Utilizing the Calm Before the Storm: An exploration of disaster mitigation in Princeville, North Carolina By Alexandra Stevenson July 2023 Director of Thesis: Cynthia Grace-McCaskey, Ph.D. Major Department: Anthropology ABSTRACT Climate change increases flooding events globally, and affluent populations typically have more access to recovery resources than marginalized communities. Literature regarding the recovery of Princeville, North Carolina (NC) from flooding caused by Hurricane Floyd in 1999 and Hurricane Matthew in 2016 states disaster management fails to address residents’ needs and concerns about future flood risks. Furthermore, current disaster anthropology literature emphasizes the need for disaster mitigation strategies that consider local context and knowledge. This research employs a political ecology lens to identify Princeville community members’ concerns about future floods and any challenges or barriers that may impact the extent to which decision-makers consider local knowledge and context when developing and implementing mitigation strategies that address their concerns. To do so, I utilized participant observation and conducted 20 semi-structured interviews with community members, local officials, and state and federal management agency personnel. Results suggest there is a limited capacity to adapt to future disasters, not only at the community level but within state and federal emergency management institutions as well. Utilizing the Calm Before the Storm: An exploration of disaster mitigation in Princeville, North Carolina A Thesis Presented to the Faculty of the Department of Anthropology East Carolina University In Partial Fulfillment of the Requirements for the Degree Master of Art in Anthropology By Alexandra Stevenson July 2023 Director of Thesis: Cynthia Grace-McCaskey, Ph.D. Thesis Committee Members B. Blakely Brooks, Ph.D. Ryan Schacht, Ph.D. © Alexandra Stevenson, 2023 ACKNOWLEDGEMENTS I would like to thank the local leaders and residents of Princeville, North Carolina for sharing their stories with me. I would also like to thank my family, friends, and colleagues who supported me through this research and provided the necessary support that I needed to succeed. Finally, I would like to acknowledge my mentors, past and present, who continue to equip me with the essential tools to be able to conduct research in the applied anthropological setting. TABLE OF CONTENTS LIST OF TABLES ............................................................................................................ vi LIST OF FIGURES ............................................................................................................ vii CHAPTER 1: INTRODUCTION ......................................................................................... 1 Statement of the Problem ........................................................................................... 1 Overview of the Research .......................................................................................... 2 Structure of Thesis ..................................................................................................... 3 CHAPTER 2: REVIEW OF THE LITERATURE ................................................................ 5 Chapter Overview ...................................................................................................... 5 Disaster Anthropology ............................................................................................... 6 Culture and Disaster ................................................................................................... 10 Environmental Justice and Political Ecology ............................................................ 15 CHAPTER 3: BACKGROUND OF FLOOD EVENTS AND ASSOCIATED DISASTERMITIGATION STRATEGIES ............................................................................ 21 Chapter Overview ...................................................................................................... 21 Location ............................................................................................................ 21 Historical Context ...................................................................................................... 22 History of Storm-related Flood Events ...................................................................... 24 Disaster Recovery and Mitigation ............................................................................. 26 Recovery and Mitigation in Princeville ..................................................................... 31 CHAPTER 4: RESEARCH DESIGN AND METHODOLOGY ......................................... 33 Chapter Overview ..................................................................................................... 33 Study Site ............................................................................................................ 33 Sampling Strategy and Description of the Sample .................................................... 35 Research Design......................................................................................................... 36 Data Collection .......................................................................................................... 37 Qualitative Data Analysis ......................................................................................... 41 CHAPTER 5: RESULTS AND DISCUSSION .................................................................... 44 Chapter Overview ...................................................................................................... 44 Presentation of Free List Data ................................................................................... 45 Concerns about Future Floods ................................................................................... 49 Discussion ...................................................................................................... 60 Trust in Decision-makers .......................................................................................... 63 Discussion ..................................................................................................... 68 Challenges and Barriers to Inclusive Disaster Mitigation ......................................... 69 Challenges and Barriers Identified by Community Members ...................... 70 Discussion ......................................................................................... 81 Challenges and Barriers Identified by Decision-Makers .............................. 84 Discussion ......................................................................................... 102 Emergent Themes Between Research Questions .......................................... 105 CHAPTER 6: CONCLUSION ............................................................................................. 107 Chapter Overview ..................................................................................................... 107 Research Questions and Results ............................................................................... 107 Recommendations for Policy and Future Research .................................................. 112 Dissemination of Thesis Results ............................................................................... 114 Limitations of This Research .................................................................................... 114 REFERENCES ............................................................................................................ 116 APPENDIX A: INFORMED CONSENT ............................................................................. 125 APPENDIX B: INTERVIEW GUIDE 1 ................................................................................ 126 APPENDIX C: INTERVIEW GUIDE 2 ................................................................................ 127 APPENDIX D: UMCIRB APPROVAL MEMO ................................................................... 128 LIST OF TABLES Table 3.1. Types of assistance when a major disaster declaration is declared ...................... 30 Table 3.2 Factors FEMA considers for assistance ................................................................ 31 Table 4.1 Princeville demographic information .................................................................... 34 Table 4.2 Linking interview questions to research questions ............................................... 40 Table 5.1 Free list word count: Community members .......................................................... 47 Table 5.2 Free list work count: Decision-makers ................................................................. 49 LIST OF FIGURES Figure 3.1 Location of Princeville in the Tar-Pamlico watershed ......................................... 21 Figure 3.2 Original levee with US 64 road alignments ........................................................ 23 Figure 3.3 Levee system post-Hurricane Floyd .................................................................... 25 Figure 3.4 Steps taken to recover from a hurricane .............................................................. 27 Figure 3.5 Sequence of assistance through FEMA’s Individual Assistance program ........... 29 Figure 5.1 Word cloud from community members’ responses .............................................. 46 Figure 5.2 Word cloud from decision-makers’ responses ..................................................... 48 Figure 5.3 Concerns community members have about the future risks of flooding ............. 50 Figure 5.4 Depiction of the interconnection between community members’ concerns and the reproduction of disasters ........................................................................................................ 62 Figure 5.5 Challenges and barriers identified by community members ............................... 70 Figure 5.6 Challenges and barriers identified by decision-makers ....................................... 84 Figure 5.7 Link between challenges and barriers, and the capacity of institutions .............. 105 Chapter 1: Introduction 1.1 Statement of the Problem Climate change increases the frequency and severity of storms, hurricanes, and other global disasters (Bin et al. 2007; Van Aalst 2006). Economically developed countries within the Global North, such as the United States (US), are assumed to have the resources to mitigate and recover from disasters with limited lasting impacts. However, this perspective overlooks marginalized communities within these countries that may not have access to the resources necessary to withstand a devastating disaster (Miller Hesed & Ostergren 2017). Historical, social, and political processes may influence vulnerability to environmental hazards (Barrios 2016; Maldonado 2016; Miller Hesed & Ostergren 2017). Socially and economically marginalized coastal communities typically suffer more losses in a disaster and have a harder time recovering from their losses when compared to more affluent communities (Greer 2020). As researchers continue to explore the impacts of disasters on racially and socioeconomically marginalized communities within the US, we gain insight into how climate change has become inherently unjust. For example, Miller Hesed and Ostergren (2017) explain that relatively affluent communities emit higher levels of greenhouse gases, which contribute more to climate change than other communities. Nevertheless, marginalized communities remain more vulnerable to the long-lasting impacts of climate change. Additionally, research shows that environmental injustice is perpetuated more in communities that have dealt with other types of injustice and inequity such as racism and discrimination (Miller Hesed & Ostergren 2017). Research also shows that Black Americans have historically contributed less than other Americans to climate change, yet they are more likely to suffer at a disproportion rate due to a lack of access to resources and little-to-no decision-making power throughout disaster mitigation processes (Congressional Black Caucus Foundation, Inc. 2004; Miller Hesed & Ostergren 2017). Decision-making bodies within state and federal government agencies tend to view culture and communities as static, unchanging entities, which may lead to disaster mitigation strategies that do not account for local context and knowledge. Even so, evidence suggests that policymakers, and in turn policies themselves, constantly influence and reshape culture and communities around the world (Maldonado 2016). Without the consideration of local context, knowledge, and concerns, disaster mitigation strategies will only perpetuate environmental 2 injustice. Moreover, vulnerable communities remain at risk for destructive disasters in the future. For this reason, throughout the entirety of this thesis, “ineffective disaster mitigation” will be used to encompass mitigation strategies that do not consider and address the concerns community members have about future environmental risks. Princeville, North Carolina (NC), is one such community facing long-lasting impacts from disasters and ineffective disaster mitigation strategies. This thesis explores the disaster mitigation strategies in Princeville, the first town chartered by Black Americans in the US. Since its incorporation, Princeville has experienced many flood events. Even with assistance from state and federal agencies, such as North Carolina Emergency Management (NCEM) and Federal Emergency Management Agency (FEMA), the residents of the town continue to experience hazardous conditions with the constant threat of a devastating disaster. This leads many community members to rely on external, non-government agencies for assistance with recovery. 