[20-Feb-2024] Jarugula, S., Fournier, S., Reager, J.T., and Pasoclini-Campbell, M.
Presented at the 2024 Ocean Sciences MeetingObserving the global ocean sea surface salinity (SSS) from remote sensing satellites over the past decade to an accuracy of 0.2 pss has advanced our understanding of the salinity variability in response to the hydrological cycle, ocean circulation, land-sea exchanges, air-sea interaction. It remains a huge challenge to obtain accurate satellite SSS measurements close to the coast owing to land-contamination issues. Our study uses more than two million World Ocean Database (WOD) salinity profiles along with gridded Argo data in the vicinity of 40-500 km from the global coastline to perform an intercomparison of the existing satellite SSS products: NASA's Multi-Mission Optimally Interpolated Sea Surface Salinity (OISSS), NASAs' Soil Moisture Active Passive (SMAP) from by Remote Sensing Systems (REMSS) and Jet Propulsion Laboratory (JPL), ESA's Soil Moisture Ocean Salinity (SMOS) and Climate Change Initiative (CCI) salinity. We find that the mean bias (0-2 pss), root-mean-squared error (0.5-4 pss) and signal-to-noise ratio (1-3.8) in satellite SSS decrease exponentially from 40-100 km distance to the coast and the curve flattens past 100 km, with OISSS, CCI and SMOS having minimum error compared to other products. Using empirical orthogonal function (EOF) analysis, we find that the satellite products capture modes-1 and 2 seasonal variability very well but have notable differences in reproducing the mode-3 interannual variability in coastal SSS, especially over the Maritime continent and at major river mouths like Congo, Ganga-Brahmaputra. Remote sensing of SSS close to river mouths needs improvement to better understand the hydrological cycle and coastal processes.