Quaternary grounding-line fluctuations in Antarctica

Abstract

The West Antarctic Ice Sheet may be prone to rapid collapse under climates warmer than today due to a dynamic instability at the grounding line, where the ice sheet goes afloat over seawater. However, there is to-date no conclusive evidence that the ice sheet has gone away in last few million years. Thus, characterizing and understanding the transitions between the glacial and interglacial states of the ice sheet is a fundamental step towards predicting its response to future warming. Here, I investigate the history of ice sheet fluctuations in the Ross and Weddell Sea sectors of Antarctica over thousand- to million-year timescales, using cosmogenic nuclide analysis of glacial deposits and glaciated bedrock surfaces, ice-penetrating radar surveys, and numerical modeling of radar waveforms and ice flow. I have mapped and dated glacial deposits from Darwin and Hatherton glaciers, which have been used to constrain the last deglaciation in the Ross Embayment. I find that these glaciers thinned slowly and steadily through the Holocene, thousands of years later than other glaciers in the region. I use ice flow models to show (1) that their thickness changes require changing catchment boundaries upstream, and (2) that ice thickness changes at the glacier mouth are not a simple proxy for grounding line position. Next, I present new ice-penetrating radar surveys from Crary Ice Rise, a promontory in the Ross Ice Shelf that provides stability to portions of the West Antarctic Ice Sheet. I find that the ice rise contains large amounts of marine ice that accreted in basal crevasses and rifts before or during ice rise formation. Marine ice could have strengthened the damaged ice shelf, facilitating ice rise formation. Finally, I use cosmogenic nuclide concentrations in a subglacial bedrock core and a large ensemble of ice sheet model simulations to investigate the long-term stability of the West Antarctic Ice Sheet. The concentrations in the core preclude 150 m of ice sheet thinning at the Pirrit Hills since at least 2 Ma. The ice sheet model results show that continuous burial of the bedrock core requires a stable Filchner-Ronne ice shelf.