Near-Surface Energy Balance on an Alpine Rock Glacier: Murtèl-Corvatsch

Abstract: This project investigates the near surface energy balance on the Murt`el-Corvatsch rock glacier in the Upper Engadine, Swiss Alps, using the 1D physical SNOWPACK model. A correct representation of the near surface energy balance is important to predict the long term evolution of permafrost below rock glaciers. This is of interest in the context of future water availability and management of water resources in a changing climate and also in the context of natural hazards. Some difficulties in modelling the thermal regime of rock glaciers are related to the large pore spaces between the blocks, which allow for different modes of heat transport. With this in mind, different modelling approaches were investigated: using the standard SNOWPACK (without advective heat flux, ventilation or canopy module), adding an advective heat flux, using the ventilation and canopy modules. The most promising results, i.e. the best match between measured and modelled temperatures, were obtained from the ventilation parameterisation. This parameterisation accounts for boundary-layer air penetrating into the blocky layer. Furthermore it was found that the most important input variables are the thickness of the the blocky layer, since this is where the additional modes of heat exchange take place, and the ice and void volume fraction together with the field capacity in the icy layer. The latter are particularly relevant for long term modelling as they determine the amount of ice melt and water transport in the icy layer. Measured and modelled temperatures at depths of 0.5 m, 2.5 m, 3.5 m and 7.5 m were compared. Generally good agreements between modelled and measured temperatures were obtained for the depths 0.5 m, 3.5 m and 7.5 m. The slight warming trend at the end of the modelled period (2012- 2016) that can be observed in the borehole data is also present in the modelled temperatures. The depth of 2.5 m shows the least agreement between modelled and measured temperatures with and overestimation during the snow free period and an underestimation during the snow covered period. However, agreement between modelled and measured temperatures improves for the snow covered period after a simulation period of about ten years.