An algorithm for the retrieval of oceanic wind vectors from the simulated SASS normalised radar cross-section measurements

Abstract

A new algorithm has been proposed to recover wind vectors from the simulated Seasat A satellite scatterometer (SASS) normalized radar cross-section (NRCS) measurements exploiting the fact that all the dual polarized noise-free NRCS measurements taken in any azimuth direction should provide the same wind speed in the true direction of wind. The algorithm deals with the computation of wind speeds from SASS NRCS measurements for the entire range of wind direction. The direction for which the standard deviation of wind speeds is minimum, is considered to be the true direction, and the corresponding average wind speed as true speed. The performance of the algorithm is evaluated and compared with the officially adopted SASS-I algorithm (SA) under noise-free and noise (standard deviation of 0.13 to 0.40 dB with zero mean) added conditions for wind speeds ranging from 3 to 24 m/s and for all directions from 0° to 360°. When the direction corresponding to the lowest minima of standard deviation of wind speed is taken as true wind direction (case 1), an average improvement of 6.0±1.6% is obtained. When the direction corresponding to the two lowest minima of standard deviation of wind speed, which is closest to the true wind direction, is considered as true wind direction (case 2), an average improvement of 4.0 ± 1.3% is obtained.