Figures

top

Introduction

The ongoing decrease in the age, thickness and extent of Arctic sea ice cover is projected to strongly impact climate, weather, ecosystems, matter fluxes and human activities in the near future (IPCC, 2022

IPCC (2022) Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Pörtner, H.-O., Roberts, D.C., Tignor, M., Poloczanska, E.S., Mintenbeck, K., Alegría, A., Craig, M., Langsdorf, S., Löschke, S., Möller, V., Okem, A., Rama, B. (Eds.)]. Cambridge University Press, in press.

). The transition from a perennial ice cover with major sea ice transport systems to a seasonally ice-free ocean with isolated ice fields and floes of different origin is already impacting the distribution of gaseous, dissolved, and particulate matter, with far reaching consequences for Arctic ecosystems and biogeochemical cycles (e.g., Krumpen et al., 2019

Krumpen, T., Belter, H.J., Boetius, A., Damm, E., Haas, C., Hendricks, S., Nicolaus, M., Nöthig, E.-M., Paul, S., Peeken, I., Ricker, R., Stein, R. (2019) Arctic warming interrupts the Transpolar Drift and affects long-range transport of sea ice and ice-rafted matter. Scientific Reports 9, 1–9. https://doi.org/10.1038/s41598-019-41456-y

). Accurate knowledge of sea ice origin and drift is therefore required to understand the changing role of the marine cryosphere in regulating matter fluxes and ecological processes.

Satellite-based sea ice motion products are currently the only available resource for reconstructing sea ice origin and drift trajectories (Krumpen et al., 2021

Krumpen, T., von Albedyll, L., Goessling, H.F., Hendricks, S., Juhls, B., Spreen, G., Willmes, S., Belter, H.J., Dethloff, K., Haas, C., Kaleschke, L., Katlein, C., Tian-Kunze, X., Ricker, R., Rostosky, P., Rückert, J., Singha, S., Sokolova, J. (2021) MOSAiC drift expedition from October 2019 to July 2020: sea ice conditions from space and comparison with previous years. The Cryosphere 15, 3897–3920. https://doi.org/10.5194/tc-15-3897-2021

). However, these mainly provide information about ice and atmospheric conditions, while the water mass distribution during ice growth and drift remains unexplored. Biological and biogeochemical processes in sea ice are, however, tightly linked to the parental seawater composition. Geochemical provenance tracers have the potential to reveal this composition as well as biogeochemical atmosphere-ice-ocean exchange during drift, thus ideally complementing satellite-based observations.

Radiogenic neodymium (Nd) isotopes (expressed as ɛ_Nd = (¹⁴³Nd/¹⁴⁴Nd)_sample/(¹⁴³Nd/¹⁴⁴Nd)_CHUR − 1} × 10⁴, with CHUR = 0.512638; Jacobsen and Wasserburg, 1980

Jacobsen, S.B., Wasserburg, G.J. (1980) Sm-Nd Isotopic Evolution of Chondrites. Earth and Planetary Science Letters 50, 139–155. https://doi.org/10.1016/0012-821X(80)90125-9

) are a powerful tracer of water mass mixing and ocean circulation (Frank, 2002

Frank, M. (2002) Radiogenic isotopes: tracers of past ocean circulation and erosional input. Reviews of Geophysics 40, 1–38. https://doi.org/10.1029/2000RG000094

), but their application as a tracer of the water masses from which sea ice forms is novel. This potential results from the incorporation of small amounts of Nd from the ocean surface layer into sea ice during sea ice growth (Laukert et al., 2017c

Laukert, G., Frank, M., Hathorne, E.C., Krumpen, T., Rabe, B., Bauch, D., Werner, K., Peeken, I., Kassens, K. (2017c) Pathways of Siberian freshwater and sea ice in the Arctic Ocean traced with radiogenic neodymium isotopes and rare earth elements. Polarforschung 87, 3–13. https://doi.org/10.2312/polarforschung.87.1.3

). Due to salt-proportional rejection, the concentrations of Nd and other rare earth elements (REEs) in sea ice drop below those of seawater. This makes precise and accurate ɛ_Nd analysis difficult but can be compensated for by larger sample volume.

