Hitting rock bottom - Experimental study of bedrock erosion by debris flows

While there are field observations and studies on erosion of debris flows over loose sediment, bedrock erosion by debris flows has not yet been studied comprehensively. Nevertheless, traces of erosion in the bedrock channel after debris-flow events, such as scars, ridges, grooves or individual impacts, indicate a non-negligible entrainment of material as a result of process-related impact and shear forces. In the Alps, such erosion phenomena are often found in the upper steep and inaccessible parts of the catchment areas and are thus difficult to analyse. In this study we therefore investigate the potential erosion capacity of debris flows of different rheological characteristics on immobile channel beds with a small-scale physical model. We try to understand how bedrock strength influences erosion and if shear or impact forces dominate bedrock erosion by debris flows. To this end, erosion rates in terms of volume and the forces causing erosion are examined in over 100 laboratory experiments. We also compare and scale these rates to natural bedrock erosion caused by debris flows. In our small-scale laboratory investigation, polyurethane foam boards act as bedrock surrogates. This material has already been used as a bedrock simulant in studies into fluvial bedrock erosion. The boards were installed in the lower 2.5 m of the flume channel, which has a total length of 5.6 m. Different board strengths indicating different erosion susceptibility were tested with three different debris-flow mixtures. The slope of the channel was kept constant at 34. After each debris flow, the change in bed elevation was measured with a laser scanner to determine erosion rates at submillimeter accuracy. Laser distance sensors, pore water pressure sensors, a load cell and a geophone were used to quantify debris-flow dynamics and different erosion forces, including shear and impact. Our results show an exponential increase in erosion with a decreasing tensile strength of the bedrock simulant. While impact and shear forces both influence erosion rates, the decisive erosion force component appears to depend on the proportion of gravel and clay in the debris-flow mixture. The results of this study serve to deepen our understanding of the debris-flow process and expand our knowledge of erosion processes in the upper reaches of debris-flow catchments. In addition, our results can be used to tune existing models for longer-term landscape evolution.