We investigated a small-scale laboratory model of a talus slope evolution. Five different size classes of basaltic rock were selected and marked with different colours. Homogenized mixtures of grains of different sizes were dropped from a fi xed height onto a tilted experimental board covered with a loose granular layer. This was conducted in a series of regular sequences, and the resulting distribution on the board was studied after each sequence. At the beginning of the experiment, the grains developed a longitudinal gradation similar to natural talus slopes, where small grains settle at the top while the large ones roll down to the distal part. However, after a transient period dominated by single-particle dynamics on the inert granular medium, the evolution proved to be more variable than expected. Due to the continuous shower of falling grains, the shear stress at the bottom of the upper granular layer increased. This resulted initially in a slow creep down slope that fi nally collapsed in large avalanches homogenizing the material. The slides occurred at the boundary between a weaker layer created by migration of small grains through the interstices, and marked by a vertical transition between small and large grains. We compare the experimental fi ndings with observations from natural talus slopes, and suggest that similar experiments may be helpful in understanding the evolution of taluses.

We present experimental results for the rolling of spheres on a granular bed. We use two sets of glass and steel spheres with varying diameters and a high-speed camera to follow the motion of the spheres. Despite the complex phenomena occurring during the rolling, the results show a friction coefficient nearly independent of the velocity (0.45-0.5 for glass and 0.6-0.65 for steel). It is found that for a given sphere density, the large spheres reach a longer distance, a result that may also help explain the rock sorting along natural stone accumulations at the foot of mountain slopes.

No abstract