Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Civil and Environmental Engineering

First Advisor

Roy K. Dokka


This dissertation describes the results of two projects that demonstrate the power of band ratioing of remote sensing data. The first method, Band Ratio Differencing (BRD), challenges the more traditional techniques for detecting environmental disturbance of the surface in and terrains. These traditional techniques are inadequate because they cannot identify the nature of the change (physical, chemical, biological) or the processes (natural, anthropogenic) involved in causing the change. In contrast, BRD of Landsat TM data has the advantage of providing information about changes in the composition of alluvial surfaces. These compositional changes, when combined with empirical field data provides insight into the specific causes and effects of environmental change. BRD was applied at Fort Irwin, California, the U.S. Army's National Training Center (NTC) for desert warfare. Two dates of Landsat TM imagery (1985 and 1995) were acquired and BRD applied. A total of 514.5 square kilometers, or 42% of the alluvial surface at Fort Irwin has been affected by off road vehicular traffic. Two major types of change were identified: Disturbed Alluvial Surface and Continuous Dust Mantle, which can be further subdivided into six separate classes of change, related to either the cause or the effect of the disturbance. Hyperspectral Assisted Mapping (HAM) was developed as an alternative to Spectral Signature Mapping (SSM) techniques typically applied to AVIRIS hyperspectral data. Our analysis suggests that SSM-based techniques when applied to AVIRIS provide little to no additional information about sedimentary and igneous rocks. This is because many rock forming minerals of sedimentary and igneous rocks possess little to no diagnostic spectral absorption features in the range measured by the AVIRIS sensor. The HAM method integrates the geornorphic information provided by a Digital Elevation Model (DEM) and the lithologic discrimination, but not identification, provided by hyperspectral data to produce a surface materials map. HAM was successfully applied at the North Alvord Slope study area, on the southern margin of Fort Irwin, California. Nine distinct surface materials classes were identified by the technique. All nine classes were discriminated on the basis of minor mineral constituents in the materials, such as micas and clays.