Wind-vector estimation for RADARSAT-1 SAR images: Validation of wind-direction estimates based upon geometry diversity

In this letter, a new wind-vector algorithm is presented that uses radar backscatter σ^o measurements at two adjacent subscenes of RADARSAT-1 synthetic aperture radar (SAR) images, with each subscene having slightly different geometry. Resultant wind vectors are validated using in situ buoy measurements and compared with wind vectors determined from a hybrid wind-retrieval model using wind directions determined by spectral analysis of wind-induced image streaks and observed by colocated QuikSCAT measurements. The hybrid wind-retrieval model consists of CMOD-IFR2 [applicable to C-band vertical-vertical (VV) polarization] and a C-band copolarization ratio according to Kirchhoff scattering. The new algorithm displays improved skill in wind-vector estimation for RADARSAT-1 SAR data when compared to conventional wind-retrieval methodology. In addition, unlike conventional methods, the present method is applicable to RADARSAT-1 images both with and without visible streaks. However, this method requires ancillary data such as buoy measurements to resolve the ambiguity in retrieved wind direction.