Estimation of wind friction velocity and direction at the ocean surface from physical models and space-borne radar scatterometer measurements

Abstract

This paper presents a new technique for estimating the wind friction velocity at the ocean surface from C-band radar scatterometer measurements. This technique uses physical models of ocean surface waves and electromagnetic backscattering from a rough surface at intermediate angles of incidence to generate predictions of the normalized radar cross section (NRCS, or σ0) of the ocean surface for a given wind friction velocity and observational geometry. The ocean spectral model used in this technique has been developed specifically for this application. It combines in situ wave measurements at low wave numbers with the Phillips [1985] equilibrium spectral model. This choice of ocean wave model is supported by a set of open ocean wave measurements summarized in this paper. A suite of models, derived from both in situ and remote measurements of the sea surface, is used to characterize the directional spreading of ocean waves relative to the wind direction. The resulting two-dimensional ocean wave spectra are used with a composite surface model to predict radar backscattering from the ocean surface at C-band. These radar cross-section predictions are combined with ERS-1 scatterometer measurements in a cost function minimization scheme to yield estimates of the friction velocity vector at the ocean surface. We present examples of this technique and compare friction velocity retrievals obtained via this scheme with buoy-based measurements under a variety of wind and wave conditions. On the basis of the analysis of a limited number of cases, this technique yielded friction velocity estimates for which the magnitude was within 22% and the direction was within ±25°. Given that scientific applications require magnitude estimates within 10-15% and directional estimates within ±20° of in situ measurements, these preliminary results suggest that this is a promising approach to wind retrieval. Copyright 2001 by the American Geophysical Union.