Assessing avalanche risk by terrain analysis : an experimental GIS-approach to The Avalanche Terrain Exposure Scale (ATES)

Abstract: This study addresses the complex nature of snow avalanches and how their location can be predicted, and thus avoided, when navigating through mountainous terrain during winter recreations. The avalanche risk in the ski area of Nuolja, in northern Sweden, was evaluated by an experimental implementation of the Avalanche Terrain Exposure Scale (ATES) which is a model for pre-trip planning and assessment of exposure to avalanches. The model includes a high level of subjectivity and thus requires some background knowledge in avalanche science. The parameters in ATES consider terrain characteristics and avalanche history to determine the exposure. Analysis was carried out by the use of GIS software which is a departure from the intended usage method. The model was implemented two times: Firstly, according to the original model and, secondly, through modifying ATES by including an altitude parameter which is a terrain factor that is not included in the original model. The results indicated that the highest exposure mainly residues in areas that have experienced avalanches in the past. Including altitude did not show any clear improvements, mainly due to difficulties in evaluating the results from a model that already includes the “ground truth”, which in this case were previous avalanches. It was concluded that ATES can be a valuable tool for avalanche prediction but the current model is not very well suited for GIS. Furthermore, the subjectivity in the model makes the results highly dependent on the producer.