In the Arctic Ocean, the ɛ_Nd signature of surface waters reflects contributions from water masses and major rivers and ranges between −17 and −5.5 (see Fig. S-1 and description of marine ɛ_Nd systematics in the Supplementary Information; Andersson et al., 2008

Andersson, P.S., Porcelli, D., Frank, M., Bjork, G., Dahlqvist, R., Gustafsson, O. (2008) Neodymium isotopes in seawater from the Barents Sea and Fram Strait Arctic-Atlantic gateways. Geochimica et Cosmochimica Acta 72, 2854–2867. https://doi.org/10.1016/j.gca.2008.04.008

; Porcelli et al., 2009

Porcelli, D., Andersson, P.S., Baskaran, M., Frank, M., Bjork, G., Semiletov, I. (2009) The distribution of neodymium isotopes in Arctic Ocean basins. Geochimica et Cosmochimica Acta 73, 2645–2659. https://doi.org/10.1016/j.gca.2008.11.046

; Laukert et al., 2017a

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Rabe, B., von Appen, W.-J., Wegner, C., Zieringer, M., Kassens, H. (2017a) Ocean circulation and freshwater pathways in the Arctic Mediterranean based on a combined Nd isotope, REE and oxygen isotope section across Fram Strait. Geochimica et Cosmochimica Acta 202, 285–309. https://doi.org/10.1016/j.gca.2016.12.028

, 2017b

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Gutjahr, M., Janout, M., Hölemann, J. (2017b) Transport and transformation of riverine neodymium isotope and rare earth element signatures in high latitude estuaries: a case study from the Laptev Sea. Earth and Planetary Science Letters 477, 205–217. https://doi.org/10.1016/j.epsl.2017.08.010

, 2017c

Laukert, G., Frank, M., Hathorne, E.C., Krumpen, T., Rabe, B., Bauch, D., Werner, K., Peeken, I., Kassens, K. (2017c) Pathways of Siberian freshwater and sea ice in the Arctic Ocean traced with radiogenic neodymium isotopes and rare earth elements. Polarforschung 87, 3–13. https://doi.org/10.2312/polarforschung.87.1.3

, 2019

Laukert, G., Makhotin, M., Petrova, M.V., Frank, M., Hathorne, E.C., Bauch, D., Böning, P., Kassens, H. (2019) Water mass transformation in the Barents Sea inferred from radiogenic neodymium isotopes, rare earth elements and stable oxygen isotopes. Chemical Geology 511, 416–430. https://doi.org/10.1016/j.chemgeo.2018.10.002

; Paffrath et al., 2021

Paffrath, R., Laukert, G., Bauch, D., Rutgers van der Loeff, M., Pahnke, K. (2021) Separating individual contributions of major Siberian rivers in the Transpolar Drift of the Arctic Ocean. Scientific Reports 11, 8216. https://doi.org/10.1038/s41598-021-86948-y

). Significantly less radiogenic ɛ_Nd signatures (<−17) are only introduced via discharge from Greenland and the Canadian Arctic Archipelago (Filippova et al., 2017

Filippova, A., Frank, M., Kienast, M., Rickli, J., Hathorne, E.C., Yashayaev, I.M., Pahnke, K. (2017) Water mass circulation and weathering inputs in the Labrador Sea based on coupled Hf–Nd isotope compositions and rare earth element distributions. Geochimica et Cosmochimica Acta 199, 164–184. https://doi.org/10.1016/j.gca.2016.11.024

; Laukert et al., 2018

Laukert, G., Dreyer, J., Frank, M., Hathorne, E.C., Meulenbroek, K. (2018) Greenland-sourced freshwater traced by radiogenic neodymium isotopes and rare earth elements on the North-East Greenland Shelf. Goldschmidt Abstracts 2018, 1419. https://goldschmidtabstracts.info/2018/1419.pdf

; Grenier et al., 2022

Grenier, M., Brown, K.A., Colombo, M., Belhadj, M., Baconnais, I., Pham, V., Soon, M., Myers, P.G., Jeandel, C., François, R. (2022) Controlling factors and impacts of river-borne neodymium isotope signatures and rare earth element concentrations supplied to the Canadian Arctic Archipelago. Earth and Planetary Science Letters 578, 117341. https://doi.org/10.1016/j.epsl.2021.117341

). Four ice floes sampled in the central Arctic Ocean in 2012 had ɛ_Nd compositions similar to their parental waters (Laukert et al., 2017c

Laukert, G., Frank, M., Hathorne, E.C., Krumpen, T., Rabe, B., Bauch, D., Werner, K., Peeken, I., Kassens, K. (2017c) Pathways of Siberian freshwater and sea ice in the Arctic Ocean traced with radiogenic neodymium isotopes and rare earth elements. Polarforschung 87, 3–13. https://doi.org/10.2312/polarforschung.87.1.3

), pointing to the incorporation of seawater ɛ_Nd signatures during sea ice growth. However, these samples were unfiltered and direct contributions from particulate phases could not be excluded.

