Snow Avalanche Prediction in Grasdalen, Norway: Application of Statistical Methods

Snow avalanches are a common hazard in many mountainous environments. As such, it is often necessary to be able to predict these events. In this study detailed analysis of weather parameters in western Norway has established common triggering factors for avalanches in Grasdalen. Two large sets of weather data have been used. A number of wind drift factors have been derived, as several authors consider these to be important. Statistical procedures were applied, including cumulative probability plots and classification trees. The latter highlighted the most important weather parameters used to classify data in terms of dry avalanche days or non avalanche days. Results of these indicate important splitting factors to be various sums of precipitation and the maximum temperature. The combined wind drift factors were considered less important. These results are not in total agreement with previous authors, however, this study serves as a critical appraisal of current statistical methods of avalanche prediction and would be useful for avalanche forecasters and researchers.