the where Baterris the terrigenous barium, Alsampleis the excess barium: aluminum contentin the sample,and (Ba/A1)crustis the barium to aluminum ratio of the crust. Baexcess=Batot- (A1 x 0.0067) (2) 598 NORNBERGET AL.:BARIUM ACCUMULATION IN THESOUTHERN OCEAN
Table 1. Sur•ce SedimentSampling Positions Broadway, Boulder, CO, 80303, phone 303 497 6280, fax 303 497 6513, Internet e-mail: [email protected]). Sample LSR MAR AR Baexcess Within the polar frontal zone (PS1756-5) the excessbarium cm kyr-l g cm-2 kyr-l lagcm -2 yr-• concentrationsare around500ppm and do not vary significantly with depth, whereas excess barium exhibits PS1654-1 60.0 36.0 71.68 drastic downcore variations in all other cores. Within the core PS 1751-2 0.5 0.3 1.04 sections characterized by diatomaceousooze, the aluminum PS1752-5 0.5 0.3 1.17 concentrationsare significantly lowered (<1%). All other PS1754-2 1.5 0.9 2.79 PS1755-1 2.0 1.2 5.63 sectionsexhibit aluminumconcenl•rations up to 7%. PS1756-6 7.0 1.9 4.36 On the basisof the 23øThex normalizationapproach to PS1764-2 41.0 24.6 46.56 correct for focusing and winnowing [Frank et al., 1996], PS1765-1 25.0 5.5 5.15 excess barium accumulation rates were converted to vertical PS1768-1 15.0 7.0 7.81 PS1651-2 19.0 11.4 8.48 rain rates (Figure 5 and Table A2). Downcore vertical rain PS1652-2 55.0 33.0 20.99 rates of excess barium vary between 0.1 and 4.8 gg cm- PS1771-4 25.0 15.0 10.53 2 yr-l. CorePS1768-8, located north of today'saverage sea PS 1649-1 0.8 0.5 0.56 ice maximum within the Antarctic zone, shows the highest PS 1772-6 0.5 --- 0.4 0.25 PS1773-2 0.3 0.2 0.30 vertical rain rates of 4.2 and 4.8 gg cm-2 yr -l at the termination oxygen isotope chronozone 2 to 1 and during Abbreviations are LSR, linear sedimetationrate; MAR, mass oxygen isotope chronozone 5.5, respectively. In the accumulationrate; and AR Baexcessaccumulation rate of Baexcess Antarctic zone with winter sea ice (PS1772-8), high vertical rain rates of excess barium exclusively occur during oxygen isotopechronozone 5.5 (3.9 gg cm-2 yr-•). CorePS1756-5, where Baexcess is the excess barium, Batot is the total, locatedin the polar frontal zone, showsmaximum rates during (measured) barium, and A1 is the aluminum content of the oxygen isotope chronozone 2, 3 and 5.5 of about sample. The calculated (Ba/A1)crustratio matchesthe global 2.6 gg cm-• yr -•. North of the Subantarcticfront (Ba/A1)aluminosilicateratio of 0.0065 given by Wedepohl (PS2082-1), the pattern of excess barium accumulation [1991]. changes substantially. The highest vertical rain rates occur
Results duringglacials (1.0 to 1.3gg cm-2 yr-•).
