Volume 32, Issue 1
Climate
Free Access

A 220 kyr record of Pb isotopes at Dome C Antarctica from analyses of the EPICA ice core

P. Vallelonga

P. Vallelonga

Department of Applied Physics, Curtin University of Technology, Perth, Western Australia, Australia

Search for more papers by this author
P. Gabrielli

P. Gabrielli

UMR CNRS/Université Joseph Fourier 5183, Laboratoire de Glaciologie et Géophysique de l'Environnement, St. Martin d'Hères, France

Department of Environmental Sciences, University of Venice, Venice, Italy

Search for more papers by this author
K. J. R. Rosman

K. J. R. Rosman

Department of Applied Physics, Curtin University of Technology, Perth, Western Australia, Australia

Search for more papers by this author
C. Barbante

C. Barbante

Department of Environmental Sciences, University of Venice, Venice, Italy

Institute for the Dynamics of Environmental Processes, CNR, Venice, Italy

Search for more papers by this author
C. F. Boutron

C. F. Boutron

UMR CNRS/Université Joseph Fourier 5183, Laboratoire de Glaciologie et Géophysique de l'Environnement, St. Martin d'Hères, France

Unité de Formation et de Recherche de Physique et Observatoire des Sciences de l'Univers, Université Joseph Fourier, Grenoble, France

Search for more papers by this author
First published: 13 January 2005
Citations: 40

Abstract

[1] Pb isotopic compositions and Pb and Ba concentrations are reported in EPICA Dome C ice core samples dating to 220 kyr BP, indicating that Pb isotopic compositions in Antarctic ice vary with changing climate. 206Pb/207Pb ratios decrease during glacial periods, with the lowest values occurring during colder climatic periods (stages 2, 4 and 6) and the Holocene. Low Pb concentrations (<1 pg/g) were found during the Holocene and the last interglacial (climate stage 5.5) while higher Pb concentrations (>10 pg/g) were found during cold climatic periods. Ba, a proxy for mineral dust, was used to determine that dust usually accounts for ∼70% of Pb in Dome C ice, while the remaining ∼30% was attributed to volcanic emissions. Pb isotopic compositions at Dome C differ from those reported in pre-industrial ice from other Antarctic locations, due to greater proportions of dust Pb at Dome C.

1. Introduction

[2] The deep ice cores from East Antarctica have provided an insight into climatic conditions and aerosol fallout fluxes and provenance during the Holocene and previous glacial and interglacial periods [EPICA Community Members, 2004]. High Pb concentrations [Hong et al., 2003, and references therein] observed during cold climate stages 2, 4 and 6 in the Vostok [Petit et al., 1999] and “old” (1978) Dome C [Lorius et al., 1979] ice cores were attributed to increased dust levels, which have been determined to most probably originate from the loess deposits of South America [Delmonte et al., 2004].

[3] Rosman et al. [1999] reported Pb isotopic compositions in the earlier Dome C ice core and noted that Holocene Pb isotopic compositions were more-radiogenic (higher 206Pb/207Pb and 208Pb/207Pb ratios) than those of the last glacial maximum (LGM), suggesting that Pb isotope ratios in Antarctic ice may be affected by changing climate. Recent analyses of Pb isotopes at coastal locations in Antarctica have determined pre-industrial Holocene Pb isotopic signatures [Matsumoto and Hinkley, 2001; Planchon et al., 2003; Vallelonga et al., 2002b] and revealed that proportions of dust and volcanic Pb vary across Antarctica. The provenance of Antarctic dust (by Pb isotopes) is restricted by a dearth of Pb isotope data representing dust sources in the Southern Hemisphere.

[4] We present the first Pb isotopic compositions from the EPICA Dome C ice core and the longest Pb isotopic record available from an ice core. These data1 allow variations in Pb deposition and dust Pb content at Dome C to be evaluated for the past two glacial cycles.

