A New Multi-Mission Sea Surface Salinity Optimum Interpolation (OISSS) Analysis for Ocean Research and Applications

Abstract

We introduce a new version of the multi-mission sea surface salinity (SSS) optimum interpolation analysis (OISSS) which combines observations from NASA’s AQUARIUS/SAC-D and SMAP (Soil Moisture Active-Passive) satellite missions into continuous and consistent SSS data record. The dataset covers the period from September 2011 to present. Measurements from ESA’s SMOS (Soil Moisture and Ocean Salinity) satellite are used to fill gaps in SMAP observations during June-July 2019 and August-September 2022, when the SMAP satellite was in a safe mode and did not deliver scientific data. The analysis is based on Optimum Interpolation (OI), utilizes Level-2 (swath) data, and uses satellite-specific bias-correction algorithms to correct the satellite retrievals for large-scale biases.  The dataset includes uncertainty estimates, both formal and empirical. We use this dataset as an example to discuss requirements for the multi-mission SSS data products.To demonstrate its utility, the new dataset is used to characterize spatial patterns of SSS variability in the global ocean and on different time scales. The spatial pattern of the regional SSS trends show that the subtropical North Pacific is becoming fresher while the subtropical South Indian Ocean is becoming saltier. This is seemingly a part of a longer term oscillation as the trends are reversed compared to the preceding decade (2005-2015) estimated from Argo data. In particular, abrupt changes occurred during 2015, related, presumably, to a strong El Nino event of 2015-2016. The annual cycle is a dominant signal globally and can nicely be described by two leading empirical orthogonal functions (EOFs) explaining more than 35% of the total SSS variance. Except for the Indian Ocean, the oscillations are out of phase in the Northern and Southern Hemispheres and describe poleward propagation away from the Equator driven, presumably, by Ekman dynamics. The intra-seasonal signal is strongest in the tropics, particularly in the quasi-zonal bands associated with the Inter-tropical convergence zone (ITCZ) and South Pacific convergence zone (SPCZ), but also near outflows of major rivers, including the Amazon, Congo, Mississippi, Plata, Ganges and Brahmaputra.  Another region of interest is the northern North Atlantic, where satellite observations during the last decade have provided an unprecedented resource to study the spatial distribution and temporal evolution of SSS, allowing to observe areas typically not available by in-situ components of the ocean observing system. Here, the multi-mission SSS dataset is examined in its accuracy and appropriateness for studying SSS variability in high latitudes and marginal seas.