one 3208 m by dissolution. e for calculating to calendar l rates. Our from

Transcription

1 A METHOD FOR QUANTIFYING DEEP-SEA CARBONATE i4c DISSOLUT ION DATING USING S. A. VAN KREVELD,1 G. M. GANSSEN,1 J. E. VAN HINTE,1 M. M. MELKERT 1 S. R. TROELSTRA,1 K. VAN DER BORG2 A. DE JONG2 ABSTRACT. We quantified the rate carbonate dissolution with increasing water bonate depth by mass taking accumulation the difference rate in the deep car- ( m) core top sediments from assumed the shallowest was unaffected one 3208 m by dissolution. which w This method depends on high quality 14C ( e for calculating dates that we calibrated sedimentation to calendar l rates. Our years results show low (ranging from 0 tp, 3 g cm ka-1) high (ranging g cm-2 ka-1) carbonate dissolution from 1.5 to rates, above below 4000 m, respectively. increase Therefore, in the carbonate we interpret dissolution the sudden rate at 4000-m water depth to mark the lysocline. INTRODUCTION Calcium selective carbonate deposition in the world's oceans is an important component cycle. the Because global carbon changes in the rate carbonate dissolution affect carbonate accumulation, be quantified. they must Previous attempts to estimate carbonate dissolution by means (Balsam er 1983; Farrell Pre111989), dissolution foraminiferal (Berger Berger 1970; Parker 1971: Berger von Rad 1972; Berger 1976), coccolith assemblages McIntyre (McIntyre 1971; Ber g er 1973a Roth Berger 1975; Schneidermann 1977), tonic benthic foraminifera to plankratios, percent fragments or a combination these methods (Arrhenius 1 Thunell 1976; Malm 952; gr en 1983) were qualitative. Francois, Bacon Suman bonate (1990) dissolution quantified carusing 230Th; de Vernal et al. (1992) used the relative abundance organic CaCO3 linings shells benthic foraminifera. We present here a simple method for bonate quantifying dissolution carusing the decrease in the carbonate mass accumulation rate with depth. increasing Note that water our carbonate dissolution rate estimates reflect only the amount calcite not aragonite. dissolved A slight change in sedimentation rate affects the carbonate mass rate accumulation eventually the carbonate dissolution rat e estimate. 14C Thus, it is essential to dates. have high-quality METHODS We used six box-core sediment tops recovered in the Northeast Atlantic between W N from mwater depths (Fig. 1, Table 1). These cores were JGOFS, collected Leg during 4 cruise aboard R/V Tyro in June C Carbonate dissolution quantification dating involves two sediment q samples taken from the top from a deeper level in respectively each box (Fig. core, 2, Table 2). These ages were calibrated to calendar dates using adaldata the marine bi - set (Stuiver g dec Braziunas 1993 CALIB 3.0 (Stuiver Reimer includes 1993), which already a modeled, time-dependent reservoir correction ca. 402 yr for the global ocean Braziunas (Stuiver 1993). We assume that our Northeast Atlantic planktonic foraminifera from samples an environment come similar to the model world ocean thus the regional reservoir age (z R) correction = 0 (Stuiver Braziunas 1993). 1Center for Marine Earth Sciences, Free University, De Boelelaan 1085, NL-1081 HV Amsterdam, The Netherls 2Robert J. van de Graaff Laboratorium, State University Utrecht, Box , NL-3508 TA Utrecht, The Netherls Proceedings the 15th International 14C Conference, edited by G. T. Coop D. D. Harkness, B. F. Miller E. M. Scott. RADIocA teow, Vol. 37, No. 2, 1995, P

