| Knowledge of microwave scattering from the ocean surface is necessary for remote sensing of ocean surface winds, waves, and other ocean surface phenomena. Empirical models of microwave scatter from the ocean have been used successfully, but are limited to the ocean conditions and parameters observed. Here, we investigate the physics of microwave scattering from the ocean for its own sake and to understand the limitations of empirical models.;In this work we examine the validity of the aforementioned scattering approximations by comparing their predictions with calculations done using the method of moments, which is an exact numerical method. Next, a new large wavenumber spectrum for the ocean is proposed. The spectrum has a physical interpretation, and slope variances calculated using it are in good agreement with observations. Our moment method calculations show that the two-scale approximation should be accurate for an ocean described by this spectrum, and, in fact, cross section calculations do compare well with observations. Swell is included in the model, and we find that its effect on radar measurement of surface winds is minimized by use of large incidence angles, high frequencies, and vertical polarization. Finally, the theory of cross section modulation by long ocean waves is extended to the case of two-scale scattering. Calculations of the modulation transfer function show that this extension to two-scale scattering is primarily important for small incidence angles.;The physics of scattering from rough surfaces such as the ocean has been analyzed using the small perturbation, physical optics, and two-scale approximations, although their ranges of valid application have not been well established. The best previous calculations of scattering from the ocean were achieved using the two-scale approximation and an ocean surface described by a modified Pierson-Moskowitz spectrum at small wavenumbers and a spectrum found from wave tank observations at large wavenumbers. Agreement of calculated and observed cross sections is good, but surface slope variances calculated from the spectrum are much smaller than independently measured slope variances. |
