Soil Moisture Ocean Salinity (SMOS) salinity data validation over Malaysia coastal water

The study of sea surface salinity (SSS) plays an important role in the marine ecosystem, estimation of global ocean circulation and observation of fisheries, aquaculture, coral reef and sea grass habitats. The new challenge of SSS estimation is to exploit the ocean surface brightness temperature (Tb) observed by the Microwave Imaging Radiometer with Aperture Synthesis (MIRAS) onboard the Soil Moisture Ocean Salinity (SMOS) satellite that is specifically designed to provide the best retrieval of ocean salinity and soil moisture using the L band of 1.4 GHz radiometer. Tb observed by radiometer is basically a function of the dielectric constant, sea surface temperature (SST), wind speed (U), incidence angle, polarization and SSS. Though, the SSS estimation is an ill-posed inversion problem as the relationship between the Tb and SSS is non-linear function. Objective of this study is to validate the SMOS SSS estimates with the ground-truth over the Malaysia coastal water. The LM iteratively determines the SSS of SMOS by the reduction of the sum of squared errors between Tb SMOS and Tb simulation (using in-situ) based on the updated geophysical triplet in the direction of the minimum of the cost function. The minimum cost function is compared to the desired threshold at each iteration and this recursive least square process updates the SST, U and SSS until the cost function converged. The designed LM's non-linear inversion algorithm simultaneously estimates SST, U and SSS and thus, map of SSS over Malaysia coastal water is produced from the regression model and accuracy assessment between the SMOS and in-situ retrieved SSS. This study found a good agreement in the validation with R square of 0.9 and the RMSE of 0.4. It is concluded that the non-linear inversion method is effective and practical to extract SMOS SSS, U and SST simultaneously.