1.2 Overview of the Research Data for this thesis were collected between May and December 2022. The goal of this research is to identify Princeville community members’ concerns about the future risk of flooding and any challenges or barriers that may impact the extent to which mitigation strategies consider and address these concerns. To achieve this goal, I developed three key research questions: RQ1: What are the main concerns community members have about the future risks of flooding? RQ2: Do community members trust that decision-making bodies—including Princeville’s Board of Commissioners and state and federal agencies—will consider these concerns when implementing post-disaster policies? RQ3: What factors impact the extent to which local officials and state and federal agencies utilize local context and knowledge when implementing disaster mitigation policies? To answer these questions, I utilized participant observation by attending Princeville’s Board of Commissioners meetings from May 2022 to December 2022, and Farmer’s Markets on the second and fourth weekends of the month from August 2022 to November 2022. In addition, 3 I conducted 20 semi-structured interviews with Princeville community members, local leaders, and state and federal agency personnel. I was introduced to the community via my thesis advisor, who established rapport with community members and local leaders during previous community engagement work in the area. Additional participants were selected with the use of snowball sampling, a non-probability sampling strategy (Bernard 2011). When possible, interviews were recorded and transcribed verbatim to maintain participants’ true meaning and context (Bernard 2011). Transcriptions were then analyzed through an iterative process that used grounded theory to allow for the emergence of relevant themes (Bernard 2011). Previous recovery literature for Princeville explicitly states that disaster mitigation thus far has not considered the concerns of those impacted by flood events (Princeville Recovery Plan 2018). Therefore, this research aims to encourage decision-makers to further identify challenges and barriers that negatively impact their ability to consider and address community members’ concerns when developing new disaster mitigation strategies. The results from this thesis align with relevant literature that suggests that effective disaster mitigation must include strategies that not only address community members’ concerns, but they must also provide the stability necessary for Princeville to grow economically and allow for more people to learn about the historical significance of the town. 1.3 Structure of Thesis In Chapter 2, I present relevant literature from the field of disaster anthropology that informed the initial development of this thesis. This includes a brief overview of the field itself, as well as the theoretical frameworks that I utilize to examine the disaster mitigation strategies in Princeville. I introduce research that highlights the culture-disaster relationship, and I discuss key concepts in disaster anthropology such as resilience, vulnerability, and adaptive capacity. Additionally, I present environmental justice and political ecology frameworks frequently utilized to examine differential power relationships throughout disaster response, recovery, and mitigation. In Chapter 3, I present background information on Princeville including the historical significance of the town’s founding. I further discuss the impacts of Hurricane Floyd in 1999 and Hurricane Matthew in 2018. I also introduce typical disaster recovery and mitigation strategies 4 implemented by state and federal emergency management agencies in North Carolina. I close the chapter with an explanation of the state centric approach that was developed by the North Carolina Department of Public Safety (NCDPS) and the Federal Emergency Management Agency (FEMA) and implemented in Princeville throughout the data collection phase of this research. In Chapter 4, I introduce the main research questions and associated goals that motivated this research. I discuss the sampling strategy and introduce the participants. I review the qualitative data collection and analysis methods that I used and provide context for the reasoning behind these methodological choices. In Chapter 5, I present the data associated with each research question and discuss the ways in which the results relate to the literature presented in Chapter 2. Finally, Chapter 6 contains the conclusion of this thesis as well as any limitations for this research. I also introduce potential areas for future research in similar realms of disaster anthropology. Chapter 2: Review of the Literature 2.1 Chapter Overview In this thesis, I explore the disaster mitigation process in Princeville, North Carolina (NC) to identify community members' concerns about future flood risks and how current strategies address those concerns. Furthermore, I examine state, federal, and non-profit entities' roles in disaster recovery and mitigation processes. To understand disaster mitigation from an anthropological perspective, I discuss three key themes present in the current literature: (1) disaster anthropology, (2) culture and disaster, and (3) environmental justice and political ecology. Examining disaster through an anthropological lens provides policymakers with the information necessary to consider cultural practices and the well-being of impacted individuals throughout the mitigation process. Current literature provides case studies that explore the link between culture and disaster, which is necessary when attempting to understand the impacts disasters have on a community or culture. Anthropological investigation of recovery and mitigation processes provides insight into the practices that may perpetuate environmental injustices due to differential power relationships between policymakers and community members. After situating the anthropological investigations of disasters and describing the link between culture and disaster, I emphasize the roles environmental justice and political ecology play in examining disaster mitigation strategies. In this section, I describe several injustices that are perpetuated throughout decision-making processes in marginalized communities. I explain the need for the co-production of knowledge, policies, and mitigation strategies to address the root causes of environmental injustice. This chapter aims to orient my current research through anthropological epistemologies and frameworks, to provide insight into the urgent nature of climate change research. Furthermore, I hope to emphasize the need for truly collaborative research between policymakers, community members, and other stakeholders to achieve effective and successful disaster mitigation that promotes the capacity to adapt and withstand the impact of future disasters. 6 2.2 Disaster Anthropology Anthropology is broadly defined as the study of humankind, both past and present, and is divided into four subfields: biological, cultural, linguistic, and archaeology. This research utilizes frameworks present in environmental and disaster anthropology which are subdisciplines of cultural anthropology (McGee and Warms 2020). Anthropologists have unique ways of understanding disasters and lived experiences of individuals impacted by disasters through holistic approaches and methodologies, which I will discuss further in chapter 4. Environmental anthropologists are concerned with the human-environment relationship and how culture is shaped by the environment and vice versa. Furthermore, disaster anthropologists are concerned with the human-disaster relationship; specifically, how culture is affected by disasters, how humans intensify impacts brought on by disasters, the causes of specific disasters, etc. (Kopnina & Shoreman-Oiment 2017). This review sheds light on the ways in which disaster mitigation efforts may inadvertently exacerbate environmental justice concerns and perpetuate the unequal impacts of disasters. Disasters are unique to each cultural and environmental context, making it difficult to define “disaster.” Numerous disaster anthropologists argue that disasters should not be considered “natural” because of the intricate nature of the human-environment relationship; rather, disasters are a result of the interaction between environmental hazards and human populations and development (Oliver-Smith 1999; Barrios 2016; Graef 2017). Disasters are characterized by external variability and internal complexity. External variability refers to objective phenomena (natural or technological) that cause disasters and produce various levels of impact, whereas internal complexity refers to the ways in which unique compounding processes or events intersect and affect the level of impact within social, environmental, cultural, political, physical, or technological domains (Oliver-Smith 2019, p. 30). There is a lot to learn about societies and groups within the context of disasters because they bring about rapid social and environmental change. Crises and disasters allow policymakers, practitioners, and researchers to evaluate political and social structures and identify issues with current disaster risk-reduction strategies. Furthermore, researchers can explore cultural and societal structures through the lens of crisis and change, providing special opportunities for the development of social theory and applied methodological approaches (Faas & Barrios 2015, p. 7 290). For example, political ecology is utilized as a tool to understand disasters as impending historical processes brought on by differential power relationships between disparate groups of people within their unique environmental contexts. Although environmental risk is produced by culture, researchers identified the numerous strategies implemented by traditional and indigenous groups to reduce the risk of disasters. This is a basic example of a cultural adaptation known as disaster risk reduction (DRR). Many traditional forms of DRR were undermined by colonial expansion and most modern attempts fail to address the root causes of disasters (Faas & Oliver-Smith 2016; Hope et al. 2019). While DRR is used in theoretical