1. Introduction
Damages from natural hazards are on the rise in the United States, and the processes of recovery tend to exacerbate existing societal inequalities (Howell and Elliott 2019). Hazard vulnerability is often compounded by population characteristics such as poverty and social status (Burton et al. 2018; Elliott and Pais 2006; Tanim and Tobin 2018), minority racial or ethnic status (Peacock et al. 1997), older age (Aldrich and Benson 2008; Tanim and Tobin 2018), and lack of access to resources such as transportation (Litman 2006), healthcare (Rickless et al. 2021) and safe housing (Rumbach et al. 2020). These characteristics increase vulnerability in all phases of a disaster, from preparedness to recovery (Finch et al. 2010; Masozera et al. 2007). South Florida is a prime example of a social landscape riddled with inequalities that magnify its hurricane hazard vulnerability (Tobin 1999).
A combination of high natural exposure, persistent coastal development, population growth, and changing coastal demographics have made the U.S. coastline increasingly vulnerable to hurricanes, particularly south Florida (John 2015; Prasad and Stoler 2016). Yet south Florida did not experience a major hurricane landfall from 2005 to 2017 (Mitsova et al. 2018). Hurricane droughts can contribute to underpreparedness, particularly among those who have not previously endured a storm. In 2016 south Florida experienced a near-miss in Hurricane Matthew, and in 2017, a state of emergency was declared for 67 counties after Hurricane Irma devastated the region and ended the region’s decade-long hurricane drought (Mitsova et al. 2018; Zolnikov 2018). As Irma demonstrated, hurricanes remain a major threat to south Florida’s prosperity. Residents of mobile home parks (MHPs), communities that house some of Florida’s lowest-income households, bear a disproportionate burden of hurricane risk (Mitsova et al. 2018).
MHP residents face a mix of “social, legal, geospatial, and market forces” that produce housing insecurity for a predominantly low-income residential population (Rumbach et al. 2020; Sullivan 2018). Mobile housing communities are also particularly vulnerable to natural disasters such as wildfires or hurricanes. The weaker physical structure of mobile or manufactured housing, especially older homes, is extremely susceptible to hurricane damage from winds and flooding. After extensive damage from Hurricane Andrew in 1992, Florida strengthened the building codes for mobile homes in 1994, which has somewhat reduced the risk of damage for homeowners who can afford the upgrades or for buyers of new homes (FEMA 2018; Simmons and Sutter 2008). Despite these code changes, mobile housing remains one of the least safe housing types during a hurricane due to these structures and their placement (Schreiber 2005). Reports estimate that of the 828 000 mobile homes in Florida, only about one-half are insured and up to 600 000 are older models that do not meet current hurricane standards (Danielle 2017; Tobin and Burton 2009). In south Florida, around 72% of all mobile housing units were built before 1994 when the new building codes were implemented (GeoPlan Center University of Florida 2017). These mobile homeowners have generally been less likely to allow an intensive home inspection that could help them gain relevant risk information and facilitate any necessary home risk-mitigation action (Chatterjee and Mozumder 2014).
In addition to dealing with inferior structural vulnerability of their housing, MHP residents tend to be less economically resilient, with limited access to resources and less socially integrated into the community (Edwards et al. 1973; Howell and Elliott 2019; MacTavish 2007). Exposure to natural hazards has been linked to wealth inequalities along racial, educational, and home ownership differentials (Howell and Elliott 2019). MHP residents are generally more likely to evacuate from an approaching hurricane than other residents, as only 20% of MHPs in the southeastern United States include storm shelters (Huang et al. 2016; Strader et al. 2019). Even when given mandatory evacuation orders, a sizable proportion of the mobile housing community may be unwilling or unable to evacuate (Hudak 2017; Kusenbach 2017a,b; Strader et al. 2019; Tanim and Tobin 2018). In south Florida, MHP residents tend to be lower-income, older adults who are either long-term residents or retiree migrants from other parts of the country, both permanent and seasonal. The combination of unsafe housing and socioeconomic and demographic vulnerability poses serious challenges for disaster planning and mitigation efforts.
The challenges associated with storm surges and inland flooding can have serious implications for mobile homeowners and for emergency management (Ash et al. 2020; Ash 2017; Schmidlin et al. 2009; Schreiber 2005; Strader and Ashley 2018). High flood risk areas, such as storm-surge zones and floodplains, are vulnerable to hurricane-induced flooding, and hurricane-force winds pose tornado threats to inland areas (Rumbach et al. 2020; Schreiber 2005). Flooding can cause extensive damage to the interior and exterior of mobile homes from strong pressure on foundations and piers, as well as from floating debris (Schreiber 2005). As Florida’s sea levels rise (Balaguru et al. 2016; Frazier et al. 2010), higher and more extensive storm surges will not only increase the risk of damage to mobile homes in the current surge zone but will also include more inland areas, and therefore more mobile homes, into the surge zone. Larger storms are likely to expand the surge zone farther still (Irish et al. 2008).
