Radar backscatter radiance has considerable potential for discrimination 

 among soil and rock types, and geobotanical features. Topographical effects 

 are a confusion factor for this application so that intermediate-look angles 

 (30°-35°) are preferred. Theory and field studies highlight the importance 

 for discrimination based upon backscatter radiance of acquiring both like- and 

 cross-polarized data. Radar backscatter radiance varies with surface geometry 

 and moisture content, while infrared reflectance varies primarily with surface 

 chemistry. The essential independence of these two processes suggests that 

 radar and infrared reflectances should be combined for multicomponent 

 analyses. The experiment would be further enhanced by a second radar 

 wavelength to permit microwave as well as infrared spectral discrimination. 



B. KEY RADAR PARAMETER RESEARCH ISSUES 



A mission requirements specification for a SAR satellite must include 

 the desirable f requency(ies ), angle(s) of incidence, polar iza tion( s ), 

 resolution(s ), number of looks, and revisit interval(s). Other radar 

 parameters of particular importance to the geologist include swath width, 

 calibration, dynamic range, registration, and multiple looks. 



In order to specify these parameters for a meaningful satellite radar 

 geology experiment, the following research issues must be addressed: 



1. Sensitivity to topography, vs. frequency, polarization, resolution, 

 and angle of incidence. 



2. Sensitivity to surface roughness and vegetation cover , vs. 

 frequency, polarization, resolution, and angle of incidence. 



3. Sensitivity to soil moisture , vs. frequency, resolution, and angle 

 of incidence. 



It is stressed that these issues can only be addressed with high quality 

 (calibrated and registered) multiparameter SAR imagery over wide swaths. From 

 a practical viewpoint, some of this work can be done using airborne 

 multiparameter SARs and, indeed, specific experiments can be proposed to 

 utilize airborne SAR data. But even the best airborne SAR data suffers from 

 a wide variation in incidence angle over the swath width so that suturing 10- 

 20-km wide images to form a 100-km mosaic presents formidable problems when 

 large-swath regional context images are needed. This serious angle-dependence 

 of airborne SAR data means that only space-borne SAR data over 75-150-km swath 

 widths, with a relatively constant angle of incidence, are adequate to address 

 the utility of SAR for regional geologic mapping applications. 



3-2 



