Empirical and analytical analyses of laboratory granular flows to investigate rock avalanche propagation

Laboratory experiments which consist of releasing dry rigid non-cohesive grains or small bricks on an unconfined chute have been designed to investigate rock avalanche propagation mechanisms and to identify parameters influencing their deposit characteristics. Factors such as volume, fall height, basal friction angle, material used, structure of the material before release, i.e. bricks randomly poured into the reservoir before failure or piled orderly one on top of the other, and type of slope break, i.e. curved or sharp angular, are considered and their influence on apparent friction angle, travel angle of the centre of mass, deposit length and runout is analysed. Results highlight the influence of the structure of the material before release and of the type of transition at the toe of the slope on the mobility of granular avalanches. The more angular and sharp is the slope break, the more shearing (friction) and collisions will develop within the sliding mass as it changes its flow direction, the larger will be the energy dissipation and the shorter will be the travel distance. Shorter runout is also observed when bricks are randomly poured into the reservoir before release compared to when they are piled one on top of the other. In the first case more energy is dissipated all along the flow through friction and collisions within the mass. Back-analysis with a sled block model of experiments with a curved slope break underlines the importance of accounting centripetal acceleration in the modelling of the distance travelled by the centre of mass of a granular mass. This type of model though is not able to assess the spreading of the mass and its total runout because it does not take into account the internal deformation and the transfer of momentum within the mass which, as highlighted by the experimental results, play an important role in the mobility of rock avalanches.