Numerical modeling of mixed two-phase in long runout flow-like landslide using LPF3D

Under extreme climate conditions, such as heavy precipitation, glacial lake outbursts, and ice and snow melting, the long runout flow-like landslide with mixed solid-liquid two-phase has become one of the most disastrous types in the world. Numerical simulation is one of the most important means in disaster prevention and mitigation work for this type of landslide. In this study, a new full three-dimensional landslide post-failure numerical platform (LPF3D) was proposed. This method serves as a bridge between continuum-medium algorithms and discrete-medium algorithms based on the same theoretical numerical framework of smoothed particle hydrodynamics (SPH). This computational model employs an elastic-viscoplastic constitutive law and basal resistance of frictional model for the solid grain phase, and the fluid phase is treated as a Newtonian fluid. A drag law describes the interaction stress between fluid and grain. The boundary normal forces were applied using the penalty function method to ensure the stability of the algorithm. Comparison analyses of small-scale flume experiments and realistic-scale landslide cases and different numerical methods were performed to show that this numerical modeling is capable of simulating mixed two-phase flow. This new method realizes the actual physical and mechanical action process of mixed two-phase flow with higher computational efficiency, and is an appropriate benchmark for the post-failure risk forecasting and assessment of long runout flow-like landslide.

Automated summary

A new full three-dimensional landslide post-failure numerical platform (LPF3D) is proposed to simulate the actual physical and mechanical action process of mixed solid-liquid two-phase flow, enabling the prediction and assessment of post-failure risk for long runout flow-like landslides.