Mechanics-informed machine learning for geospatial modeling of soil liquefaction: global model map products for LPI, LPIish, and LSN

This dataset contains a suite of raster files (geotiffs) which comprise a global geospatial liquefaction model. This geospatial liquefaction model is designed to predict manifestation of earthquake-induced soil liquefaction at the ground surface through geospatial surrogate models of three standard-of-practice geotechnical models: the liquefaction potential index (LPI); a modified LPI, termed LPIISH; and the liquefaction severity number (LSN). The dataset includes a folder for each continent, with a subdirectory for each surrogate geospatial liquefaction model (LPI, LPIISH, and LSN) which contain subfolders for the (1) model parameters (“Model”); and (2) the model variance classifications (“Variance”). Model users are referred to the journal publication provided in the project description for details regarding model development, performance, and use.

Sanger, M.D., Geyin, M., & Maurer, B.W. (2024). Mechanics-informed machine learning for geospatial modeling of soil liquefaction: global and national surrogate models for simulation and near-real-time response. In Review.