research, disaster anthropologists utilize it as a framework to develop policies that result in a decline in economic damages and losses of life by reducing vulnerability and achieving and maintaining resilience (Faas & Oliver-Smith 2016; Barrios 2016). Vulnerability, Resilience, and Adaptive Capacity Anthropologists utilize vulnerability, resilience, and adaptive capacity to understand the best mechanisms for coping and adapting to future environmental threats (Young 2010; Newman et al. 2014). Vulnerability, defined as the potential to experience harm, damage, or loss due to climate change (Turner 2006, p. 28; Mason & Riggs 2019, p. 10), is used as a tool for identifying a community’s susceptibility to harm that arises with the absence of the capacity to adapt (Adger 2006). To develop successful disaster mitigation strategies, policymakers must understand the drivers of vulnerability in a particular community. For example, historical, political, spatial, and temporal factors regularly influence vulnerability (Ribot 2013; Mason & Riggs 2019). Nonetheless, it is not simply enough to understand this for one community; researchers and practitioners must understand the drivers of vulnerability for all surrounding communities to ensure that measures taken to lessen vulnerability in one place do not, inadvertently, promote vulnerability in another (Mason & Riggs 2019). Although the term “resilience” has many different interpretations, the field of disaster anthropology views resilience as the ability of a community to withstand the shock of a disaster or flood event and bounce back to a “new normal” in a reasonable amount of time (Barrios 2016; Princeville Recovery Plan 2018). An emphasis is placed on the ability to establish a “new normal” because returning to the original state may only perpetuate environmental inequities. The common misinterpretation of this term by policymakers allows infrastructure to hinder the 8 ability to achieve resilience within marginalized communities (Barrios 2016). It is important that researchers clearly define resilience because misuses of this term disregard the fact that marginalized communities are already resilient in other ways. Moreover, communities that are under constant environmental threat continue to return and rebuild despite the challenges brought on by climate change impacts (Barrios 2016). Adaptive capacity refers to the ability to prepare for, respond to, or minimize the destructiveness of climate change impacts and is often a measure of vulnerability; meaning that a community with greater adaptive capacity is theorized to have a lower vulnerability (Young 2010; Newman et al. 2014; Miller Hesed & Ostergren 2017). It is in this same way that I define successful disaster mitigation: mitigation strategies are only successful if they promote adaptive capacity and lessen the community’s vulnerability, without increasing the vulnerability of any nearby communities. This includes the ability to anticipate, respond to, and recover from disasters while simultaneously taking advantage of new opportunities for growth (Cinner et al. 2018). Adaptive capacity is often discussed in terms of communities that are impacted by environmental hazards or disasters; however, it is also important to discuss the adaptive capacity of the institutions aiding in disaster mitigation strategies. For the purposes of this discussion, institutions encompass all state, federal, and non-profit organizations that attempt to assist communities in reducing their vulnerability, recovering from a disaster, or building the capacity to adapt to future storms. O’Riordan and Jordan (1999) state that while institutions have the means to hold society together, there is a fundamental gap in the interpretations and resolutions of climate change between impacted communities and the institutions themselves. In an attempt to mitigate disasters in the long run, institutions may increase a community’s short-term vulnerability to environmental hazards (Adger 2000). Nonetheless, institutions are an important aspect of governing the human-environment relationship in the face of climate change (Brown et al. 2010). It is important to understand that adaptive capacity requires a willingness to transform current responses to disaster within both the affected communities, and the institutions that assist in recovery and mitigation efforts. However, this goal is achieved through much more than standard disaster mitigation strategies. Cinner et al. (2018) discuss five key domains through 9 which adaptive capacity could be established: assets that people can utilize in times of need, flexibility to adjust and change current mitigation strategies, the capacity for collective action, knowledge to respond to climate change, and the agency to decide to change or not (p. 2–3). People tend to adapt more easily in times of stress and disaster when they have assets they can utilize in times of need. The assets required to achieve adaptive capacity can take the form of financial, technological, or service resources (Brooks et al. 2005). Service assets refer to public resources such as hospitals, and for the purposes of this review, the institutions that assist in disaster recovery and mitigation. Flexibility is established through a continuous review of current mitigation strategies with frequent adjustments when needed (Ciner et al. 2018). Establishing flexibility in the face of the ever-changing impacts of climate change will allow for more opportunities to build and improve the adaptive capacity of communities and institutions. Learning requires access to information, therefore, gaining the level of knowledge necessary to respond to disasters and build adaptive capacity, requires institutions at the local, state, and federal levels to provide supporting processes that enable community members to recognize change, identify the causes of environmental change, and assess different levels of responses (Cinner et al. 2018). Lastly, agency refers to the power and freedom to mobilize all components of adaptive capacity to actively shape future outcomes. This is achieved through the inclusion of local knowledge, context, skills, and management throughout mitigation efforts. In this realm institutions can empower communities through participatory processes, the co-production of disaster policy, and the removal of barriers that hinder people’s ability to exercise agency (Cinner et al. 2018). The five domains discussed by Cinner et al. (2018) provide insight into the ways in which institutional adaptive capacity influences and shapes the adaptive capacity of communities themselves. Oriangi et al. (2021) explore perceptions of resilience to climate-induced disaster in the Mbale municipality of Uganda. In this area, climate change has led to an increase in the frequency and severity of droughts, which lead to a need to enhance resilience. The authors utilize the Community Based Resilience Analysis Approach to analyze components of resilience that are perceived as important and the extent to which these components have been achieved. The community members identified access to education, healthcare, employment, and peace and security as the most important aspects of resilience. Urban areas within developing countries are likely to suffer the worst from the impacts of climate change due to the prominent level of 10 poverty, the inadequate socio-economic infrastructure, and the limited-service programs available. For example, the respondents who lived in marginalized parts of the municipality reported a decreased level of resilience, whereas other residents from more affluent areas reported an increased level of resilience. To achieve an equal level of resilience in municipalities like Mbale, practitioners and policymakers must close the gap between local knowledge and global frameworks. The Community Based Resilience Analysis approach allowed researchers to understand the most and least achieved resilience components, which aids in identifying and prioritizing sectors that have not achieved enhanced resilience. This also provides practitioners and policymakers with insight into the different perceptions held by community members of different ages and genders (Oriangi et al. 2021). Barrios (2014) uses a case study of two post-Hurricane Mitch housing reconstruction sites in Southern Honduras—Limón de la Cerca and Marcelino Champagnat—to demonstrate the need for resilience-cultivating relationships. In this article, Barrios explains that anthropologists should reframe the way they think of communities. Current research depicts communities as fixed, unchanging entities when they are, in fact, the exact opposite. Communities are shaped and reshaped by their relationships with government and non-government agencies and the environment (Barrios 2014; Maldonado 2016). Therefore, collaboration is one of the keys to successfully improving resilience. The need for collaboration between state and local officials is further emphasized when Barrios (2016) asks an important question regarding resilience in post- disaster literature: rebuilding back to what? Communities that are vulnerable to increasing flood risks and coastal hazards are often politically and economically disadvantaged, so rebuilding the community to post-disaster times would mean returning to the same disadvantaged state they were in before the disaster occurred. The article also explains that difficulties in the recovery process are not caused by a lack of resilience on the part of community members, but rather by post-disaster decision-making processes that do not rely on the useful insight of community members (Barrios 2016). The literature presented above was chosen to depict how useful it is to examine disasters and their impacts through an anthropological lens. 2.3 Culture and Disaster 11 Humans interact with nature in a variety of unique ways. For instance, subsistence strategies are developed through a set of socially constructed and meaningful behaviors (hunting, fishing, trading, building homes, etc.). Furthermore, human interactions with nature contain deep-rooted and persistent histories that diversely influence the culture-disaster relationship. Before colonization and globalization, the unique knowledge and interactions each group had with their environment allowed for some level of adaptation to hazards. Physical disturbances were perceived as a part of their existence and experience rather than as disastrous events (Hoffman 2017). Therefore, researchers must understand each disaster through the context of the community or culture that is impacted. Humans inhabit four distinct types of environments; basal terrain, modified, built, and culture. The basal terrain refers to the physical environment in which people live, however, people rarely live in a place without altering it in some way. Nevertheless, alterations to the modified environment can often increase vulnerability to environmental hazards. Many disasters are primarily caused by the built environment, or the structures erected by humans. The built environment encourages people and communities to relocate or expand (Hoffman 2017). Hoffman (2017) argues that the core of each disaster is culture, the fourth environment in which humans inhabit. Culture dictates how the physical environment is utilized, modified, and built upon, which supports the claim that all disasters are caused by humans. As discussed above, without the presence of humans, environmental phenomena are just naturally occurring events with no disastrous impacts. However, it is important to note that disasters tend to manifest in areas with historically produced patterns of vulnerability—the core element of disasters—and conditions the behaviors of individuals and organizations (Hoffman 2017). In this sense, disasters are made through miscalculations, misperceptions, lack of foresight, etc. Even though disasters occur through a pattern of events and processes within specific cultural contexts, a