Numerical simulation of rapid flow-like landslides and some related phenomena based on improved depth-averaged models

Numerical simulation is an efficient tool for the run-out analysis of rapid flow-like landslide. In this thesis, I address three topics related to the modeling of flow-like landslide which were not sufficiently investigated in the previous studies. Three improved depth-averaged models are used to simulate the selected typical flow-like landslides and related phenomena. In the first topic, a two-layer depth-averaged model is proposed to simulate the frontal plowing phenomenon in some rapid flow-like landslides. The propagation process of a loess landslide in Shaanxi Province, China and its interaction with the terrace material is analyzed. The second topic is related to the influence of the slope gradient and gully channel on the run-out behavior (especially the entrainment and deposition characteristics) of rockslide-debris flow. The run-out process of the Verghereto landslide in Italy is analyzed by using an improved depth-averaged model. The third topic is related to the numerical assessment of the impeding effect of check dam on debris flow. Another improved depth-averaged model, which takes both entrainment and the impeding effect of check dam into account, is proposed and adopted to analyze the interaction of debris flow and check dams in a debris flow gully in Sichuan province, China. The model is then used to assess the efficiency of the actual check dams in this debris flow gully. The main purpose of this thesis is to investigate the capacity of the improved depth-averaged models on the simulation of rapid flow-like landslides and the related phenomena (frontal plowing, entrainment, and interaction between debris flow and check dam). The simulation results of the case studies in this thesis show that these improved models perform well in simulating the rapid flow-like landslides and the phenomena mentioned above, which demonstrates the potential application ability of these models for the risk assessment of rapid flow-like landslides