2. Methods

[5] Thirty samples have been analysed from 26 sections of the EPICA ice cores drilled at Dome C (75°06′S, 123°21′E; 3233 m asl; mean temperature −54°C, Figure 1). From the surface to 763 m depth, samples were from the EDC96 core (drilled 1996–1999), while deeper samples were from another EPICA Dome C core (drilled 1999–2003). Sections analysed were from depths of 229 to 2193 m, corresponding to the period 6.9 to 217 kyr BP using the EDC2 timescale [EPICA Community Members, 2004]. Core sections were typically 55 cm long and 5 cm in radius. About 30% of the cross-section of each core section (cut longitudinally) was available for Pb isotope analyses. While two samples were usually produced from a 55 cm core section, a preliminary data set of thirty samples is presented here.

Details are in the caption following the image
Map of Antarctica indicating the locations of Dome C and other sites for which Pb isotope data have been reported: Law Dome (LD; 66°46′S, 112°48′E), Taylor Dome (TD; 77°48′S, 158°43′E) and Coats Land (CL; 77°34′S, 25°22′W).

[6] Samples were prepared (decontaminated) at the Laboratory of Glaciology and Geophysics of the Environment in Grenoble, France, in a clean bench supplied with High Efficiency Particulate (HEPA) filtered air and located within a cold room (Temperature −15°C). The decontamination procedure [Candelone et al., 1994; Gabrielli et al., 2004] involved the chiselling of concentric layers of an ice core to remove external contamination introduced during drilling, transport and/or storage. Decontaminated samples were melted and aliquotted into LDPE bottles that were then sent, frozen, to Curtin University in Perth, Australia, for Pb and Ba analysis.

[7] The preparation and analysis of ice core samples by thermal ionization mass spectrometry has been described in detail by Vallelonga et al. [2002a, and references therein]. Samples were melted and ∼5 g was transferred to a Teflon beaker. HNO3, HF, H3PO4 and an isotopically enriched tracer solution (containing 205Pb and 137Ba for isotope dilution) were added before the sample was evaporated to dryness. The sample was then transferred in 4 μl of silica gel to a zone-refined rhenium filament and again evaporated to dryness. Samples were then loaded into the mass spectrometer.

[8] Samples were analysed using a thermal ionization mass spectrometer (model VG354, Fisons Instruments). All Pb isotopes (204Pb, 205Pb, 206Pb, 207Pb and 208Pb) were measured using a Daly detector, with ∼70 isotope ratios collected per sample. 205Pb, which is not present in terrestrial Pb, was used as an isotopic tracer as it enabled the determination of Pb concentrations and Pb isotope ratios in one analysis. Ba can be used as a proxy for mineral dust because it is much more abundant in Earths' crust compared to volcanic emissions [Matsumoto and Hinkley, 2001]. All Pb/Ba ratios were calculated by weight.

[9] The accuracy of the reported Pb and Ba concentrations is estimated to be ±15% (95% confidence interval). This relatively large uncertainty (for isotope dilution) is a result of the ±7% (2σ) precision to which μL-quantities of the tracer solution could be dispensed. The accuracy of the isotope ratio measurements was monitored by routine analysis of a 200 pg sample of Standard Reference Material 981 (NIST, Gaithersburg, USA). A correction for isotopic fractionation of 0.24 ± 0.06 % per mass unit was applied to the measured ratios. Pb isotope ratio uncertainties (Table 1) account for the sampling and storage of ice cores as well as sample analysis.

