SAR imagery analysis of vegetation in desert environment

Orbital synthetic aperture radar (SAR) C-band data acquired by ERS-1/2 in vv-polarization and Radarsat in hh-polarization during the period from 1996 to 1999 were used to evaluate their combined information potential for classification of land cover in the arid environment of Kuwait. Individual SAR scenes were orthorectified using a digital elevation model (DEM) of Kuwait, radiometrically adjusted for incidence angle effects, and mosaics were generated for the whole country. Thirteen classes of the joint ERS-1/2 and Radarsat images were identified based on Bhattacharya distance and geospatial pattern. The high degree of correlation between the C-band radar backscatter observed by ERS and Radarsat made unambiguous classification of surface material difficult when using C-band data alone. However, the physical data collected at the ground verification sites were used to assign nominal categories to the radar clusters that resulted from unsupervised classification. This categorization or relabeling of the clusters then provided the basis both for the generation of thematic maps and for accuracy assessment. Backscatter is shown to be related to the percent cover by annual vegetation for ERS, while its influence appears to be somewhat less for Radarsat. Backscatter was observed to be positively related to the percent cover by perennial vegetation. Radar backscatter is more highly correlated with total vegetation volume for ERS than for Radarsat, and is probably a consequence of both the lower angle of incidence for ERS and its vv-polarization. In addition to vegetation cover, the C-band radar data were found to be related to surface roughness, percentage of coarse material in the surface layer, and moisture conditions.