Quantifying the influence of avalanche release volume on runout probability: A case study of a mountain road section

Mountain roads are frequently exposed to snow avalanches, making effective hazard and risk management strategies necessary. Historically, snow avalanche hazard assessments have been based on deterministic hazard analyses, which rely on discrete input data. In practice, the results are usually presented as hazard maps that focus on calculating the run-out length or area, including the spatial distribution of intensity, for predefined scenarios, but neglect inherent uncertainty. To examine the potential of probabilistic avalanche hazard assessment, we present and exemplify an approach that specifically considers uncertainties associated with avalanche release and apply it to a road segment in the Austrian Alps. We demonstrate that potential sources of uncertainty within the initial conditions of computational avalanche simulations result in a bias in release volume. Using probabilistic, thickness integrated avalanche simulations in three-dimensional terrain, we illustrate probability maps for showing the potential impact of avalanches on the road. This demonstrates the potential variability in the simulation results, such as peak pressure. Therefore, the developed framework allows the quantification of uncertainties inherent in snow avalanche simulations and increases the density of information for hazard assessment, particularly with regard to potential biases in runout length, width and affected area. The results showed considerable variation in both spatial coverage and linear spreading across different return periods and intensity limits, with values ranging from 0.2 ha to 2.7 ha and from 26 m to 167 m, respectively. Consequently, the limitations of the deterministic hazard assessment for traffic lines were revealed and the underlying uncertainties quantified, thereby demonstrating the value of a probability-based hazard analysis. The research outcomes demonstrate the effectiveness of a probabilistic approach for snow avalanche hazard assessment, which could assist road authorities to make informed decisions about avalanche mitigation and adaptation. • Probabilistic approach to snow avalanche hazard assessment. • Quantifying avalanche release volume on runout probability. • Addressing uncertainties in both release size (geo-spatial uncertainties) and release thickness (snow depth uncertainties). • Probabilistic avalanche simulation using AvaFrame with PDF of the avalanche release volume. • Using probability maps to represent avalanche runout.