The Tufts Geohazards Research Group has been funded to study liquefaction by the USGS and NSF several times. The overarching theme of these projects has been to constrain uncertainties associated with liquefaction risks.
Current Grants
National and Global Liquefaction Loss Model for PAGER
USGS Award #G19AP00023
Principal Investigator: Laurie Baise
One of the original goals of the GGLM was to provide rapid estimates of loss in regards to liquefaction hazards after an event. Explanatory variables of outputs from GGLM models and GIS infrastructure layers and proxies will be used in a regression to estimate areas of loss based on amounts of liquefaction documented in reconnaissance reports. Example infrastructure layers and proxies include open street map (OSM) transportation networks, population densities, and port locations.
Updating the Geospatial Liquefaction Database and Model
(January 2020 through December 2020) USGS Award No. G20AP00029
Principal Investigator: Laurie Baise
This project will update the geospatial liquefaction database with liquefaction locations from recent earthquakes. Additionally, the project will use rasters for explanatory variables that increase spatial resolution but decrease spatial extent. The model will analyze several earthquakes in the United States which produced liquefaction and assess whether regressions to predict liquefaction extent can be improved while using higher resolution explanatory variables.
Past Grants
Regional Liquefaction Hazard Mapping
USGS NEHRP Award #05HQGR0103
This project helped develop guidelines for utilizing geotechnical data in regional liquefaction mapping projects. Prior to this project, liquefaction potential on a large scale was typically determined only by surficial geology while liquefaction potential on a finer scale was analyzed with boreholes. Boreholes are expensive and time-consuming, meaning they cannot be utilized for hazard mapping on a large scale. This project uses statistical, probabilities, and geostatistical methods to estimate liquefaction potential within and across geologic units. The project uses densely collected CPT and shear wave velocity data combined with liquefaction potential index values to interpolate liquefaction potential on a regional scale.
Liquefaction Hazard Mapping for Boston, Massachusetts
USGS NEHRP Award #02HQGR0036 and #02HQGR0040
Collaborative research with William Lettis & Assoc., Inc. and Tufts University
This two-year project mapped liquefaction hazards for Boston, Massachusetts, by characterizing the distribution of potentially liquefiable sediments and artificial fill. This study encompassed downtown Boston and its surrounding communities, much of which is highly populated and industrious, and has experienced several earthquakes greater than a moment magnitude of 6.0. Fist, a digital database of 2963 boreholes was compiled to determine liquefaction susceptibility of subsurface units. Next, geotechnical analysis of this database was used in tandem with published geologic maps, aerial photographic interpretation, and soil stratigraphy from another 12,000 boreholes to understand liquefaction triggering thresholds.
Post-Liquefaction Reconnaissance Using Remote Sensing
USGS NEHRP Award #G10AP00025 and #G10AP00026
Collaborative Research with Tufts University and University of Alaska Fairbanks
As liquefaction occurs in saturated granular soils due to an increase in pore pressure, often resulting in excess water rising to the surface, liquefaction inducted terrain should have an associated increase in soil moisture with respect to the surrounding non-liquefied regions. Spectral bands and indices sensitive to soil moisture content can be used to identify areas with increased soil moisture after an earthquake. This project uses change detection with pre- and post-event thermal’sl and tasseled cap wetness images for the 2001 Bhuj earthquake in India to verify this hypothesis and helped to delineate earthquake induced liquefaction areas.
A Geospatial Liquefaction Model for Rapid Response and Loss Estimation
NSF Award #1300781
Principal Investigator: Laurie Baise; Co-Principal Investigator: Eric Thompson
Prior to this project, rapid response maps after earthquakes did not include any information related to liquefaction hazards. This project took place over four years with the aim of providing estimates of spatial liquefaction probability quickly after an earthquake. A model is produced to estimate this spatial probability with a logistic regression approach using earthquake-specific intensity measures in conjunction with possible explanatory variables such as soil saturation, vegetation, and hydrology. Global rasters were used for explanatory variables which allows this geospatial model to be implemented rapidly for an earthquake’s intensity measures in any part of the world, resulting in a Global Geospatial Liquefaction Model (GGLM).
Validation of a Geospatial Liquefaction Model for Noncoastal Regions Including Nepal
USGS Award #G16AP00014
Principal Investigator: Laurie Baise
The primary purpose of this project was to evaluate the geospatial liquefaction model produced by Zhu et al. (2017) across 53 earthquakes. This was done by summing the total expected area of liquefaction from the model for each event and comparing with levels of liquefaction documented in reconnaissance reports. Earthquakes are also analyzed in Nepal, an area which the original geospatial liquefaction model did not include. The project also compares expected amounts of liquefaction with CPT-based liquefaction potential indices for urban areas which have been studied more in-depth by other researchers. The study concludes that the GGLM produced by Zhu is appropriate at both urban and regional scales, but it also includes a minimum PGA threshold to improve liquefaction estimations.