Abstract: Data acquisition and inventory of landslide dams are the foundation of their regional research. Himalaya is located in the southwest of China. A number of landslide dams occurred here and brought serious property losses, as well as undesirable international influence. This paper is based on Remote Sensing and Geographic Information technologies. It obtains 136 landslide dams. Their spatial location, basic attributes and geometry features are obtained by RS image information extraction and field investigation. Then the landslide dams inventory is finished. It is found that the landslide dams are mainly caused by landslide, collapse and debris flow. The dams have the regional feature of concentrating distribution among Milin, Zhada, Jiacha, Cuona, Longzi and Longxian. Calculating information values of influence factors, it is found that landslide dam susceptibility increases at first, then decreases with the increase of elevation, slope and seismic acceleration. It decreases at first, then increases with the increase of seismic dot density. It also increases with the increase of tectonic line density, and decreases with the increase of distance to river. West slope, high mountain, Langxian tectonic melange and hard rock of lithology, High Himalaya tectonic region and Yarlu Zangbo tectonic region are the favorable factors of landslide dam formation. In view to the information values, among all the influence factors, elevation, geomorphic type, lithological characteristics and seismic dot density are more closely related to landslide dam susceptibility. As to the geometry features, there is a typical power function relation between dam length and dam area, as well as landslide area and dam area. Nevertheless, the relations between other geometry parameters are not striking.