Search

Upcoming submillimeter wave satellite missions require an improved understanding of the sea ice emissivity to separate atmospheric and surface microwave signals under dry polar conditions. This work investigates hectometer-scale airborne sea ice emissivity observations between 89 and 340 GHz combined with high-resolution visual imagery from two Arctic airborne field campaigns in summer 2017 and spring 2019 northwest of Svalbard, Norway. We identify four distinct sea ice emissivity spectra through K-Means clustering, which occur predominantly over multiyear ice, first-year ice, young ice, and nilas. Nilas features the highest, and multiyear ice features the lowest emissivity among the clusters. Each cluster exhibits similar nadir emissivity distributions from 183 to 340 GHz. To relate hectometer-scale airborne to kilometer-scale satellite footprints, we quantify the reduction of airborne emissivity variability with increasing footprint size. At 340 GHz, the emissivity interquartile range decreases by almost half from the hectometer scale to a footprint of 16 km, typical for satellite instruments. Furthermore, we collocate the airborne observations with polar-orbiting satellite observations. After resampling, the absolute relative bias between airborne and satellite emissivities at similar channels lies below 3 %. Additionally, spectral nadir emissivity variations on the satellite scale are low, with slightly decreasing emissivity from 183 to 243 GHz, which occurs for all hectometer-scale clusters except for predominantly multiyear ice. Our results will enable the development of microwave retrievals and assimilation over sea ice from current and future satellite missions such as Ice Cloud Imager (ICI) and European Polar System (EPS) Sterna. [ABSTRACT FROM AUTHOR]