Table 1. Lead and Ba Concentrations and Pb Isotopic Compositions in Antarctic Dome C Ice
Depth (m) Age (kyr BP) 206Pb/207Pb ±a 208Pb/207Pb ±a 206Pb/204Pb ±a Pbb (pg/g) Bab (pg/g)
229.1 6.9 1.190 0.016 2.471 0.020 19.2 1.2 0.19 7.4
229.4 6.9 1.195 0.022 2.467 0.032 19.0 1.7 0.22 6.3
316.3 10.3 1.194 0.024 2.460 0.026 19.0 1.7 0.19 6.7
379.2 12.8 1.220 0.009 2.486 0.011 19.7 0.8 0.70 13
405.9 14.2 1.217 0.009 2.481 0.011 19.3 0.5 0.82 17
432.6 15.6 1.193 0.013 2.483 0.012 18.7 1.4 0.52 16
461.5 17.4 1.226 0.002 2.488 0.009 19.0 0.4 6.1 107
471.4 18.2 1.257 0.004 2.507 0.008 19.8 0.6 29.5 176
489.0 19.7 1.190 0.011 2.493 0.065 18.3 0.8 15.0 460
515.6 21.9 1.199 0.014 2.465 0.019 18.5 0.3 14.3 467
574.2 27.0 1.207 0.008 2.465 0.014 17.7 0.4 16.1 362
598.1 29.1 1.203 0.017 2.468 0.057 18.4 1.5 8.6 219
654.0 33.7 1.205 0.003 2.477 0.013 18.7 0.3 8.2 185
709.0 38.2 1.226 0.007 2.470 0.019 17.6 1.6 2.7 37
818.1 47.2 1.206 0.004 2.467 0.008 18.6 0.3 5.4 127
983.4 60.7 1.214 0.006 2.484 0.004 18.3 0.6 9.8 225
1010.9 63.5 1.196 0.003 2.464 0.006 18.5 0.4 13.9 360
1093.1 71.6 1.227 0.013 2.504 0.024 18.1 2.2 1.4 19
1258.4 86.0 1.222 0.006 2.477 0.007 19.0 0.5 3.5 61
1423.4 102.3 1.225 0.003 2.477 0.010 18.9 0.5 3.2 43
1533.1 114.1 1.229 0.010 2.487 0.009 19.3 0.3 0.36 11
1643.1 122.8 1.216 0.014 2.497 0.023 18.9 1.0 0.73 16
1643.4 122.8 1.208 0.010 2.475 0.018 19.2 0.7 0.52 14
1753.4 131.9 1.237 0.006 2.488 0.010 18.6 1.2 2.6 30
1863.1 152.2 1.204 0.003 2.468 0.006 18.6 0.2 18.3 370
1863.4 152.2 1.193 0.003 2.459 0.009 18.6 1.0 18.9 447
1973.1 174.4 1.200 0.004 2.470 0.009 18.7 0.3 7.3 192
1973.4 174.5 1.192 0.005 2.462 0.009 18.6 0.2 9.7 150
2094.4 198.9 1.199 0.018 2.489 0.025 19.4 1.5 0.27 9.5
2193.4 217.1 1.214 0.009 2.476 0.006 19.4 1.0 2.0 32
  • a Uncertainties in the isotope ratios are 95% confidence intervals.
  • b Concentrations are accurate to ±15% (95% confidence interval).

3. Results and Discussion

[10] Pb and Ba concentrations in ice from the EPICA Dome C core have varied by almost two orders of magnitude over the past 220 kyr with the greatest concentrations constrained to cold climatic periods (Table 1). Lower Pb and Ba concentrations were observed during the Holocene (0.4 pg Pb/g, 11 pg Ba/g) and the last interglacial (0.5 pg Pb/g, 14 pg Ba/g), while higher Pb and Ba concentrations occurred during climate stages 2 (LGM, 14 pg Pb/g, 380 pg Ba/g), 4 (63 kyr BP, 12 pg Pb/g, 290 pg Ba/g) and 6 (the penultimate glacial maximum, 11 pg Pb/g, 240 pg Ba/g), as shown in Figure 2. The results presented here agree with those reported in other Antarctic ice cores [Hong et al., 2003, and references therein], which were linked to enhanced aerosol production by such mechanisms as lowered sea levels, stronger winds and reduced hydrological washout of aerosols, favouring the long-distance transport of dust from continents of the Southern Hemisphere to Antarctica [Delmonte et al., 2004].

