Application of geographic detector in identifying influencing factors of landslide stability: A case study of the Jiangda County, Tibet

The high mountain and gorge area is an typically area, where geological disasters happen frequently. Especially landslide is one of the most serious geological disasters. Recently relative researches on geological disasters showed that landslides had an increasing trends due to the impacts of both climate change and human activities. In this study, Jiangda County in Tibet Autonomous region was selected as our study area, which located in the high mountain and gorge area of the Hengduan Mountain Region. In addition, using the landslide data for 85 sites based on field survey, choosing Slope, River density, Geomorphic type, Precipitation, the distance from the fault, Road density, the ground motion peak acceleration and Lithology as 8 influencing factors on landslide, and then employing the Geodetector model to analyze the impact of various variables on landslide stability. The results showed that 1) according to the volume of landslide, medium and small landslides are main types in Jiangda County; According to its stability, more than 50% of the landslide is in a stable state; according to the danger level, they are mainly divided into Ⅲ, Ⅳ. In space, it is mainly along rivers and roads in Jiangda County, which caused by the limitations of the field survey besides physical factors. Because all 85 landslide survey sites located near rivers, and more than 71% sites are distributed on both sides of the road. 2) The geomorphic type, River density, Road density and the distance from the fault are major factors to affect the stability of the landslide in Jiangda County, its contribution rate are 0.501, 0.477, 0.465 and 0.332, respectively. When the influence factors interact in pairs, the explanatory power of factors is always greater than that of a single factor to the stability of landslides. In other words, when the two influencing factors interact, they always promote the instability of landslides.