Integrating Regionalized Geotechnical Information Into the U.S. Geological Survey’s Liquefaction Product Within a Bayesian Framework
Description:
The U.S. Geological Survey’s (USGS) Ground Failure (GF) product team has been tasked with estimating the extent of liquefaction (LF) and landslides (LS) following significant earthquakes worldwide since 2018. Because these models were calibrated against global LF and LS inventories, they are generalized and applicable to any earthquake globally. However, the USGS GF models are based only on geospatial proxies, and when regionalized geological and geotechnical information is available, better estimates of ground failure can be obtained by combining the GF models with this localized geotechnical information in a Bayesian manner. We expand upon a model previously developed to update liquefaction probabilities using such geotechnical information. This previous framework involved: (1) estimating the distribution of Liquefaction Potential Index (LPI), which is based on subsurface geotechnical measurements, 2) using a function defining liquefaction probability in terms of LPI, and (3) updating the USGS global model’s liquefaction probability estimates given the distribution of LPI for areas within artificial fill. Having made several improvements, we present the enhanced updating framework and demonstrate the utility of integrating this LPI-based geotechnical information for the 2019 Anchorage, Alaska and 1989 Loma Prieta, California earthquakes (for which liquefaction inventories are available for model comparison and framework validation). This model, which can update liquefaction probabilities in several geologic units (including nonsusceptible hard rock units), provides improved estimates of liquefaction probability where units were reliably mapped. This framework represents a novel means of integrating information not present in the USGS GF model to get improved liquefaction estimates.
Session: Coseismic Ground Failure: Advances in Modeling, Impacts and Communication
Type: Oral
Date: 4/20/2023
Presentation Time: 09:00 AM (local time)
Presenting Author: Davis T. Engler
Student Presenter: Yes
Invited Presentation:
Authors
Davis Engler Presenting Author Corresponding Author dengler@usgs.gov U.S. Geological Survey |
Eric Thompson emthompson@usgs.gov U.S. Geological Survey |
Mertcan Geyin mertcan.geyin@ngi.no Norwegian Geotechnical Institute |
Brett Maurer bwmaurer@uw.edu University of Washington |
Paula Burgi pburgi@usgs.gov U.S. Geological Survey |
Kishor Jaiswal kjaiswal@usgs.gov U.S. Geological Survey |
Kate Allstadt kallstadt@usgs.gov U.S. Geological Survey |
|
|
Integrating Regionalized Geotechnical Information Into the U.S. Geological Survey’s Liquefaction Product Within a Bayesian Framework
Category
Coseismic Ground Failure: Advances in Modeling, Impacts and Communication