Estimating Antarctic surface melt rates using passive microwave data calibrated with weather station observations

Abstract. We present a dataset of Antarctic annual surface melt rates (6.25 km resolution, 2011–2021) from 19 GHz Special Sensor Microwave Imager/Sounder (SSMIS). First, melt occurrence is detected via thresholds for brightness temperature, diurnal variation, and winter anomaly, calibrated with Automatic Weather Station (AWS) data. Second, AWS-driven surface energy balance modeling yields an empirical relation between annual melt days and water-equivalent melt volume. SSMIS-derived melt volumes correlate well with AWS-based melt estimates (R2=0.83). Compared to QuikSCAT and RACMO2.4p1 outputs, SSMIS captures a similar spatial melt pattern but estimates a total melt volume approximately 15 % lower than RACMO2.4, on the decadal average.