disaster in one place can impact, or even trigger a different disaster, in another place (Faas & Oliver-Smith 2016). The same is said for adaptations to disasters; an adaptive measure in one area can increase the vulnerability of another area. Disaster anthropologists identified key factors to consider when conducting research following a disaster. First, one must consider the differential impacts of a disaster: was the disaster in question small-, medium-, or large-scale? What are the demographic characteristics of 12 the impacted community (race, gender, socio-economic status (SES))? Which impacts are being felt most by survivors—financial, physical, mental, or emotional (Cajilig 2017; Barrios & Batres 2019)? Second, the perceptions individuals have of the environment, risks, and disasters dictate how the environment is used and decisions are made (Hoffman 2017). Lastly, it is important to consider how sense of place and place attachment influence the way people perceive disasters related to their environment (Stedman 2003; Greer 2020). Perception of Risk Individual cultural groups have unique perceptions of the human-environment relationship. For example, Western societies tend to form finite distinctions between the areas, or structures, in which they live and the natural environment in a way that indigenous or traditional societies typically do not (Hoffman 2017). In addition, the same can be said about people’s perceptions of risk and disasters; local risk perception and interpretation are influential aspects of disaster risk reduction (Faas & Oliver-Smith 2016). Disaster anthropologists emphasize that risk and danger are not the same. Danger refers to a hazard that continuously exists in a particular environmental setting, whereas risk acts as a calculation that affects the extent to which people or communities acknowledge and address existing dangers. Hoffman (2017) likens this phenomenon to a water faucet: people choose whether to open the faucet all the way, which would consist of acknowledging all or most of the dangers around them; they can open the faucet a small amount by only acknowledging some of the dangers around them and ignoring others; or people can keep the faucet completely closed by not acknowledging any existing hazards. The calculation and perception of risks are linked to the social, cultural, and political norms of the community. For example, Eriksen and Simon (2017) explain that socio-historical and political factors may influence the lived experiences and perceptions of individuals and communities, further shaping their perceptions of risks and hazards. This perception then influences the way individuals and communities prepare for and respond to disasters, which directly affects levels of vulnerability. Decision-making and planning are linked to people’s perceptions of climate risks. Lazrus (2015) utilizes the Cultural Theory of Risk (CTR) as a framework to understand why people in Tuvalu—a low-lying Pacific Island nation—are concerned about specific risks as well as how they believe risks should be addressed. CTR prioritizes the role of culture in disaster risk 13 perception by identifying how people believe society should function and how society and nature should interact, then identifying threats to those held beliefs. Lazrus (2015) argues that adaptive decisions that address community risk concerns, worldviews, knowledge, and experiences are likely to be more successful when compared to efforts that falsely claim community engagement but do not include local context and knowledge (52). Worldviews emerge as distinct cultural priorities and preferences, which influence the diverse ways people perceive risk (Lazrus 2015, p. 53). Lazrus (2015) describes four idealized worldviews about how societies should function and interact with nature: the individualist, the fatalist, the hierarchist, and the egalitarian. The individualist worldview is characterized by weak social bonds and a minimal social structure. This worldview perceives nature as benign and able to adjust and adapt to human alterations. The fatalist worldview has weak social bonds and a stratified society governed by a set of rules. Unlike the individualist, however, the fatalist acknowledges the unpredictable nature of the environment. The hierarchist worldview has strong social bonds that are governed by numerous rules. This worldview characterizes nature as manageable and able to handle some human-induced changes. Additionally, this worldview expects scientists and practitioners to identify the point at which the environment can no longer handle anthropogenic alterations and plan accordingly. Lastly, the egalitarian worldview consists of strong social ties between people with internally agreed-upon rules. The egalitarian characterizes the nature-culture relationship as fragile, meaning the environmental balance can be easily tipped (Lazrus 2015, p. 54). Determining the worldview of affected communities will allow researchers to gain a deeper understanding of the way environmental risks are perceived in specific cultural contexts. Community perceptions, observations, and priorities are not sufficient solutions to climate change impacts on their own, but they are necessary when implementing adaptive strategies for marginalized communities. Identifying and analyzing people’s perceptions of environmental risks can also shed light on systemic vulnerabilities that leave particular groups of people disproportionately affected by the impacts of climate change (Lazrus 2015). Lazrus (2015) explains that the ways in which people perceive risk influence the adaptive actions they take to address the issue. 14 Oriangi et al. (2021) conduct a unique study that compares different sectors’ status of marginalization and perceptions of resilience in Mbale Uganda. The authors place an emphasis on identifying different perceptions of risk among gender categories, as well as age groups. The findings support the claim that marginalized communities’ perceptions and knowledge should contribute to the development of policies, and the different perceptions among different people within the communities should be considered (Oriangi et al 2021). Doing so may allow for improved responses to environmental hazards and disasters and may allow for the development of mitigation strategies that promote equity and sustainability, especially in marginalized communities. Sense of Place and Place Attachment Current research in the field of disaster anthropology shows that people may not want to move away from risk-prone areas despite being able to do so; recent literature identifies sense of place and place attachment as two contributing factors to this trend (Stedman 2003; Greer 2020; Swapan & Sadeque 2020). Place attachment is defined as the subjective bonds people develop with the particular places that they deem important or meaningful (Greer 2020, p. 307). Meaning can be derived from social, spiritual, ecological, or cultural values. Sense of place is a seemingly rhetorical feeling that people gain when a large portion of their emotional real estate and self- identity is tied to an area or community (Phillips 2012; Grace-McCaskey et al. 2021; Adkins et al. 2021). Both concepts influence a person’s decision to migrate to an area of lower risk or stay in an environmentally threatened place (Greer 2020). Disasters are place-based phenomena that alter the physical and social landscape, which in turn, challenges how people understand where they live and their bonds with their environment. The physical environment can play a role in the development of sense of place and place attachment, which makes understanding these concepts crucial to successful disaster mitigation that supports community members. The place, or object of attachment, encompasses both physical and social characteristics that influence the development of place attachment (Greer 2020). The physical characteristics are tangible components that make the place practically desirable such as jobs, transportation, resource access, etc. Social characteristics include the people who live in the area and the social networks people develop and share within the community. Developing relationships with fellow community members has a positive relationship with the development of place attachment (Greer 15 2020). Additionally, an individual’s perception of place is affected by their individual experiences in that place. For example, Greer (2020) states that lower-income individuals may have a stronger attachment to place because of the social ties formed through shared coping strategies to economic stressors. The same can be said regarding coping strategies developed by a marginalized community threatened by environmental hazards and disasters (Swapan & Sadeque 2020). Marginalized communities typically suffer the most losses in a disaster because they do not have access to the resources needed to recover. High levels of place attachment may lead people to underestimate the potential for damage and harm caused by disasters and influence peoples’ decisions to return to an area after being temporarily displaced. Furthermore, individuals with elevated levels of place attachment tend to suffer increased distress after a disaster. This may affect one’s proclivity to resist change and push to restore the place to the way it was before the disaster (Greer 2020). 3.4 Environmental Justice and Political Ecology Environmental justice refers to “the fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income, with respect to development, implementation, and enforcement of environmental laws, regulations, and policies” (EPA 2023, para. 1). This includes equitable protection from environmental and health hazards, as well as equal access to decision-making processes regarding the environment in which people live and work (EPA 2023). However, even though the Environmental Protection Agency (EPA) established an environmental justice department, impacts brought on by disasters are still exacerbated in areas with existing vulnerabilities, such as poverty, poor infrastructure, and limited access to resources. Furthermore, equity is defined by the US Congress and House Committee on Homeland Security as “recognizing and responding to different needs to ensure everyone can recover with dignity” (US House of Representatives 2021, p. 1). Hoffman (2017) labels inequity in the disaster recovery and mitigation sectors as unequal shares of safety, where less culturally valued segments of the population experience disproportionate impacts from disasters. Less culturally valued segments of the population could be groups based on gender, class, or ethnicity, and are made more vulnerable by their “place in the environment, economics, politics, social structure, ethos, and prejudice” (Hoffman 2017, p. 196). Similarly, Emrich et al. (2021) label social vulnerability—a set of pre-existing social conditions at the individual or 16 community levels that hinder disaster preparedness, recovery, and management efforts (p. 3)—as a factor that influences the pace and progression of post-disaster recovery processes. Miller Hesed and Ostergren (2017) explain that climate change is not equally contributed to by all communities, not all groups of people are equally exposed to the impacts of climate change, and not all community members are equally empowered to contribute to post-disaster decision-making processes. This supports the idea that climate change is inherently unjust. For example, relatively affluent communities emit higher amounts of greenhouse gases, which, in turn, contribute to climate change more than others; however, marginalized communities remain more vulnerable to the long-lasting effects of climate change. Environmental injustices, or “the disproportionate exposure of certain populations to environmental hazards” (Maantay 2002; p. 161), are perpetuated more in communities that have dealt with other types of inequalities in the past, such as racism and discrimination. Research shows that Black Americans have historically contributed less than other Americans to climate change and continue to do so. For instance, Black households emit twenty percent less carbon dioxide than White households, yet they are more likely to suffer from environmental threats at a disproportionate rate due to a lack of resources and little-to-no decision-making power when it comes to disaster mitigation (Congressional Black Caucus Foundation, Inc. 2004; Barrios 2016; Miller Hesed & Ostergren 2017). One of the key factors contributing to environmental injustice is the lack of inclusion of community members in decision-making. Excluding community members from decision-making processes will only perpetuate environmental injustices and communities will remain at risk for destructive environmental events in the future (Miller Hesed & Ostergren 2017; Maldonado 2016). Miller Hesed and Ostergren (2017) introduced an environmental justice workshop to communities on the coast of Maryland. Their goal was to create space for multidisciplinary discussions on justice and adaptation that led to recommendations for increasing justice in this area. They engaged community members in hopes of increasing shared knowledge and trust in researchers and practitioners. Like Maldonado (2016) and Oriangi et al. (2021), Miller Hesed and Ostergren (2017) argue that community members should be included as equal stakeholders with equal decision-making power. 