Please join us for a UW Data Science Seminar on Thursday, March 7th from 4:30 to 5:20 p.m. PST. The seminar will feature Morgan Sanger, a PhD scholar in geotechnical engineering at the University of Washington.
The seminar will be held in the Electrical & Computer Engineering Building (ECE), Room 105
“Mechanics-informed, geospatial machine learning for natural hazard planning and response”
Abstract: Hazard-resilient communities and infrastructure networks rely on hazard predictions that can be made accurately, quickly, at regional scale, and at high resolution. In earthquake-prone regions, earthquake-induced soil liquefaction is one of the most relevant and consequential geotechnical hazards. Accurate soil liquefaction hazard analyses require geotechnical testing, which cannot be continuously performed across large areas, thus presenting the need for “geospatial” liquefaction models. This project employs supervised machine learning to extend engineering mechanics and sparse geotechnical testing to map-scale using publicly available geospatial variables. In doing so, this model can be used in network analysis for emergency response planning, evaluating community impacts, and identifying mitigation priorities.
Bio: Morgan Sanger, P.E., is a PhD student scholar in geotechnical engineering at the University of Washington. Her doctoral research involves applying machine learning to large geospatial and geotechnical data sets for improved natural hazard modeling and risk management. Morgan is a 2023-2024 Herbold Data Science Fellow for the College of Engineering.
The 2023-2024 seminars will be held in person, and are free and open to the public.