Contents I. Introduction
Sea surface wind speed is a crucial parameter in the studies of many oceanic applications, i.e. sea mapping, meteorological forecasting, oil slick observation, ship detection, wind source assessment, etc. Among the different available wind sources (i.e. measured data from meteorological buoys, numeric weather model data, etc.), the retrieval of wind speed from Synthetic Aperture Radar (SAR) data is particularly preferred due to many interests of SAR systems such as stable operations in most meteorological conditions, revisit period, and high resolution. In fact, wind speed retrieval was performed from the data of different SAR systems [1]–[2], i.e. JERS-1, Envisat, Radarsat-1/2, TerraSAR-X, and Cosmo-SkyMed. The common point of these studies is the use of empirical (EP) models to estimate wind speed. In such models, the dependency of radar backscattering on wind speed and the geometry of observations, i.e. wind direction relative to radar look, incident angle, is described by the Geophysical Model Functions (GMFs) [3]–[6]. Such GMFs have been constructed and validated by the series of satellite scatterometer missions. For instance, the C-band GMFs, known as CMOD.4 [3], CMOD.IFR2 [4], CMOD.5 [5], and CMOD.5N [6], were derived using the data from scatterometers onboard the European Remote Sensing 1 and 2 satellites (ERS-1/2).
