Item: Comparing Models of Different Levels of Complexity for the Prediction of Wet-Snow Avalanches
-
-
Title: Comparing Models of Different Levels of Complexity for the Prediction of Wet-Snow Avalanches
Proceedings: International Snow Science Workshop 2014 Proceedings, Banff, Canada
Authors:
- Christoph Mitterer [ Bavarian Avalanche Warning Service, Munich, Germany ] [ WSL Institute for Snow and Avalanche Research SLF, Davos Dorf, Switzerland ]
- Jürg Schweizer [ WSL Institute for Snow and Avalanche Research SLF, Davos Dorf, Switzerland ]
Date: 2014-09-29
Abstract: The amount and timing of infiltrating water plays an important role in determining wetsnow instability. However, the nature of water flow in snow is very complex and heterogeneous. In addition, the mechanical interaction of water and snow is poorly understood. Both facts complicate wet-snow avalanche forecasting. Due to this complexity and the poor understanding, statistical rather than physical models have been used in the past to explain wet-snow avalanche activity. Various multivariate statistical models used different input parameters with varying complexities ranging from simple air temperature recordings to modeled energy balance. We tested these statistical models for the region of Davos. Meteorological data from different automatic weather stations at three elevations (below treeline, at treeline and in the alpine) for the winter periods 2004-2005 to 2012-2013 were used and results compared to wetsnow avalanche activity data. Results show that the forecasting models using only air temperature, or which were trained with other local data, had only limited success, i.e. transferability was limited. To improve forecasting skills, it seems crucial to include information on both, boundary conditions (e.g. air temperature) and the snowpack itself (e.g. snow surface temperature). Best predictive power was obtained when we accounted for aspect, energy fluxes and state of infiltration. The results suggest that a certain complexity is necessary to correctly depict the processes favoring wet-snow avalanche activity with a model. Already now, parts of the results are in operational use to better predict days with wet-snow avalanche activity.
Object ID: ISSW14_paper_O1.02.pdf
Language of Article: English
Presenter(s):
Keywords: Wet-snow avalanches, avalanche forecasting, danger assessment.
Page Number(s): 9-14
-