Late Pliocene growth of the West Antarctic Ice Sheet to near-modern dimensions

doi:10.21203/rs.3.rs-3894377/v1

Download PDF

Physical Sciences - Article

Late Pliocene growth of the West Antarctic Ice Sheet to near-modern dimensions

https://doi.org/10.21203/rs.3.rs-3894377/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Accurate prediction of West Antarctic Ice Sheet (WAIS) collapse and its impact on sea level in future warmer climate states remains uncertain¹. Understanding the Pliocene-Pleistocene ice sheet behaviour provides an opportunity to evaluate the role of forcing factors for ice mass gain/loss, determine the tipping points of WAIS growth/collapse and its impact on future sea level rise. Here, we provide evidence for the transition from a dynamic² and smaller-sized WAIS during the warm Pliocene to a relatively stable and expanded ice sheet closer to its modern configuration during the Pleistocene based on the longest sedimentary geochemical records from Amundsen Sea, in proximity to modern maximum ice loss³. This transition coincided with initiation of orbitally paced 40-ka glacial-interglacial (G-I) cycles and atmospheric pCO₂ drop, thus suggests that major tipping points for the growth and stability of the WAIS lies within atmospheric-oceanic conditions encountered during the Pliocene-Pleistocene transition. Despite large G-I variations in global temperature (+4 to -10 °C) and pCO₂ (~180 – 280 ppm)⁴, the observation of relatively stable Pleistocene WAIS is contrary to model observations⁵^,⁶ and questions its contribution to late Pleistocene sea level change. These findings have significant implications for predicting future ice sheet behavior and its impact on sea level rise.

Earth and environmental sciences/Climate sciences/Palaeoclimate

Earth and environmental sciences/Solid Earth sciences/Geochemistry

The West Antarctic Ice Sheet (WAIS) is largely marine-based (grounded below sea level) and therefore highly sensitive to climatic and oceanographic changes⁷. Over the past decades, the WAIS has been undergoing dramatic mass loss due to subglacial melting and contributed to sea-level rise at a faster rate than any other continental ice sheet on Earth⁷^,⁸. Model-based simulations predict a complete collapse of the WAIS in the near future, that would result in a global sea-level rise of 3.3-4.3 meters⁶^,⁹. However, these model predictions differ for both past and future stability of ice sheets and have large uncertainties associated with the timing of the ice sheet collapse and resulting sea-level rise¹⁰. Thus, these numerical model-based projections require validation. The alternative approach based on geological evidence from proximal West Antarctic ice sheet margins can be utilized to constrain ice sheet sensitivity to climate forcing(s) particularly during geologic times when climatic conditions were similar to modern day and/or the near-future. Arguably the best analogue in Earth’s history for near-future global climate is the Pliocene (5.33–2.58 Ma)¹¹. This epoch experienced 2-3°C higher global mean temperatures¹² and atmospheric carbon dioxide concentrations (pCO₂) between 350 and 450 ppm, that are comparable to the early 21^st century and 25–60% higher than pre-industrial values^13-16. The transition period from the warm Pliocene to a colder Pleistocene is a critical interval to investigate the dynamics of ice sheet behavior, given its analogous ice sheet boundary conditions of the Pleistocene to the modern period, as well as those of the Pliocene to likely near-future conditions in terms of atmospheric pCO₂, sea surface temperature and sea-level¹⁷^,¹⁸. A recent study has demonstrated that Antarctic Ice Sheet dimensions are tied to a multitude of temperature thresholds beyond which ice loss is irreversible due to hysteresis behaviour of the ice sheet: that is, the currently observed ice-sheet volume and extent is not regained even if temperatures are reversed to present-day levels¹⁹. ²¹Reconstructions of WAIS collapse or substantially smaller WAIS in the geological past will enable us to test ice sheet models that aim to predict future hysteresis behaviour¹⁹. Few attempts have been made earlier based on sediment cores retrieved from the western Ross Sea under the Antarctic Drilling (ANDRILL) project. However, these ANDRILL results²⁰ are not representative of the WAIS in the Amundsen Sea sector where maximum mass loss is taking place today²¹ (Fig. 1). The Amundsen Sea in West Antarctica is a key region for monitoring ice-sheet stability (Fig. 1), since the continental ice sheet in this sector is currently being lost at an accelerated pace, leading to rapid ice sheet retreat of Pine Island-, Thwaites-, and neighboring glaciers²¹^,²². Numerical model-based studies suggested that the Amundsen Sea sector is a key region for major West Antarctic Ice Sheet (WAIS) retreat or potential collapse during the most intense warm periods of the Plio-Pleistocene over the past 5 Ma⁶^,⁹. However, geological evidence for such a collapse or smaller dimensions of the WAIS is, so far, indirect and mainly based on far-field data, such as sea-level records²³. Direct geological evidence from ice-proximal records to WAIS is very limited.