Another challenge is associated with rebuild versus relocate decision-making after a hurricane. The Federal Emergency Management Agency does not require mobile homeowners to buy flood insurance (FEMA 2018). Because insurance for a mobile home covers only the depreciated value and not the total replacement cost, mobile homeowners—especially those with the oldest and most vulnerable homes—often choose not to buy insurance coverage (FEMA 2018; Murphy 2017). In south Florida, where the risk of hurricane-induced displacement is one of the highest in the country (Esnard et al. 2011) and with one-third of mobile homes in the storm-surge zone (Balaguru et al. 2016), the lack of insurance can impede rebuilding and trigger displacement. Renters are most likely to be displaced and face challenges in recovery (Pais and Elliott 2008; Rumbach et al. 2020), as are racial/ethnic minorities, the less educated, and the poor (Esnard et al. 2011; Myers et al. 2008).
Despite higher levels of vulnerability, research on MHP residents has been somewhat limited due to multiscale challenges associated with reaching out to this subpopulation, which sometimes results in stigmatized, insular communities (Kusenbach et al. 2010; Kusenbach 2017b; Prasad 2016; Prasad and Stoler 2016). Much of the hurricane research on the mobile housing community has focused on vulnerabilities during and in the aftermath of an event. Other studies have documented the post-Irma experiences of Florida residents across sociodemographic characteristics, but not all distinguished between conventional homes and mobile homes (Chakalian et al. 2019; Wynter-Minott 2017). Few studies have investigated hurricane vulnerability and storm-related behavior for MHP residents using mixed methods. To address the social structural factors that may limit the options of MHP residents, the study is also guided by constrained choice theory (CCT).
a. Constrained choices
Decision-making and social action have traditionally been studied by social scientists using economic orientations that operate under the notion of individual rational choice without considering socially constructed patterns of decision-making. Sociologists, behavioral geographers, public health practitioners, and policy makers have slowly moved toward incorporating theoretical and empirical explanations based on social and structural mechanisms that constrain spatial-, climate-, and health-related behaviors (Desbarats 1983; LaGory 1982; Pescosolido 1992; Stuart and Schewe 2016; Vuolo et al. 2016). This theoretical shift lifts the sole responsibility of individuals in the decision-making process and contextualizes the social forces at hand. CCT elaborates on rational choice theory’s contention of an individual’s rationality, intentional action, and self-interest to consider a broader social context that includes event- and network-based factors (Collins 2017, 2018) that may impede or facilitate individual behaviors.
Disaster research can highlight many of the social inequalities and structural vulnerabilities across different groups and their social positions in society (Myers et al. 2008; Reid 2013). The interactions of socioeconomic status, race, gender, age, and number of social networks have theoretical and methodological implications about social vulnerabilities and inequalities for disaster research (Myers et al. 2008). Studies have documented the impact of these differential hurricane experiences using the cases of Hurricanes Andrew and Rita in Florida and Katrina in Louisiana (Cutter and Emrich 2006; Haney et al. 2007; Lovekamp 2008; Myers et al. 2008; Peacock et al. 1997; Reid 2013). Thus, CCT provides a useful lens to study the differential outcomes of social vulnerability across groups in MHPs for hurricane disaster preparedness, how disaster events and recovery transpire and their implications for subsequent hurricane risk and vulnerabilities (Fucile-Sanchez and Davlasheridze 2020; Rumbach et al. 2020).
Rieker and Bird’s constrained choice model (CCM) explains the processes by which social contexts, competing priorities and constraints cumulatively limit options and opportunities for prioritizing individual health and behavior (Bird and Rieker 2008). Although social vulnerability models attempt to include some social components impacting disaster outcomes (Rufat et al. 2019), there is a paucity of scholarship and a need for studies that capture social constraints on hurricane-related decision-making (Ersing et al. 2020; Rickless et al. 2021). No hurricane or climate-related research has applied the CCM to understand hurricane hazards or vulnerability. This model can explain how hurricane vulnerability is a product of government-, community-, work-, and family-level constraints. Rather than conceptualize resident complacency, our use of CCM allows for a multidimensional analysis of the structural and social inequalities that restrict behavior and social action. The condition of place, space, and housing contribute to both people’s and communities’ physical and mental well-being (Fitzpatrick 2002). If hurricane vulnerability is determined by these same conditions, then it is also associated with physical and mental well-being (Karaye et al. 2019), and thus, a CCM can explain resident hurricane vulnerability. When applied to hurricane vulnerability, the CCM helps conceptualize choice in the context of social, geospatial, legal, and market forces that are connected by larger systemic, societal, and structural determinants that drive behavior, decisions, priorities, and risk perceptions for hurricane safety and well-being (Rieker and Bird 2005).