17 Through an exploration of post-disaster contexts in New South Wales, Australia, Matthews et al. (2020) found that marginalized groups are disproportionately impacted by extensive flooding events and are at a greater risk of displacement and poor mental health. The authors aim to establish the ways, if any, social capital provides benefits to marginalized groups in post-disaster contexts. Matthews et al. (2020) utilize a cross-sectional survey to quantify the associations between various flood impacts, individual social capital, and psychological distress. The authors discuss exposure to a flood event harms the mental health of the entire community, but those with more social capital are not as severely impacted. Aboriginal and financially disadvantaged respondents reported lower levels of social capital than the general community respondents, yet informal social connectedness and belonging were crucial factors for all participants and were associated with risk reduction (Matthews et al. 2020). Furthermore, anthropologists Reinke and Eldridge (2020) explore the role of bureaucratic violence in perpetuating environmental injustices in post-disaster contexts. Bureaucratic violence is a form of everyday violence in which lengthy bureaucratic processes and “red tape” perpetuate environmental inequities (Reinke and Eldridge 2020). For instance, the temporality of bureaucratic violence in times of crisis is a form of bureaucratic violence that exacerbates suffering. The authors utilize ethnographic methodologies to explore the perception of hurricane recovery in Cumberland County, NC, an area along the Cape Fear River that was impacted by Hurricane Matthew in 2016 and Hurricane Florence in 2018. This case study identifies the ways in which bureaucratic decisions may impact disaster survivors. Moreover, bureaucracies, such as local, state, and federal disaster recovery and mitigation agencies, are interactive socio-cultural entities that influence and shape the everyday realities of community members. When decisions are made without including local perceptions, new knowledge is produced, and power is shifted away from the impacted individuals (Reinke and Eldridge 2020). Unfortunately, broader institutions, such as the Federal Emergency Management Agency (FEMA), often shape recovery processes. According to Reinke and Eldridge (2020), FEMA representatives are often not well-versed in the needs of local people, and, all too often, survivors are left with unwieldy and time-consuming paperwork that they do not understand. Any money received by survivors for repairs, clothing, food, shelter, etc. must be tracked carefully with receipts. If proper steps are not followed by the survivors, then they may have difficulty 18 receiving funds in the future (Reinke and Eldridge 2020). Emrich et al. (2021) state that agencies may not account for social vulnerabilities, which perpetuates inequity in disaster recovery, response, and mitigation processes. Thomas (2022) provides further evidence of this claim through their comparison of flood-mitigation strategies implemented in marginalized and wealthy communities. For example, small towns along the Atlantic and Gulf coasts of Florida— whose population is 90% White Americans despite the White population of Florida only being 53%—have received 10% of the $58 million that has been allocated to the entire state for elevations (Thomas 2022). When emergency management agencies prioritize wealthy, predominantly white communities, environmental justice concerns grow within marginalized communities. Disaster recovery and mitigation strategies that do not account for social vulnerabilities reduce the pace and progression of recovery and mitigation efforts, which leaves little room for long-term disaster mitigation strategies that consider community members’ concerns and the future trends of climate hazards (Reinke and Eldridge 2020; Emrich et al. 2021). Political Ecology Political ecology is a common theoretical framework implemented by disaster anthropologists to identify inequities related to the impact, response, and recovery of a disaster (Barrios 2016). Peluso and Watts (2001) explain that political ecological research focuses on differential power relationships to provide “tools for thinking about conflicts and struggles engendered by the forms of access to, and control over, resources” (p. 24-25). Political ecology has shaped the understanding of disasters as historical processes that involve co-constitutive interactions and relationships between people and the material environment in which they live. This theoretical framework aligns the goals of disaster anthropology with the core tenants of cultural anthropology by placing a focus on core interpretations of cultural interactions with disasters (Faas & Barrios 2015). Maldonado (2016) acknowledges how historical processes, differential power relationships, and disasters shape and reshape culture. For instance, policymakers and media outlets reshape culture in similar ways; with the development of catchphrases that become “new truths” that overshadow and overlook the survivors of a disaster and its underlying causes. This is an important aspect of political ecology because catchphrases and headlines act as imposed 19 boundaries and definitions that create or perpetuate harmful power relationships. This can also act as a barrier to achieving equity. Throughout the case studies presented by Maldonado (2016), one theme remains present, victim blaming. Victim-blaming not only places the burden and responsibility of the disaster on the victims but also deprives survivors of agency and social cohesion. Maldonado utilizes political ecology to argue for a better understanding of culture as fluid and ever-changing. Similarly to Oliver-Smith (1999; 2019) as well as other sources, she acknowledges the ways in which culture is influenced by differential power relationships and long-lasting historical processes. In 2017, Miller Hesed and Ostergren assessed the vulnerability of three Black communities along the coast of Maryland. This evaluation required the identification of each community’s sensitivity and risk of exposure to disaster impacts, as well as an analysis of their adaptive capacity. Many of the environmental justice issues in these communities were not properly addressed because of the different perspectives held by community members, decision- makers, and other stakeholder groups regarding the timeframe of recovery, vulnerability, environmental conservation, and knowledge. Policymakers often have different goals for disaster mitigation than those of community members. This creates obstacles to participation, such as a lack of financial resources, historic prejudices, and little-to-no access to decision-making processes, which exacerbates issues related to environmental injustice. If true collaboration with community members is prioritized, there will be more successful disaster mitigation strategies put in place (Miller Hesed & Ostergren 2017). Co-Production of Knowledge and Disaster-Risk Management Strategies The literature presented above suggests the importance of considering local knowledge and context within the realm of policy and strategy development. However, researchers must strive for the co-production of disaster-risk management strategies and related knowledge. Co- production is broadly defined as different people deliberately working together to achieve a common goal (Bremer and Meisch 2017). Environmental justice frameworks expand on this simple definition, stating that co-production is the “process by which established hierarchical, power relationships between experts and lay communities, and service providers and users, are broken down, so that citizens, experts, and decision-makers are more equal partners in co- producing programming and policy” (Hope et al. 2019, p. 1159). Some researchers advise that 20 current attempts to co-produce disaster-related knowledge focus too much on the individual and less on the systemic, structural changes that need to take place to promote equity in decision- making and address the root causes of poverty, climate change, and vulnerability (Hope et al. 2019; Maldonado 2016). Hope et al. (2019) argue that power, wealth, energy, and information drive unequal access to the resources necessary to mitigate environmental hazards and disasters. Because of the rapid social change associated with disasters, the recovery and mitigation processes provide unique opportunities to address the root causes of climate change inequity. It is in this way that state, federal, and non-governmental agencies incite social change and perpetuate environmental injustices that are felt by so many marginalized communities. In this chapter, I presented a variety of literature to demonstrate the importance of exploring disaster mitigation through the lens of anthropology. Anthropologists can define key terms and concepts holistically while considering culture and community. However, scientists and practitioners cannot tackle extreme climate threats on their own. The case studies described above demonstrate the need for collaborative mitigation strategies that consider local context and knowledge. Furthermore, successful disaster mitigation requires the co-production of knowledge and policies, including all community members, resource users, and stakeholders. Disaster research cannot and should not be done in an academic vacuum. Disasters influence culture and vice versa; therefore, disaster-related policy influences culture and how communities interact with their surrounding environment. In the next chapter, I discuss the role of the above concepts within the context of Princeville: the first town chartered by Black Americans in the US. Chapter 3: Background of Flood Events and Associated Disaster Mitigation Strategies 3.1 Chapter Overview To understand the current concerns and dilemmas regarding the recovery of Princeville, North Carolina (NC), it is important to comprehend the historical processes that influence the town’s current vulnerability, resiliency, and hazard mitigation strategies. This chapter illustrates how the location of Princeville and its historical context have led to increased vulnerability to the impacts of flood events. Furthermore, I describe the role that state and federal institutions have in the disaster recovery and mitigation processes at the national level. Lastly, I explain the steps that local, state, and federal decision-makers have implemented in Princeville to mitigate future disasters. Data for this chapter were provided by several government websites and planning documents created to promote recovery and resilience in Princeville. 