Journal of the Korean Society of Mineral and Energy Resources Engineers ISSN:2288-0291(Print) 2288-2790(Online) 한국자원공학회지

Rainfall is one of the main factors which induce landslides. In this study, logistic regression analysis was conducted to quantify the dependency of landslide occurrences on rainfall as a trigger by using rainfall data and landslide event records during the last seven years. This method employed daily maximum rainfall and cumulative rainfall for three days as explanatory variables. Through this procedure, the relative weights of the degree of influence in inducing landslides were determined. Furthermore, the probabilities that might be occurred by expected rainfalls were derived. According to the results, daily maximum rainfall is more influential to landslide occurrences than 3-day-cumulative rainfall. But, landslide occurrence data include the only events that were reported in the main mass media, which means landslide induced the damage of social system or the injury of human lives. Despite of this limitation, the results can make useful linkage with the landslide forecasting prediction system. Furthermore they showed the potential that additional explanatory variables and high-quality data acquisitions enable quantitative result to be basis for landslide hazard mapping by using computer-based tools, such as GIS.