In surface sediments, the excess barium concentrations Excess Barium Distribution in Surface Sediments range from a few ppm to about 4700 ppm in this study The modern Southern Ocean is a high-nutrient (Figure 2 and Table A1; the data of Table A1 have been low-chlorophyll (HNLC) area, where iron limitation may archived,and are available in a digital form, at the World Data causethe incomplete consumptionof the nutrients[Martin et Center-A for Paleoclimatology,NOAA/NGDC 325 Broadway, Boulder, CO, 80303, phone 303 497 6280, fax 303 497 6513, Internet e-mail: [email protected]). Within the Antarctic Circumpolar Current, the excessbarium contents west of the Greenwich meridian are lower than 2000 ppm. The highest excessbarium concentrations(>3000 ppm) occur in the polar frontal zone of the eastern South Atlantic. 5OOO Sediments along the coastal area west of 10øW, near islands (Falkland, SouthOrkney, SouthSandwich, South Georgia, and 4OOO 6O Bouvet), in the Weddell Basin, and in the Argentine Basin show very low values (<500 ppm). •- 3000 The regionally varying accumulationof biogenic sediment 4O components has to be considered when interpreting their m 2000 distribution patterns. Recent accumulation rates of excess • v 2O 1 ooo barium (AR Baexcess)were determinedfor sites lying on a .<3 NE-SW trending transect that crosses the oceanographic fronts at 12øE to 0 ø in the easternSouth Atlantic (Figure 4 and 0 Table 1). The dominant portion of excess barium is 46 48 50 52 54 56 58 accumulatedwithin the Antarctic zone just south of the polar Degree southern latitude front, whereasareas commonly coveredby winter sea ice and areas north of the polar front exhibit very low accumulation Figure 4. Concentrations of excess barium and accumulation rates of excess barium in surface rates of excess barium. sediments from a NE-SW trending transect crossing The downcore records in cores PS2082-1, PS1756-5, the oceanographic fronts at 12øE to 0 ø in the eastern PS1768-8, and PS1772-8 reveal excess barium concentrations South Atlantic. The polar frontal zone and the mean up to 2700 ppm (Figure 3 and Table A2; the data of Table A2 winter sea ice are shaded. Stratigraphical data are have been archived, and are available in a digital form, at the according to A. Abelmann et al. (unpublished report, World Data Center-A for Paleoclimatology,NOAA/NGDC 325 1992). NORNBERG ET AL.: BARIUM ACCUMULATION IN THE SOUTHERN OCEAN 599
PS2082-1 PS1756-5 PS1768-,8 PS1772-8 Verticalrain rate of Baexcess[!•g cm-2 yr-1 ] Isotope Age 0 I 2 3 I 2 3 I 2 3 I 2 3 Stage [kyr]
12 20 24
40'
60' 59
74 ..• 8o.
100'
120'
130 140
160
180
N Subantarctic Polar Frontal Antarctic Antarctic Zone with Zone Zone Zone winter sea ice
Figure 5. Vertical rain rates of excess barium (VRRBaexcess) down to oxygen isotope chronozone 6 in cores of the southern South Atlantic. The interglacial isotope stages are shaded. The dashed line connects the first core sample with the corresponding surface sample. Stratigraphy is according to Mackensen et al. [1994], Niebler [1995], Frank et al. [1996], G. Fischer et al. (unpublished manuscript, 1997), and A. Abelmann et al. (unpublished report, 1992). al., 1990; Kurnar et al., 1995]. The Antarctic surface water is rapidly. This pattern is also reflected by the accumulationof enrichedin inorganicnutrients, especially in the area southof opal [DeMaster, 1981; Charles et al., 1991; Mortlock et al., the polar front. Owing to upwelling at the Antarctic 1991]. Investigationsof Trdguer and Van Bennekorn[1991] divergence,a continoussupply of nutrientsoccurs. show that the highest production of biogenic silica occurs The recent distribution of excess barium in the sediments within the surface layer of the ice free and seasonally ice (Figure 2)is also reflected by the biogenic silica content. coveredAntarctic zone, whereaswithin the polar frontal zone, [DeMaster, 1981; M.SchlQteret al., Silica cycle in surface' low biogenic silica values