To further explore the suitability of Nd isotopes for tracing the marine origin of sea ice, we determined dissolved Nd isotopes along with salinity, temperature, stable oxygen isotopes and the complete set of REEs in different sea ice cores collected from ice floes in Fram Strait during cruise PS85 of RV Polarstern in 2014. Comparison with snow cover and surface seawater properties as well as ice drift trajectories reconstructed with IceTrack (Krumpen et al., 2019

Krumpen, T., Belter, H.J., Boetius, A., Damm, E., Haas, C., Hendricks, S., Nicolaus, M., Nöthig, E.-M., Paul, S., Peeken, I., Ricker, R., Stein, R. (2019) Arctic warming interrupts the Transpolar Drift and affects long-range transport of sea ice and ice-rafted matter. Scientific Reports 9, 1–9. https://doi.org/10.1038/s41598-019-41456-y

) allows us to significantly expand our knowledge of the mechanisms controlling ɛ_Nd in the marine cryosphere.

top

Tracer Data and Satellite-Based Sea Ice Origin

The tracer data were determined in sea ice and snow samples collected from nine sea ice floes of different sizes, ages, and origins across Fram Strait as well as in nearby surface seawater (Fig. 1; see Supplementary Information for methodology and data). Based on visual inspection, the ice cores and the filters used for filtration of the melted sea ice samples were free of ice rafted detrital material. The dissolved ɛ_Nd of the sea ice ranges between −31.6 and −10.1, while less variability is observed in snow and surface seawater (average ɛ_Nd = −14.7 ± 1.0 and −11.1 ± 0.7, respectively; 1 s.d., n = 7). The least radiogenic ɛ_Nd compositions were determined in the lowermost ice core intervals at stations 419, 454 and 472 and correspond to highest Nd concentrations ([Nd]) and the lowest heavy (H) to light (L) REE ratios normalised to PAAS (Post Archaean Australian Shale × 10⁶; see Supplementary Information) (Fig. 1). A low HREE/LREE ratio generally indicates exchange with freshly weathered rock material and is characteristic of freshwater that has been recently discharged to the ocean (e.g., Laukert et al., 2017a

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Rabe, B., von Appen, W.-J., Wegner, C., Zieringer, M., Kassens, H. (2017a) Ocean circulation and freshwater pathways in the Arctic Mediterranean based on a combined Nd isotope, REE and oxygen isotope section across Fram Strait. Geochimica et Cosmochimica Acta 202, 285–309. https://doi.org/10.1016/j.gca.2016.12.028

). [Nd] and HREE/LREE in our sea ice correlate well with ɛ_Nd (R² = 0.75 and 0.86, respectively; Fig. 2a,b). A correlation is also observed between HREE/LREE and ɛ_Nd (R² = 0.7) for surface seawater despite the smaller range in ɛ_Nd compositions. In contrast, no significant correlations between any of these parameters exist for the snow samples. On average, REE patterns from sea ice are similar in the dissolved (<0.45 μm) and truly dissolved (<3 and 30 kDa) size fractions, exhibiting LREE depletion and negative Ce anomalies, whereas REE patterns in snow are much flatter, consistent with the origin of REEs from atmospheric aerosols (Figs. S-2, S-3). The truly dissolved REE concentrations of sea ice average ∼85 % of the dissolved REEs, except at station 481 where only ∼55 % of the dissolved REEs were present in the truly dissolved pool. In snow, the truly dissolved concentrations average 50–60 % of the dissolved REEs. The dissolved [REE] in sea ice on average correspond to only 10–30 % of the dissolved [REE] of Arctic surface seawater.

