Surface melting occurs across many of Antarctica’s ice shelves, mainly during the austral summer. The onset, duration, area and fate of surface melting varies spatially and temporally, and the resultant surface meltwater is stored as ponded water (lakes) or as slush (saturated firn or snow), with implications for ice-shelf hydrofracture, firn air content reduction, surface energy balance and thermal evolution. This study applies a machine-learning method to the entire Landsat 8 image catalogue to derive monthly records of slush and ponded water area across 57 ice shelves between 2013 and 2021. We find that slush and ponded water occupy roughly equal areas of Antarctica’s ice shelves in January, with inter-regional variations in partitioning. This suggests that studies that neglect slush may substantially underestimate the area of ice shelves covered by surface meltwater. Furthermore, we found that adjusting the surface albedo in a regional climate model to account for the lower albedo of surface meltwater resulted in 2.8 times greater snowmelt across five representative ice shelves. This extra melt is currently unaccounted for in regional climate models, which may lead to underestimates in projections of ice-sheet melting and ice-shelf stability.

Because Antarctic surface melt is mostly driven by local processes, its simulation necessitates high-resolution regional climate models (RCMs). However, the current horizontal resolution of RCMs (≈25–30 km) is inadequate for capturing small-scale melt processes. To address this limitation, we present SUPREME (SUPer-REsolution-based Melt Estimation over Antarctica), a deep learning method to downscale surface melt to 5.5 km resolution using a physically-informed super-resolution model. The physical information integrated into the model originates from observations tied to surface melt, specifically remote sensing-derived albedo and elevation. These remote sensing data, in addition to a Regional Atmospheric Climate Model (RACMO) run at 27 km resolution, account for the diverse drivers of surface melt across Antarctica, facilitating effective generalization beyond the training region of the Antarctic Peninsula. A comparison of SUPREME with a dynamically downscaled RACMO run at 5.5 km over the Antarctic Peninsula shows high accuracy, with average yearly RMSE and bias of 5.5 mm w.e. yr⁻¹ and 4.5 mm w.e. yr⁻¹, respectively. Validation at five automatic weather stations reveals SUPREME's marked improvement with substantially lower average RMSE (81 mm w.e.) compared to RACMO 27 km (129 mm w.e.). Beyond the training region, SUPREME aligns more closely with remote sensing products associated with surface melt than super-resolution models lacking physical constraints. While further validation of SUPREME is needed, our study highlights the potential of super-resolution techniques with physical constraints for high-resolution surface melt monitoring in Antarctica, providing insights into the impacts of localized melting on processes affecting ice shelf integrity such as hydrofracturing.