3.2 Location A floodplain is a flat, or nearly flat, parcel of land that experiences flooding due to its proximity to a lake, river, or stream (Alvarez 2010). Princeville, NC is situated in a floodplain along the Tar River in Edgecombe County (Figure 3.1). While floodplains often provide many Figure 3.1. Location of Princeville (star) in the Tar-Pamlico watershed (Map provided by John E. Sabin III ) 22 environmental benefits, in this context, an unmanaged floodplain has led to continuous environmental risks for the residents of Princeville. When a human settlement is established inside a floodplain, it prompts the development of flood management strategies such as levees and dikes (Opperman et al. 2017). I will discuss this further in sections 3.5 and 3.6 of this chapter. 3.3 Historical Context During the Civil War, Union Troops occupied a small area of land along the southern part of the Tar River in Edgecombe County, NC. Many enslaved people moved to the area in search of freedom and federal protection. After the war ended, in 1865, and enslaved peoples were freed, Black Americans began to settle in swampy lands across the Tar River, just south of the nearby town of Tarboro. Those who settled in the area began to erect structures and named the area Freedom Hill, after the area of land where Union soldiers announced the Emancipation Proclamation (Mobley 1986). The deed to the land, however, belonged to two white farmers. It is thought that they chose not to evict the new residents because of the presence of Union soldiers and the location of the land. The floodplain in which Freedom Hill was established affected the utility and subsequent value of the land. This worked in the favor of the new residents because the land was inexpensive and easy to obtain; many newly liberated Black Americans were able to become landowners in a time of racial violence (Mobley 1986). The proximity of Freedom Hill to Tarboro—approximately 1 mile—was mutually beneficial for the residents of both towns; Freedom Hill residents could work and earn money close to home, while Tarboro residents profited from various forms of labor provided by the residents of Freedom Hill. Tarboro residents were also pleased with the “social distance” they maintained from the Black residents in Freedom Hill (Mobley 1986). While the Civil War was over, racial tensions were still very high. Shortly after the founding of Freedom Hill, residents experienced political and economic growth by becoming legal landowners, establishing small businesses, and gaining further employment in Tarboro. In 1885, citizens of Freedom Hill petitioned for incorporation of the town for multiple reasons; legal incorporation of the town would offer protection from white hostility and racial discrimination, residents would gain the ability to elect town officials and contribute to local politics—something previously denied to 23 Black Americans, and Black residents would have the freedom to succeed without white interference (Mobley 1986, 350). Freedom Hill was incorporated as Princeville in 1885, making history as the first town chartered by Black Americans in the United States (US) (Mobley 1986). Residents chose the name “Princeville” to honor Turner Prince, one of the town’s most influential citizens who was one of the first to purchase land and construct a dwelling (Naylor et al. 2020). This incorporation allowed Princeville to be governed by the same provisions as Tarboro, including the number and structure of governing bodies: one Mayor, five Commissioners, a Clerk, a Treasurer, and a Constable (Mobley 1986). Although Princeville was growing economically and politically, the proximity to the Tar River continued to be an environmental threat. Princeville residents were continuously impacted by flood events in 1887, 1919, 1924, 1928, 1940, and 1958 (Naylor et al. 2020). Every time the area was impacted by a flood event, residents would clean up and rebuild, which continues to be a source of pride for Princeville residents today (Naylor et al. 2020; Grace-McCaskey et al. 2021). The US Army Corps of Engineers (USACE) began developing a dike in 1967 in an attempt to protect the town from a 500-year flood event, which is determined to have a 0.2% chance of impacting any given area, annually (USACE 2015a, 1; Naylor et al. 2020; FEMA 2023). The Federal Emergency Management Agency (FEMA) categorizes flood events based on the National Flood Insurance Figure 3.2. Original levee with US 64 road alignments (Appendix B, USACE 2015a, 1) 24 Program (NFIP) Terminology Index (FEMA 2023). The original construction of the dike and levee structure in 1967 (Figure 3.2) consisted of a levee system, divided into two segments that ran parallel to the Tar River, and a segment of US Highway 64 (USACE 2015a, p. 1) This separated the town from the river and provided protection from smaller flood events. Because of the assurance the US ACE and local leaders provided, residents believed this would solve the problems associated with flooding and more people would move to Princeville (Phillips et al. 2012). However, in 1999, a series of small storms, coupled with category 4 Hurricane Floyd, would prove the dike to be ineffective against high floodwaters (Phillips 2012; USACE 2015a, p. 3). 3.4 History of Storm-related Flood Events On September 16, 1999, Hurricane Floyd made landfall over eastern North Carolina (ENC). The hurricane along with five previous rain events caused the Tar River to crest over 43 ft, breaking the levee and flooding Princeville (USACE 2015a). The flood waters entered the town beyond the southern segment of the levee, however, as flooding increased, water continued to breach the northern section of the levee as well. Fortunately, no lives were lost, but Hurricane Floyd destroyed over 1,000 residential and business structures in the town. The town remained underwater for approximately two weeks (Yoon 2009; Phillips 2012; USACE 2015a; Grace- McCaskey et al. 2021). Several public figures including Jesse Jackson, Prince, and former President Bill Clinton made public appeals to bring awareness to Princeville in the wake of Hurricane Floyd. President Clinton issued Executive Order 13146 to “repair and rebuild Princeville, and to the extent practical, protect Princeville from future floods” (USACE 2015a, 3). Furthermore, officials concluded that the levee should be rebuilt and extended at both ends (Figure 3.3). FEMA responded to the flood by offering a total acquisition, or buyout, of all residential and business structures within the 100-year floodplain. A 100-year flood, or base flood, is defined as a flood that has a 1% chance of impacting this area annually (FEMA 2023). Acquisition of all residential structures would mean the displacement and relocation of all residents. Princeville’s Board of Commissioners voted to decline the acquisition proposal, as they felt that leaving the area would mean giving up the legacy their ancestors worked so hard to establish (Phillips 2012; Grace-McCaskey et al. 2021). Some commissioners also believed that 25 residents did not have the financial means to start a new life somewhere else and that the town’s tax base would be negatively impacted by the outflow of citizens. With the denial of the acquisition, USACE offered to rebuild the levee (Grace-McCaskey et al. 2021). According to residents, they believed that the levee would protect them from storms for the next 100 years since they were told by officials that the levee was built to withstand a 100-year flood event (Grace-McCaskey et al. 2021). However, some residents believed this provided a false sense of security to those who chose to return to the area (Naylor et al. 2020; Grace-McCaskey et al. 2021). Just seventeen years later, on October 8, 2016, Hurricane Matthew again brought extensive rain to eastern NC, causing the Tar River to crest over 36 ft. The rising water went around the levee and flooded the town, leaving more than 80% of Princeville under water (Naylor et al. 2020; Adkins et al. 2021; Grace-McCaskey et al. 2021). FEMA offered acquisitions, but this time, individuals were able to accept or deny the offer as they deemed fit. Figure 3.3 Levee system post Hurricane Floyd (Appendix F, USACE 2015b, 3) 26 Some property owners impacted by both hurricanes accepted the buyouts and relocated elsewhere. Between 2016 and 2018, the population of Princeville declined by about 10%. While some residents did accept the acquisition, for many, the decision to stay in the area despite flooding still outweighed the decision to relocate (Adkins et al. 2021; Grace-McCaskey et al. 2021). However, the recovery and mitigation processes in Princeville are slow-moving, to say the least; Hurricane Matthew impacted Princeville over 6 years ago, yet community members are still displaced or living in damaged houses. 3.5 Disaster Recovery and Mitigation This thesis focuses on the hazard mitigation strategies employed in Princeville, NC, however, it is important to broadly understand the procedures and strategies implemented by state and federal agencies in the US for recovery and mitigation following a disaster. FEMA is responsible for responding to federal disasters, while each state has its own branch of emergency management. For instance, North Carolina Emergency Management (NCEM) is responsible for handling disasters at the state level in NC. All of the information presented in the following section was pulled directly from the official website for FEMA. When a disaster impacts a community, the county and state governments are responsible for funding recovery and mitigation efforts unless the damages exceed the resources available to the state. Directly following a disaster, the state is responsible for contacting the FEMA Regional Office for their area to request a joint, state and federal, Preliminary Damage Assessment (PDA). In large-scale disasters, this is done using the flyover method and the development of flood maps. The PDA team, comprised of local leaders and state and federal government personnel, assesses the extent of the damage, the impact the disaster has on individuals and public facilities, and the types of federal assistance that may be needed. The governor is responsible for including this information in a declaration request for federal assistance, to show that the extent of the damage exceeds the capabilities of the state due to a lack of resources. Even though the governor may submit a declaration request to the president through the proper FEMA Regional Office, they are still responsible for taking the proper steps laid out in the state’s emergency plan. In doing so, the governor must provide the FEMA Regional Office with a description of the state and local government efforts and resources available for use during recovery efforts, a description of other federal agency efforts and resources utilized, and a 27 description of the type and extent of additional resources needed to alleviate the impact of the disaster. This information is collected during the PDA and included in any requests for federal assistance (FEMA 2023). If the disaster is severe enough to require federal assistance, the president may issue one of two declarations: emergency declarations and major disaster declarations (see Figure 3.4). The president can declare an emergency declaration any time they think federal assistance is needed for an area to recover from a disaster. This declaration aims to supplement the recovery efforts made by state and local governments to ensure the protection of lives, property, public health, and safety. While the emergency declaration is designed to lessen the threat of a disaster in any part of the US, the total funds of each declaration must not exceed $5 million (FEMA 2023). To meet the requirements for an emergency declaration, the governor must submit a request to the