Comment on egusphere-2026-662

<strong class="journal-contentHeaderColor">Abstract.</strong> The accuracy of remotely-sensed Antarctic sea-ice thickness is limited by assumptions of snow properties and processing choices that are often informed from the Arctic. To quantify retrieval-driven spread, we compare winter radar freeboard and snow-corrected sea-ice freeboard from four CryoSat-2&ndash;era products for 2013&ndash;2018. All products reproduce the large-scale spatial structure, yet each shows systematic offsets relative to CCI that persist across multiple spatial scales. Both the Western Weddell and Ross Sea sectors display interannual variability, while correlation, bias and RMSE exhibit sector- and variable dependent performance relevant to thickness.