A quantitative framework for evaluating marine-based geologic constraints on Antarctic Ice Sheet retreat

Reconstructing Antarctic Ice Sheet evolution through the last deglaciation is crucial for quantifying global sea level rise and solid Earth loading changes. This endeavor requires both empirical geologic observations and numerical ice sheet models: geologic data offer valuable but spatially and temporally fragmented snapshots of past glacial configurations, whereas numerical models provide three-dimensional physically based ice evolution but generate a wide range of plausible deglacial histories. Together, these approaches enable process-level insights into ice sheet sensitivity to warmth. Here we focus on the marine seafloor, combining geologic information gleaned from sediment cores and mapped seafloor landforms with numerical ice sheet modeling. We develop model-data comparison metrics that assess both the timing of grounding-line recession and its spatial pattern across the Amundsen and Ross Sea continental shelves. The application of these metrics to a targeted ensemble of ice sheet model simulations demonstrates the power of geologic data to constrain the chronology and amplitude of modeled deglacial ice loss. Marine radiocarbon and seafloor landform observations are translated into glaciologically consistent reconstructions of regional ice sheet deglaciation, lending quantitative context for the reconstruction of low-slope ice sheets that remain pinned on seafloor highs as deglaciation progresses. This work highlights the use of physically self-consistent ice sheet models as a powerful tool to contextualize geologic records of Antarctic deglaciation.