are found. Suggestedby remote sediments of the South Atlantic, submitted to Deep Sea sensingdata, the primary productionrate within the Antarctic Research, 1996] with maxima on the Weddell Sea shelf, in the CircumpolarCurrent is no morethan 40-50 g C m'2 yr '1 Scotia Sea, and in the eastern Atlantic sector of the Antarctic [Antoine et al., 1996]. Nevertheless, the high opal Circumpolar Current. High excessbarium concentrationsin accumulationoccurs south of the polar front, probablydue to a the Weddell shelf depositsare reflected by high productionof low dissolution rate of siliceous material in that region biogenic silica [Schliiter, 1991]. In the deep centralWeddell [Nelson et al., 1995]. North of the polar front and within Sea, instead, very low excess barium concentrations were seasonallyice covered areas, only low amountsof siliceous observed. This is in accordance with low benthic fluxes of material are apparently exported, mainly due to a high biogenic silica [Schliiter, 1991] and with the smallest annual dissolution rate of silicate in surface waters. particle flux in that region yet observedin the world ocean [Wefer and Fischer, 1991]. In the ScotiaSea, the high excess Barium as a Paleoproductivity Indicator barium concentrations correlate with high integrated production rates of biogenic silica, whereas south of that Within the northern Antarctic zone south of the polar front, region within the marginal ice zone of the northern Weddell peak vertical rain rates of excessbarium were reachedduring Sea, lower biogenic silica values are determined[Qudguiner et the last deglaciation and the following Holocene climatic al., 1991]. The maximum excess barium accumulation within optimum (circa 14 - 8 ka B.P.) and oxygen isotope the Antarctic Circumpolar Current occursin the Antarctic zone chronozone 5.5 (Figure 5). The high productivity interval between the polar front and the winter sea ice boundary during oxygen isotope chronozone5.5 is also documentedat (Figure4). North of the polar front and within the ice covered sites PS1756-5 and PS1772-8. Core PS1772-8 was recovered Antarctic zone, excess barium accumulation decreases from the southern Antarctic zone where productivity is- 600 NORNBERG ET AL.' BARIUM ACCUMULATION IN THE SOUTHERN OCEAN
Table 2. AverageVertical Rai_n Rates of ExcessBarium in theEastern South Atlantic
Average Vertical Rain Rate Baexcess, gg cm-2 yr-1
Core Core Core Core PS2082-1 PS1756-5 PS1768-8 PS1772-8
Holocene (0-10 kyr) 0.5 + 0.1 2.1 1.9 + 0.5 0.8 + 0.1 Last glacial maximum (16-26 kyr) 0.6 + 0.2 1.4 + 0.2 0.9 + 0.2 0.3 + 0.01
recently low due to annual sea ice cover (Figure 4). Since sea periods. During the Holocene,high averagevertical rain rates ice cover leads to diminished new productivity, the high of excessbarium of about2.1 and 1.9gg cm -2 yr-• occurin excessbarium vertical rain ratesof up to 3.9 ggcm -2 yr-] cores PS1756-5 and PS1768-8, respectively (Table 2). during interglacial oxygen isotope chronozone5.5 can only Farther north, the average vertical rain rates decrease, be explained by a retreat of the sea ice edge during this resultingin a very low value of 0.5 gg cm-2 yr-• in the climatic optimum. Sea ice cover, instead, expanded Subantarctic zone (43øS; PS2082-1). significantly during glacial periods [e.g., Cooke and Hays, During the LGM, the highest vertical rain rates of excess 1982; Olaussen, 1988] as deduced from the lower vertical rain barium are found at 49øS in core PS1756-5 rates at site PS1768-8 during oxygen isotopechronozone 2-4 (1.4 gg cm-2 yr-•). The verticalrain rates in corePS2082-1 (Figure 5). During glacial periodsnorth of the polar frontal are somewhat higher during the LGM than during the zone (PS2082-1), vertical rain rates of excess barium were Holocene. Summersea surfacetemperature estimates based on considerablyhigher than during interglacials,pointing to an radiolarian transfer functions in core PS2082-1 were 4ø-5øC enhancednew