Full size image

Full size image

The stable oxygen isotope signatures (δ¹⁸O) of the sea ice range from −5.8 to −0.7 ‰ at an average salinity of 3.6 ± 0.8 (1 s.d., n = 29). Melting of sea ice is evidenced by near zero salinities only for the upper 50 cm of station 481 and the uppermost 10 cm of stations 461 and 472 (Fig. S-4). The snow samples have more negative δ¹⁸O values reaching ∼−23 ‰ and salinities approaching zero, consistent with atmospheric deposition. Neither salinity nor δ¹⁸O in sea ice or snow correlate with [Nd], HREE/LREE or ɛ_Nd (Fig. 2c,d). Surface seawater δ¹⁸O values are not available for our samples but likely range between −2.8 and +0.3 ‰ (Laukert et al., 2017a

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Rabe, B., von Appen, W.-J., Wegner, C., Zieringer, M., Kassens, H. (2017a) Ocean circulation and freshwater pathways in the Arctic Mediterranean based on a combined Nd isotope, REE and oxygen isotope section across Fram Strait. Geochimica et Cosmochimica Acta 202, 285–309. https://doi.org/10.1016/j.gca.2016.12.028

).

Tracking of sea ice drift based on satellite data (see Supplementary Information) reveals that the ice floes of stations 419 and 426 originated from the nearby Norske Øer Ice Barrier (NØIB), a fast ice field that forms along East Greenland’s coast and was sampled directly at station 408 (Fig. 1). The common origin of this ice is confirmed by similar temperature and salinity profiles, except at station 408 where near-freezing temperatures reflect direct sampling from the NØIB (Fig. S-4). According to IceTrack, all other ice floes originated from ice fields of the central Arctic Ocean transported via the Transpolar Drift (Fig. S-5). Exactly where and when the individual ice floes were formed along the drift route cannot be determined from IceTrack alone. However, the different ice thicknesses of these floes indicate that they must have formed at different times and places along the route.

top

Reconstructing Marine Provenance of Sea Ice and Atmosphere-Ice-Ocean Exchange

Our data show that the dissolved ɛ_Nd signature of surface seawater is incorporated into sea ice during growth and preserved during transport despite rejection of up to ∼90 % of the seawater REEs. Most upper ice core intervals have an ɛ_Nd signature between −15 and −10 and a salinity normalised [Nd] of ∼1 pmol, which is within the compositional range of surface seawater in the central Arctic Ocean (Fig. 3a) and hence in agreement with ice growth along the drift route reconstructed for the pack ice floes (Fig. S-5). We use the ratio of [Nd] and salinity ([Nd]/S) to correct for salt-proportional rejection of Nd, which enables direct comparison between sea ice and seawater (Laukert et al., 2017c

Laukert, G., Frank, M., Hathorne, E.C., Krumpen, T., Rabe, B., Bauch, D., Werner, K., Peeken, I., Kassens, K. (2017c) Pathways of Siberian freshwater and sea ice in the Arctic Ocean traced with radiogenic neodymium isotopes and rare earth elements. Polarforschung 87, 3–13. https://doi.org/10.2312/polarforschung.87.1.3

). The least radiogenic ɛ_Nd signatures reported to date for waters circulating in the vicinity of Greenland (∼−24; Laukert et al., 2018

Laukert, G., Dreyer, J., Frank, M., Hathorne, E.C., Meulenbroek, K. (2018) Greenland-sourced freshwater traced by radiogenic neodymium isotopes and rare earth elements on the North-East Greenland Shelf. Goldschmidt Abstracts 2018, 1419. https://goldschmidtabstracts.info/2018/1419.pdf

) are more positive than the values determined in the lower ice core intervals at stations 419, 454, and 472 (reaching ∼−32). However, despite the lack of published seawater data with similarly low signatures, these likely were introduced from Greenland through erosion and weathering of metamorphic rocks with even lower ɛ_Nd reaching values of ∼−42 (Laukert et al., 2017a

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Rabe, B., von Appen, W.-J., Wegner, C., Zieringer, M., Kassens, H. (2017a) Ocean circulation and freshwater pathways in the Arctic Mediterranean based on a combined Nd isotope, REE and oxygen isotope section across Fram Strait. Geochimica et Cosmochimica Acta 202, 285–309. https://doi.org/10.1016/j.gca.2016.12.028

and references therein). Dissolved ɛ_Nd signatures of Canadian rivers draining the same lithologies in a similar glacial environment are as low as those in our ice cores and, due to the very high [Nd] and despite estuarine REE removal, likely extend well beyond coastal areas once introduced into the surface ocean (Grenier et al., 2022

