LANDSLIDE SUSCEPTIBILITY ANALYSIS FOR PIPELINE ROUTE OPTIMIZATION IN WASHINGTON’S ISSAQUAH ALPS

2024-12-26
Öney, Murat
This study focuses on assessing landslide susceptibility and optimizing pipeline routing in the Issaquah Alps region, between Newcastle and Preston, Washington State. Using logistic regression (LR), landslide susceptibility maps (LSMs) were generated at both regional and corridor-specific scales to evaluate three proposed pipeline routes. A comprehensive inventory of 192 mapped landslides, covering approximately 10 km², served as the basis for the regional LSM across a 177.5 km² study area. Additional corridor-specific LSMs were created within 1 km and 500 m buffer zones along each route to refine localized risk assessments. The analysis incorporated ten conditioning factors, including slope, geology, land cover, elevation, and proximity to faults, roads, and waterways, as well as topographic wetness and stream power indices. Results revealed distinct risk profiles for each route: Route 1, the shortest and flattest, presents moderate landslide susceptibility but requires stabilization near existing infrastructure; Route 2, traversing mountainous terrain, faces the highest risk and demands significant mitigation; Route 3, following valley regions, exhibits moderate risks but requires careful hydrological management near streams. Corridor-specific analyses showed improved accuracy, with higher AUC values (up to 0.841 for 1 km buffers) and success rates (up to 79.3% for 500 m buffers), highlighting the precision of localized assessments compared to the regional LSM. This study underscores the value of combining regional and corridor-specific LSMs for infrastructure planning in landslide-prone areas. The findings provide a methodological framework for pipeline routing, balancing safety, cost, and environmental concerns.
Citation Formats
M. Öney, “LANDSLIDE SUSCEPTIBILITY ANALYSIS FOR PIPELINE ROUTE OPTIMIZATION IN WASHINGTON’S ISSAQUAH ALPS,” M.S. - Master of Science, Middle East Technical University, 2024.