Development of hazard map for deep-seated landslide using stochastic response surface method

This paper presents a numerical procedure to develop a hazard map for a deep-seated landslide using the stochastic response surface method considering the uncertainty of strength parameters. The following conclusions are drawn from this study; (1) Based on the parametric study of 3D slope stability analysis, the cohesion of 110kPa and the frictional angle of 35^o are estimated for the deep-seated landslide at Tateno area in Kumamoto prefecture caused by the 2016 Kumamoto earthquake, Japan. (2) The failure probability of seep-seated landslide for the foreshock and main shock of the 2016 Kumamoto earthquake are estimated to be 30% and almost 100% respectively. (3) Under the situation having the uncertainty of material parameters, the stochastic response surface method using the polynomial chaos expansion is useful for developing a hazard map for the failure probability and the arrival risk of a large-scale landslide such as a deep-seated landslide.