Pou ití logistické regrese pro hodnocení náchylnosti k sesouvání ve st ední ásti Ch ib ( eská republika)

The paper presents use of logistic regression and analysis tools implemented in the Geographic Information System (GIS) environment for landslide susceptibility assessment. As a case study, the middle part of the Chriby Mountains (Czech Republic) widely affected by landslides was selected, with a total area of 147 km2. The landslide susceptibility assessment is based on the comparison and statistical elaboration of territorial dependence on input factors influencing the instability of the slopes. In this study, the lithology, slope angle, distance to stream, hypsographic level, and the current land cover were evaluated as the factors influencing slope stability and their susceptibility to sliding. The comparison shows that the high susceptibility to sliding is primarily in the areas of potential landslides and active, while 62 % of existing slope failures are created in areas highly susceptible to sliding, 32 % are in areas susceptible to moderate sliding. The remaining 6 % of the landslides are located in low areas prone to sliding. This may be the accumulation parts of landslide bodies. An overall assessment of the 33 % of the area can be classified as having a high degree of susceptibility to the sliding surface and 35 % as moderately susceptible to landsliding. Low degree of susceptibility to sliding represents the remaining 32 % area of the territory.