A comparison of reconstructed Ku-band scatterometry, C-band scatterometry, and SSM/I imagery for tropical vegetation classification

A new multivariate image reconstruction technique (Long et al., 1993) has improved the resolution of imagery produced by SASS (Seasat A scatterometer), the ERS-1 scatterometer (Active Microwave Instrument), and SSM/I to resolutions between 5 and 20 km, making them candidates for medium-scale monitoring of cloudy global areas. Furthermore, successful efforts have been made to remove atmospheric distortion in the SSM/I data (Daum, 1994). One goal of the research reported in this paper is to combine these reconstructed image data sets with AVHRR vegetation index images for discriminating between vegetation formations in central South America. This also provides an opportunity to compare the utility of the various image types for equatorial vegetation discrimination in general.