Modelling debris flow runout considering grain size distributions of debris flows at the Illgraben, Swiss Alps
Abstract. Debris flows are important processes for the assessment of natural hazards due to their damage potential. To assess the impact of a potential debris flow, parameters such as the flow velocity, flow depth, maximum discharge and the volume are of great importance. This study uses data from the Illgraben observation station to explore the relationships between these flow parameters and the debris flow dynamics. The flows were simulated with the RAMMS debris flow runout model, which is based on a numerical solution of the shallow water equations for granular flows using the Voellmy friction relation. Here, the events were modeled in an effort to explore the friction parameters μ and ξ, which describe the basal friction and the viscous turbulent friction, respectively, in the model. Additionally, sediment samples from levee deposits were analyzed for their grain size distributions (14 events) and their mineralogical properties (four events) to explore if the properties of the fine-grained matrix have an influence on the debris flow dynamics. Finally, field data from various debris flows such as the flow velocities and depths were statistically compared with the grain size distributions, the mineralogical properties, and the simulation results to (ii) identify the key variables controlling the kinematics of these flows. The simulation results point to the existence of several ideal solutions, with the friction parameters μ and ξ showing a strong dependency on the Froude number of the flow. It is also shown that no statistically significant correlation exists between the grain size distribution, the mineralogical composition of the matrix, and the debris flow properties, confirming the notion that a fixed debris-flow friction relation or rheology is a limiting assumption, at least for the Voellmy relation. Rather, the flow properties appear to be determined by the flow volume, from which most other parameters can be derived, consistent with common engineering practice.
Viewed (geographical distribution)