production. However, the glacial vertical rain lower than today [Brathauer, 1996], indicating conditions rates of the Subantarctic zone are not comparable to similar to the recent southern Subantarctic zone, i.e., a interglacial rates southof the polar front. northward migration of the Subantarcticfront. The vertical To explain regional patterns during glacial/interglacial rain rates of excess barium in core PS 1768-8 are similar to the changes in the Atlantic sector of the Southern Ocean, we Holocenerates in core PS1772-8 (0.9ggcm-2yr-•), averagedvertical rain rates of excessbarium for the Holocene indicating that this site location was densely covered by (0-10 ka) and the last glacial maximum ((LGM) 16-26 ka; winter seaice duringthe LGM. Accordingto Gersondeet al. Table 2). Significant variations between 0.3 and [1994], the mean winter sea ice boundary was located near 2.1 gg cm-2 yr-] are apparentduring glacial/interglacial 50øS. Summer sea surface temperaturesof-1.3øC basedon
Holocene Last glacial maximum
10øW 0 ø 10 ø 20øE 10øW • 10 ø 20øE 30 ø
Cape Basin
40 ø
F;ont PS2082-1ß
PS175&1
PS1756-5
50 ø
I• PS1768-8ß
60øS
10øW 0 ø 10 ø 20øœ 10øW 0 ø 10 ø 20øœ
Figure 6. Schematic diagram showing the sea ice distribution and the location of the oceanographic fronts in the eastern sector of the South Atlantic during the Holocene and the last glacial maximum according to our interpretation. The zone of maximum export production is between the polar front and the sea ice which is shaded. NORNBERG ET AL.: BARIUM ACCUMULATION IN THE SOUTHERN OCEAN 601 diatom transfer functions further indicate that the summer sea region is also characterizedby high opal accumulationrates. ice boundary was situated south of the site location of In conclusion,our investigationssupport the applicability of PS1768-8 [Zielinski, 1993]. Frank [1996] investigated a barium as a productivityindicator within the SouthernOcean. sediment core on the same transect (PS1754-1; 46ø46'S, 2. During the interglacial oxygen isotope chronozones1 007ø37'E; Figure 6) and observedhigh vertical rain rates of and 5, the area south of the Subantarctic zone was excessbarium during the LGM (1.9gg cm -2 yr-1), which characterized by high excess barium vertical rain rates. wereas high as duringthe Holoceneat the polarfront. Niebler Especially within the Antarctic zone, maximum vertical rain [1995] reconstructedtemperatures based on oxygen isotope rates of excess barium were observed during the last recordsof planktic foraminifersfor this core, which provided deglaciation and following climatic optimum and the oxygen evidencefor temperatureconditions in the modern southern isotope chronozone 5.5, pointing to a more expanded zone polar frontal zone. We thereforesuspect that the polar front with maximum export to the south than today. During migrated by about 3ø latitude northward during the LGM. glacials, within the Subantarctic zone, enhanced excess Consequently,core PS1756-5 was locatedsouth of the polar barium accumulation is observed, but the vertical rain rates of front. This pattern is in accordancewith observationsof excessbarium are as high as south of the Subantarcticfront. Morley and Hays [1979], who postulateda northwardshift of Therefore export production was not higher in the glacial the polar front during the LGM by about30-5 ø latitudein the SouthernOcean. In conclusion,changes in productivityseem western Atlantic and 1 ø-3ø latitude in the eastern Atlantic. to be controlled by sea ice fluctuationswithin the Antarctic Figure6 illustratesthe sea ice distributionand the location zone and expansionsto the north. of the oceanographicfronts in the easternAtlantic sectorof the SouthernOcean during the Holoceneand the LGM. Recent Acknowledgments.For intensivediscussion and improvementof the oceanographicinvestigations by Lutjeharrnsand Valentine manuscriptwe thank D. Niirnberg and R. Keir. For critical reviews we are grateful to R.W. Murray, J. Dymond, and