Development of wind speed retrieval model using RISAT-1 SAR cross-polarized observations

In this study, a method for retrieving ocean surface wind speed using C-band cross-polarization SAR observations has been outlined. A linear least square technique has been used to develop a Geophysical Model Function (GMF), C2P. The GMF was derived using NRCS observations from RISAT-1 and wind-speed observations from ASCAT. The correlation between observed and simulated NRCS values obtained from C2P was 0.66, with a negative bias of 0.01 dB and the corresponding root mean square difference of 1.13 dB. Subsequently, the developed GMF was tested with 774 RISAT-1 MRS datasets to retrieve wind speed along the Indian coast and also of the tropical cyclone ‘Megh’. The measured intensity and radius of maximum wind speed were 30 m s−1 and 16.65 km respectively. Subsequently, the retrieved wind speed was validated with ASCAT wind-speed observations. The statistical comparison of RISAT-1 and ASCAT observed wind speed showed negative biases of 0.90 and 0.34 m s−1 with the corresponding RMSD of 2.11 and 1.77 m s−1 respectively, for CMOD5.N and C2P. The developed GMF C2P showed 16% more accuracy than that of CMOD5.N.