b. Objectives
This paper presents the results of a mixed-methods, cross-sectional study of residents of MHPs in Broward and Miami-Dade Counties in Florida. In this study, mobile home is used as an umbrella term that refers to manufactured homes, mobile homes, and trailer homes, all of which are used to describe structures that are affixed to a metal chassis within a mobile home park and that are often immobile (Sullivan 2017). We use a constrained choice theoretical framework to analyze hurricane vulnerabilities given the premise that individual choices affecting human well-being are altered by individual, family, community, government, policy, and structural interactions before, during, and after hurricane events.
This study was initiated in July 2016, just months before Hurricane Matthew, and continued in May 2018, 8 months after Hurricane Irma. Our objectives were to answer two primary research questions:
-
What were the hurricane-related knowledge, attitudes, and practices of mobile home residents related to storm-surge awareness, resettle versus relocate decision-making, and climate change beliefs during “hurricane drought” conditions in 2016?
-
What were the most salient social and structural determinants that shaped hurricane preparedness and decision-making?
The results of this case study add to the literature on mobile home parks and hurricane vulnerability among mobile home residents across the region, with implications for building resilience, coping strategies, and adaptive capacity among similar communities with low participation in local government and high social inequalities.
2. Data and methods
This study used a return-by-mail survey to assess mobile home residents’ prestorm hurricane-related knowledge, attitudes, and practices in 2016, and then used focus groups and in-person surveys to assess Hurricane Irma experiences and perceptions in 2018. The purpose of this study was to examine baseline hurricane-related attitudes, risk perception, and behaviors of select mobile housing communities during “hurricane drought” conditions in our 2016 phase, and lay the groundwork for returning after a major storm, as we did in 2018 after Hurricane Irma.
Few MHPs were initially willing to participate in this study. By engaging MHP managers in 2016, we were able to start by distributing a return-by-mail survey, but a low, nonrepresentative response rate limited any substantive conclusions with these data. The study team was not positioned to return in 2016 after Hurricane Matthew, which made landfall in south Florida just months after our initial survey. But we reconnected with MHP managers in 2018 and proceeded with focus groups and in-person surveys after Hurricane Irma prompted renewed community interest in learning from residents’ experiences and improving preparedness. All study instruments and procedures were approved by the Institutional Review Board at the University of Miami. All surveys and focus group scripts are available in the online supplemental material.
a. Study area
Miami-Dade and Broward Counties in southeast Florida (Fig. 1) have some of the highest exposure to hurricanes and tropical storms in the United States (Pielke et al. 2008; Zandbergen 2009). The region’s flat, low-lying coastal terrain and poor natural drainage make it especially susceptible to storm flooding from coastal storm surges and inland water body overflows. Elevated sea levels and the ongoing subsurface saltwater intrusion will further exacerbate the risk of coastal inundation as well as inland flooding (Zhang 2011).
The study region in Miami-Dade and Broward Counties, with the distribution and relative size of all mobile home parks, and corresponding storm-surge zones.
Citation: Weather, Climate, and Society 15, 3; 10.1175/WCAS-D-22-0122.1
Local population vulnerability compounds the natural exposure of the study area to hurricanes and sea level rise. According to data from the U.S. Census Bureau’s 5-Year American Community Survey (2012–16), Miami-Dade and Broward were the two most populous counties in Florida with a combined population of around 4.5 million people. Roughly 14% of this population is over 65 years of age. Among Florida counties, Miami-Dade ranks first, and Broward third, in total population above 65 years of age. Approximately 50.4% of this population is of Hispanic ethnicity. Miami-Dade leads Florida in total Hispanic population (1.8 million) as well as for Hispanics as a percent of total population (66%). Around 20% of the population in Miami-Dade, and 14.4% in Broward, live below the federal poverty threshold. Mobile homes have a strong presence in the study area with approximately 34 665 mobile housing units, most of which are located within MHPs. Our digital map overlay of mobile home locations with surge zones and floodplain locations revealed that 54 of the 168 MHPs (32%) in Miami-Dade and Broward counties are in the coastal storm-surge zone, and 63 (37.5%) are in the 100-yr floodplain. Mobile home residents in south Florida are typically older retirees who may be longtime residents, recent permanent transplants, or seasonal “snowbirds” from northern U.S. states and are often of lower socioeconomic status (Prasad 2016).
