Integration of quantitative geophysical information from optical and radar remotely sensed data to characterize mineralogy and morphology of surfaces

The integration of geophysical information from remotely sensed optical and radar data provided a more complete understanding of geologic surfaces in Death Valley, California. Geophysical inversion models were used to extract quantitative information about surface mineralogy from the optical data and morphology from the radar data. Radar inversion methods allowed for the calculation of rms surface roughness, fractal dimensions, and dielectric constants. The fractal dimension and surface roughness were used to generate synthetic surfaces that represent the morphology of field sites. Frequency-polarization data cubes were generated from the radar data and used in classification techniques to produce maps of surface backscatter variation. Spectral unmixing techniques were used to produce mineral abundance maps from the optical data. Water vapor and surface liquid water maps were made from the optical data using atmospheric models.; The optical and radar data were used synergistically to study two sites in Death Valley in detail. Integrating the inversion results from the radar and optical data for an alluvial fan site allowed for a better understanding of the surface morphological variations. The data integration showed that the fan morphology was affected by the mineralogy of the surface. A more accurate geologic map was produced and the relative ages of the fan surfaces were determined. A better understanding of the hydrology and morphology of a portion of the salt flats in Death Valley was obtained when the optical and radar data inversion results were used synergistically. The salt flats showed morphological and mineralogical variations that were caused by surface washing and flooding. A more complete understanding of geologic surfaces was obtained by integrating optical and radar data than could be obtained using one data type alone. The integration of diverse data types should be the future emphasis in remote sensing studies.