Here, we provide a new record that constrains the behaviour of the WAIS in the Amundsen Sea sector during the late Pliocene and Pleistocene, based on sedimentological and geochemical data from marine sediment core Site U1532 (68°36.683′S, 107°31.500′W, 3961 m water depth) retrieved during IODP Expedition 379 in 2019²⁴ (Fig. 1, Extended Data Fig. 1, Supplementary Data). This site is located in close proximity to the WAIS, having recorded processes directly related to past WAIS dynamics in the form of sediment deposition after downslope transport via deep-sea channels. Based on the age-depth model, the cores of the present study represent the depositional history of the Pliocene-Pleistocene interval (see Methods, Extended Data Fig. 2, 3). To constrain the dynamics of the West Antarctic Ice Sheet, we produced isotope geochemical records providing key information on climate-state dependent sediment sourcing from the late Pliocene into the Pleistocene based on detrital marine sediment signatures from IODP Site U1532. We used radiogenic isotope compositions of neodymium (¹⁴³Nd/¹⁴⁴Nd, expressed as ε_Nd, which is the deviation of measured ¹⁴³Nd/¹⁴⁴Nd from the Chondritic Uniform Reservoir in parts per 10,000) and lead isotope ratios (²⁰⁸Pb/²⁰⁴Pb and ²⁰⁸Pb/²⁰⁷Pb) to trace sediment provenance. In conjunction, Nd and Pb isotopic signatures of the bulk sediments contain unique information about the continental source areas and are relatively conservative during transport and deposition. The detrital e_Nd record based on a total of 140 sedimentary samples at IODP Site U1532 shows a major shift from less radiogenic values (-7±1) during the late Pliocene to more radiogenic values (-3±1) during the Pleistocene. It is noteworthy to highlight that except four anomalous data points (three being slight more radiogenic, one being less radiogenic, Fig. 2a), all the detrital e_Ndvalues (n=38) were stable at -3 ± 1 across all sampled Pleistocene glacial-interglacial (G-I) cycles (Fig. 2a). The detrital Pb isotope records (²⁰⁸Pb/²⁰⁴Pb and ²⁰⁸Pb/²⁰⁷Pb) also followed similar trends with higher ratios during the late Pliocene and lower ratios in the Pleistocene (Fig. 2b, c). A previous report of detrital e_Nd records (~1100 ka) from two core sites²⁵, i.e. PC493a and PS58/254 in close proximity to our core sites (Fig. 1b) show almost uniform values during the late Pleistocene G-I periods and are consistent with our detrital e_Nd record (Fig. 3). The overall trend derived from the compiled e_Nd records of these three core sites clearly demonstrate a gradual and distinct shift from the supply of substantially less radiogenic Pliocene sediments to more radiogenic Pleistocene sediments and thereafter largely invariant sediment sourcing during the entire Pleistocene. The physical properties, lithological and geochemical parameters of bulk sediments exhibit consistent changes with the radiogenic isotope profiles, especially during the Pliocene-Pleistocene transition (Fig. 2).

West Antarctic geology in the vicinity of the core site encompasses a diverse range of lithologies and ages. Pine Island Glacier (PIG) and Thwaits Glacier (TG) are the two major ice streams draining into the southeastern Amundsen Sea Embayment (ASE) and supply glaciogenic detritus to the core sites (Fig. 1b,). Glacially eroded fine-grained detritus delivered offshore by PIG is characterised by lower ε_Nd value (~-9) and distinct from the sediments supplied by the TG (~-4)²⁶ (Extended Data Fig. 4a, b). Therefore, the substantial shift in the e_Nd, ²⁰⁸Pb/²⁰⁴Pb and ²⁰⁷Pb/²⁰⁸Pb records during the Pliocene-Pleistocene climate transition clearly implies changes in regional sediment sourcing over this key interval (Fig. 2a, b, c, Extended Data Fig. 5a, b, c, Extended Data Table 2). The prominent shift in the radiogenic isotope profiles can either be explained by (i) mineralogical and grain size sorting, (ii) sediment dispersal and transport associated with ocean currents and (iii) changes in sediment provenance and erosion regime associated with ice-sheet dynamics and/or combination of both these factors. Detrital e_Nd measurements in the bulk samples of the present core U1532 and fine fraction (< 63 μm) of the samples from the proximal core sites PS58/254 and PC493²⁵ show similar values within their measurement uncertainty (Fig. 3a) and thus, rules out potential effects of mineralogical or grain size sorting on the isotope compositions. Further, we argue against the possible role of coastal currents in sediment transport by driving significant changes in the detrital Nd and Pb isotope record (see Supplementary Information) and suggest that a shift in the sediment provenance for the supply of glacial detritus to the core site (Extended Data Fig. 5) caused the distinct shift in the detrital e_Nd and Pb isotope profiles during the PPT.

The large variations in TOC/TN ratio (Fig. 2d) suggest mixed input of terrestrial- and marine-derived matter, with a higher ratio indicating a greater contribution of terrestrial or regenerated organic matter during the Pliocene²⁷. Higher sedimentation rates (Fig. 2f) during the Pliocene warm period together with higher TOC/TN, suggests enhanced erosion from the continental hinterland due to extended warming associated with an ice sheet margin further inland and supply of more terrestrial sediments from the interior. Based on seismic correlation of sediment layers from the drill sites to the continental shelf, grounding zone wedges have been identified in the shelf sediments that predominantly formed during ice sheet retreat within an extended Pliocene warm period from ∼4.2– 3.2 Ma². These findings from the seismic study indicate that the WAIS was highly dynamic and a regionally much less extensive ice sheet along the Amundsen Sea margin during the Pliocene. Warmer regional climatic boundary conditions during the Pliocene would have resulted in a smaller and less extensive WAIS, which is consistent with the ice sheet model prediction for the Pliocene⁹. This situation gradually changed with expansion of the WAIS preceding the transition to the early Pleistocene cold climate. The Pleistocene expansion of the ice sheet restricted sediment supply by almost fivefold compared to the Pliocene as reflected in the sedimentation rates (Fig. 2f). Therefore, reduced erosion flux during the Pliocene-Pleistocene transition is primarily attributed to change in ice sheet extent and glacial dynamics.