president within thirty days of the disaster and the situation must surpass the capabilities of the local and state governments. Under an emergency declaration, Public Assistance (PA) and Individual Assistance (IA) are provided to aid in recovery and mitigation efforts. Figure 3.4: Steps taken to recover after a hurricane For more information on the programs available through a Major Disaster Declaration, see Table 3.1 below 28 PA programs provide additional funds to “state, tribal, or territorial governments…so communities can respond to and recover from major disasters or emergencies” (FEMA 2023). In an emergency declaration, PA provides the resources to remove debris and take emergency protective measures, however, the cost is split between federal agencies which are responsible for 75% of the total costs, and non-federal agencies which supply the remaining 25%. The only form of IA available under an emergency declaration is known as the Individuals and Household Program (IHP), which is typically completely funded by federal agencies (FEMA 2023). According to FEMA, “the president can declare a major disaster for any natural event, including any hurricane, tornado, storm, high water, wind-driven water, tidal wave, tsunami, earthquake, volcanic eruption, landslide, mudslide, snowstorm, or drought, or, regardless of cause, fire, flood, or explosion, that the president determines has caused damage of such severity that it is beyond the combined capabilities of state and local governments to respond” (FEMA 2023, para 15). Like an emergency declaration, this type of declaration requires the governor to submit a request to the president within thirty days of the disaster. This request must show that the impact of the disaster is beyond the capabilities and resources of the state. Furthermore, it must also include the following: confirmation that the governor has taken the appropriate action described in the state emergency plan, an estimate of the amount and severity of damage to public and private sectors, a description of the efforts taken by state and local governments to mitigate the disaster, estimates of the type and amount of assistance needed, and a certification by the governor that the state or local governments will comply with all applicable cost-sharing requirements (FEMA 2023). The assistance provided by federal agencies is dependent on the type of assistance requested by the governor. Unlike an emergency declaration, a major disaster declaration includes Hazard Mitigation Assistance which assists in any actions taken to prevent or reduce long-term risk to life and property from environmental hazards (see table 1). IA programs are a bit more extensive than that of an emergency declaration, consisting of the IHP, crisis counseling, disaster case management, disaster unemployment assistance, disaster legal services, and disaster supplemental nutrition assistance program (FEMA 2023). The process of IA is outlined in Figure 3.5. FEMA assesses the needs of each impacted individual to see what resources are required to provide housing assistance, replace personal property, and assist in other areas of need such as 29 medical or dental expenses. If further assistance is needed outside of the scope of financial resources that FEMA can provide, the individual is referred to the Small Business Administration (SBA) to apply for a loan. Any unmet needs are then addressed via the long-term recovery division of emergency management. FEMA is responsible for evaluating declaration requests and sending recommendations to the president about how to proceed. Six factors are assessed when considering the amount and type of IA provided to any given area. These include: the state fiscal capacity and resource availability, the number of uninsured homes and personal property lost or impacted, a profile of special populations that may need special or long-term care that is difficult to maintain during a disaster, the amount of impact to the local infrastructure, the number of casualties, and any disaster-related unemployment patterns (FEMA 2023). Figure 3.5 Sequence of assistance through FEMA’s IA program (FEMA 2023) 30 The PA program under a major disaster declaration is described as “assistance to State, Tribal, and local governments and certain private nonprofit organizations for emergency work and the repair or replacement of disaster-damaged facilities, which may include the following categories: (A) debris removal, (B) emergency protective measures, (C) roads and bridges, (D) water control facilities, (E) buildings and equipment, (F) utilities, and (G) parks, recreational, and other facilities” (FEMA 2023). Table 3.2 presents the factors FEMA considers when providing different forms of assistance to impacted areas. The six factors considered for IA are straightforward and need no explanation, however, the factors that FEMA personnel consider when providing PA to a community are more in-depth. PA considerations begin with analysis of the estimated cost of the requested assistance. To gain a measure of per capita impact, FEMA measures the estimated cost of federal and non-federal PA against the affected population. Per capita amounts are used by FEMA as an indicator that the disaster is so severe that it warrants federal support. It is important to note that this figure changes annually based on the Consumer Price Index created by the US Bureau of Labor Statistics (BLS) (FEMA 2023). FEMA then considers the localized impacts; specifically, they evaluate the impact of a disaster at the state and local government levels to see if any extraordinary circumstances warrant federal involvement, even if the statewide, per capita quota, is not met. Table 3.1: Types of Assistance When a Major Disaster Declaration is Declared Individual Assistance Public Assistance Hazard Mitigation IHP A- Debris removal Assistance for actions taken to prevent or reduce long- term risk to life and property from environmental hazards. Crisis Counseling B- Emergency Protective Measures Disaster case management C- Roads and bridges Disaster unemployment assistance D- Water control facilities Disaster Legal Services E- Buildings and equipment Disaster supplemental nutrition assistance program F- Utilities G- Parks, recreational, and other facilities (FEMA 2023) 31 The next consideration is insurance coverage in force. In other words, FEMA estimates the amount of active insurance coverage that is required by law at the time of the disaster and then reduces their assistance by that amount. Hazard mitigation is considered to encourage steps to mitigate future damage brought on by disasters. This consideration is extremely important when discussing communities or towns that do not meet the estimated cost assistance per capita consideration (FEMA 2023). FEMA considers the history of disaster impacts within twelve months, to get a better sense of the overall assistance needed in a particular area. Lastly, FEMA considers any additional federal programs that may aid the affected area. There may be different agencies that can meet different needs created by disaster impacts. 3.6 Recovery and Mitigation in Princeville The recovery and mitigation strategies outlined in the previous section are similar to the strategies implemented in Princeville, NC. However, the development and implementation of the state centric pilot program established by FEMA HMGP is new and relatively unique to Princeville. Throughout the course of the data collection phase of this research, North Carolina Emergency Management (NCEM) and FEMA developed and introduced the state centric approach to disaster mitigation in Princeville. According to the North Carolina Department of Public Safety (NCDPS), this program allows the state to provide funds to finance contract work for mitigation purposes and complete all grants that have been previously awarded. This program was designed to remove the financial burden from counties and smaller municipalities when Table 3.2: Factors FEMA considers for assistance Individual Assistance Public Assistance State fiscal capacity and resource availability Estimated cost of assistance Uninsured home and personal property losses impacted Localized impacts Disaster impacted population profile special populations Insurance coverage in force Impact to community infrastructure Hazard mitigation Casualties Recent multiple disasters Disaster-related unemployment Other Federal agency assistance programs (FEMA 2023) 32 attempting to initiate mitigation efforts such as elevations (ncdps.gov, accessed on 05/01/2023). This program also attempts to hire local contractors to decrease the costs of materials and ensure that funds are redistributed throughout the applicant’s community. Financial responsibility, as well as the coordination of labor, materials, and project management, lies with the state. Additionally, Princeville has been the recipient of many grants to mitigate the impacts of future disasters. In 2018, Princeville, along with twenty-nine other communities affected by Hurricane Matthew, received the Community Development Block Grant-Disaster Relief (CDBG- DR) in the form of $625,000. The CDBG-DR program was initiated by the US Department of Housing and Urban Development (HUD) and administered in part by the NC Commerce Rural Economic Development Division (NC Department of Commerce 2022). This grant attempts to promote affordable housing and infrastructure development that addresses the needs of the overall Princeville community. Because the funds are utilized to promote disaster mitigation on the broader community level, they are categorized as Public Assistance (FEMA 2023; NC Department of commerce 2022). Furthermore, to maintain Princeville’s original historical significance and ensure the community's safety, the state funded the annexation of 53-acres of land outside of the 100-year floodplain. While the town plans to relocate essential services and construct affordable housing, the original location of Princeville will continue to hold historically significant structures such as the town hall, senior center, museum, Princeville Elementary School, and Mt. Zion Primitive Baptist Church (Grace-McCaskey et al. 2021). An additional 88 acres were acquired by the town to further develop ancillary facilities and mitigate the destructive nature of future disasters (Grace-McCaskey et al. 2021). This chapter briefly outlined the historical processes that led to an increase in Princeville’s vulnerability to disasters as well as the recovery and mitigation efforts implemented to ensure the future safety of Princeville citizens. This highlights the complex nature of recovery and mitigation efforts in this area while explaining the steps community members and local leaders have taken to protect themselves from future harm. Chapter 4: Research Design and Methodology 4.1 Chapter Overview Data for this thesis were collected using a variety of qualitative methods between May 2022 and December 2022. I utilized participant observation to gather contextual information and gain rapport with potential participants. Semi-structured interviews were conducted with community members and local leaders of the town of Princeville in Edgecombe County, North Carolina (NC). Additionally, I interviewed state and federal agency personnel who have worked with the local leaders in Princeville during the recovery and disaster mitigation processes. This chapter explains the study site, specifically the demographic and economic characteristics of the area. I discuss the sampling strategy and the research design beginning with an explanation of the research questions and associated goals. Lastly, I describe the qualitative methods I used for data collection and analysis. 