5:15 PM - 7:15 PM
[MIS01-P06] Prototype of probabilistic avalanche hazard maps reflecting snow cover uncertainties in winter conditions
Keywords:uncertainty, haard map, snow avalanche
Snow avalanches are the most common gravity-driven mass flow in snowy regions. They can become natural disasters when their flow paths intersect with human activity zones. One effective measure to mitigate avalanche damage is developing hazard maps that indicate avalanche runout areas. To illustrate avalanche runout areas, depth-averaged avalanche flow models are widely used. However, these models require uncertain input variables, such as initial flow thickness and parameters related to basal friction.
In this study, the uncertainty in input variables is represented by probability distributions. This uncertainty propagates through the flow models to their outputs, enabling the development of probabilistic hazard maps. Here, these probabilistic hazard maps indicate runout areas with associated probabilities, such as the exceedance probabilities of a model output (e.g., maximum flow thickness) or the expected values of the output.
As probabilistic evaluations require numerous numerical simulations, a polynomial emulator based on the polynomial chaos quadrature method was developed (Tanabe et al., 2025). A specific release area was targeted, and probability distributions were estimated based on data from an automated weather station near the release area. Two types of probability distributions were considered: one based on average values during the winter period and the other reflecting current weather and snow conditions. Using the polynomial emulator, hazard maps corresponding to these different probability distributions were generated at a low computational cost. In this presentation, a prototype of probabilistic avalanche hazard maps that illustrate temporal variations in runout areas and probability values based on changing weather conditions will be introduced.
Reference
Tanabe, T., Tsunematsu, K., and Nishimura, K. (2025). Quantitative evaluation of probabilistic hazard mapping with polynomial chaos quadrature and its practical application. Journal of Geophysical Research: Earth Surface,130, e2024JF007970. https://doi.org/10.1029/2024JF007970
In this study, the uncertainty in input variables is represented by probability distributions. This uncertainty propagates through the flow models to their outputs, enabling the development of probabilistic hazard maps. Here, these probabilistic hazard maps indicate runout areas with associated probabilities, such as the exceedance probabilities of a model output (e.g., maximum flow thickness) or the expected values of the output.
As probabilistic evaluations require numerous numerical simulations, a polynomial emulator based on the polynomial chaos quadrature method was developed (Tanabe et al., 2025). A specific release area was targeted, and probability distributions were estimated based on data from an automated weather station near the release area. Two types of probability distributions were considered: one based on average values during the winter period and the other reflecting current weather and snow conditions. Using the polynomial emulator, hazard maps corresponding to these different probability distributions were generated at a low computational cost. In this presentation, a prototype of probabilistic avalanche hazard maps that illustrate temporal variations in runout areas and probability values based on changing weather conditions will be introduced.
Reference
Tanabe, T., Tsunematsu, K., and Nishimura, K. (2025). Quantitative evaluation of probabilistic hazard mapping with polynomial chaos quadrature and its practical application. Journal of Geophysical Research: Earth Surface,130, e2024JF007970. https://doi.org/10.1029/2024JF007970