Sediments deposited at three proximal sites (U1532, PC493 and PS58/254) show similar e_Nd values to modern sediments deposited in the eastern and western Amundsen Sea shelf over the past ~1.1 Ma (Fig. 3a). This could be either due to uniform sediment provenance in the hinterland or sediment provenance did not undergo significant changes during the entire Pleistocene. Given that regional Nd isotope signatures of the Pine Island (e_Nd ~-9) and Thwaites (e_Nd ~-4) basins are clearly distinct (Extended Data Fig. 4), the lack of substantial Nd isotopic variability over the late Pleistocene clearly indicates stable sediment provenance. Therefore, our observation based on three records does not support any larger scale retreat or collapse of the WAIS and questions its contribution to sea level change during Pleistocene super interglacials²⁸ (e.g. MIS 5e, 9, 11, 31) (Fig. 3). We support the hypothesis of a stable WAIS during the Late Pleistocene, with no substantial ice retreat towards the interior of the continent sufficient to alter isotopic provenance signatures. With regard to glacial-interglacial ice sheet dynamics between the WAIS and EAIS during the Pleistocene, broader consensus based on modeling studies and geological records is that the EAIS underwent significant changes with several collapse events during several super interglacials. In the light of traceable Pleistocene EAIS retreat events and modern-day rapid WAIS mass loss, our finding of a stable WAIS during the Pleistocene is unexpected. The ice-sheet model prediction for the Pliocene is consistent with our proposition, however, contradicts the predictions for the Pleistocene particularly for the last interglacial period⁹^,²⁹.

To assess the role of forcing factors and determine the threshold of the boundary conditions for the retreat and growth of the WAIS, we compare the erosion records with reconstructions of greenhouse/radiative forcing, specifically atmospheric pCO₂, and orbital forcing (solar insolation) (Fig. 4b, c). The WAIS was substantially smaller compared with its modern configuration at a pCO₂ of ~350 – 500 ppm during the Pliocene, and a stable ice sheet persisted throughout the Pleistocene in spite of large variations in temperature (Extended Data Fig. 6, 7) and atmospheric pCO₂ during the glacial-interglacial cycles (~180 to 280 ppm). The change point detection analysis in the temperature record revealed a clear cooling trend between 3.25 Ma to 2.75 Ma (Extended Data Fig. 8) with an onset of ~40 ka glacial-interglacial cycles paced by the orbital forcing (sharp increase in temperature variance at 32-64 ka band, Fig. 4d, Extended Data Fig. 7) that resulted in changes in WAIS dimensions from a dynamic smaller ice sheet to a more expanded and stable ice sheet afterwards. This observation is consistent with previous findings reported from the Ross Ice Shelf based on Drilling Program’s (ANDRILL) AND-1B sediment core records that suggested 40-ka cyclic variations in ice-sheet extent during the Pliocene were linked to obliquity cycles²⁰. Further, the AND-1B core record provide evidence for a smaller WAIS³⁰, prolonged marine conditions in the Ross Embayment, and reduced sea ice extent prior to 3.3 Ma. Subsequent Southern Ocean cooling of ∼2.5 °C in concert with a stepwise expansion of sea ice in the Ross Sea between 3.3 and 2.75 Ma coincided with drawdown of atmospheric pCO₂ from ∼400 to ∼280 ppm (Extended Data Fig. 9). We note that the major expansion of an ice sheet in the Ross Sea also began at ∼3.3 Ma³⁰. This cooling increased the seasonal persistence of Antarctic sea ice and expanded the westerly wind belt, resulting in the northward migration of ocean fronts in the Southern Ocean³¹^,³². The Pliocene records from the proximity of the WAIS margin in the Amundsen Sea and the Ross Sea clearly provide direct evidence for orbitally induced oscillations in the WAIS. The evidence is consistent with a new ice-sheet/ice-shelf model⁹ that simulates fluctuations in Antarctic ice volume associated with the loss of the WAIS, in response to ocean-induced melting paced by obliquity.

Surprisingly, no substantial changes were observed even during late Pleistocene super interglacials (e.g. last interglacial period) when global mean temperatures were 2 -3°C higher and atmospheric pCO₂ was on the order of ~300 ppm for the duration of 2-6 ka during the most pronounced interglacials MIS 5e, 9 and 11 (Fig. 2). This suggests that these extreme super interglacial conditions of the Pleistocene were insufficient for the collapse of the WAIS, which is contrary to previous reports⁵ from the Ross Sea Embayment. Further, our observation of a stable Pleistocene ice sheet is also in contrast to the findings from the East Antarctic ice sheet¹⁸, which shows ice margin retreat or thinning in the vicinity of the Wilkes Subglacial Basin of East Antarctica during warm late Pleistocene interglacial intervals. Therefore, we suggest that present mass loss of the West Antarctic Ice sheet is not comparable to the long-term climate cycles. Instead, it rather appears to be a recent phenomenon resulting from the very rapidly progressing concomitant changes in atmospheric circulation and changes in the incursion strength and pattern of the circumpolar deep water circulation.

In conclusion, our data suggest a relatively stable WAIS during the Late Pleistocene, without substantial ice sheet margin retreat further towards the interior of the continent even during the most intense interglacials of the late Pleistocene. The ice-sheet model prediction for the Pliocene is consistent with our proposition. However, it contradicts the predictions for the Pleistocene particularly for the last interglacial period⁶ and does not support the earlier suggestion of WAIS contribution to the Pleistocene sea level change. With regard to glacial-interglacial ice sheet dynamics between the WAIS and EAIS during the Pleistocene, broader consensus based on modeling studies and geological records is that EAIS underwent significant changes with several collapse events at least during super interglacials. In the context of present-day rapid mass loss from the West Antarctic ice sheet and mass gain of the East Antarctic ice sheet, the findings in the present study are unexpected as it has been traditionally suggested that the West Antarctic Ice Sheet is more vulnerable to basal melting compared with the East Antarctic Ice Sheet. Our findings suggest that the tipping point for the stability of the WAIS lies within oceanic and atmospheric boundary conditions of the Pliocene-Pleistocene transition. These findings have significant implications for predicting future ice sheet behavior and its contribution to sea level rise in a warming world.

