SELECTING ANTARCTIC SITES FOR SUBGLACIAL BEDROCK RECOVERY TO TEST FOR PAST ICE SHEET COLLAPSE, AN EXAMPLE FROM THE PIRRIT HILLS

Large portions of the Antarctic Ice Sheet overlie deep marine basins and may be susceptible to unstable retreat. There is strong but indirect evidence for a smaller ice sheet during one or more late Pleistocene interglacials; however, the timing and magnitude of past deglaciations remain uncertain. Establishing whether and when deglaciations occurred would be invaluable for understanding, and potentially predicting, the stability of marine-based portions of the ice sheet. Because most cosmic radiation is absorbed by as little as 5-10 m of ice, measurements of cosmogenic nuclides in subglacial bedrock have the potential to determine whether that rock was exposed by past ice thinning. Drilling technology coming online promises increased access to subglacial bedrock for cosmogenic nuclide analyses. Careful site selection will be critical for maximizing the information obtained from future bedrock coring.

Sites must meet three important criteria: (i) changes in local ice levels should reflect changes in the size of the broader ice sheet; (ii) the bedrock should contain minerals that allow cosmogenic nuclides with different decay rates (e.g. ¹⁰Be, ²⁶Al, ³⁶Cl, ²¹Ne, ¹⁴C) to be measured; (iii) the cosmogenic nuclide record, which is produced within a few meters of the surface, must have been continuously protected from erosion.

As an example, in 2017 we obtained an 8 m bedrock core from below 150 m of ice near the Pirrit Hills, a small group of nunataks in West Antarctica. We chose this site, located midway between the divide and grounding line, because we expect ice thickness here to reflect glacial-interglacial changes in ice-sheet size; this is supported by a published ice-sheet model. Weathered bedrock descends below the modern ice surface and exposure ages within 50 m of the surface range from 0.2 to 1.2 Myr, both indicating prolonged exposure, lower ice levels in the past, and preservation of the cosmogenic nuclide record. Similar relations are common elsewhere in Antarctica. We aimed for rock within ~200 m of the surface to ensure continuous protection by cold-based ice, targeting the subglacial ridge of a small nunatak beneath flat-lying and slow-flowing (< 0.5 m/yr) ice. Our first hole was abandoned because we inadvertently hydrofractured the basal ice; the second was successful.