Graduation Year

2026

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Geosciences

Major Professor

Yi Qiang, Ph.D.

Committee Member

Joni Firat, Ph.D.

Committee Member

Barnali Dixon, Ph.D.

Committee Member

Xiao Huang, Ph.D.

Keywords

Disaster Resilience, Human Mobility, Natural Disaster, Spatial Analysis

Abstract

Disasters increasingly affect population movement in the United States, but their relationships with mobility and migration at various spatio-temporal scales are not fully understood. Disaster-related movement may appear as short-term evacuation during hazardous events, long-term migration after major or repeated disaster exposure, or selective redistribution across socio-economic groups. Therefore, this dissertation aims to address three main questions: 1) What are the spatial patterns of evacuation, and how do evacuation rates vary across space during a major disaster? 2) To what extent are natural disasters associated with long-term human migration patterns in the United States, and how can their contributions be quantified relative to socio-economic drivers? 3) How are disasters related to bi-directional migration flows in the United States, and is there socio-economic selectivity in the migration flows? To answer the first question, human mobility data are used to examine evacuation-related movement during Hurricane Ian. To answer the second question, county-level population data, disaster data, and socio-economic variables are used to analyze the association between disasters and long-term migration patterns from 2000 to 2020. To answer the third question, origin-destination migration flows are used to examine aggregate migration volume and migrant composition changes from 2012 to 2021. These analyses employ various spatial analysis, statistical modeling and machine learning methods. The findings show that disaster-related population movement is temporally layered, spatially uneven, and socially differentiated. Short-term mobility responses during Hurricane Ian exhibit clear evacuation patterns, distance-decay effects, and spatial variation in responses to evacuation orders. For long-term migration, socio-economic conditions remain dominant predictors, but disaster-related factors still show measurable associations, varying by disaster type, severity, and frequency. Additionally, disasters may be associated with bidirectional migration flows and selective redistribution of population across income, age, education, and employment groups, particularly under extreme disaster conditions. Overall, this dissertation shows that disasters do not simply produce temporary displacement or uniform out-migration. Instead, they are associated with multiple forms of population change, including evacuation, relocation, and selective demographic change. By examining disaster-related movement across different scales, this dissertation provides a broader understanding of how disasters reshape population movement and offers insights for evacuation planning, recovery, and equity-centered climate adaptation.

Download

Share

COinS