b. Data collection
1) Mail-in survey
After consultation with MHP management, we received permission to distribute paper-based surveys to approximately 900 households in select MHP (see Fig. 1). The MHPs that were selected were based on a convenience sample for those MHP in south Florida whom we were able to contact their management office and obtain consent for participation. The surveys were left at the front door of each home (specifically not in mailboxes, which is illegal) in July 2016 in a university-branded envelope containing copies of the survey in English and Spanish, and a preaddressed stamped envelope to return the survey to our university. We anticipated a response rate near 15%, which has been demonstrated by many return-by-mail surveys conducted by nonprofit organizations (Hager et al. 2003). But just 68 households responded by mail between July and September, for a response rate of 7.5%. Of these 68 responses, 9 (0.5%) were empty envelopes or blank surveys, and one was excluded because it was authored by a 13-yr-old. An additional response was excluded because it was the sole response to indicate being a seasonal resident (all others were permanent residents, consistent with summer). We thus analyzed a sample of 57 responses in this initial wave of the study.
The 3-page return-by-mail survey contained 25 items drawn from the hurricane preparedness, evacuation, and climate change perceptions literatures, and was designed to be completed in less than 10 min. The survey collected demographic and housing characteristics; asked about hurricane experience, behavior, and preparedness; and asked about risk perception and concern about sea level rise and flooding.
2) Focus groups
In May 2018, we conducted six focus groups at five different MHPs where return-by-mail surveys had been distributed in 2016. We chose focus groups to efficiently foster rich discussions and gain a wider scope of responses on the subject matter (Barbour 2008). We designed our study to elicit community perceptions, and moderators facilitated interaction between respondents through active probing and effective listening (Morgan 1997), aware that participants are often more comfortable addressing each other in their responses, rather than moderators.
The focus groups were organized by contacting regional and on-site MHP management and designed as a convenience sample of MHP locations that participated in the return-by-mail surveys. Upon making contact with MHP management, the on-site managers helped us to recruit participants for the six discussions. All focus groups were held in the respective community, with participants rarely having to walk farther than the equivalent of one city block. The discussions were held in the MHP’s community building or in a covered space next to the management office and conducted during the early evening in a combination of English and Spanish by bilingual moderators. Participant compensation was limited to the dinner served at all focus group discussions, generally pasta or pizzas with cold beverages (soft drinks and water), with enough for participants to bring extra food home for family members. All discussions were audio recorded, transcribed, and translated by study assistants, and transcriptions were compared with field notes for consistency. Pseudonyms were used for participants and for the mobile home parks to protect confidentiality.
The focus group discussion topics followed the chronology of the Hurricane Irma experience. We began by asking about Irma preparedness and evacuation behavior to understand their experiences regarding travel, expenses, anxiety, etc. associated with their storm planning and discussed how participants perceived Hurricanes Matthew and Irma before they made landfall. We then discussed the aftermath, such as how long it took to get back to normal at home and at the MHP, the extent to which participants felt they had recovered (physically, emotionally, financially), and whether they had applied for assistance from FEMA or other groups. We also discussed risk perceptions and any mitigation steps that participants engaged in post-Irma, including home insurance and inspection issues, and expectations around rebuilding versus relocating after a hypothetical category 5 storm.
3) In-person survey
Focus group participants were invited to complete a hurricane vulnerability questionnaire following each focus group session, and 44 participants completed this anonymous in-person survey. This survey was available in Spanish and English and took approximately 15 min for participants to complete. The survey followed the same chronology of events as the focus group, such as measuring perceptions of hurricane risk, inventorying evacuation behaviors, and assessing potential behavior changes before, during, and after Hurricane Irma. This survey also probed issues that participants would be less likely to have discussed in a group setting, such as preexisting medical conditions; levels of stress and anxiety; and lifestyle changes regarding sleep, diet, exercise, and alcohol consumption.
c. Data analysis
Our analysis of the return-by-mail survey began by examining the descriptive statistics for all study items to better understand the variation in the respondent sample. We first assessed the bivariate relationships between survey items, focusing on demographic, knowledge, and behavioral measures. We then used multivariable logistic regression to fit models of four indicators, consistent with hazard vulnerability theory: 1) knowledge accuracy regarding their storm-surge risk, 2) likelihood of rebuilding or resettling in the same place after a catastrophic storm, 3) concern about higher storm surges and future floods, and 4) concern about the future impacts of storms on themselves and their children. Our full modeling procedures are available in the online supplemental material (in the “Data analysis of return-by-mail survey” section).
