Retrieval of Sea Surface Salinity from The NASA Soil Moisture Active Passive and Aquarius Mission Data
[16-Dec-2016] Yueh, S.H., Fore, A., Tang, W., and Hayashi, A.
Presented at the 2016 AGU Fall Meeting
NASA's Soil Moisture Active Passive (SMAP) mission was launched in January 2015 to provide global mapping of soil moisture. SMAP has two instruments, a polarimetric radiometer and a multi-polarization synthetic aperture radar. The radar stopped operation on 7 July 2015. Both instruments operate at L-band frequencies and share a single 6-m rotating mesh antenna, producing a fixed incidence angle conical scan at 40° across a 1000-km swath. We have analyzed all available SMAP and Aquarius data to improve the geophysical model functions, relating the L-band radar and radiometer data to ocean surface wind speed, wind direction, significant wave height and sea surface temperature. We find that it is necessary to account for the fourth order harmonics for wind direction effect and SST influence on sea surface scattering. The SMAP SSS retrieval algorithm developed at the Jet Propulsion Laboratory leverages the QuikSCAT and Aquarius algorithms to account for SMAP's two-look geometry for retrieval of SSS and wind speed. The retrieval algorithm has been applied to more than one year of SMAP radiometer data. We have also applied the Combined Active Passive (CAP) algorithm to about three months of SMAP data from April to early July 2015. The spatial patterns of the SMAP SSS agree well with climatological distributions, but exhibit several unique spatial and temporal features. The SMAP SSS reveals the temporal evolutions of freshwater plumes from several major rivers, consistent with the timing of rainy and dry seasons, indicated in the SMAP's soil moisture and Global Precipitation Missions's daily rain products. Accuracy assessment has been performed by comparison with in situ SSS data from buoys and ARGO floats. The accuracy of monthly averaged SMAP product is about 0.2 psu for tropics and mid-latitudes. The improved roughness model function has also been applied to the Aquarius data, resulting in reduction in standard deviation errors and seasonal bias at high latitudes.