Measuring sea surface salinity from an airborne SAR in the Gironde region, France

Radar is being developed in many work fields essentially in coastal areas and mounted on satellites (ERS, Seasat). Only a few experiments have been made these last decades in order to study sea surface salinity from microwave radiation (Miller et al., 1998). It has become a real key in coastal area remote sensing. The tool we are working with is a P-band airborne synthetic aperture radar (SAR) called RAMSES made by the ONERA (Paillou et al., 2001). It is a multiband frequency radar which works for all four-polarizations. The active P-band challenge is that such an airborne radar has never been experimented before this day for the study of salinity. This is particularly surprising because of its rising sensibility with salinity gradient. The major aim is to discuss how we can discern bathymetry effects in the Gironde mouth from salinity response into the radar signal. Adding to the radar data, simulations of salinity fields (MARS 3D model from IFREMER) over the Atlantic platform were run for this date contributing to ensure what we suppose with the radar signal. Then we modeled the intensity of the radar (σ₀) found on a dielectric constant model (Klein & Swift, 1977) and on Bragg diffusion theory from sea surface. The model simulates a variability of less than 1 dB with the signal intensity for a 1% gradient considering a P-band radar and three time less for a L-band one. This range is ten times less than what is observed from the data. From a qualitative point of view, the intensity of the retrodiffusion globally increases along the saline gradient from low to high salinities. Moreover, the VV polarization is getting more dependent to salinity with higher incidence angles. The study of the saline plume with chlorophyll and suspended matter has permitted to make coincided the 10 to 20 dB drop in the signal with the turbidity limit and phytoplanktonic bloom.