Â鶹´«Ã½

Debris flows and debris avalanches are rapid to extremely rapid landslides that tend to travel considerable distances from their source areas. One of the critical challenges to overcome with respect to debris flow risk is, therefore, the credible prediction of their size, travel path, runout distance, and depths of erosion and deposition. In this study, a quantitative modelling approach based on cellular automata methods, wherein individual cells move across a digital elevation model (DEM) landscape following behavioral rules defined probabilistically, is proposed and tested. The adopted model is able to estimate erosion and deposition soil volumes along a debris flow path by deploying at the source areas autonomous subroutines, called agents, over a 5 m spatial resolution DEM, which provides the basic information to each agent in each time-step.