Intercomparison of In Situ and Satellite Sea Surface Salinity Products for Global Coastal Ocean Studies

Abstract Observing coastal ocean salinity is crucial for understanding the physical and biological processes impacting ocean circulation, ecosystems, and the hydrological cycle. We perform a thorough comparison between in situ and satellite sea surface salinity (SSS) products with a focus on global coastal ocean. Between 40 and 500 km from the coastline, we intercompare SSS from more than two million World Ocean Database (WOD) salinity profiles, gridded Argo data, and satellite products: NASA Multimission Optimally Interpolated Sea Surface Salinity (OISSS), 40- and 70-km Remote Sensing Systems (REMSS) Soil Moisture Active Passive (SMAP), Jet Propulsion Laboratory (JPL) SMAP, Laboratory of Oceanography and Climate: Experiments and Numerical Approaches (LOCEAN) Soil Moisture Ocean Salinity (SMOS), and European Space Agency (ESA) Climate Change Initiative (CCI) salinity. We find good agreement between the satellite products and in situ data with satellite-in situ mean differences (0.5–1.7 pss, where pss is the Practical Salinity Scale), root-mean-square differences (0.5–2.4 pss), and signal-to-noise ratios (1–3.8) generally decreasing at a faster rate from 40- to 100-km distance to the coast and remaining constant past 100 km. NASA OISSS, ESA CCI, 70-km REMSS SMAP, and LOCEAN SMOS have minimum differences with in situ observations. Empirical orthogonal function (EOF) analysis shows that all the satellite and in situ SSS compare well in capturing mode-1 and mode-2 seasonal variability but have notable differences in the amplitude of the mode-3 semiannual variability in coastal SSS. This study has implications for validation and improvement of satellite SSS and emphasizes the need for designing future satellite missions to better resolve the coastal salinity.