D. Lea. We thank [1984] deducedthat the polar front variesbetween 49ø39'S and G. Kuhn for providing unpublishedcarbonate data. This study was 50ø47'S due to a recent location of core PS1756-5 within the funded by German Science Foundation (DFG) grant Bo1049-1. We polar frontal zone. However,the extremelyhigh verticalrain acknowledge the assistanceof R/V Polarstern's crews and masters rates of excess barium in the surface sediment of core during numerouscruises to the SouthernOcean. PS1756-5 (48ø54'S) dated to 0.5 ka B.P. point to a location at the polar front. Thereforeit is apparentthat the front was References mainly located 1ø-2øfarther north during the entire Holocene Antoine, D., J.-M. Andr6, and A. Morel, Oceanicprimary production, time span(Figure 6). The exportproduction was probablynot 2, Estimation at global scale from satellite (coastal zone color higherin the glacialSouthern Ocean, because during the LGM scanner)chlorophyll, Global Biogeochern.Cycles, 10, 57-69, 1996. the vertical rain rates of excessbarium are as high as in the Bareille, G., M. Labracherie, P. Bertrand, L. Labeyrie, G. Lavaux, and M. Dignan, Glacial-Interglacialchanges in the accumulationrates modem zone of maximum export production. Also, the zone of major biogeniccomponents in southernIndian Oceansediments, of maximumexport productionexpanded by about6 ø during J. Mar. Syst.,in press,1997. the Holoceneand by about4 ø during the LGM, pointingto a Berger,W. H., R. C. Finkel, J. S. Killingley, and V. Marchig,Glacial- lessextended high-productivity belt. This interpretationis in Holocenetransition in deep-seasediments: Manganese-spike in the accordancewith biogenic silica productivity estimates of east-equatorialPacific, Nature, 303, 231-233, 1983. Bernstein, R. E., R. H. Bryne, P. R. Betzer, and A.M. Greco, Mortlock et al. [ 1991] and Charles et al. [ 1991]. Charles et al. Morphologiesand transformationof celestitein seawater:The role [1991] explain the higher opal burial rates in the southern of acantharians in strontium and geochemistry, Geochim. Subantarctic zone (46ø-49øS) with a northward shift of the Cosrnochirn.Acta, 56, 3273-3279, 1992. "silica front." Also, Bareille et al. [1997] and Francois et al. Bishop,J. K. B., The barite-opal-organiccarbon association in oceanic particulatematter, Nature, 332,341-343, 1988. [1995] do not observe any increase in export production Bonn, W. J., Biogenopal und biogenesBarium als Indikatoren fiir during glacial times in the Indian sector of the Southern sp•itquart•ire Produktivit•its•inderungen am antarktischen Ocean. In contrast,Kumar et al. [1995] postulateda Fe-based Kontinentalhang,atlantischer Sektor., Ber. Polarforsch.,180, 1-186, enhancedglacial productivity north of the presentpolar front 1995. in the Atlantic sector of the Southern Ocean. This is Bostr6m, K., O. Joensuu, C. Moore, B. Bostr6m, M. Dalziel, and A. Horowitz, Geochemistryof barium in pelagicsediments, Lithos, 6, supportedby more recent investigationsat the Antarctic polar 159-174, 1973. front of the Atlantic sector showing that higher values of Boyle, E. A., Cadmium' Chemical tracer of deepwater dissolved iron significantly increase primary production paleoceanography,Paleoceanography, 3, 471-489, 1988. north of the Antarctic zone [De Baar et al., 1995; Ltscher et Boyle,E. A., Cadmiumand d13Cpaleochemical ocean distributions al. , 1997]. during the stage2 glacial maximum,Annu. Rev. Earth Planet. Sci., 20, 245-287, 1992. Conclusion Brathauer, U., Rekonstruktion quart•irer Klima•inderungen im atlantischenSektor des Siidpolarmeeresanhand von Radiolarien, Bet. Polarforsch.,216, 1-163, 1996. The investigations of excess barium and the subsequent Broecker,W. S., and T.-H. Peng, Tracersin the Sea, 690 pp., Lamont- estimationof vertical rain rates supportprevious descriptions DohertyGeol. Obs.,Columbia Univ., Palisades,N.Y., 1982. of the Quaternarydepositional environment in this area. The Brumsack,H.