Grenier, M., Brown, K.A., Colombo, M., Belhadj, M., Baconnais, I., Pham, V., Soon, M., Myers, P.G., Jeandel, C., François, R. (2022) Controlling factors and impacts of river-borne neodymium isotope signatures and rare earth element concentrations supplied to the Canadian Arctic Archipelago. Earth and Planetary Science Letters 578, 117341. https://doi.org/10.1016/j.epsl.2021.117341

). Similarly unradiogenic ɛ_Nd signatures are therefore expected for surface waters near Greenland that were incorporated into sea ice at the end of the Transpolar Drift at stations 419, 454, and 472. Alternatively, the partial dissolution of small labile particulate phases originating from Greenland and incorporated during sea ice formation and drift or during sample processing may explain the very negative signatures. However, the ice core intervals with these signatures have [Nd]/S and HREE/LREE/S values for a given ɛ_Nd similar to Greenland coastal waters or sea ice expected to form from them (Fig. 3; the slight offset between actual and expected HREE/LREE/S values in sea ice reflects a process discussed below), which strongly suggests that these signals reflect direct incorporation of Greenland meltwater. Besides, a contribution of Greenland-sourced particles to the sea ice signal would not change our interpretation, given that their distribution in surface waters will not be significantly different from that of Greenland-sourced meltwater.

Full size image

At station 419, despite the proximity to the Greenland coast, surface seawater ɛ_Nd signatures were around −11 and thus differed markedly from the sea ice ɛ_Nd of −27.6. The ice floes from stations 454 and 472 originated from the central Arctic Ocean and have similar ice core lengths and ɛ_Nd distributions. They likely acquired their unradiogenic signatures from coastal waters advected to the area northeast of Greenland via the North-East Greenland Coastal Current (Laukert et al., 2017a

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Rabe, B., von Appen, W.-J., Wegner, C., Zieringer, M., Kassens, H. (2017a) Ocean circulation and freshwater pathways in the Arctic Mediterranean based on a combined Nd isotope, REE and oxygen isotope section across Fram Strait. Geochimica et Cosmochimica Acta 202, 285–309. https://doi.org/10.1016/j.gca.2016.12.028

). The multi-year ice at station 419 thus likely originated from the same area before it attached to the NØIB prior to sampling. In contrast to ɛ_Nd, δ¹⁸O is insensitive to Greenland meltwater incorporation due to indistinguishable δ¹⁸O values in surface waters along the East Greenland coast and Arctic open ocean waters (> ∼−3 ‰; Laukert et al., 2017a

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Rabe, B., von Appen, W.-J., Wegner, C., Zieringer, M., Kassens, H. (2017a) Ocean circulation and freshwater pathways in the Arctic Mediterranean based on a combined Nd isotope, REE and oxygen isotope section across Fram Strait. Geochimica et Cosmochimica Acta 202, 285–309. https://doi.org/10.1016/j.gca.2016.12.028

).

The lower ice core interval at station 481 has some of the highest HREE/LREE of all cores, which argues against incorporation of Greenland meltwater. The ɛ_Nd signature of −15.5 thus can neither be entirely explained by meltwater incorporation nor by infiltration of snow meltwater (ɛ_Nd = −13.7). Instead, it most likely reflects the uptake of Nd from Lena River freshwater characterised by an ɛ_Nd of ∼−17 to ∼−16 (Laukert et al., 2017b

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Gutjahr, M., Janout, M., Hölemann, J. (2017b) Transport and transformation of riverine neodymium isotope and rare earth element signatures in high latitude estuaries: a case study from the Laptev Sea. Earth and Planetary Science Letters 477, 205–217. https://doi.org/10.1016/j.epsl.2017.08.010

). This is supported by a relatively low δ¹⁸O value of −5.8 ‰ in the upper ice core interval and differences in the REE distribution between the different size fraction pools. The incorporation of a higher fraction of colloidal REEs (>30 kDa) at station 481 (∼45 %) than at all other stations (∼15 %) is consistent with incorporation of river-borne REEs and colloids from the Lena River (Laukert et al., 2017b

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Gutjahr, M., Janout, M., Hölemann, J. (2017b) Transport and transformation of riverine neodymium isotope and rare earth element signatures in high latitude estuaries: a case study from the Laptev Sea. Earth and Planetary Science Letters 477, 205–217. https://doi.org/10.1016/j.epsl.2017.08.010

). The near zero salinity in the upper ice core interval at this station indicates advanced melting from the previous year, also supporting the formation of this multi-year ice in the Laptev Sea and its long distance transport across the central Arctic Ocean.

