Large-scale risk assessment on snow avalanche hazard in alpine regions

Abstract

Abstract. Snow avalanches are recurring natural hazards that affect the population and infrastructure in mountainous regions, such as in the recent avalanche winters of 2018 and 2019, when considerable damage was caused by avalanches throughout the Alps. Hazard decision makers need detailed information on the spatial distribution of avalanche hazards and risks to prioritize and apply appropriate adaptation strategies and mitigation measures and thus minimize impacts. Here, we present a novel risk assessment approach for assessing the spatial distribution of avalanche risk by combining large-scale hazard mapping with a state-of-the-art risk assessment tool, where risk is understood as the product of hazard, exposure and vulnerability. Hazard disposition is modeled using the large-scale hazard indication mapping method RAMMS::LSHIM (Rapid Mass Movement Simulation::Large-Scale Hazard Indication Mapping), and risks are assessed using the probabilistic Python-based risk assessment platform CLIMADA, developed at ETH Zürich. Avalanche hazard mapping for scenarios with a 30-, 100- and 300-year return period is based on a high-resolution terrain model, 3 d snow depth increase, automatically determined potential release areas and protection forest data. Avalanche hazard for 40 000 individual snow avalanches is expressed as avalanche intensity, measured as pressure. Exposure is represented by a detailed building layer indicating the spatial distribution of monetary assets. The vulnerability of buildings is defined by damage functions based on the software EconoMe, which is in operational use in Switzerland. The outputs of the hazard, exposure and vulnerability analyses are combined to quantify the risk in spatially explicit risk maps. The risk considers the probability and intensity of snow avalanche occurrence, as well as the concentration of vulnerable, exposed buildings. Uncertainty and sensitivity analyses were performed to capture inherent variability in the input parameters. This new risk assessment approach allows us to quantify avalanche risk over large areas and results in maps displaying the spatial distribution of risk at specific locations. Large-scale risk maps can assist decision makers in identifying areas where avalanche hazard mitigation and/or adaption is needed.

Climate change impacts on snow avalanche risk in alpine regions

“…Furthermore, lower snow depths would result in lower avalanche volumes. This might result in lower avalanche intensities and shorter run-out distances at lower elevations (Ortner et al, 2023a).…”

“…3.2). Station 2ME0 in Meien at 1320m above sea level was used as the base for the authors first study (Ortner et al, 2023a). It belongs to the SLF (= WSL Institute for Snow and Avalanche Research SLF) observer network (SLF-Observer, 2023) where manual daily snow measurements are carried out by an SLF observer.…”

“…These scenarios were defined according to the maximum 3-day snow depth accumulation and directly define the mean avalanche fracture depths, therefore a correction procedure for the release volumes was applied (Salm et al, 1990). The same methodology was used to generate three different avalanche hazard scenarios for risk analysis and has been established and described by Bühler et al (2022) and Ortner et al (2023a). In scenarios with return periods of 300-years, we have assumed higher avalanche release volumes compared to a 30-year return period scenarios (see Fig.…”

“…(Schneebeli and Bebi, 2004;Bebi et al, 2009;Teich et al, 2012a;Brožová et al, 2021). To consider protective forests in avalanche simulations, we have created a binary protective forest layer by using an approach suggested by Bebi et al (2021) and applied for large scale avalanche hazard mapping in (Bühler et al, 2022) and (Ortner et al, 2023a). We have used this GIS layer in the selection of potential release areas and used it in the avalanche simulation software "RAMMS" with further detailed properties to describe the friction for avalanche flow in forested areas.…”

“…presented an approach to identify potential avalanche release areas on a large scale using a high resolution digital terrain model and a protection forest layer. This approach was used by Bühler et al (2022) for large-scale avalanche hazard mapping and applied by Ortner et al (2023a) for large-scale avalanche risk mapping. In this study we use the same approach and applied it to consider scenario dependent release area sizes.…”