Variational retrieval of sea surface wind vectors using a polarimetric approach

Abstract

This study presents one-dimension variational retrieval of sea surface wind vectors using simulated microwave polarimetric measurements at 10.7, 18.7, and 37 GHz. The algorithm only requires four Stokes components at a single frequency; and therefore runs very efficiently. Under non-precipitation conditions, the errors of wind direction and speed are about 10° and 0.6 m/s, respectively, if the random noises of 0.1 K for vertical and horizontal polarization states and 0.15 K for the third and fourth Stokes components are taken into account. In hurricane conditions, polarimetric measurements at 10 GHz, especially for the 3rd and 4th Stokes components, vary within a few degree of Kelvin and provide unique information on surface wind direction.