Slope stability and failure dynamics of rainfall-induced landslide: Algorithm and applications

Abstract

Rainfall-induced landslides severely threaten to engineering facilities and the safety of people’s lives. However, accurately capturing the complex interactions between hydrology and geomaterials remains challenging, so further research on the mechanism and failure dynamics of rainfall-induced landslides can provide effective support for engineering design and disaster prevention. Due to the insensitivity to mesh distortion, the material point method (MPM) can effectively simulate the large deformation process of geomaterials, but it still faces limitations in computational efficiency and in handling large-scale simulations. In this study, based on the GPU-accelerated MPM software CoSim-MPM, the dynamic grid algorithm (DGA) is developed to overcome the limitations of calculation scale. Then the rainfall hydrological process in slope unit and evolution of mechanical behaviors of the geomaterials are coupled to analyze the influence of rainfall on slope stability. The proposed method can simulate the hydrological process of surface runoff and subsurface infiltration in slope unit, and consider the interaction mechanism between hydrology and geotechnics. Taking a rainfall-induced landslide as an example, which is at a highway dumpsite in Yunnan, China, its stability under natural conditions and failure process under heavy rainfall are simulated. The results indicate that the slope remains stable under natural conditions but suffers failure and large deformation under heavy rainfall. The numerical results are corresponding well with field investigations, which demonstrates that the provided method can well be used to simulate the whole process of rainfall infiltration, moisture migration, stability, deformation and failure of slopes.