Data availability

All data Nd and Pb isotope generated in this study can be found in the Supplementary Data. The other data sets can be found here: The IODP Exp. 379 datasets of Paleo-magnetic intensity, magnetic susceptibility, colour reflectance, bio-stratigraphy and trace elements based on shore based XRF analyzed for this study can be found in LIMS Reports (https://web.iodp.tamu.edu/ LORE/).

Acknowledgements

The authors are grateful to the International Ocean Discovery Program (IODP) organizations, in particular the JOIDES Resolution Science Operator (JRSO) as well as the master and his ship and drill crews of RV JOIDES Resolution for their support in conducting IODP Expedition 379. This research used samples and data provided by the IODP. We sincerely thank Alexander von Humboldt Foundation for providing fellowship and GEOMAR Helmholtz Centre for Ocean Research Kiel for supporting Pb isotope measurements.

Author contributions

WR designed this study in collaboration with MG. WR participated in the IODP379 expedition to collect the sediment core samples. PP and WR analysed detrital Nd isotopes. WR and MG analysed Pb isotopes. WR wrote the first draft. All the authors contributed in interpreting the data and writing the manuscript.

Competing interests

Authors declare no competing interest.

DeConto, R. M. et al. The Paris Climate Agreement and future sea-level rise from Antarctica. Nature 593, 83-89, doi:10.1038/s41586-021-03427-0 (2021).
Gohl, K. et al. Evidence for a Highly Dynamic West Antarctic Ice Sheet During the Pliocene. Geophysical Research Letters 48, e2021GL093103, doi:https://doi.org/10.1029/2021GL093103 (2021).
Rignot, E., Mouginot, J. & Scheuchl, B. Antarctic grounding line mapping from differential satellite radar interferometry. Geophysical Research Letters 38, doi:https://doi.org/10.1029/2011GL047109 (2011).
Lüthi, D. et al. High-resolution carbon dioxide concentration record 650,000–800,000 years before present. Nature 453, 379-382, doi:10.1038/nature06949 (2008).
Scherer, R. P. et al. Pleistocene Collapse of the West Antarctic Ice Sheet. Science 281, 82-85, doi:doi:10.1126/science.281.5373.82 (1998).
Pollard, D. & DeConto, R. M. Modelling West Antarctic ice sheet growth and collapse through the past five million years. Nature 458, 329-332, doi:10.1038/nature07809 (2009).
Joughin, I. & Alley, R. B. Stability of the West Antarctic ice sheet in a warming world. Nature Geoscience 4, 506-513, doi:10.1038/ngeo1194 (2011).
Frezzotti, M. & Orombelli, G. Glaciers and ice sheets: current status and trends. Rendiconti Lincei 25, 59-70, doi:10.1007/s12210-013-0255-z (2014).
DeConto, R. M. & Pollard, D. Contribution of Antarctica to past and future sea-level rise. Nature 531, 591-597, doi:10.1038/nature17145 (2016).
Dolan, A. M., de Boer, B., Bernales, J., Hill, D. J. & Haywood, A. M. High climate model dependency of Pliocene Antarctic ice-sheet predictions. Nature Communications 9, 2799, doi:10.1038/s41467-018-05179-4 (2018).
IPCC, A. Climate change 2013: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change 1535 (2013).
Dowsett, H. J., Robinson, M. M. & Foley, K. M. Pliocene three-dimensional global ocean temperature reconstruction. Climate of the Past 5, 769-783, doi:10.5194/cp-5-769-2009 (2009).
Foster, G. L., Royer, D. L. & Lunt, D. J. Future climate forcing potentially without precedent in the last 420 million years. Nature Communications 8, 14845, doi:10.1038/ncomms14845 (2017).
Pagani, M., Liu, Z., LaRiviere, J. & Ravelo, A. C. High Earth-system climate sensitivity determined from Pliocene carbon dioxide concentrations. Nature Geoscience 3, 27-30, doi:10.1038/ngeo724 (2010).
de la Vega, E., Chalk, T. B., Wilson, P. A., Bysani, R. P. & Foster, G. L. Atmospheric CO2 during the Mid-Piacenzian Warm Period and the M2 glaciation. Scientific Reports 10, 11002, doi:10.1038/s41598-020-67154-8 (2020).
Seki, O. et al. Alkenone and boron-based Pliocene pCO2 records. Earth and Planetary Science Letters 292, 201-211, doi:https://doi.org/10.1016/j.epsl.2010.01.037 (2010).
Cook, C. P. et al. Dynamic behaviour of the East Antarctic ice sheet during Pliocene warmth. Nature Geoscience 6, 765-769, doi:10.1038/ngeo1889 (2013).
