干旱区生态环境知识资源中心

The present study deals with the estimation of the advanced spaceborne thermal emission and reflection radiometer (ASTER) data for geological application. Visible, near-infrared and short wave infrared reflectance data (9 ASTER bands) have been processed and interpreted in framework of a mapping project concerning the western margin of the Kalahari desert. A methodology has been applied to apparent reflectance images corrected to illumination effects. A Principal Component Analysis (PCA) has been realised on the 9 ASTER bands in order to reduce the redundancy information in highly correlated bands. PCA results enable the validation and the revision of the lithological boundaries defined on previous geological map and provide information for characterising new lithological units corresponding to surficial formations previously unrecognised. A supervised classification realised from PCA results and based on the geological map shows that lithological units may have similar spectral signature, a strong spectral variability or a spectral homogeneity. The processing of ASTER remote sensing data set can thus be used as a powerful tool for geological mapping. (C) 2004 Elsevier Ltd. All rights reserved.