Microscopic fossils show that, from 10,400 to 7,500 years ago, upwelling of a water mass called Circumpolar Deep Water destabilized Antarctic ice shelves — a finding that advances our understanding of ice-sheet retreat. See Article p.43
Understanding the past drivers of Antarctic ice-sheet retreat is key to recognizing the constraints on present and future ice-sheet stability. The Amundsen Sea Embayment (ASE) region of West Antarctica is a dominant contributor to the present mass loss from the West Antarctic Ice Sheet1, containing enough ice to raise the global sea level2 by 1.2 metres. In the ASE, ocean-driven melting of the undersides of ice shelves, which restrain the flow of ice from the ice sheet, is mainly caused by the inflow of a relatively warm water mass, Circumpolar Deep Water (CDW), onto the continental shelf beneath the ice shelf3 (Fig. 1). This inflow is thought to have led to ice-sheet retreat in the past. On page 43, Hillenbrand et al.4 provide the first definitive evidence that enhanced upwelling of CDW forced deglaciation of the ASE from at least 10,400 years ago until 7,500 years ago, when ice-shelf collapse could have caused rapid ice-sheet thinning. The authors also suggest that this process has been responsible for ice loss in the region since the 1940s.
The persistence of the modern West Antarctic Ice Sheet relies on the stabilizing influence of its ice shelves5. During the last glacial period (between 110,000 and 11,700 years ago), cooling or decreased presence of CDW, or both, would have reduced melting beneath the Antarctic ice shelves, contributing to ice-sheet growth and stability6. One study has suggested that the postglacial retreat of the ice sheet to modern levels in the ASE occurred about 8,000 years ago, and its authors speculated that the inflow of warmer ocean waters led to ice-shelf instability, which in turn drove ice-sheet retreat7.
Although scientists can use satellites and other instrumentation to investigate the modern West Antarctic Ice Sheet, ice shelves and surrounding seas, they must rely on proxy measurements from archives to reconstruct past changes. Hillenbrand and colleagues studied sediment cores recovered from the Amundsen Sea to reconstruct CDW upwelling onto the ASE continental shelf and to determine the role of CDW upwelling in driving ice-sheet retreat over the past 11,000 years. The authors analysed the chemical composition and assemblage of the microscopic fossil shells of organisms known as planktic and benthic foraminifera, which live in the upper ocean and on the sea floor, respectively.
Hillenbrand et al. used the ratios of chemical elements in benthic foraminifera as a proxy to show that relatively warm bottom water persisted on the ASE continental shelf before 7,500 years ago, suggesting that warm CDW flooded the region until that time. Subsequently cooler temperatures indicate that CDW inflow was reduced until modern times. The authors confirmed these findings using measurements of a water-mass tracer on planktic and benthic foraminifera — CDW has a distinct chemical signature, and the authors found that its presence on the ASE continental shelf is recorded until about 8,000 years ago, after which the pure CDW signature is reduced as a result of the mixing of CDW with other water masses.
The authors then used variations in the assemblage (species composition) of benthic foraminifera to infer the presence or absence of an ice shelf covering the ASE. They found that species indicative of a sub-ice-shelf environment dominate the assemblage until 7,500 years ago, when a distinct change — to an assemblage dominated by species attributed to an ice-shelf edge environment — occurred.
Taken together, these lines of evidence provide strong support for an oceanic driver of ice-shelf collapse in the ASE between about 8,000 and 7,500 years ago. Specifically, the enhanced inflow of warm CDW onto the ASE continental shelf, beginning at least 10,400 years ago, probably contributed to the melting of the undersides of ice shelves, leading to their collapse. Hillenbrand et al. attribute the intensification of CDW inflow before 7,500 years ago to a southerly position of the Southern Hemisphere westerly wind belt8. A major strength of the authors' study is the multi-proxy approach taken, because it allows independent validation of data.
Hillenbrand and colleagues' findings provide a crucial oceanic link to a previous study that found that Pine Island Glacier, one of two main glaciers that drain the West Antarctic Ice Sheet into the ASE, experienced rapid thinning about 8,000 years ago7. This ice-sheet retreat coincided with the strengthened CDW inflow on the ASE continental shelf found by Hillenbrand and collaborators. Such inflow probably caused the ice shelf to collapse, reducing its stabilizing effect on the ice sheet and leading to increased rates of ice-sheet retreat.
The authors also reconstructed CDW inflow to the ASE over the past century, and although their work is based on limited data, they found a renewed strengthening of CDW inflow onto the ASE continental shelf since the 1940s. If that is the case, it would confirm oceanic forcing and CDW inflow as the main drivers of ice-shelf collapse9 and ice-sheet retreat in the ASE region7 in the past few decades.
Hillenbrand and colleagues' findings are limited by the length of the sediment records and the low resolution of the bottom-water temperature data. Because the sediment records extend back about 11,000 years, when CDW was already present on the ASE continental shelf, it is difficult to determine when the enhanced CDW inflow began. Constraining this timing could help to confirm whether the southerly shift of the Southern Hemisphere westerly wind belt led to the intensification of CDW inflow. Furthermore, the bottom-water temperature data are of lower resolution than the other data used, and do not extend back beyond about 8,000 years. Additional analyses from new sediment cores would help to confirm the authors' findings. Nevertheless, their study represents a major advance in our understanding of the drivers of ice-shelf collapse.Footnote 1
Mouginot, J., Rignot, E. & Scheuchl, B. Geophys. Res. Lett. 41, 1576–1584 (2014).
Rignot, E., Mouginot, J., Morlighem, M., Seroussi, H. & Scheuchl, B. Geophys. Res. Lett. 41, 3502–3509 (2014).
Jacobs, S. S., Jenkins, A., Giulivi, C. F. & Dutrieux, P. Nature Geosci. 4, 519–523 (2011).
Hillenbrand, C.-D. et al. Nature 547, 43–48 (2017).
Pollard, D. & DeConto, R. M. Nature 458, 329–332 (2009).
Alley, R. B. et al. Annu. Rev. Earth Planet. Sci. 43, 207–231 (2015).
Johnson, J. S. et al. Science 343, 999–1001 (2014).
McGlone, M. S., Turney, C. S. M., Wilmshurst, J. M., Renwick, J. & Pahnke, K. Nature Geosci. 3, 622–626 (2010).
Paolo, F. S., Fricker, H. A. & Padman, L. Science 348, 327–331 (2015).
About this article
Transboundary atmospheric pollution in Southeast Asia: current methods, limitations and future developments
Critical Reviews in Environmental Science and Technology (2018)
The role of land bridges, ancient environments, and migrations in the assembly of the North American flora
Journal of Systematics and Evolution (2018)