Abstract: The microwave radiation imager (MWRI) aboard the Fengyun-3D satellite passively acquires microwave data, from which the area and edge line of Arctic sea ice are derived. MWRI data have seldom been used in related scientific studies, therefore their accuracy has not been sufficiently evaluated. This study combines sea ice concentration (SIC) results from six MWRI data processing algorithms to extract Arctic sea ice area and edge line, which are compared with results from the special sensor microwave imager/sounder (SSMIS) system and from the advanced microwave scanning radiometer 2 (AMSR2). Then, the SIC product derived from merged Moderate Resolution Imaging Spectrometer (MODIS) and AMSR2 observations (from 26 August 2019 to 31 December 2019, aboard NASA Terra and Aqua satellite) is used to evaluate MWRI data accuracy. The temporal variation trend of MWRI sea ice area is consistent with that of the merged MODIS-AMSR2 product. The sea ice edge line extracted from MWRI data is in better agreement with the merged MODIS-AMSR2 SIC product, with a root mean square error of 20~30 km, than with the SSMIS product. These differences are mainly caused by spatio-temporal resolution discrepancies between the datasets. In conclusion, the MWRI shows better performance than comparable instruments and is best suited for Arctic sea ice research and monitoring.

Key words: Fengyun-3D, special sensor microwave imager/sounder (SSMIS), MODIS, Arctic sea ice, sea ice area, sea ice edge line