Wilson, D. J. et al. Ice loss from the East Antarctic Ice Sheet during late Pleistocene interglacials. Nature 561, 383-386, doi:10.1038/s41586-018-0501-8 (2018).
Garbe, J., Albrecht, T., Levermann, A., Donges, J. F. & Winkelmann, R. The hysteresis of the Antarctic Ice Sheet. Nature 585, 538-544, doi:10.1038/s41586-020-2727-5 (2020).
Naish, T. et al. Obliquity-paced Pliocene West Antarctic ice sheet oscillations. Nature 458, 322-328, doi:10.1038/nature07867 (2009).
Turner, J. et al. Atmosphere-ocean-ice interactions in the Amundsen Sea Embayment, West Antarctica. Reviews of Geophysics 55, 235-276, doi:https://doi.org/10.1002/2016RG000532 (2017).
Rignot, E. et al. Four decades of Antarctic Ice Sheet mass balance from 1979–2017. Proceedings of the National Academy of Sciences 116, 1095, doi:10.1073/pnas.1812883116 (2019).
Dutton, A. et al. Sea-level rise due to polar ice-sheet mass loss during past warm periods. Science 349, aaa4019, doi:doi:10.1126/science.aaa4019 (2015).
Gohl, K. W. J. S. K. A. & the Expedition, S. Amundsen Sea West Antarctic Ice Sheet History. Proceedings of the International Ocean Discovery Program. Vol. 379 (International Ocean Discovery Program, 2021).
Pereira, P. S. Insights Into West Antarctica's Geology and Late Pleistocene Ice Sheet Behaviour from Isotopic Sedimentary Provenance Studies. (Imperial College London, 2018).
Simões Pereira, P. et al. The geochemical and mineralogical fingerprint of West Antarctica's weak underbelly: Pine Island and Thwaites glaciers. Chemical Geology 550, 119649, doi:https://doi.org/10.1016/j.chemgeo.2020.119649 (2020).
Meyers, P. A. Preservation of elemental and isotopic source identification of sedimentary organic matter. Chemical Geology 114, 289-302, doi:https://doi.org/10.1016/0009-2541(94)90059-0 (1994).
Melles, M. et al. 2.8 Million Years of Arctic Climate Change from Lake El'gygytgyn, NE Russia. Science (New York, N.Y.) 337, 315-320, doi:10.1126/science.1222135 (2012).
Lau, S. C. Y. et al. Genomic evidence for West Antarctic Ice Sheet collapse during the Last Interglacial. Science 382, 1384-1389, doi:10.1126/science.ade0664 (2023).
McKay, R. et al. Antarctic and Southern Ocean influences on Late Pliocene global cooling. Proceedings of the National Academy of Sciences 109, 6423-6428, doi:doi:10.1073/pnas.1112248109 (2012).
Bard, E. & Rickaby, R. E. M. Migration of the subtropical front as a modulator of glacial climate. Nature 460, 380-383, doi:10.1038/nature08189 (2009).
Sijp, W. P. & England, M. H. Southern Hemisphere Westerly Wind Control over the Ocean's Thermohaline Circulation. Journal of Climate 22, 1277-1286, doi:https://doi.org/10.1175/2008JCLI2310.1 (2009).
Fretwell, P. et al. Bedmap2: improved ice bed, surface and thickness datasets for Antarctica. The Cryosphere 7, 375-393, doi:10.5194/tc-7-375-2013 (2013).
Lisiecki, L. E. & Raymo, M. E. A Pliocene-Pleistocene stack of 57 globally distributed benthic δ18O records. Paleoceanography 20, doi:https://doi.org/10.1029/2004PA001071 (2005).
Gohl, K., Wellner, J., Klaus, A. & Expedition 379 Scientists. Vol. 379 (International Ocean Discovery Program, College Station, TX , USA, 2021).
Lisiecki, L. & Raymo, M. Lisiecki, L. E. & Raymo, M. E. A Pliocene-Pleistocene stack of 57 globally distributed benthic 18O records. Paleoceanography 20, PA1003. Paleoceanography 20, doi:10.1029/2004PA001071 (2005).
Rae, J. W. B. et al. Atmospheric CO2 over the Past 66 Million Years from Marine Archives. Annual Review of Earth and Planetary Sciences 49, 609-641, doi:10.1146/annurev-earth-082420-063026 (2021).
Laskar, J. et al. A long-term numerical solution for the insolation quantities of the Earth. A&A 428, 261-285 (2004).
Jouzel, J. et al. Orbital and Millennial Antarctic Climate Variability over the Past 800,000 Years. Science 317, 793-796, doi:doi:10.1126/science.1141038 (2007).
Rohling, E. J. et al. Sea level and deep-sea temperature reconstructions suggest quasi-stable states and critical transitions over the past 40 million years. Science Advances 7, eabf5326, doi:doi:10.1126/sciadv.abf5326 (2021).

