Historical Calibration of Basal Melt Parameters in a circum-Antarctica Ice-Sheet Model

The Antarctic Ice Sheet (AIS) is a major contributor to future global sea level rise. Approximately half of the surface mass gain is offset by ocean-induced basal melting, highlighting the critical role of ice-ocean interactions. Uncertainty in projections of AIS evolution remains strongly linked to how basal melting is represented and calibrated in ice-sheet models, together with divergent future climate forcing scenarios.

In this study, we use a circum-Antarctic high-resolution configuration of the Úa ice-sheet model to conduct a series of 360 hindcast simulations (spanning 2000-2020) to quantify uncertainties and sensitivities in modelled ice-shelf melt. The ensemble covers a range of ice rheology and basal sliding parameters, as well as multiple basal melt parameterizations (quadratic, PICO and plume) and a physically plausible range of parameter choices for each parameterization.

Whereas previous studies have calibrated basal melt parameters using fixed ice-sheet geometries or relied primarily on basal melt observations alone, this study presents two advances: 1) ice-ocean feedbacks were included in the calibration through temporally evolving basal melt rates, and 2) simulated changes in ice velocity and thickness over the hindcast period were validated against remote-sensing data.

After calibration, model performance improves in the representation of both basal melt rates and ice-dynamic response patterns. For most basal melt parameters, the posterior distributions exhibit clear localization relative to the prior, indicating well-defined optimal parameter values. The resulting calibrated parameter ranges therefore provide a more robust foundation for future long-term projections of AIS evolution and its contribution to global sea-level rise. Notably, these optimal parameter values differ from those obtained using calibration approaches based on fixed ice-shelf cavities or basal-melt observations alone. We also examine regional variability in calibration results. The relative performance of basal melt parameterizations differs between Antarctic sectors, while optimal parameter ranges within each parameterization remain broadly consistent with the Antarctic-wide calibration.