Abstract

This paper presents a comprehensive and well-documented landslide case study involving detailed site investigation, field tests, seismic signal analyses, and advanced numerical analyses using the material point method (MPM) to investigate the post-failure process and kinematic behavior of two landslides. The slope failures were triggered by Typhoon Jangmi in September 2008 and had impacts on the downslope residential areas. The background information of two landslides, including geological and hydrological conditions obtained from the detailed site investigation and field tests, was first introduced. The MPM analyses were then conducted and validated by comparing measured and predicted results for the final landslide profiles and duration. The numerical results, including variations in deviatoric strain, volumetric strain, and excess pore water pressure with time, were further examined to determine the post-failure mechanism of the landslides. The effects of soil drainage conditions during landslides on the final landslide profile were also compared and discussed. This study demonstrated that MPM analyses could accurately predict the runout distance of both landslides. The numerical results indicate that the landslides occurred through a complex process of soil failure that involved the stages of shearing, translational movement, compression, and deposition. When the soil was subjected to different loading conditions in different stages, the corresponding post-failure mechanisms are also different. High positive excess porewater pressure could develop near the sliding surface and has a significant influence on landslide kinematics and final landslide profiles. Compared with undrained analysis, fully coupled analysis, which permitted porewater pressure dissipation during a landslide event, produced more accurate predictions of landslide kinematics.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call