The Soil Moisture and Ocean Salinity (SMOS) satellite mission has been measuring sea surface salinity (SSS) for over 6 years with about 5-day global ocean coverage and a spatial resolution of about 50 km. In rainy regions, at local and short time scales, the spatio-temporal variability of SSS is dominated by rainfall. The relationship between surface freshening and rain rate (RR) has been highlighted in the Pacific intertropical convergence zone (Boutin et al., 2014). In this context, this study investigates the rainfall characteristics that may be inferred from SMOS SSS based on statistical approach. Salinity anomalies associated with rainfall events are first estimated. In order to do so, a reference salinity (i.e. with no rain-induced signal) is computed for each pixel of the SMOS observation using the statistical distribution within 3°x3° region of SMOS SSS. In case the distribution is asymmetrical toward low values, suggesting a rain influence, a mean ‘non-rainy’ SSS corresponding to a Gaussian distribution fitted onto the highest part of the distribution (quantile>0.8) is computed. Rain rate probability associated with SSS anomalies are then inferred from a probabilistic approach. It also enables us to separate the rain intensity depending on the SSS anomaly. Finally, a RR retrieval algorithm based on SSS is developed combining this dependence with the SSS-RR relationship described in Boutin et al. (2014) and a spatial association index (spatial correlations of SSS anomalies within 100 km). SMOS-derived RRs are then collocated with various radiometers and CMORPH RR datasets. Their consistency is assessed. A particular focus will be put on RRs estimates derived during the Salinity Processes in the Upper Ocean Regional Study (SPURS-2, http://spurs2.jpl.nasa.gov) from a near real time implementation of rain retrieval from SMOS SSS. Boutin et al. (2014), Sea surface salinity under rain cells: SMOS satellite and in situ drifters observations, JGR: Oceans, doi:10.1002/2014JC010070