2 586 S. A. van Kreveld et al. Fig. 1. Physiographic map the Northeast Atlantic Ocean showing the position the box cores between N W. These cores were recovered from m water depths. Depth contours are given in meters. TABLE 1. Location, Water Depth Length the Recovered Box Cores length Box core Latitude Longitude depth (m) (cm) T90-11B ' N ' W T90-8B ' N ' W T90-4B ' N ' W T90-13B ' N ' W T90-15B ' N ' W T90-16B ' N ' W We calculated sedimentation rate by dividing the sediment thickness by the difference in the calendar calibrated 14C ages for each box core, therefore assuming a constant sedimentation rate between these age-control points. Carbonate mass accumulation rate was calculated using where cmar= SR x DBD x %CaCO3 x 0.01 (1) cmar = carbonate mass accumulation rate (g cm-2 ka'1) SR = sedimentation rate (cm ka-1) DBD = dry bulk density (g cm-3). Assuming that the initial carbonate deposition rate preserved in the shallowest core is the same for all cores, the difference in the carbonate accumulation rate with the deeper cores will give the rate carbonate dissolution as a function depth expressed as cdr = cmar(shallowest core) - cmar(deeper core). (2)

3 Quantifying Deep-Sea Carbonate Dissolution 587 These assumptions are viable because the cores were recovered erred by from the same water masses ecological regimes. pods Also, the juvenile shallowest foraminifera, core contains indicating terothat it lies p above has the undergone ara gonte little, compensation if any, carbonate depth dissoluti on. Therefore, its tion rate carbonate can be taken (calcite) as mass the original accumulacarbonate (calcite) deposition negligible rate. i n We our considered shallowest aragonite core because as we found only a few the carbonate ptero od consists fragments calcite. Our carbonate dissolution rates estimate only the amount calcite We determined dry bulk density by weighing 12 cm3 wet sediment invent was then wet-sieved through 32,um mesh screen both the fine (<32 Mm the coars (>32,um) fractions were collected weighed. The percent carbonate by weight the bulk sample the Scheibler-type fine asometric technique (Bruin 1937), with a surements precision ± 2%. the Based weights on these the different meafractions, the carbonate calculated. content Using accelerator the coarse mass fraction spectrometry was (AMS) 14C measurements, samples consisting we dated hpicked, 12 >250 excellently ;um preserved, mixed p lanktoni c foraminifera (Table 2). TABLE 2.14C Dates 12 >250,um Samples Mixed Planktonic Fo raminifera Depth in core Lab no Box c ore (cm) (UtC) (%opdb)* 14C (BP)t T90-11B ± 50 BP$ cal BP T90-8B T90-4B T90-4B 16-, T90-13B T90-13B T90-15B T B T90-16B T90-16B *Values measured at the Geol ogy D epartment, Utrecht t Age in years before present for 14C activity after normalization to $Co nverte S13C = -25% d to calendar dates using the marine bidecadal data set with (Stuiver dr=o Reimer (Stuiver 1993) Braziunas 1993) CALIB 3,0 RESULTS AND DISCUSSION The sediments are calcareous oozes marls, with the three nic dee rock pest cores fragments containing reaching some a volcamaximum diameter 4 mm (Fig. bonate 2). oozes The surface with sediments a coarse carbonate are carfraction lanktonic amounts P foramini fera benthic pteropods foraminifera with minor ostracods, a fine carbonate aminiferal fraction fragments. Only the shallowest surface sample tained collected fragments at 3208-m pteropods. water depth The conbulk carbonate content 90% the box down to 77% by weight, with a higher fine than coarse carbonate content (Fig. y ght bulk carbonate 3). The generally ercent decreases gradually p with is increasin no distinct wat trend for the individual g er depth, though there fine coarse carbonate fractions (Fig. 3).

