2019 CSDMS meeting-057


Log in (or create account for non-CSDMS members)
Forgot username? Search or email:CSDMSweb@colorado.edu

Browse  abstracts

A new approach to mapping landslide hazards: a probabilistic integration of empirical and process-based models

Erkan Istanbulluoglu, University of Washington Seattle Washington, United States. erkani@uw.edu
Ronda Strauch, Seattle City Lights Seattle Washington, United States. Ronda.Strauch@seattle.gov
Jon Riedel, National Park Service Sedro-Woolley Washington, United States. Jon_Riedel@nps.gov

A new approach for mapping landslide hazard is developed by combining probabilities of landslide probability derived from a data-driven statistical approach and process-based model of shallow landsliding. Our statistical approach integrates the influence of seven site attributes on observed landslides using a frequency ratio method. Influential attributes and resulting susceptibility maps depend on the observations of landslides considered: all types of landslides, debris avalanches only, or source areas of debris avalanches. For each landslide type the frequency ration (FR) classification is converted into a Stability Index (SI), mapped across our study domain in the North Cascades National Park, WA. Using distributed landslide observations a continuous function is developed to relate local SI values to landslide probability. This probability is combined with spatially distributed probability of landsliding obtained from Landlab using a two-dimensional binning method that employs empirical and modeled based probabilities as indices and calculates empirical probability of landsliding at the intersections of bin ranges of the empirical and process-based probability domains. Based on this we developed a probabilistic correction factor to modeled local landslide probability. Improvements in distinguishing potantially unstable domain with the proposed model is quantified statistically.