Articles | Volume 13, issue 6
Nat. Hazards Earth Syst. Sci., 13, 1655–1667, 2013
Nat. Hazards Earth Syst. Sci., 13, 1655–1667, 2013

Research article 22 Jun 2013

Research article | 22 Jun 2013

A novel approach to evaluate and compare computational snow avalanche simulation

J.-T. Fischer J.-T. Fischer
  • Federal Research and Training Centre for Forests, Natural Hazards and Landscape – BFW, Department of Natural Hazards, Rennweg 1, 6020 Innsbruck, Austria

Abstract. An innovative approach for the analysis and interpretation of snow avalanche simulation in three dimensional terrain is presented. Snow avalanche simulation software is used as a supporting tool in hazard mapping. When performing a high number of simulation runs the user is confronted with a considerable amount of simulation results. The objective of this work is to establish an objective, model independent framework to evaluate and compare results of different simulation approaches with respect to indicators of practical relevance, providing an answer to the important questions: how far and how destructive does an avalanche move down slope. For this purpose the Automated Indicator based Model Evaluation and Comparison (AIMEC) method is introduced. It operates on a coordinate system which follows a given avalanche path. A multitude of simulation runs is performed with the snow avalanche simulation software SamosAT (Snow Avalanche MOdelling and Simulation – Advanced Technology). The variability of pressure-based run out and avalanche destructiveness along the path is investigated for multiple simulation runs, varying release volume and model parameters. With this, results of deterministic simulation software are processed and analysed by means of statistical methods. Uncertainties originating from varying input conditions, model parameters or the different model implementations are assessed. The results show that AIMEC contributes to the interpretation of avalanche simulations with a broad applicability in model evaluation, comparison as well as examination of scenario variations.