-J., Geochemistryof recentTOC-rich sedimentsfrom the sedimentarysignals can be summarizedas follows: Gulf of California and the Black Sea, Geol. Rundsch., 78, 851-882, 1989. 1. The distribution of excess barium in the Weddell Sea and Chan, L. H., D. Drummond,J. M. Edmond,and B. Grant, On the barium the Antarctic CircumpolarCurrent reflectsthe distributionof data from the Atlantic GEOSECS Expedition, Deep Sea Res., 24, the production of biogenic silica. Maximum accumulation 613-649, 1977, rates of excess barium occur within the Antarctic zone. This Charles,C. D.,P. N•Froelich, M. A. Zibello,R. A. Mortlock,and J. J. 602 NORNBERG ET AL.: BARIUM ACCUMULATION IN THE SOUTHERN OCEAN
Morely, Biogenic opal in southernocean sedimentsover the last and export productionin the glacial SouthernOcean, Nature, 378, 450,000 years: Implications for surface water chemistry and 675-680, 1995. circulation,Paleoceanography, 6, 697-728, 1991. Lea, D., and E. Boyle, Barium content of benthic foraminifera Chow, T. J., and E. D. Goldberg, On the marine geochemistry of controlledby bottom-watercomposition, Nature, 338, 751-753, 1989. barium, Geochim. Cosmochim. Acta, 20, 192-198, 1960. Lea, D. W., and E. A. Boyle, Foraminiferal reconstructionof barium Church,T. M., Marine barite, in Reviewsin Mineralogy, 6, editedby R. distributions in water masses of the glacial oceans, G. Burns, pp. 175-209, Mineral. Soc. of Am., Washington,D.C., Paleoceanography,5, 719-742, 1990. 1976. Lea, D. W., and E. A. Boyle, Barium in planktonic foraminifera, Church, T. M., and K. Wolgemuth, Marine barite saturation, Earth Geochim. Cosmochim.Acta, 55, 3321-3331, 1991. Planet. Sci. Lett., 15, 35-44, 1972. Lea, D. W., G. T. Shen, and E. A. Boyle, Coralline barium records Cooke, D. W., and J. D. Hays, Estimatesof Antarctic Ocean seasonal temporalvariability in equatorial Pacific upwelling,Nature, 340, sea ice cover during glacials intervals, in Antarctic Geoscience, 373-376, 1989. edited by C. Craddock, pp. 1017-1025, Univ. of Wisc. Press, Li, Y. H., T. L. Ku, G. G. Mathieu, and K. Wolgemuth,Barium in the Madison, 1982. Antarctic Ocean and implications regarding the marine De Baar, H.J.W., J.T.M. de Jong, D.C.E. Bakker, B.M. L6scher, C. geochemistryof Ba and 226Ra,Earth Planet.Sci. Lett., 19, 352-358, Veth, U. Bathmann, and V. Smetacek, Importance of iron for 1973. plankton blooms and carbon dioxide drawdown in the Southern L6scher, B. M., J. T. M. de Jong, H. J. W. de Baar, C. Veth, and F. Ocean, Nature, 373, 412-415, 1995. Dehairs, The distribution of Fe in the Anatarctic Circimpolar Dehairs, F., and L. Goeyens, The biogeochemistryof barium in the Current,Deep Sea Res., Part II, 44, 143-187, 1997. Southern Ocean, Belg. Sci. Res. Program Antarct., Sci. Results Lutjeharms,J. R. E. and H. R. Valentine, SouthernOcean thermal fronts Phase One (Oct 85-Jan 89), II, 1-99, 1989. southof Africa, Deep Sea Res., 31, 1461-1475, 1984. Dehairs, F., R. Chesselet,and J. Jedwab,Discrete suspendedparticles Mackensen, A., H. Grobe, H.-W. Hubberten, and G. Kuhn, Benthic of barite and the barium cycle in the open ocean,Earth Planet. Sci. foraminiferalassemblages and the dl3C-signalin the Atlanticsector Lett., 49, 528-550, 1980. of the SouthernOcean: Glacial-to-interglacialcontrasts, in Carbon Dehairs, F., L. Goeyens, N. Stroobants, P. Bernard, C. Goyet, A. Cycling in the Glacial Ocean: Constraintson the Ocean2v Role in Poisson, and R. Chesselet, On suspendedbarite and the oxygen Global Change,edited by R. Zahn, T. F. Pedersen,M. Kaminzki,and minimumin the SouthernOcean, Global Biogeochem.Cycles, 4, 85- L. Labeyrie, NATO ASI Ser., Ser. I: Global Environ. Change, 102, 1990. 105-144, 1994. Dehairs, F., N. Stroobants, and L. Goeyens, Suspendedbarite as a Martin, J.H., S.E. Fitzwater, and R.M. Gordon,Iron deficiencylimits tracer of biological activity in the SouthernOcean, Mar. Chem., 35, phytoplanktongrowth in Antarctic waters, Global Biogeochem. 