Graduate Thesis Or Dissertation
 

Towards a national map of soil liquefaction susceptibility : modeling with heuristic and geotechnical methods

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/cf95jh275

Descriptions

Attribute NameValues
Creator
Abstract
  • This study investigates the use of predictive mapping techniques as well as geotechnical criteria in developing a multiregional soil liquefaction model and subsequent maps. The maps were produced using National Cooperative Soil Survey data, in the gSSURGO format, combined with soil liquefaction data gathered from studies, articles, and traditional seismic hazard maps. Geotechnical liquefaction studies are typically conducted at depths greater than three meters. This study and maps are not intended to replace detailed, site-specific investigations, but rather to provide regional interpretations of soil liquefaction susceptibility in surface soils of zero to three meters depth. The common digital soil mapping (DSM) models, Adaptive Boosting, Random Forest and CART were applied to the data, with boosting providing greatest accuracy. A second model, composed of simplified recent geotechnical liquefaction criteria was also fit to the data, to provide a comparator. Study areas include Washington State (WA), Oregon (OR), Arkansas (AR), Missouri (MO), and South Carolina (SC), with liquefaction susceptibility maps produced for OR & SC. Results indicate that a boosting model composed of WA & OR data adequately describes liquefaction data from AR, MO & SC with F-measures ranging from 0.92-0.97, while performing less well for OR & WA (F-measures ~0.69). The simplified geotechnical model had moderate to substantial prediction agreement with the DSM model when considered at the map unit level. For rapid, initial, liquefaction susceptibility assessment, the application of simplified geotechnical liquefaction criteria, with the addition of parent material and soil moisture status, to soil survey data is effective. For surface investigation in greater detail, the boosting model provides a more nuanced view of liquefaction susceptibility.
License
Resource Type
Date Available
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Non-Academic Affiliation
Subject
Rights Statement
Publisher
Peer Reviewed
Language
Replaces

Relationships

Parents:

This work has no parents.

In Collection:

Items