Abstract: Hydrothermal deposits are often accompanied by a certain type of mineralized alteration, which can indicate potential favorable metallogenic areas. The Heishan-Xianquanzi deep fault at Liuyuan area in Gansu Province can be regarded as one of the best demonstration areas for alteration information extraction based on remote sensing data because of its numerous types of hydrothermal deposits, abundant alteration, and well exposed bedrocks. In this paper, with the ASTER images, principal component analysis(PCA) was used to extract the anomalous information of iron-stain, hydroxyl alteration and carbonate, and the quantitative inversion of SiO₂ content of surface rock was used to extract silicified alteration. Comparing with the existing geological data, the field verification shows that the extracted information is in good agreement with the actual situation,and the relevant factor reaches 88.9%. Based on the known mineralization in the mining area, regional structure, remote sensing alteration information and rock distribution, two favorable prospecting areas were successfully delineated, which provides great support for exploration in this area. This study proves the reliability of the alteration anomaly extraction by the method of PCA and the quantitative inversion of SiO₂ content based on ASTER data, which can be extended to the alteration information extraction of similar areas.