4 588 S. A. van Kreveld et al. LTTHOLOGIC AND WEIGHT PERCENT CARBONATE PROFILES T90-11B T90-8B T90-4B T90-13B T90-15B T90-16B 3208 m 3393 m 3945 m 4016 m 4177 m 4375 m Z marl calcareous ooze volcanic rock fragments burrows * 14C-dated interval size y Fig. 2. Lithologic percent carbonate by weight priles the box cores...t.,..*.. Carbonate mass accumulation rate decreases with increasing water depth (Fig. 4), with a distinct drop at 4000 m. Here, the drop in the coarse carbonate accumulation rate is about twice that the fine carbonate implying that dissolution affects the former more than the latter. However, we cannot say this for certain because we find fragments planktonic foraminifera incorporated in the fine fraction. Carbonate dissolution rate increases with increasing water depth (Fig. 5). These rates are very low, ranging from 0 to 0.3 g cm-2 ka between m suddenly increasing to 1.5 g cm-2 ka at 4016 m reaching a maximum 1.7 g cm'2 ka in the deepest sample (4375 m) when calculated using sedimentation rates based on calendar ages. In relative percentages, the carbonate loss increases from 0-7% to 42-48%. The carbonate dissolution rate shows a remarkable increase at 4000-m water depth, which we interpret to mark the lysocline. This is in contrast to previous studies based on percent carbonate by weight, which show the lysocline to be several hundred meters deeper. Previous studies e.g., Heath, Moore Dauphin 1976; Pisias 1976; Thiede, Suess Muller 1982, ; Kuehl Fuseth Thunell 1993) calculated sedimentation rates without converting the 14C dates to calendar years. Our data show that these sedimentation rates can be about 14-20% higher. These differences are reflected in the carbonate mass accumulation rate eventually in the dissolution rate estimates. The carbonate dissolution rate shows a remarkable increase at about 4000-m depth, which we interpret to mark the lysocline. This depth coincides with Berger's (1977) lysocline position for this area. However, other studies based on percent carbonate by weight (Biscaye, Kolla Turekian 1976) degree planktonic foraminifera fragmentation, absence fragile forms, pitted tests, an eroded apertures or internal walls, large numbers benthic foraminifera (Kipp 1976; see also Broecker Takahashi 1978) show the lysocline at 4900 m. Crowley (1983) estimated the lyso-

5 Quantifying Deep-Sea Carbonate Dissolution 589 WEIGHT PERCENT CARBONATE i\'s5 r+ti+tir Fig. 3. The sediment core tops are composed calcareous oozes with carbonate content varying from 90% down to 77% by weight CARBONATE MASS ACCUMULATION RATE 4, ACD 'Jtir' ii rf l till ACD 0.ti ;tirti:+tir Water depth (m) Aragonite compensation depth Coarse carbonate (>32 run) Fine carbonate (<32 µm) Fig. 4. Coarse (>32 um) fine (<32 um) mass accumulation rate given in g cm2 ka-1 plotted against water depth cline depth at 4500 m in the Canary Basin, based on the planktonic foraminifera fragmentation percentage benthic foraminifera. The different methods used to estimate carbonate dissolution give various lysocline depths. The one based only on percent carbonate by because weight for underestimates sediments with high the amount initial carbonate carbonate dissolved weight occurs content, little until change about in 50% the percent carbonate the carbonate b by has dissolved (Morse 1973; Morse Mackenzie 1990). From our data, we observe only a 13% decrease in percent carbonate by weight when 48%

6 590 S. A. van Kreveld et al. CARBONATE DISSOLUTION RATE w Water depth (m) ACD Aragonite compensation depth Uncertainty in the carbonate dissolution rate estimates 4400 Fig. 5. Carbonate dissolution rate shows a remarkable increase at 4000-m water depth, which we interpret to mark the lysocline. The uncertainty in the carbonate dissolution rate estimate is calculated using the 1 a range the calendar-calibrated age (Table 2) to calculate sedimentation rates errors in the percent carbonate by weight dry bulk density measurements. the carbonate has been dissolved (compare Figs. 3 5), consistent with Morse's (1973) estimates. Likewise, the benthic to planktonic foraminifera ratio may underestimate carbonate dissolution. Benthic foraminifera are generally more resistant than planktonic foraminifera (Arrhenius 1952; Berger 1973b); therefore, an increase in the former may indicate enhanced dissolution. However, laboratory experiments show that some calcareous benthic foraminifera are less resistant than their most resistant planktonic counterparts (Adelseck 1978; Boltovskoy Totah 1992). Additionally, Corliss Honjo (1981) found varying dissolution susceptibilities among different benthic foraminiferal taxa by suspending specimens various species in a deep-sea mooring. The use the benthic to planktonic foraminifera ratio as a dissolution index probably accounts, in part, for the difference in our lysocline depth with those Crowley (1983) Kipp (1976). Past carbonate dissolution estimates using percent carbonate, solution susceptibility planktonic foraminifera coccoliths, benthic-to-planktonic foraminifera ratios, degree fragmentation were qualitative (Ruddiman Heezen 1967; McIntyre McIntyre 1971; Berger 1976; Tappa Thunell 1984; Peterson Pre111985; Naidu, Malmgren Bornmalm 1993). We expect our results to allow for the calibration these qualitative estimates in the study area. This calibration has to be repeated for each ecological province because carbonate dissolution is species-dependent each zone has a distinct faunallfloral composition. CONCLUSION Deep-sea carbonate dissolution quantification is possible using the methodology presented here provided the cores are recovered from a small geographic area, so that the initial carbonate deposition rate can be assumed the same for all cores. It is also essential to have sufficient 14C dates for calculating sedimentation rates. Our carbonate dissolution rate estimates show a remarkable increase at the 4000-m water depth, which we interpret to be the lysocline.

