Abstract
In unconsolidated glacial sediments and exhumed bedrock of Minnesota, USA, landslides may occur on low-relief steep slopes generated by post-glacial fluvial network development and drainage of glacial lakes. As part of an investigation to understand the link between various environmental variables, post-glacial landscape development, and a previously mapped landslide inventory, we performed multivariate logistic regression analyses to create landslide susceptibility maps of five landslide-prone regions in Minnesota. The regions are in northwestern, northeastern, southern, and southeastern parts of the state, denoted as Red River, Lake Superior, Metro, Minnesota River, and Southeastern Region, respectfully. The results from this analysis denote the relative likelihood that a landslide will occur at the mapped locations when considering the contributing environmental variables. The susceptibility modeling results and maps offer a broad assessment of landslide hazards as it used data across large regions of the state. Site specific assessments may be warranted for further risk and hazard investigations. Funding for this project was provided by the Minnesota Environment and Natural Resources Trust Fund recommended by the Legislative-Citizen Commission on Minnesota Resources (LCCMR), USGS Landslide Hazards Program, the Mary T. Dooley and James Goff endowments at the Minnesota State University, Mankato, and the Environmental Studies Program at Gustavus Adolphus College.