399-410, 1991. Cycles,4, 5-12, 1990. DeMaster, D. J., The supply and accumulationof silica in the marine Morley,J. J., andJ. D. Hays,Comparison of glacialand interglacial environment, Geochim. Cosmochim. Acta, 45, 1715-1732, 1981. oceanographicconditions in the South Atlantic from variationsin Dymond, J., E. Suess,and M. Lyle, Barium in deep sea sediment:A cal-cium carbonateand radiolariandistributions, Quat. Res., 12, geochemicalindicator of paleoproductivity,Paleoceanography, 7, 396-408, 1979. 163-181, 1992. Mortlock, R. A., C. D. Charles, P. N. Froelich, M. A. Zibello, J. Elderfield, H., Tracers of oceanpaleoproductivity and paleochemistry: Saltzmann, J. D. Hays, and L. H. Burckle, Evidence for lower An introduction,Paleoceanography, 5, 711-717, 1991. productivity in the Antarctic Ocean during the last glaciation, Francois,R., M. A. Altabet, M.P. Bacon,E. F. Yu, D. S. Sigman,and L. Nature, 351,220-223, 1991. Labeyrie, Export flux and nutrientutilization in the SouthernOcean Murray, R. W., M. Leinen, and A. R. Isern, Biogenic flux of A1 to duringthe last glacial maximum(abstract), International symposium sediment in the Central Equatorial Pacific Ocean: Evidence for on carbonfluxes and dynamicprocesses in the SouthernOcean: increasedproductivity during glacial periods.,Paleoceanography, 8, Presentand past, 53-54, Brest,France, 1995. 651-670, 1993. Frank, M., Reconstructionof Late Quaternaryenvironmental conditions Nelson, D. M., P. Tr6guer, M. A. Brzezinski, A. Leynaert, and B. applyingthe naturalradionuclides 23øTh, løBe, 231pa and 238U: A Qu6guiner, Production and dissolution of biogenic silica in the studyof deep-seasediments from the easternsector of the Antarctic ocean:Revised global estimates,comparison with regionaldata and CircumpolarCurrent System, Ber. Polarforsch.,186, 1-136, 1996. relationshipto biogenicsedimentation, Global Biogeochem.Cycles, Frank, M., R. Gersonde,M. Rutgers van der Loeff, G. Kuhn, and A. 9, 359-372, 1995. Mangini, Late Quarternary sediment dating and quantification of Niebler, S., Rekonstruktion yon Pal•io-Umweltparameternanhand yon lateralsediment redistribution applying 230Thex: A studyfrom the stabilen Isotopen und Faunenvergesellschaftungenplanktischer eastern Atlantic sector of the Southern Ocean, Geol. Rundsch., 85, Foraminiferen,Ber. Polarforsch.,167, 1-198, 1995. 554-566, 1996. Ntirnberg,C. C., Bariumflugund Sedimentationim stidlichenStidatlantik Gersonde,R., A. Abelmann,G. Bohrmann,Southern high latitude - Hinweise auf Produktivit•its•inderungenim Quart'•, Geomar Rep., paleo-environmentalevolution during the last deglaciationand the 38, 1-105, 1995. Holocene (Atlantic sector) (abstract),Berichte, Fachbereich Olaussen, E., The Southern Ocean - the Antattic: Present and past, Geowissen-schaften,Univ. Bremen,52, 51-52, 1994. Paleogeogr.Paleoclimatol. Paleoecol., 67, 1-179, 1988. Goldberg, E. D., and G. Arrhenius, Chemistry of Pacific pelagic Peterson,R. G., and L. Stramma, Upper-level circulationin the South sediments,Geochim. Cosmochim.Acta, 13, 153-212, 1958. Atlantic Ocean,Prog. Oceanogr.,26, 1-73, 1991. Gooday, A. J., and J. A. Nott, Intracellular barite crystals in two Prahl, F., L. Muehlhausen, and M. Lyle, An organic geochemical Xenophyophores,Aschemonella ramuliformis and Galatheammina assessmentof oceanographicconditions at MANOP Site C over the sp. (protozoa:rhizopoda) with commentson the taxonomyof A. past26,000 years,Paleoceanography, 4, 495-510, 1989. ramuliformis,J. Mar. Biol. Ass.,U. K., 62, 595-605, 1982. Qu6guiner,B., P. Tr6guer, and D. M. Nelson, The productionof Hellmer, H. H., M. Bersch, E. Augstein, and I. Grabemann, The biogenic silica in the Weddell and Scotia Seas, Mar. Chem., 35, Southern Ocean: A survey of oceanographic and marine 449-459, 1991. meteorologicalresearch work, Ber. Polarforsch.,26, 1-115, 1985. Rau, G. H., P. N. Froelich, T. Takahashi, and D. J. Desmarais, Does Jumars,P. A., A. V. Altenbach, G. J. DeLange, S. R. Emerson, B. T. sedimentaryorganic dl3C recordvariations in Quaternaryocean Hargrave,P. J. Mtiller, F. G. Prahl, C. E. Reimers,T. Steiger,and E. [CO2(aq)]?,Paleoceanography, 6, 335-347, 1991. Suess, Transformation of seafloor arriving fluxes into the Revelle, R., M. Bramlette,G. Arrhenius,and E. D. Goldberg,Pelagic sedimentaryrecord, in Productivityof the Oceans:Present and past, sedimentsof the Pacific, Spec. Pap. Geol. Soc. Am., 62, 221-235, edited by W. H. Berger, V. S. Smetacek,and G. Wefer, pp. 291- 1955. 