Snow cornices up to several metres in height bear witness to the immense influence exerted on the snowpack in the Alps by winds. In a wind tunnel and at measuring sites we are investigating where the wind erodes and deposits snow and how the snow changes during transport.
Storm-force winds occur frequently in the Alps. The wind creates ripples and swirls, as well as snow cornices that overhang leeward slopes (Fig. 1). In irregular terrain it transports snow from peaks and crests, completely fills gullies and bowls, and can make entire transportation routes impassable.This process changes not only the local snow quantities, but also the layering and stability of the snowpack. The snow crystals themselves also undergo transformation during transport; they become smaller as a consequence of evaporation and colliding with the snow surface. Furthermore, snowdrift accumulations often give rise to slab avalanches. For this reason experts sometimes refer to the wind as the 'architect of avalanches'.In various projects we examine how snow is transported and where erosion and deposits occur in the terrain. The measurement results are integrated in computer models that simulate the properties of the snowpack and can assist in the development of measures to protect against snow transport.
In two wind tunnels we are analysing the transport and sedimentation of snow in controlled conditions. The velocity, height, trajectory and flight distance of single snow particles are among the variables we observe and measure (Fig. 2). We also examine the formation of wind crusts on the snowpack in the wind tunnel. In view of the large gap that exists in the understanding of how snow responds to changeable wind forces, we investigate in another project the dynamics of drifting snow and its interaction with turbulent currents. Alongside other outcomes, results of the research conducted in the wind tunnel help in the development of suitable measures to protect against drifting snow and in the endeavour to improve snow transport models.
Experiments conducted in the wind tunnel provide penetrating insights into the fundamentals of the drifting snow process, but can only replicate to a limited extent the complex conditions which exist in the field. In nature, for example, the snow surface is very irregular and snow properties are highly variable. For example, in some places there might be light powder snow which can be easily transported by light winds. At the same time a wind crust or wafer surface might have formed in a different aspect which prevents the snow from eroding. Furthermore, winds in the mountains are very gusty and variable. At the Weissfluhjoch test site (Fig. 3) we are thus investigating how, and how much, snow is transported over short distances by turbulent eddies. For this purpose we measure the amount of snow that is transported by the wind close to the ground, up to a height of 2 metres, and record wind speed with a resolution of 0.05 seconds. The results are to be used to validate and improve the snowpack models SNOWPACK and Alpine3D. The findings relating to snow transport by gusty winds are ultimately to be applied in avalanche danger forecasting. In an additional project we examine how a wind crust is formed. Different types of wind crusts appear to exist. Some are very soft and up to one metre thick, while others are very hard and just a few centimetres thick. We are attempting to identify the physical processes which occur and the conditions required for such crusts to form.
Since winter 2016/17 two automatic measuring stations have been supplying us with weather and snow transport data in East Antarctica. We are thereby able, for the first time, to record the snow transporting processes which take place in the Antarctic winter. The data can be used to estimate how much snow has been deposited over the course of a year, and to what extent the snow transport process has diminished the effects of the snowfall. The amount of snow is reduced not only by wind erosion, but also and in particular because the snow particles sublimate while airborne. Until now this sublimation process could be modelled only in a very simplified form. With the aid of our new measurements it should be possible to reflect the impact of sublimation more accurately in future.