Report of Investigations--KGS
Abstract
Physics based landslide modeling can be difficult and data intensive. Generating high quality and practical map results is often not feasible outside of small, thoroughly characterized study areas. Using the physics based program Probabilistic Infinite Slope Analysis (PISA m), users can perform expedient assessments of landslide hazard over large study areas where comprehensive geotechnical data may be lacking, but other data inputs are robust. PISA m uses an infinite slope equation and spatial layers, such as a digital elevation model (DEM), a lidar derived forest cover layer, and a soil map, to calculate the probability that the factor of safety ( or FS ) for an area will be less than or equal to one. F actor of safety values less than or equal to one often infer slope instability. This investigation considers two landslide inventories one reflecting the assumed background climatic conditions seen over a decade and the other gathered following an extreme rainfall event. These two weather scenarios were approximated with parameter specifications and used over four models based on different soil unit input s : shale beds with alluvium and colluvium, 1:24,000 scale bedrock formations, United Soil Classification System (USCS) distributions with geotechnical values derived from drilling re- ports, and USCS distributions with generalized geotechnical values. Model results were symbolized as five susceptibility groups based on equal intervals of the probability of FS ≤ 1. The model results were compared to landslides that post date the lidar DEM and non landslide locations to evaluate the program’s accuracy as a regional landslide susceptibility tool. PISA m results indicating a high probability (0.50 1.0) of FS ≤ 1 around a landslide were considered true positives, while lower probabilities (0 0.50) for non landslide areas were considered true negatives. Model accuracies varied across the models and study areas, averaging 78% for the background climatic conditions proxy and 82% for the extreme rainfall event area, with the best model accuracy of 84% for the shale bed case in the extreme event specifications. While these practical, first order landslide susceptibility model results are promising, these outcomes rely on effective use of high resolution input data and expert knowledge of ground characterization to bolster the lack of precise geotechnical descriptions.
Publication Date
Spring 3-2025
Series
14
Report Number
1
Digital Object Identifier (DOI)
doi.org/10.13023/kgs.14.ri.1.2025
Funding Information
Preliminary efforts with program funded by Federal Emergency Management Agency Pre-Disaster Mitigation grant, project number PDMC-PL04-KY-2017-00, titled “Landslide Assessment and Mitigation Plan for the Big Sandy Area Development District.”
Repository Citation
Koch, H.J., Haneberg, W.C., Crawford, M.M., Dortch, J.M., and Swallom, M.L., 2025, Evaluating the physics-based slope stability program PISA-m at a regional scale: Kentucky Geological Survey, ser. 14, Report of Investigations 1, 22p.