Details are in the caption following the image
220 kyr record of δ deuterium [EPICA Community Members, 2004] and Pb concentration (a), Ba concentration (b) and 206Pb/207Pb (c) from the EPICA Dome C ice core. Marine isotope stage numbers [Delmonte et al., 2004] are also shown in Figure 2a.

[11] If it is assumed that dust deposited in Antarctic ice has a Pb/Ba ratio identical to that reported for the upper continental crust (∼0.03 [McLennan, 2001]), then the average Pb/Ba ratio observed at Dome C (∼0.05) indicates that mineral dust contributes ∼70% of Pb in ice. Higher Pb/Ba ratios indicate that Pb from other sources, most likely volcanic emissions, occasionally contributed up to 65% of Pb at Dome C. More-radiogenic Pb isotopic compositions occur in samples with higher Pb/Ba ratios, suggesting that these samples contain greater proportions of volcanic Pb. The clearest sign of volcanic Pb in Dome C ice is an LGM sample (471.4 m, 18.2 kyr BP) with exceptionally high 206Pb/207Pb and Pb/Ba ratios, indicating a dust Pb proportion of ∼20%. This sample corresponds to a fluoride peak in the EDC96 core [Schwander et al., 2001, and references therein] which matches a larger, contemporaneous, fluoride peak identified in the Byrd ice core. A contemporaneous fluoride peak in the two cores indicates a volcanic emission source local to Antarctica.

[12] Pb isotopic compositions observed at Dome C over the past 220 kyr vary with changing climatic conditions (Figure 2). During the last glacial period, 206Pb/207Pb ratios decreased from 1.22 (average 71.6–102 kyr BP) to 1.21 (average 33.7–63.4 kyr BP) and then averaged 1.20 (19.6–29.1 kyr BP) during the LGM. They were greater during the glacial-interglacial transition (1.22, 12.8–18.2 kyr BP) compared to the LGM (1.20, 19.6–29.1 kyr BP), but they are not precise enough to determine any change since the end of the last transition. These trends also appear likely to have occurred during the penultimate glacial period, with ratios lower during the penultimate glacial period (1.20, 152–217 kyr BP), high during the transition (1.24, 132 kyr BP) and lower during the last interglacial (1.22, 114–123 kyr BP).

[13] The Pb isotope variations observed in Dome C ice appear to be climate related but the relationship between 206Pb/207Pb and temperature (δD) is not simple. Pb in Dome C ice originates from mineral dust and volcanism. If there is more than one source of mineral dust deposited in Antarctica, the Pb isotopic composition in the ice is likely to change as different source regions are diversely affected by climate change. For example, Delmonte et al. [2004] noted that the higher latitudes of southern South America, New Zealand and Australia were more likely to have experienced a colder and drier climate during the LGM, compared to interglacial conditions, while the low latitudes of Chile and southwestern southern Africa experienced a wetter climate compared to interglacial conditions. If all dust sources were found to have identical Pb isotopic compositions, then changing proportions of dust sources would not alter the isotopic composition of Pb in Dome C ice.

[14] Pb isotopic compositions in EPICA Dome C ice are within the range 206Pb/207Pb = 1.19–1.26 and 208Pb/207Pb = 2.46–2.51 (Figure 3), representing a mixture of Pb from mineral dust and volcanism. Data reported for Dome C [Rosman et al., 1994, 1999], Taylor Dome [Matsumoto and Hinkley, 2001] and Coats Land [Planchon et al., 2003] snow and ice are also shown in Figure 3. Most of the EPICA Dome C samples display lower 206Pb/207Pb and 208Pb/207Pb ratios than those observed at Taylor Dome and Law Dome. At the latter locations, where the proportion of dust Pb in ice is consistently less than 30%, Pb isotopic compositions are similar to those reported for local volcanic Pb sources such as Antarctic Peninsula basalts [Hole et al., 1993], Ross Island basanitoids [Sun and Hanson, 1975] and Marie Byrd Land volcanics [Hart et al., 1997]. While Matsumoto and Hinkley [2001] noted that Taylor Dome ice displays similar Pb isotopic compositions to some ocean island volcanoes of the Southern Hemisphere, the extent to which such volcanoes influence Antarctica is still to be determined and so only Antarctic volcanic systems have been shown in Figure 3.

