Improved method for retrieving Arctic summer sea ice velocity based on FY-3D/MWRI brightness temperatures

Rapid decline in Arctic summer sea ice cover has brought new opportunities for shipping and resource exploitation in the Arctic, and accurate determination of summer sea ice conditions has become increasingly important. Knowledge of sea ice velocity (SIV) is crucial for ice engineering and navigation in the Arctic. This study retrieved SIV in summer based on the polarization difference (PD) of brightness temperatures (TBs) at 36 GHz derived from the recalibrated FY-3D satellite microwave radiometer data FY-3D/MWRI. Compared with swath TB data, swath PD data can better suppress the noise associated with atmospheric wetness (cloud liquid water and water vapor). Moreover, different from the arbitrary template selection scheme used in current algorithms, we propose an optimized template selection scheme based on the image texture, which considers the frequent deformation processes of sea ice in summer. Determination of SIV using templates with irregular distribution presents lower uncertainty and higher computational efficiency compared with templates with regular distribution. Based on the improved algorithm, this study obtained either swath irregular SIV fields with nominal spatial resolution of 30 km or daily gridded SIV fields with 50-km spatial resolution in summer 2019 using 36-GHz PDs.