Effects of ground freezing and snow avalanche deposits on debris flows in alpine environments

Debris flows consist of a mixture of water and sediments of various sizes. Apart from few exceptions, the water is usually contributed directly from precipitation. In a high mountain environment like the Alps, it appears necessary to consider infiltration of water into the ground during rainfall events, the runoff characteristics and the potential supply of sediment as a function of a multitude of climatic and hydrogeological factors. This paper outlines several new processes – either linked to ice formation in the ground before an event, or to the presence of snow avalanche deposits – that change the probability of observing an event. These processes were identified during field observations connected with extreme weather events that occurred recently in the Valais Alps (south-western Switzerland): they can be seen as factors either amplifying or reducing the potential of slope instability caused by the precipitation event. An intense freezing of the ground during the week preceding the exceptional rainfall event in mid-October 2000 amplified the probability of triggering debris flows between roughly 1800 and 2300 m asl. Both growth of ice needles and superficial ground freezing destroyed soil aggregates (increasing the availability of sediments) and/or, a deeper ground freezing resulted in decreased infiltration rate (increased runoff) during the first hours of heavy rainfall. The presence of snow avalanche deposits in a gully could be simultaneously an amplifying factor (the snow deposits increase the base flow and create a sliding plane for the sediments, mainly at the time of summer storms) or a reducing factor (reduction in the impact energy of the raindrops, mainly at the time of winter storms) of the risk of triggering debris flows. If it is not currently possible to establish rainfall threshold values for debris flow triggering, the knowledge and the implementation of these processes in the analysis of the potential triggering (for example by comparing the catchment hypsometric curve with the meteo-climatic situation) would nevertheless make the analysis of debris flows and forecasting more efficient. Correspondence to: E. Bardou ([email protected])