Age model of IODP Site U1532

Age constraints used in this study were taken from the shipboard age-depth models, which were established based on biostratigraphy age datums in correlation with magnetostratigraphic polarity zones¹. Based on shipboard analyses of Holes U1532A–U1532G, an age-depth model for the Site U1532 was constructed using biostratigraphic age datums (Extended Data Table 1) to constrain correlation of magnetostratigraphic polarity zones at Site U1532 correlated to the Gradstein et al. (2012) geological timescale (GTS2012)² (see Table T5 in the Expedition 379 methods chapter³). Biostratigraphic age control for the Site U1532 is based on diatom and radiolarian datums³. Sufficient microfossils for biostratigraphic age assignment were only present in the upper ~10 m of the section (i.e. from 0 to 92 m; Lithostratigraphic Subunit IA) which provides an age of middle Pleistocene to recent (0–0.60Ma). Based on diatom and radiolarian biostratigraphy, the interval between ~92 and 156 m is assigned a mid-to-late Pliocene age of 3.2–3.8 Ma. Using these biostratigraphic age control points, interpreted correlation of magnetostratigraphic polarity reversals identified at Site U1532 (Table T16R)⁴ to the GTS2012 is relatively unambiguous. The interval between 0 and ~45 m is assigned a Pleistocene age, and the interval between ~45 and 150 m represent Pliocene age. Because of an absence of siliceous microfossils, there is no biostratigraphic age control between ~10 and 92 m in Hole U1532A-B⁴. For Hole U1532A, a reliable shipboard magnetostratigraphy consisting of four normal and four reversed polarity intervals was obtained. These intervals are the Brunhes–Matuyama polarity transition (0.781 Ma), the termination and beginning of the Olduvai Subchron (1.778 and 1.945 Ma), the Matuyama–Gauss polarity transition (2.581 Ma), the termination and beginning of the Kaena Subchron (C2An.1r; 3.032 and 3.116 Ma) and the termination and beginning of the Mammoth Subchron (C2An.2r; 3.207 and 3.330 Ma, respectively) (see Extended Data Table 1). Paleomagnetic measurements in Hole U1532B identified the beginning of the Mammoth Subchron (C2An.2r; 3.330 Ma) and the Gauss–Gilbert polarity transition (3.596 Ma). Total eight magnetostratigraphic tie points were found between 0 and ~100 CSF-A(m) and two tie points between 101 and 146 CSF-A(m) for the basis of the chronology (Extended Data Table 1, Extended Data Fig. 3a). To improve the shipboard chronology further, we have identified more tie points in the present study based on the correlation between geochemical parameters and global benthic δ¹⁸O curve (LR04)⁵ (Extended Data Fig. 2). Elemental ratios of (a) Ba/Al, (b) Ba/Rb, (c) PC1 (first principal component) derived from Principal Component Analysis (PCA) of multiple elemental ratios (Extended Data Fig. 2a, b, c) were compared with the LR04 curve. In addition to previously determined tie points, nine additional new tie-points were identified based on the excursions in the geochemical parameters and its correspondence with the glacial-interglacial signals in the LR04 δ¹⁸O curve (shown in Extended Data Fig. 2). Finally, MatLab™-based program “Undatable”⁶ was used to generate an age-depth model with an uncertainty. The advantage of using Undatable over other available age-depth modelling programs is that it allows for the input of uncertainties in both age and in depth, and for a series of different types of age control points to be incorporated. Undatable was run using 100,000 simulations⁶. A probability density cloud illustrating the uncertainty envelope around the modeled age‐depth points was computed using Monte Carlo iterations. Blue and black dotted lines show 1σ and 2σ uncertainty, respectively. This revised model is based on sixteen tie-points within the study interval (0–3.5 Ma, Extended Data Fig. 3b, Extended Data Table 1). The age-depth model for the Pleistocene interval is better constrained than the Pliocene interval due to the availability of more tie points/ages and therefore Pleistocene glacial-interglacial variations can be resolved with confidence.

Detrital sediment Nd and Pb isotope measurements

Nd isotopes were measured in the detrital phases of the bulk sediments following the method adopted from earlier studies⁷^,⁸. Authigenic fractions were extracted and removed from the bulk sediments using the leaching method adopted from previous studies^9-11. To ensure complete removal of the authigenic fractions, all samples were leached twice, first for ten seconds followed by a second leach lasting 24 hours. Approximately 0.3 g of powdered and homogenized sediment samples were agitated with a mixture solution of 0.005M hydroxylamine hydrochloride, 1.5% acetic acid, and 0.003M Na-EDTA, buffered to pH 4 with NaOH. Then leachates were removed by centrifugation and washed with Milli-Q water thoroughly. Then residual sediment samples were decarbonated by treating with 0.6 N HCl at 80 °C for ~30 minutes and then supernatant was discarded, and the acid‐free residue was washed with Milli‐Q water. The above steps were repeated to ensure complete decarbonation of the samples and washed with Milli‐Q water and checked with pH paper to ensure that carbonates are removed from the sediments. Subsequently, samples were washed thoroughly with Milli-Q water and dried in the oven at 80° C. Dried samples were then ashed at 600°C to remove organic matter. Approximately 50 mg of ashed samples were digested in precleaned Teflon vials using HF-HNO₃-HCL mixture at 120°C. Standard Reference Materials BCR-2 and BHVO-2 were also digested along with the samples. The digested samples were divided in two aliquots for the column chromatography of Nd and Pb. The dissolved samples were subjected to purification by column chromatography. They were passed through columns filled with cation-exchange resin AG50W-X8 (200–400 micron) to separate Rare Earth Elements (REE). Then REE fractions were then passed through the column filled with Eichrom LN spec^TM (50 – 100 micron) resin to separate Nd from the REEs¹².

Nd isotope Analysis

Nd isotope ratios were measured using Multi-collector Inductively Coupled Plasma Mass Spectrometer (Thermo Fischer Scientific, Neptune Plus) at National Centre for Polar and Ocean Research, Goa. Instrumental mass fractionation was corrected using the ratio of ¹⁴⁶Nd/¹⁴⁴Nd = 0.7219. The mass bias corrected ¹⁴³Nd/¹⁴⁴Nd ratios were normalised to the reported JNdi-1 standard ¹⁴³Nd/¹⁴⁴value of 0.512115¹³. A correction for direct ¹⁴⁴Sm interference was also applied, with all samples which was below (<0.1% of the ¹⁴⁴Nd signal). To ensure the quality of measurement, the international standard JNdi-1 was matrix matched and measured at every five samples. The obtained average ratios were 0.512115 ± 10 ppm (2σ, n= 51). The external reproducibility calculated for each session was 0.12–0.27 e_Nd (2σ) units based on the repeated measurements of JNdi-1. If the internal error (2σ) is larger than the external error, the internal error is reported as the final uncertainty associated with the individual measurements. Several procedural blanks were also processed along with the samples and ascertained an average blank of ~70 pg for the samples (n=9), which is several orders of magnitude lower than the total Nd typically analysed in samples. Hence, no blank correction was applied. Replicates have an average e_Nd variation of ±0.17 (n= 16). Over the course of analyses, measurements of rock standard BCR-2 and BHVO-2 processed with the samples gave ¹⁴³Nd/¹⁴⁴Nd = 0.512641 ± 0.000010 (2s, n = 7) and ¹⁴³Nd/¹⁴⁴Nd = 0.513008 ± 0.000012 (2s, n = 7), in excellent agreement with reported values¹⁴.

