关键词: 高位远程滑坡, 动力侵蚀, 数值模型, DisasterFlow V1.0, 量化预测

Abstract: After the rapid disintegration of long run-out accumulation landslides, debris flows are often formed. They dynamically erode the underlying soft materials along their sliding paths and change the dynamic characteristics of landslide movement, which will lead to a significant increase in landslide volume. Long run-out accumulation landslides have strong disaster effects, which has become the focus of disaster prevention and mitigation research in recent years. To address the issue of quantitative prediction of dynamic erosion process for potential long run-out accumulation landslides, a continuum numerical model is constructed based on the depth-integrated hydrodynamic calculation theory and solid-liquid two-phase shear stress erosion model framework. The software program DisasterFlow V1.0 with second-order calculation accuracy is compiled for numerical realization. Through two classic case studies of dam-break flow, the results calculated by the developed model program are highly consistent with the actual situations. The developed numerical solution scheme is correct and has a good total variation diminishing (TVD) behavior. The prediction of dynamic erosion process for the long run-out accumulation landslide in Baiyanglin shows that the developed numerical model and the conventional sled model basically agree with the prediction results of the landslide sliding velocity. Compared with the conventional sled model, the developed numerical model has better adaptability to the terrain of the sliding path, resulting in less oscillation in the predicted values of the landslide sliding velocity. In addition, during the numerical prediction and analysis of the dynamic erosion process for the long run-out accumulation landslide in Baiyanglin, the developed numerical model can quantify the landslide dynamic erosion process in the time and space dimensions, such as quantifying the sliding depth, velocity and accumulation depth in each duration, which are not available in the conventional sled model. This paper provides a feasible quantitative prediction solution of the dynamic erosion process for long run-out accumulation landslides, which could enriches the quantitative prediction theory and technical system for long run-out landslides.

Key words: long run-out landslides, dynamic erosion, numerical model, DisasterFlow V1.0, quantitative prediction