7 Quantifying Deep-Sea Carbonate Dissolution 591 ACKNOWLEDGMENTS This manuscript benefitted from the critical review G. S. Burr. We thank R. van Els P. H. as, Willekes M. Konert for measuring g the carbonate content the samples. Ocean The Leg 4 Flux Joint Study Global (JGOFS) expedition was sponsored by the Dutch Council This is Sea publication Research the Netherls Research School Sedimentary Marine Geology, Earth Center Sciences, for Free University, Amsterdam. REFERENCES Adelseck, C. G., Jr Dissolution deep-sea carbonate: Preliminary calibration preservational morphologic aspects. Deep-Sea Research 25: Arrhenius, G Sediment cores from the East Pacific. Reports the Swedish Deep-Sea Expedition, : Balsam, W. L Carbonate dissolution on the Muir Seamount (Western North Atlantic): Interglacial/glacial changes. Journal Sedimentary Petrolology 53 (3): Berger, W. H Planktonic foraminifera: Selective solution the lysocline. Marine Geology 8: a Deep-sea carbonates: Evidence for a coccolith lysocline. Deep-Sea Research 20: _1973b Deep-sea carbonates: Pleistocene dissolution cycles. Journal Foraminiferal Research 3(4): Biogenous deep-sea sediments: Production preservation. In Riley, J. P. Chester, R., eds., Chemical Oceanography. London, Academic Press: Carbon dioxide excursions the deep-sea record: Aspects the problem. In Andersen, N. R. Malahf, A., eds., The Fate Fossil Fuel CO2 in the Oceans. New York, Plenum Press: Berger, W. H. von Rad, U Cretaceous Cenozoic sediments from the Atlantic Ocean. Deep-Sea Drilling Project Initial Reports 14: Biscaye, P. E., Kolla, V. Turekian, K. K Distribution calcium carbonate in surface sediments the Atlantic Ocean. Journal Geophysical Research 81: Boltovskoy, E. Totah V. I.1992 Preservation index preservation potential some foraminiferal species. Journal Foraminiferal Research 22(3): Broecker, W. S. Takahashi, T The relationship between lysocline depth in situ carbonate ion concentration. Deep-Sea Research 25: Bruin, P Enige ervaringen bij de bepaling van het gehalte van grond aan koolzure kalk volgens de methode Scheibler. Chemisch Weekblad 34: Corliss, B. H. Honjo, S Dissolution deep-sea benthonic foraminifera. Micropaleontology 27: Crowley, T. J Calcium-carbonate preservation pat- terns in the central North Atlantic during the last 150,000 years. Marine Geology 51:1-14. de Vernal, A., Bilodeau, G., Hillaire-Marcel, C. Kassou, N Quantitative assessment carbonate dissolution in marine sediments from foraminifer linings vs. shell ratios: Davis Strait, northwest North Atlantic. Geology 20: Farrell, J. W. Prell, W. L Climatic change CaCO3 preservation: An 800,000 year bathymetric reconstruction from the central equatorial Pacific ocean. Paleoceanography 4(4): Francois, R., Bacon, M. P. Suman, D Thorium 230 priling in deep-sea sediments: High resolution records flux dissolution carbonate in the equatorial Atlantic during the last 24,000 years. Paleoceanography 5(5): Heath, G. R., Moore, T. C., Jr, Dauphin, J. P Late Quaternary accumulation rates opal, quartz, organic carbon, calcium carbonate in the Cascadia Basin area, Northeast Pacific. In Cline, R. Hays, J., eds., Investigation Late Quaternary Paleoceanography Paleoclimatology. Geological Society America Memoir 145: Kipp, N. G New transfer-function for estimating past sea-surface conditions from sea-bed distributions planktonic foraminifera) assemblages in the North Atlantic. In Cline, R. Hays, J., eds., Investigation Late Quaternary Paleoceanography Paleoclimatology. Geological Society America Memoir 145: Kuehl, S. A., Fuglseth, T. J. Thunell, R. C Sediment mixing accumulation rates in the Sulu South China Seas: Implications for organic carbon preservation in deep-sea environments. Marine Geology 11: Malmgren, B Ranking dissolution susceptibility planktonic foraminifera at high latitudes the South Atlantic Ocean. Marine Micropaleontology 8: McIntyre, A. McIntyre, R Coccolith concentrations differential solution in oceanic sediments. In Funnel, B. M. Riedel, W. R., eds., The Micropaleontology the Oceans. London, Cambridge University Press: Morse, J. W. (ms.) 1973 The dissolution kinetics calcite: A kinetic origin for the lysocline. Ph.D, dissertation, Yale University.

