Validation of Sentinel-3A SLSTR derived Sea-Surface Skin Temperatures with those of the shipborne M-AERI

Abstract

The Skin Sea-Surface Temperature (SSTskin) derived from satellite measurements is one of the critical factors for determining ocean-atmosphere interactions in climate prediction and ocean research. Sentinel-3A is a European Earth Observation satellite mission developed to support ocean, land, and atmospheric applications. We compare three years of SSTskin from the Sea and Land Surface Temperature Radiometer (SLSTR) on board the Sentinel-3A satellite from July 2017–March 2019 with independent data from Marine-Atmospheric Emitted Radiance Interferometers (M-AERIs) deployed on research cruises. The comparison results show an average difference of −0.098 K and Robust Standard Deviation of 0.296 K between SLSTR and M-AERI SSTskin values; the average value meets the pre-launch scientific objective of a mean accuracy <0.1 K. Differences between four SSTskin retrieval algorithms have been investigated, the results show that the Dual-view 3 channels night time algorithm (D3) is better than other algorithms. This study involves determining the error and uncertainty characteristics of the SST retrievals in terms of factors that influence the accuracies of the satellite SST's, and we hope this information will be useful to improve the SLSTR atmospheric correction algorithms.