Application of " Rare Events Logistic Regression" in Landslide Susceptibility Mapping

Landslide susceptibility maps are of high importance in identifying and separation of landslide-prone areas and safe places for future economic development of a region located in a country such as Iran with a diversity if seismic and geological and topographical conditions. In this study by investigated the landslides occurrence influencing factors, RELR statistical model was used to preparing a susceptibility map in Izeh-Baghmalek region. Because of the fact that the influencing area of occurrence of landslides is very small in compared to the total area of the region and in practice, we can consider them as points, the RELR statistical model was used. Landslide occurrence as dependent variable and geology, land use, vegetation, elevation, slope, aspect and distance from road, settlement, rivers, and faults and etc. were considered as the independent variables in the RELR model. The effect of each parameter was calculated using the model and the probability value obtained from RELR model are presented in as susceptibility maps. The quality of the landslide susceptibility map was evaluated and according to which, it can be used for planning, management and cost reduction planning in detailed landslide studies in the study area