Pb isotope analysis

Pb isotope was analysed in the same aliquot in which Nd isotope was measured. The Pb cuts in the detrital dissolved were purified by column chromatography filled with ∼80 μL AG1-X8 resin. Pb isotopes were measured using a Thermo Scientific Neptune Plus MC-ICP-MS at GEOMAR Helmholtz Centre for Ocean Research Kiel. Mass bias correction during Pb isotope measurements was performed externally using the Tl-doping technique⁹^,¹⁵ with added NIST997 Tl standard solution and a Pb/Tl ratio of ~4. The total procedure blanks were ~500 pg which represents less than 0.1% of the sample signal. Blank corrected and uncorrected ratios do not show significant difference and hence blank uncorrected data is reported in this study. The reproducibility of the rock standard BCR-2 and BHVO-2 is listed in Supplementary Table 2. As shown in the table, all measured standard Pb isotopic ratios are within the error of published compositions. Detailed information about the measurement procedure can be found in Süfke et al. (2019)¹⁵ and Huang et al. (2021)⁹.

Robustness of the detrital sediment Nd and Pb isotope records

Unlike other radiogenic isotope systems (for example, Sr and Hf), Nd isotopes are relatively insensitive to grain size variations¹⁶, making them a robust indicator of sediment provenance. Their insensitivity to grain size variations makes them more reliable tracers of sediment provenance. However, it is crucial to consider potential complications in provenance interpretations if changes in sediment transport processes result in changing the isotope compositions. In such cases, a link between grain size and Nd isotopes could emerge, impacting the reliability of using these isotopes for provenance analysis. To assess the robustness of Nd isotopes, Pereira P.S. (2018)¹⁷ analysed Nd isotopes in fine-grained fraction (<63 μm) of sediment cores from two locations (PS58/254 and PC493) near our core site U1532 (Fig. 1b). The results of this investigation ruled out any significant influence of grain-size variations on the Nd isotope compositions, reinforcing the suitability of Nd isotopes as a dependable proxy for sediment provenance in this particular study area. Among various combinations of Pb isotope ratios, we have used ²⁰⁸Pb/²⁰⁴Pb and ²⁰⁸Pb/²⁰⁷Pb ratios as these displayed the most distinctive, clearest and smoothest signal compared to other ratios and both these ratios co-vary in concert with the e_Nd record. Observation of changes in the paired Nd-Pb isotope records is more robust to decipher the changes in the sediment provenances than either of the two trace metal isotope compositions alone. This finding further validates the robustness of using these isotopes as tracers of sediment sources and transport processes in the specific study region. In summary, the distinct insensitivity of Nd to grain size fractionation makes them reliable tools for tracing sediment provenance. Pertinent to this, it is important to mention that Pliocene erosional history and ice loss from the East Antarctic Ice Sheet (EAIS) were successfully reconstructed previously using detrital e_Nd records from the Adélie Land sector of East Antarctica⁷^,¹⁸. Distinct sediment sourcing associated with the growth and retreat of the ice sheet margin and multiple collapse events clearly establish a close link between ice sheet dynamics, erosion and variable regional sediment supply. Therefore, our extensive analysis in the present study and successful demonstration in the previous study from this region provides confidence in the applicability of detrital radiogenic isotope compositions of Nd and Pb provide confidence for deciphering sediment sources and transport processes in the studied area.

Wavelet and scaled-average variance of wavelet power analysis

Wavelet transforms decompose time series into time-frequency space and can therefore find localized intermittent periodicities. We used Continuous Wavelet Transform (CWT) i.e. Morlet wavelet to decompose the time series into time-frequency space which enable us to identify the modes of variability and how those modes vary with time¹⁹. Statistical significance is estimated against a red noise model. Monte Carlo methods were used to assess the statistical significance against red noise backgrounds²⁰. Wavelet analysis of the temperature record was performed using online Matlab codes http://grinsted.github.io/wavelet-coherence/ to identify nonstationary power at certain frequency bands¹⁹^,²⁰. Temperature variability at ~40 ka and ~100 ka bands is shown in the wavelet analysis. More detailed information on wavelet analysis is provided elsewhere¹⁹^,²⁰. Further, to examine fluctuations in power over a range of scales (a band), we used scale-averaged wavelet power which is a time series of the average variance in a certain band. This can be used to examine modulation of one time series by another, or modulation of one frequency by another within the same time series. We have performed this analysis using online Matlab codes http://paos.colorado.edu/research/wavelets/. The scaled average variance of the temperature at 32-64 ka and 64 -100 ka band captured the wavelet analysis show how the ~40 ka and ~100 ka periodities evolved with time (Extended Data Fig. 7).

Change-Point Detection using a Bayesian Ensemble Algorithm (BEAST)

Abrupt changes in the temperature time series were detected using a MATLAB based Bayesian model averaging time-series decomposition algorithm (BEAST)²¹ (Extended Data Fig. 8). This algorithm offers robust method to detect abrupt change points and nonlinear trend in the Antarctic temperature record. The code and software are available at https://github.com/zhaokg/Rbeast. The Bayesian algorithm—BEAS decomposition time series into three contrasting components: abrupt change, periodic/seasonal change, and trend. Mathematically, BEAST is rigorously formulated, with its key equations being analytically tractable. Practically, BEAST provide estimate of probabilities of change point occurrence and detect both large and low-magnitude disturbances, and uncover complex nonlinear trend dynamics. The details of the Bayesian Ensemble Algorithm (BEAST) are discussed elsewhere²¹.

