AdaSA-SD (v1.0): An Adaptive Seasonal Algorithm for Snow Depth Retrieval Over Arctic Sea Ice

Accurate estimates of snow depth on sea ice are essential for monitoring sea ice mass balance and quantifying heat exchange between the atmosphere and sea ice. However, the seasonal evolution of pan-Arctic snow depths throughout the freezing season (October–April) remains poorly understood. In this study, we developed an Adaptive Seasonal Algorithm for Snow Depth retrieval over Arctic sea ice (AdaSA-SD) based on the Advanced Microwave Scanning Radiometer 2 (AMSR2) measurements, which incorporates time-varying and ice-type dependent algorithmic coefficients. A daily pan-Arctic snow depth dataset with a spatial resolution of 25 km × 25 km was then generated for the freezing seasons from 2012 to 2023. Our results accurately reproduce the observed spatial patterns and seasonal and interannual variability of Arctic snow depths, with mean biases ranging from 0.5 to 3.0 cm and mean absolute percentage errors between 10% and 30% when validated against airborne, buoy, and transect data. The seasonal cycle of our estimated snow depths also agrees well with simulated results from state-of-the-art snow models. Furthermore, the uncertainties and limitations associated with the snow depth retrievals are thoroughly discussed. The approach and dataset proposed in this study will improve our understanding of the changing state of the Arctic snow-sea ice system across the freezing season.