311, JohnWiley, New York, 1989. Schltiter, M., Organic carbon flux and oxygen penetration into Kumar, N., R.F. Anderson, R.A. Mortlock, P.N. Froelich, P. Kubik, B. sedimentsof the Weddell Sea: Indicatorsfor regionaldifferences in Dittrich-Hannen, and M. Suter, Increasedbiological productivity exportproduction, Mar. Chem.,35, 569-579, 1991. NORNBERG ET AL.: BARIUM ACCUMULATION IN TIlE SOUTI-IERN OCEAN 603
Schmitz, B., Barium, equatorialhigh productivity,and the northward von Breymann, M. T., K. C. Emeis, and E. Suess,Water depth and wanderingof the Indian Continent,Palaeoceanogaphy, 2, 63-77, diageneticconstraints on the use of bariumas a palaeoproductivity 1987. indicator,in UpwellingSystems: Evolution Since the Early Miocene, Shimmield, G. B., N. B. Price, and T. F. Pedersen, The influence of edited by C. P. Summerhayes,W. L. Prell, and K. C. Emeis, Geol. hydrography,bathymetry and productivity on sedimenttype and Soc. Spec.Publ., 64, 273-284, 1992. compositionof the Oman Margin and in the northwestArabian Sea, Wedepohl, K. H., The compositionof the upper earth's crust in the in The Geologyand Tectonicsof the Oman Region,edited by A. H. natural cycles of selectedmetals - Metals in natural raw materials., F. Robertsen,M.P. Searle,and A. C. Ries, Geol. $oc. Spec. Publ., in Metals and Their Compoundsin the Environment,edited by E. 49, 759-769, 1990. Merian, VCH, New York, 1991. Shimmield,G., S. Derrick, A. Mackensen,H. Grobe, and C. Pudsey, Wefer, G., and G. Fischer,Annual primary productionand export flux The history of barium, biogenic silica and organic carbon in the SouthernOcean from sedimenttrap data, Mar. Chem., 35, accumulation in the Weddell Sea and Antarctic Ocean over the last 597-613, 1991. 150,000 years, in Carbon Cycling in the Glacial Ocean: Constraints Whitworth, T., III, and W. D. Nowlin Jr., Water massesand currents of on the Ocean• Role in Global Change, edited by R. Zahn, T. F. the SouthernOcean at the GreenwichMeridian, J. Geophys.Res., Pealersen,M. Kaminzki, and L. Labeyrie, NATO ASI Ser., Ser. I: 92, 6462-6476, 1987. Global Environ. Change,555-574, 1994. Wolgemuth,K., and W. S. Broecker,Barium in seawater,Earth Planet. Stroobants,N., F. Dehairs,L. Goeyens,N. Vanderheijden,and R. Van Sci. Lett., 8, 372-378, 1970. Grieken, Barite formation in the Southern Ocean water column, Zielinski,U., QuantitativeBestimmung von Pal•ioumweltparameterndes Mar. Chem., 35, 411-421, 1991. Antarktischen Oberfl•ichenwassersim Spatquart'firanhand von Tendal, O. S., A monographof the Xenophyophoria (Rhizopoda, Transferfunktionenmit Diatomeen, Ber. Polarforsch.,126, 1-148, Protozoa),Galarhea Resp., 12, 7-103, 1972. 1993. Torres, M., H. Brumsack, G. Bohrmann, and K. C. Emeis, Barite fronts in continentalmargins: A new look at barium remobilizationin the G. Bohrmann,C. C. N'timberg,and M. Schltiter,GEOMAR Research zone of sulfate reduction and formation of "heavy" barites in Center, Kiel, Germany. (e-mail: [email protected]; authigenicfronts, Chem. Geol., I27, 125-139, 1996. dnuernberg@ geomar.de; mschlueter@ geomar.de) Tr6guer, P., and A. J. Van Bennekom, The annual production of M. Frank, Department of Earth Sciences,University of Oxford, biogenic silica in the Antarctic Ocean, Mar. Chem., 35, 477-487, Oxford, England. (e-mail: [email protected]) 1991. Turekian, K. K., and D. G. Johnson, The barium distribution in sea (ReceivedApril 24, 1996; revisedMarch 27, 1997; water, Geochim. Cosmochim. Acta, 30, 1153-1174, 1966. acceptedApril 17, 1997.)