A NEW METHOD FOR MAPPING SURFACE ROUGHNESS ON ALLUVIAL FANS USING LANDSAT SATELLITE IMAGERY AND THE BIDIRECTIONAL REFLECTANCE DISTRIBUTION FUNCTION (BRDF)

The Bidirectional Reflectance Distribution Function (BRDF) is used to describe changes in reflected light from a surface based on changes in the angle of incoming light and differences in surface roughness. Using Landsat 7 satellite imagery, the changes in observed surface reflectance, resulting from seasonal changes in the angle of incoming solar radiation, can be classified and interpreted to show differences in surface roughness. This approach has been tested using imagery of late Holocene alluvial fan surfaces in Death Valley National Park, California. Field observations of surface roughness and other variables affecting reflectance (eg. color, composition) have been used to test the hypothesis that the remote sensing classifications of seasonal changes in reflectance can be used to map surface roughness. Statistical tests show that the total amount of sand found on the land surface is the most correlated variable with remote sensing class. Other variables, such as vegetation type, vegetation amount, and varnish, are shown to be directly related to total sand amounts. Classes found near alluvial fan toes have a higher sand percentage, are smoother, and have some coverage by vegetation (up to 15%). Coarser clast dominated surfaces are described by classes that have less vegetation and are more likely to have desert varnish coatings. Aerial LiDAR data of some alluvial fans in Death Valley are also used to map surface roughness using the methods of Frankel and Dolan (2007). The results of LiDAR surface roughness mapping provide another means to evaluate the Landsat imagery surface roughness classifications. Mapping surface roughness over large areas and in remote settings using visible satellite imagery has the potential to be a powerful and inexpensive tool for studying the geomorphology of both Earth and Mars.