Robustness evaluation of the probability-based HTCA model for simulating debris-flow run-out extent: Case study of the 2010 Hongchun event, China

Abstract

Probability-based cellular automaton (CA) models are useful alternative tools for simulating the extent of debris flow run-out. Despite that CA models have high computational efficiency, the simulation results are often sensitive to several key parameters, in particular, the steps of Monte-Carlo iterations (MCI), and the spatial resolution of the digital terrain model (DTM) data. This information is fundamental for evaluating the robustness of the CA model in dealing with various cases. This study improves upon the existing hydrodynamic and topography-based cellular automaton (HTCA) model. The basics of the model are briefly reviewed, and the sensitivity of the MCI value and the DTM spatial resolution are analyzed. Various practical scenarios regarding the effects of bed sediment entrainment and check dams are explored, which is beneficial for assessing the capability of the HTCA model in dealing with complex conditions. The August 14, 2010, post-seismic debris-flow event in the Hongchun catchment in the eastern Wenchuan area of China is selected as a case study to demonstrate the influence of key parameters on the simulation. The results illustrate the model's robustness in dealing with specific scenarios. Sensitivity analysis suggests that the MCI value and the DTM spatial resolution have obvious influences on the simulation results. The simulation results of the HTCA model can be improved if the MCI value is chosen based on the resolution of the input DTM data.