The European Space Agency (ESA) ERS-1 and ERS-2 C band VV polarization active microwave instrument (AMI) offers the unique ability to combine interlaced wind scatterometer and high-resolution synthetic aperture radar (SAR) wave mode imagettes. In this study, more than 2000 imagettes were considered. Each imagette has been statistically analyzed in comparison with normalized radar cross-section (NRCS) measurements from the scatterometer mode. During the ERS-1 AMI wave mode mission the incidence angle of the imagette center was modified from roughly 19.9¿ to 23.5¿. Using ERS-1 and ERS-2 NRCS, calibration has thus been completed for both configurations, which allows a better characterization of some signal data saturation effects. However, since a SAR relies on platform displacement to achieve fine resolution, surface motions reduce its nominal resolution. As the wind speed increases, scatterer motion occurring during the SAR integration time also increases, causing the characteristic large azimuth cutoff wavelength. Acknowledging the strong dependence between sea state conditions and azimuth smearing effects, our results are highlighted by the global comparison with wind estimates derived from the scatterometer. The results demonstrate the ability to define a SAR wind algorithm from a kinematic point of view. Finally, a higher-order statistical analysis shows evidence of deviation from standard Rayleigh statistics, leading to a balance between K law and lognormal distributions. This deviation is mainly due to the SAR's high-resolution properties. ¿ 1998 American Geophysical Union