Abstract: Sea ice thickness is one of the important variables of sea ice. Our knowledge of the temporal and spatial integrity of sea ice thickness is still insufficient and lags behind that of sea ice concentration, extent, and drift velocity. At present, Arctic sea ice thickness data are mostly obtained by satellite remote sensing using CryoSat-2 and SMOS, and since November 2018, ICESat-2. Few studies have evaluated the accuracy of the latest multi-source satellite Arctic sea ice thickness data or assessed the spatial and temporal differences of data from different sources. In this study, we quantified the spatial and temporal differences between the latest sea ice thickness data from ICESat-2, CryoSat-2, and CS2SMOS (merged product of CryoSat-2 and SMOS) during 2019–2020. Results show that the mean sea ice thickness is largest in CryoSat-2, second largest in ICESat-2, and smallest in CS2SMOS. There are also considerable temporal and spatial differences. For regions covered with thick sea ice, thickness is largest in ICESat-2, and the thickness in Cryosat-2 is similar to that in CS2SMOS. For marginal ice zones, thickness is largest in CryoSat-2 and smallest in CS2SMOS. The difference between the datasets is smaller over the East Siberian Sea and Beaufort Sea and larger over the Barents Sea. We compared the satellite data with field data collected during Operation IceBridge (OIB) in April 2019 and found that the difference between CS2SMOS and OIB is smaller than that between ICESat-2 and OIB.

Key words: Arctic, sea ice thickness, satellite remote sensing, Operation IceBridge (OIB)