Date of Award

Spring 2022

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Forestry and Environmental Studies

First Advisor

Farrell, Justin

Abstract

The scale of wildfire destruction has grown rapidly in recent years, destroying nearly 25,000 buildings in the U.S. during 2018 alone. Recently, scientists have developed nationwide wildfire event data, making it possible to model wildfire effects on human migration. In Chapter 1, quarterly data on residential mobility from the Federal Reserve Bank of New York/Equifax Consumer Credit Panel are paired with building destruction counts from the U.S. National Incident Management System/Incident Command System database of wildfire events to evaluate the effects of wildfire destruction on migration probabilities in the U.S. between 1999 and 2018. Findings indicate that only the most destructive wildfires increased the probability of out-migration. The majority of destructive wildfires directly damaged very few structures, and, in these cases, residents largely remained in the area rather than mi- grating. These findings suggest that direct population displacement occurs in cases of high structure loss, but, when less destruction occurs, there is no evidence that residents are preemptively leaving high fire-risk regions to avoid future wildfires. How will now routine seasons of wildfire destruction affect human settlement patterns? Chapter 2 examines wildfire damage and post-fire building reconstruction among the approximately 23,000 buildings located in the burn footprint of the 2018 Camp Fire, one of the most destructive wildfires in U.S. history. I find that mobile homes, lower-value, and renter-occupied homes were significantly more likely to be destroyed compared to single-family homes, higher-value homes, and owner-occupied homes. These findings suggest that filtering of housing took place, wherein less valuable residential buildings were physically more susceptible to hazard damage. To map post-fire building reconstruction patterns, I train a support vector machine algorithm using data from the National Aerial Imagery Program and Microsoft’s Building Footprints database. Applying this algorithm to post-fire imagery, I identify which buildings had been reconstructed 20 months after the fire, and show that reconstructed buildings were more likely to be owner-occupied and had, on average, higher pre-fire value. These findings provide evidence for the theory of cost-burden climate gentrification, wherein wealthier households have a greater capacity to remain in a hazardous area than less affluent households. Following the Camp Fire, more than 56,000 residents evacuated from communities in the foothills of Butte County, California, known as “the Ridge.” While some remained in the area, others permanently relocated. Chapter 3 draws on 24 in- depth interviews with former Ridge residents who migrated away following the fire and describes the combination of drivers that influenced their decisions to move. While the fire acted as a precipitating event, I show how limited affordable housing was the primary constraint on many residents’ ability to rebuild or remain in the area. Additionally, the concurrent loss of embedded social networks on the Ridge and the presence of social ties elsewhere inclined many to leave. After the fire, environmental conditions continued to influence migration decision-making. Radical changes to the Ridge’s landscape and concerns over future wildfires pushed residents to leave, while the prospect of environmental amenities elsewhere was a consideration for many when they chose their new destination. For some, the ability to relocate to a new community where more residents shared their conservative political views was a draw, pulling them to Idaho. This constellation of push and pull factors suggests that post-disaster migration is not exclusively involuntary and shaped by material deprivation, as writing on disaster displacement has often assumed. In the case of the Camp Fire, post-fire mobility had elements of displacement, amenity migration, and political sorting. Despite unparalleled federal spending, government safety net programs were widely critiqued for failing to fully meet households’ material needs during the first year of the Covid-19 pandemic. Prior research on disasters has shown that informal modes of support emerge during times of financial hardship. Yet the use of formal government programs and informal social support are rarely considered relative to each other, resulting in an incomplete picture of how households navigate financial shocks. Drawing on results from a representative survey of 1,009 rural western U.S. residents during one of the first major waves of the pandemic, Chapter 4 shows that informal social support was much more extensively used than formal government programs during a period of widespread employment loss and financial stress. Person-to-person assistance, such as providing money to or sharing housing with friends or family, was the most widely practiced form of support. However, residents who reported experiencing negative impacts of the pandemic on their household finances were more likely to use both formal government programs and informal social support, indicating that both play an important role for households navigating disaster impacts.

Download

Share

COinS