The salt-normalised HREE/LREE values in sea ice are slightly lower than expected for a salt-proportional REE rejection with no change in the HREE/LREE ratio (Fig. 3b). The rejection is thus stronger for the HREEs than for the LREEs, which also explains the overall lower HREE/LREE ratios in sea ice compared to surface seawater. Despite the preferential rejection of HREEs, the differences in HREE/LREE between the parental source waters are maintained, as evidenced by a trend towards lower HREE/LREE/S values with decreasing ɛ_Nd in both sea ice and seawater. Lower HREE/LREE ratios were also observed in unfiltered sea ice samples from the central Arctic Ocean and were attributed to the contribution of particulate LREEs or differences in the incorporation of distinct size fraction pools (Laukert et al., 2017c

Laukert, G., Frank, M., Hathorne, E.C., Krumpen, T., Rabe, B., Bauch, D., Werner, K., Peeken, I., Kassens, K. (2017c) Pathways of Siberian freshwater and sea ice in the Arctic Ocean traced with radiogenic neodymium isotopes and rare earth elements. Polarforschung 87, 3–13. https://doi.org/10.2312/polarforschung.87.1.3

). Our dissolved and truly dissolved REE data allow us to exclude any direct contributions from particulate phases and instead indicate the elemental fractionation of dissolved REEs resulting from differential seawater/brine speciation. Preferential HREE rejection into the water column could also account for the accumulation of dissolved HREEs in bottom waters of the Laptev Sea (Laukert et al., 2017b

Laukert, G., Frank, M., Bauch, D., Hathorne, E.C., Gutjahr, M., Janout, M., Hölemann, J. (2017b) Transport and transformation of riverine neodymium isotope and rare earth element signatures in high latitude estuaries: a case study from the Laptev Sea. Earth and Planetary Science Letters 477, 205–217. https://doi.org/10.1016/j.epsl.2017.08.010

) but this requires further investigation in future dedicated process studies.

top

Conclusions

Our data demonstrate that Nd isotopes are a powerful tracer of the marine provenance of Arctic sea ice and biogeochemical atmosphere-ice-ocean exchange in combination with other source sensitive parameters. Even if exchange between sea ice and seawater immediately before sampling cannot be fully excluded at some stations due to indistinguishable ɛ_Nd signatures, the highly unradiogenic signatures of the Greenland-influenced ice floes suggest that, despite near zero temperatures, ɛ_Nd signatures of sea ice are largely preserved during periods in which the ice does not grow and even during periods of melting. Combined with satellite-derived sea ice motion products, our new approach enables a more accurate assessment of the water mass composition during sea ice growth and drift, which is essential for studies of sea ice formation and marine biology, biodiversity and biogeochemistry.

top

Author Contributions

GL and IP conceived the study. IP took the sea ice samples. TK reconstructed sea ice trajectories. GL prepared the samples and analysed Nd isotopes together with MG and rare earth elements together with EH. DB overlooked stable oxygen isotope sampling and analysis. All authors were involved in writing the manuscript, data interpretation and discussion.

top

Acknowledgements

We thank the Captain and Crew of the RV Polarstern for their help in collecting and transporting the samples and the chief scientist Ingo Schewe for his support (AWI_PS85_04). We also thank Antje Wildau for collecting the seawater samples and Ulrike Dietrich, Lars Chresten Lund-Hansen, Brian Keith Sorrell and Bibi Ziersen for their help with the sea ice work. Jutta Heinze is acknowledged for laboratory assistance. We acknowledge financial support by the German Federal Ministry of Education and Research (Grant BMBF 03F0776 and 03G0833) and the Ministry of Education and Science of the Russian Federation. GL also acknowledges financial support from the Ocean Frontier Institute through an award from the Canada First Research Excellence Fund. The authors declare no conflicts of interest relevant to this study.

Editor: Gavin Foster

top

References

top

Supplementary Information

The Supplementary Information includes:

Neodymium Isotope Systematics in the Arctic Ocean

Sample Collection and Preparation

Laboratory Analyses

Back-tracking of Sea Ice Based on Satellite Data

Tables S-1 to S-8

Figures S-1 to S-5

Supplementary Information References

Download Tables S-1 to S-8 (Excel).

Download the Supplementary Information (PDF).