Dynamic simulation of rock-avalanche fragmentation

Fragmentation is a common phenomenon in rock avalanches with complex features. The fragmentation intensity and process determines exceptional spreading and mobility of rock-avalanches in the run-out zone. However, studies focusing on the simulation of these phenomena are still limited and no operational dynamic simulation model including the effects of fragmentation has been proposed yet. By enhancing the mechanically controlled landslide deformation model, we propose a novel, unified dynamic simulation method for rock-avalanche fragmentation during propagation. Our formally derived method relies on the continuum mechanics that is applicable to rock masses of any size. The model includes three important aspects: mechanically controlled rock mass deformation, the momentum loss while the rock-mass fiercely impacts the ground, and the energy transfer during fragmentation resulting in the generation of dispersive lateral pressure. We reveal that the dynamic fragmentation, resulting from the overcoming of the tensile strength of the rock mass by the impact on the ground, leads to spreading, thinning, and run-out of the rock avalanche, and to its hypermobility. The elastic strain energy release caused by fragmentation is an important process. Energy conversion between the front and rear parts of the mass caused by the fragmentation process results in the forward movement of the frontal material and the hindered motion of the rear portion of the rock avalanche. Our new model describes this by amplifying the lateral pressure gradient in the opposite direction: enhanced for the frontal particles and reduced for the rear particles after the fragmentation process. The main principle is the switching between the compressional stress and the tensile stress, and therefore from the controlled deformation to substantial spreading of the frontal part of the mass in the flow direction while backward stretching of the rear part of the rock mass. In principle, observations in the laboratory and field events support our simulation results.