Modeling Liquefaction-Induced Disruption to Transportation Corridors in Southcentral Alaska: Toward a Resilient and Repeatable Infrastructure Assessment Strategy

Abstract Seismically induced liquefaction poses a critical threat to transportation infrastructure in geotechnically sensitive regions such as Southcentral Alaska. This study presents a transferable, prototype modeling framework designed to assess liquefaction risk along major highway corridors using both a widely adopted logistic regression model and a custom model calibrated to regional conditions. The custom model incorporates peak ground acceleration (PGA), shear wave velocity in the top 30 m ( VS 30 ), depth-to-water table (DTW), mean grain size ( D 50 ), and the coefficient of uniformity ( C u ). Ground failure data from the 2018 M w 7.1 Point Mackenzie earthquake were used to train and validate the models. Liquefaction probabilities were estimated for five earthquake scenarios ( M w 6.5–9.5), revealing that the custom model offers superior accuracy and spatial relevance, as shown by Brier scores and receiver operating characteristic (ROC) curves. Intended as a prototype template for integration into transportation resilience planning, the framework supports data-driven decision-making and is adaptable to other seismically vulnerable regions using publicly available geospatial data. Results provide actionable insights for agencies to prioritize retrofits, guide emergency preparedness, and embed liquefaction risk into long-term infrastructure investment strategies.