Details are in the caption following the image
Pb isotopic compositions in snow and ice from Dome C and other Antarctic sites. Also shown are Pb isotopic compositions of South Atlantic and South Pacific Ocean pelagic sediments [Chow and Patterson, 1962], Antarctic Peninsula basalts [Hole et al., 1993], Ross Island basanitoids [Sun and Hanson, 1975] and Marie Byrd Land volcanics [Hart et al., 1997].

[15] Delmonte et al. [2004] reported that dust deposited at Dome C during cold climate periods originated from the same source(s), most probably the loess deposits of southern South America. Pb isotopic compositions are not yet available for southern South American loess, so we have used pelagic sediments [Chow and Patterson, 1962] of predominantly Aeolian origin to represent South American loess. Sediments were collected near South America and are likely to contain South American dust Pb. While the less-radiogenic EPICA Dome C samples have similar Pb isotopic compositions to the pelagic sediments, the provenance of Dome C dust Pb cannot be determined reliably until Pb isotope data have been reported for potential dust sources.

[16] The findings of the present study are in contrast to those of Matsumoto and Hinkley [2001] who reported high Pb concentrations (∼5 pg/g from 2.2 to 10 kyr BP) in Taylor Dome ice during the Holocene and determined that volcanic Pb emissions are the dominant source of Pb deposited in Antarctica. The results presented here differ from those of Taylor Dome in two respects: lower Pb concentrations (∼0.4 pg/g) are observed in Dome C ice from 6.9 to 15.6 kyr BP, and average Pb/Ba ratios (∼0.05) indicate that ∼70% of Pb deposited at Dome C originates from dust. These differences might be accounted for by considering ice core Pb to be a mixture of continental dust and Antarctic aerosols containing a relatively high proportion of radiogenic Pb of volcanic origin, where the proportion of volcanic Pb at Taylor Dome is greater than that at Dome C.

4. Conclusion

[17] These preliminary results support and extend existing data showing that Pb isotopic compositions and Pb concentrations in Antarctic ice vary with changes in climate. The deeper sections of the EPICA Dome C core must be analysed to confirm the cyclical patterns observed for the past 220 kyr, while Pb isotopic composition data pertaining to the loess deposits of the Southern Hemisphere are necessary to further constrain the potential sources of dust particles deposited in central East Antarctica. Finally, reliable and detailed records of Pb and Ba concentrations over the Holocene-LGM transition are required from other Antarctic locations to account for the differences observed between Taylor Dome and Dome C.

Acknowledgments

[18] This research was supported by the Australian Research Council (A39938047) and the Antarctic Science Advisory Committee (#1092,2334); the Institut Universitaire de France, the Ministère de l'Environnement et de l'Amènagemont du Territoire, the Agence de l'Environnement et de la Maîtrise de l'Energie, the Intitut National des Sciences de l'Univers; the ENEA of the Italian Antarctic National Research Program. This work is a contribution to the “European Project for Ice Coring in Antarctica” (EPICA), a joint ESF (European Science Foundation)/EC scientific programme, funded by the European Commission and by national contributions from Belgium, Denmark, France, Germany, Italy, the Netherlands, Norway, Sweden, Switzerland and the United Kingdom. This is EPICA publication no. 111. We thank all Antarctic field personnel, F. Planchon, G. Burton and L. Burn for laboratory support, and colleagues in the John de Laeter Centre of Mass Spectrometry – TIMS Laboratory for helpful discussions. T. Hinkley and an anonymous reviewer provided helpful comments and suggestions.