INFERRING ROUGH-SURFACE PARAMETERS FROM AVERAGE SCATTERING DATA USINGAPPROXIMATE SCATTERING MODELS 1 - GAUSSIAN SPECTRUM

The quasi-specular and two-scale models for the average radar cross se ction of randomly rough surfaces have seen extensive use in inversion algorithms to estimate surface roughness parameters. This paper and a companion paper report the results of a numerical study on the accurac y of these models which was initiated with the god of understanding wh at scattering mechanisms limit their applicability. In this paper, sur faces which are Gaussian distributed in height and have Gaussian spect ra are examined. It is shown that the. quasi-specular model is limited by multiple scattering and diffraction due to surface curvature. The effects of polarization and the role of the surface dielectric constan t are examined at length. Approximate conditions for which the quasi-s pecular model is valid are provided.