References

1 Wellner, J. S. et al. (2021).

2 Gradstein, F., Ogg, J. & Hilgen, F. A Geologic Time Scale. Newsletters on Stratigraphy 45, 171-188, doi:10.1127/0078-0421/2012/0020 (2012).

3 Gohl, K., Wellner, J., Klaus, A. & Expedition 379 Scientists. Vol. 379 (International Ocean Discovery Program, College Station, TX , USA, 2021).

4 Gohl, K., Wellner, J. & Klaus, A. Amundsen Sea West Antarctic Ice Sheet History. Proceedings of the International Ocean Discovery Program 379 (2021).

5 Lisiecki, L. & Raymo, M. Lisiecki, L. E. & Raymo, M. E. A Pliocene-Pleistocene stack of 57 globally distributed benthic 18O records. Paleoceanography 20, PA1003. Paleoceanography 20, doi:10.1029/2004PA001071 (2005).

6 Lougheed, B. C. & Obrochta, S. P. A Rapid, Deterministic Age-Depth Modeling Routine for Geological Sequences With Inherent Depth Uncertainty. Paleoceanography and Paleoclimatology 34, 122-133, doi:https://doi.org/10.1029/2018PA003457 (2019).

7 Wilson, D. J. et al. Ice loss from the East Antarctic Ice Sheet during late Pleistocene interglacials. Nature 561, 383-386, doi:10.1038/s41586-018-0501-8 (2018).

8 Subha Anand, S. et al. Trace elements and Sr, Nd isotope compositions of surface sediments in the Indian Ocean: An evaluation of sources and processes for sediment transport and dispersal. Geochemistry, Geophysics, Geosystems 20, 3090-3112 (2019).

9 Huang, H., Gutjahr, M., Kuhn, G., Hathorne, E. C. & Eisenhauer, A. Efficient Extraction of Past Seawater Pb and Nd Isotope Signatures From Southern Ocean Sediments. Geochemistry, Geophysics, Geosystems 22, e2020GC009287, doi:https://doi.org/10.1029/2020GC009287 (2021).

10 Blaser, P. et al. Extracting foraminiferal seawater Nd isotope signatures from bulk deep sea sediment by chemical leaching. Chemical Geology 439, 189-204 (2016).

11 Gutjahr, M. et al. Reliable extraction of a deepwater trace metal isotope signal from Fe–Mn oxyhydroxide coatings of marine sediments. Chemical Geology 242, 351-370 (2007).

12 Pin, C. & Zalduegui, J. S. Sequential separation of light rare-earth elements, thorium and uranium by miniaturized extraction chromatography: Application to isotopic analyses of silicate rocks. Analytica Chimica Acta 339, 79-89, doi:https://doi.org/10.1016/S0003-2670(96)00499-0 (1997).

13 Tanaka, T. et al. JNdi-1: a neodymium isotopic reference in consistency with LaJolla neodymium. Chemical Geology 168, 279-281, doi:https://doi.org/10.1016/S0009-2541(00)00198-4 (2000).

14 Weis, D. et al. High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS. Geochemistry, Geophysics, Geosystems 7, doi:https://doi.org/10.1029/2006GC001283 (2006).

15 Süfke, F. et al. Early stage weathering systematics of Pb and Nd isotopes derived from a high-Alpine Holocene lake sediment record. Chemical Geology 507, 42-53, doi:https://doi.org/10.1016/j.chemgeo.2018.12.026 (2019).

16 Garçon, M., Chauvel, C., France-Lanord, C., Huyghe, P. & Lavé, J. Continental sedimentary processes decouple Nd and Hf isotopes. Geochimica et Cosmochimica Acta 121, 177-195, doi:https://doi.org/10.1016/j.gca.2013.07.027 (2013).

17 Pereira, P. S. Insights Into West Antarctica's Geology and Late Pleistocene Ice Sheet Behaviour from Isotopic Sedimentary Provenance Studies. (Imperial College London, 2018).

18 Cook, C. P. et al. Dynamic behaviour of the East Antarctic ice sheet during Pliocene warmth. Nature Geoscience 6, 765-769, doi:10.1038/ngeo1889 (2013).

19 Grinsted, A., Moore, J. C. & Jevrejeva, S. Application of the cross wavelet transform and wavelet coherence to geophysical time series. Nonlin. Processes Geophys. 11, 561-566, doi:10.5194/npg-11-561-2004 (2004).

20 Torrence, C. & Compo, G. P. A Practical Guide to Wavelet Analysis. Bulletin of the American Meteorological Society 79, 61-78, doi:https://doi.org/10.1175/1520-0477(1998)079<0061:APGTWA>2.0.CO;2 (1998).

21 Zhao, K. et al. Detecting change-point, trend, and seasonality in satellite time series data to track abrupt changes and nonlinear dynamics: A Bayesian ensemble algorithm. Remote Sensing of Environment 232, 111181, doi:https://doi.org/10.1016/j.rse.2019.04.034 (2019).

There is NO Competing Interest.

SupplementaryData.xlsx
IODP_379_Site-U1532, Standard_BHVO-2_BCR-2_Value
SupplementaryInformation.docx
Late Pliocene growth of the West Antarctic Ice Sheet to near-modern dimensions
ExtendedDataFigures.docx
Extended Data Figures 1 to 9, Extended Data Table

Download PDF

Version 1

posted

You are reading this latest preprint version

Late Pliocene growth of the West Antarctic Ice Sheet to near-modern dimensions

Status:

Version 1

Abstract

Figures

Main

WAIS growth during Plio-Pleistocene transition

Evidence for a largely stable Pleistocene WAIS

Forcing modulated WAIS extent and dynamics

Declarations

References

Methods

Additional Declarations

Supplementary Files

Status:

Version 1