8 592 S. A. van Kreveld et al. Morse, J. W. Mackenzie, F. T Geochemistry sedimentary carbonates. Amsterdam, Elsevier: 707 p. Naidu, P. D., Malmgren, B. A. Bornmalm, L Quaternary history calcium carbonate fluctuations in the western equatorial Indian Ocean (Somali Basin). PaIaeogeography, Palaeoclimatology, Palaeoecology 103: Parker, W. K. Berger, W. H Faunal solution patterns planktonic foraminifera in surface sediments the South Pacific. Deep-Sea Research 18: Peterson, L. C. Prell, W. L Carbonate dissolution in Recent sediments the Eastern Equatorial Indian Ocean: Preservation patterns carbonate loss above the lysocline. Marine Geology 64: Pisias, N. G Late Quaternary sediment the Panama Basin: sedimentation rates, periodicities, controls carbonate opal accumulation. In Cline, R. Hays, J., eds., Investigation Late Quaternary paleoceanography paleoclimatology. Geological Society America Memoir 145: Roth, P. H. Berger, W. H Distribution dissolution coccoliths in the south central Pacific. In Sliter, W. V., Be, A. W. H. Berger, W. H., eds., Dissolution Deep-Sea Carbonates. Cushman Foundation for Foraminiferal Research Special Publication 13: Ruddiman, W. F. Heezen, B. C Differential solution planktic foraminifera. Deep-Sea Research 14: Schneidermann, N Selective dissolution Recent coccoliths in the Atlantic Ocean. In Ramsay, A. T. S., ed., Oceanic Micropaleontology. New York, Academic Press: Stuiver, M. Braziunas, T Modeling atmospheric 14C influences 14C ages marine samples to 10,000 BC. In Stuiver, M., Long, A. Kra, R. S., eds., Calibration Radiocarbon 35(1): Stuiver, M. Reimer, P Extended 14C data base revised CALIB 3.014C age calibration program. In Stuiver, M., Long, A. Kra, R. S., eds., Calibration Radiocarbon 35(1): Tappa, E. Thunell, R Late Pleistocene glacial! interglacial changes in planktonic foraminiferal biacies carbonate dissolution patterns in the Vema Channel. Marine Geology 58: Thiede, J., Suess, E. Muller, P Late Quaternary fluxes major sediment components to the sea floor at the Northwest African continental slope. In von Rad, U. et al., eds., Geology the Northwest African Continental Margin. Berlin, Springer-Verlag: Thunell, R. C Optimum indices calcium carbonate dissolution in deep-sea sediments. Geology 4: