Radar Reflection and Scattering Properties from Geologic Surfaces

Planetary and terrestrial surfaces are commonly examined through the use of various radar imaging techniques, sounders and reflectivity measurements. Through these measurements we can derive a better understanding of specific geologic surface processes both visually and mathematically. Using field and laboratory observations we seek to statistically infer geologic properties of rough surfaces using radar reflectivity measurements and accurate topographic data acquired from airborne LIDAR. Well established radar theory for slightly rough and random surfaces predicts a transition from a highly peaked Gaussian distribution of normal incidence backscatter amplitude for smooth surfaces, to a broader, Rayleigh distribution as roughness approaches a significant fraction of a wavelength. This transition defines changes from reflection to scattering as a surface character alters. However, the RMS height variances of most naturally occurring surfaces are considerably greater than the standard microwave wavelengths used for most contemporary Synthetic Aperture Radar (SAR) imaging systems. Our field work was conducted at a remote watershed within California's Mojave Desert where we utilized an elevated, 1.5-GHz FMCW radar to measure backscatter at normal incidence over a rough boulder surface. In addition, we used NASA's Airborne Topographic Mapper (ATM)-LIDAR and field measurements to characterize surface slopes and height variances, and GPR to measure ground dielectric permittivity. Our lab work utilizes 100-GHz Doppler FMCW radar to measure forward scatter and backscatter at low incidence angles from various surfaces on a rotating table. All of our rough backscatter cases show Rayleigh distributions, while many of our forward scatter cases are Gaussian despite the extreme roughness. In the latter case however, the distributions are not sharp, in contradiction to the specular reflections commonly believed for grazing angle incidence. Our preliminary conclusions show that it may be possible to obtain surface roughness characteristics from radar measurements and through this technique it may be possible to backwards correlate specific surface processes to produce fidelity maps of surface terrain based on these statistics.