4.2 Study Site Princeville was selected as the study site for this research because residents have been impacted by two major flood events that continue to have long-lasting impacts—Hurricane Floyd in 1999 and Hurricane Matthew in 2016. Previous disaster recovery strategies have not allowed for economic or social growth within the town; however, disasters bring about rapid social and environmental change which provides opportunities to explore recovery and mitigation strategies (Barrios & Faas 2016). Furthermore, the town’s historical significance contributes to residents’ sense of place and place attachment, which influences their decisions to return to the area after a major flood event (Grace-McCaskey et al. 2021). All demographic and economic data of the town come from United States census data published in 2021 (US Census Bureau 2021). The population of Princeville is 1,648 individuals, 93% of which are Black Americans. Fifty-two percent of residents are between the ages of 18 and 64. Of the current population, 59% are women. Nearly 20% of citizens currently live below the poverty line, which is about 10% higher than in surrounding areas (US Census Bureau 2021). Fifty one percent of homeowners are women, 4% are men, and 43% are owned by non-family or married couples. Of the 859 residential structures in the town, 31% remain vacant and 48% are 34 mobile homes; both statistics are more than double the average percentages of the rest of North Carolina (US Census Bureau 2021). Table 4.1 Princeville Demographic Information Number Percentage Race and Ethnicity Black White Multiracial Hispanic 1,534 4 4 84 93.1 1.6 0.2 5.1 Age 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80+ 248 259 213 143 116 261 250 102 56 15.1 15.7 13.9 8.7 7 15.8 15.2 6.2 3.4 Sex Female Male 967 681 58.7 41.3 Persons Below the Poverty Line 326 19.8 Population by Household Type Married couple Male householder Female householder Non-family 498 73 839 232 30.3 4.5 51.5 14.1 Number of Occupied Houses 589 68.6 Number of Vacant Houses 270 31.4 Type of Residential Structure Single unit Multi-unit Mobile home Boat, RV, van, etc. 337 112 410 0 39.2 13 47.7 0 Note. For more detailed demographic information see US Census Bureau 2021 35 4.3 Sampling Strategy and Description of the Sample This research required the recruitment of community participants who have previously been impacted by flood events. Additionally, this research required the recruitment of local, state, and federal agency personnel who work closely with residents throughout the disaster recovery and mitigation processes. To access all these groups, I utilized snowball sampling. Snowball sampling is a non-probability, convenience sampling strategy where a few participants are asked to identify others who might be interested in participating in the research project (Kalton & Anderson 1986; Thompson 2002, p. 183; Woodley & Lockard 2016, p. 322). Convenience sampling allows researchers to access a subset of a population who may otherwise be inaccessible (Heckathorn 2011; Woodley and Lockard 2016). This type of sampling strategy is also favorable when research requires a lot of data from a few participants (Bernard 2011). Furthermore, I found that many community members who have similar concerns regarding disaster mitigation are in contact and have some form of social relationships. This was beneficial when attempting to establish rapport with residents who have been impacted by flooding. At the end of each interview, I asked the participant if they knew of anyone else who would be willing to participate in this research. In some cases, I would emphasize that the participant could verify with the potential participant that they would be interested in being interviewed before I was provided with any contact information. New participants expressed that they were able to gain insight into the interview process because previous participants shared their experiences when asking for permission to provide me with their contact information. It is in this way that snowball sampling created a level of trust between me and the participants. Snowball sampling is often critiqued by social science researchers for possible selection bias, a lack of diversity of subjects, and validity (Woodley & Lockard 2016, p. 322-324). The lack of a diverse sample size is a common critique because, oftentimes, the people who are willing to discuss their experiences in a research setting share similar social circumstances. While this method does not allow for a very diverse sample size, Woodley and Lockard (2016, p. 324) emphasize that snowball sampling allows scholars to utilize social networks to conduct research within marginalized communities without further marginalizing them. Snowball sampling relies on already existing social networks, which means that any Princeville community member who wants to participate in this research can do so, on their own terms. 36 The total sample size for this research was 20 individuals (n=20). I interviewed 8 community members who have been impacted by flood events brought on by Hurricanes Floyd and Matthew. The sample for semi-structured interviews with community members consisted of 5 women and 3 men, 7 of whom were Black, and one woman who was White. This is consistent with demographic information from the area. Additionally, I formally interviewed 12 government and nongovernment agency personnel: 3 were local Princeville leaders, 4 were state- level agency personnel, 4 were federal-level agency personnel, and 1 person worked for a conservation nonprofit agency that continues to work closely with the town. The federal-level agency personnel were associated with various departments within the Federal Emergency Management Agency (FEMA). The state-level agencies included North Carolina Emergency Management (NCEM) and the North Carolina Office of Recovery and Resiliency (NCORR). This thesis research is part of a larger National Science Foundation (NSF) grant funded through the Coastlines and People program. Because of the nature of that project, IRB approval was acquired (UMCIRB 21-001363) (see Appendix D), which stated that we were not required to obtain signatures from participants. Rather, all participants received a letter of informed consent (see Appendix A). Additionally, at the beginning of each interview, I explained to the participant that their names and any identifying information would be kept confidential, they were allowed to pause or stop the interview at any moment, and they did not have to answer certain questions if they did not want to. 4.4 Research Design The overall goal of this research is to identify the role of current disaster mitigation strategies in addressing the concerns community members have about the future risks of flooding, specifically in Princeville, NC. To do so, I developed three main research questions: (1) What are the main concerns community members have about the future risks of flooding? (2) Do community members trust that decision-making bodies—including Princeville’s Board of Commissioners and state and federal agencies—will consider these concerns when developing and implementing post-disaster policies? (3) What factors impact the extent to which local officials and state and federal agencies utilize local context and knowledge when developing and implementing disaster mitigation policies? 37 The objectives of this research complement the main research questions. First, I sought to understand previous disaster mitigation strategies and the continued mitigation policies currently being developed. It is also essential to understand the ways in which flood events have impacted residents, and how they continue to be impacted. Finally, I explored the role of state, federal, and non-profit entities throughout the recovery and mitigation processes. I utilize an exploratory, nonexperimental, descriptive research design to gather enough information to characterize current disaster-related phenomena in Princeville. I designed the research in this way because I did not have the ability to manipulate any independent or dependent variables as is possible in a more formal, experimental research design (Cook & Cook 2008; Reio Jr. 2016). Instead, I chose to examine the relationship between residents’ flood-risk concerns and measures taken by government and non-government agencies to address those concerns. I aim to maximize the benefit that my research can provide to residents and policymakers while minimizing the extractive, and sometimes harmful, nature of academic research within marginalized communities. To do so, I utilized qualitative sampling and data collection methods that allowed participants to share their stories and experiences in their own words and on their own terms. 4.5 Data Collection I collected qualitative data using three main qualitative methods: participant observation, semi-structured interviews, and free lists. This section describes the methods in detail, including the justification for the use of each method, how each method was utilized, and the type of information each method provided to this research. Participant Observation Participant observation requires a researcher to take part in the activities, rituals, interactions and events of a group of people to gain insight into different aspects of their life, routines, and culture (DeWalt and DeWalt 2010, p. 13). Researchers can participate in the common and uncommon activities of the participants. Participant observation is beneficial in providing context to cultural phenomena, such as responses to hurricane impacts on the individual and governmental levels. This method requires the use of the information gained from participation and observation through recording and analysis (DeWalt and DeWalt 2010). 38 I utilized participant observation to better understand the internal structure of the decision-making process by attending meetings held by Princeville’s Board of Commissioners and taking detailed field notes, which included observed steps of the decision-making process and the ways in which community members and commissioners interact with one another. Meetings are held on the third Monday of every month. I attended meetings between May 2022 and December 2022; however, certain circumstances hindered my ability to attend every meeting in person. When this was the case, I watched the meetings on a platform where they were streamed live to the public. Of the 8 meetings I attended, 6 were in person and 2 were online. This initial phase of participant observation aided in establishing rapport with local leaders and community members who would later participate in the interview process. This method also aided in my ability to develop meaningful research questions for residents and local leaders. Additionally, I gained insight into the decision-making process in Princeville. I attended the Farmer’s Market held on the second and fourth Saturday of every month, beginning in July of 2022 and ending in November of 2022. This allowed me to gain rapport with active community members who had a vested interest in the prosperity of the town and its residents. I collected jot field notes as I walked around and spoke with community members and vendors at the market, and I further expanded on those at the end of every week. I had the opportunity to hold two informal interviews while attending the market. In this case, I was able to ask them every question present in the interview guide, however, we were not in a place where recording was feasible. Therefore, I had to take detailed notes regarding their experiences and concerns about flooding and disaster mitigation. This individual also introduced me to his wife, who was able to fill in the gaps and add to the story of their experiences during and after Hurricanes Floyd and Matthew. This interview provided much-needed context to previous flood events. Because of the nature of these conversations, I am considering them as one interview. Semi-Structured Interviews I conducted 20 semi-structured interviews that followed a lenient interview guide with open-ended and close-ended questions (see Appendices B and C) to facilitate reliable discussion and build rapport with the participants of this study (Bernard 2011). A lenient interview g