Present remote sensing estimates of sea surface salinity are based on L-band measurements from the Soil Moisture and Ocean Salinity (SMOS) and the Aquarius missions. Ocean remote sensing at L-band is challenging as the signal is affected by many external noise sources and processing issues, notably for SMOS (e.g., radio-frequency-interference sources, Land-sea contamination, latitudinal biases, etc.), that undermine the quality of the final products. Direct comparisons with in situ measurements provide information on point-wise deviations in remote sensing sea surface salinity (SSS) maps, but not on their geophysical consistence. A method to characterize geophysical structures in SSS maps in a systematic way by means of the singularity analysis technique is proposed. Singularity analysis is very sensitive to artifacts and to correlated noise, while filtering uncorrelated noise and revealing coherent structures in ocean maps of different variables. The method allows not only to obtain a qualitative assessment of the quality of SSS maps but also to quantify their closeness to an appropriate template of geophysical structures (e.g., derived from accurate high-level SST maps). The method is used both to improve the L-band processing algorithms and to perform oceanographic process studies. For example, the method is able to assess the quality associated with different ways of processing SMOS data. In particular, the distribution of SMOS measurement errors is shown to be strongly non-Gaussian and hence non-linear filtering must be applied to optimize the quality of SSS maps. Singularity-based quality assessment shows that the new SSS mapscontain geophysical coherent structures (mainly eddies and filaments), which are not present in the corresponding climatological maps and less evident in the standard SSS products. Moreover, the latter contain artifacts not seen in the new product. The new SSS maps can then be used to assess the presence of haline fronts, the position of temperature fronts, and the propagation of tropical instability waves, as well as to track strong river discharges in the ocean. These maps will also be used to discuss the evolution of the currently expected 2014 El Niño event

SMOS+SOS Ocean Salinity Science and Salinity Remote Sensing Workshop, 26-28 november 2014, Exeter, United Kingdom.-- 1 page

Peer Reviewed