Our analysis of the focus group discussion content was largely guided by the grounded theory methods of Corbin and Strauss (Corbin and Strauss 2008). We recruited 49 focus group participants, including 20 males and 29 females. Two analysts—who had served as focus group comoderators and note-takers—coded the transcripts to develop coding schemes. Coding was carried out via open coding using line by line technique to study fragments of the transcripts. Using analytic induction, we explored patterns in the data, attempted to explain such patterns, and sought to recognize and illustrate topics and categories that arose in discussions, resulting in categories that present more analytically complex narratives (Corbin and Strauss 2008; Saldaña 2009). Axial coding followed, including the collapsing of codes and constant comparison process for the development of emergent themes. The two analysts regularly met to discuss analytical memos and compare codes to ensure coder reliability for each coding phase, and they triangulated their coding methodology using cross tabulations and word frequency queries.
To further highlight the relationship between the final set of codes, which became themes, and subcodes, we generated a circle graph via a cluster analysis of the codes in NVivo. The software allows for calculation of correlation between items. A Pearson correlation coefficient measure was used to generate the graph. The graph uses a gradient to display an increasing degree of similarity between codes as increasingly thicker lines and lighter lines for stand-alone codes that have no similarity. The circle graph helps to depict similarity, extended interactions, and salience between the codes, and ultimately highlights the codes with the most connectivity having to do with hurricane vulnerability.
The in-person surveys were analyzed by computing descriptive statistics and comparing the results with those from the return-by-mail survey. We managed all survey data and conducted quantitative analyses using the IBM, Inc., SPSS statistics software, version 24. All qualitative data were managed, coded, and analyzed using QSR International NVivo, version 12.
3. Results and discussion
This study of MHP residents in Miami-Dade and Broward Counties analyzed return-by-mail surveys to assess factors associated with hurricane vulnerability in 2016 after 10 years without a major storm and focus groups discussions and in-person surveys to assess residents’ experiences after Hurricane Irma’s landfall in 2017. The findings underscore the dimensions of social identity, status and structural barriers that constrained MHP residents’ choices and contribute to their hurricane vulnerability. Although there is variation across residents from different MHPs, it is evident that government, community (local spaces including work institutions), and family constraints are present and impact the decision-making in hurricane-related experiences. These findings are supported by other recent studies conducted after Irma on family evacuation intentions (Brodar et al. 2020).
Most study participants had previously experienced a hurricane and stayed home or in the local vicinity and planned to act similarly in the future. Prior to Irma, participants demonstrated low awareness of “FL 511,” the statewide telephone service for disaster travel and evacuation information, and their storm-surge zone; counted on a variety of media sources to receive their information, both in English and Spanish; and tended to believe that climate change and sea level rise were real phenomena, but with mixed concern for their impacts. Months after Irma, most residents reported experiencing some degree of medical, financial, or psychosocial hardship, and were most concerned about the cleanup process, their exclusion from many institutional preparation and recovery resources, and their prospects for enduring future storms. The details of the results from our surveys and focus groups are reported below.
a. Mail survey
Table S1 in the online supplemental material summarizes the demographics and knowledge, attitudes, and practices of the 57 return-by-mail study responses that we analyzed, all of which came from permanent residents. More than one-half of respondents were female (54%), Hispanic or Latino (63%), owned their mobile home (86%), had dependents in the household (54%), had no more than a high school education (60%), and had lived there for fewer than 10 years (54%). Nearly one-half of respondents were over 60 years of age (mean 58.5), and 47% were either retired or unemployed.
Most participants (84%) reported previously experiencing a hurricane, and either sheltered in place (40%), or went to a local hotel or friend’s place (49%). Most expected to evacuate to a local hotel or friend’s place for a future storm (53%), emphasizing the importance of social networks and familiarity with the region among long-term permanent residents. This choice was associated with longtime residency (residing in the current mobile home for 10 years or more) [χ2 (1, N = 49) = 5.408; p = 0.020]. Nearly one-quarter (23%) of respondents indicated that they would likely “stay in place” during their next evacuation scenario.
Past behavior was also associated with planned future behavior: 95% of those who did not stay in place in the past did not plan to stay in place in the future [χ2 (1, N = 42) = 11.197; p = 0.001]. Everyone who previously evacuated to a shelter expected to go to a shelter in the future [χ 2 (1, N = 42) = 24.122; p = 0.000], and 73% of those who went to a local hotel or friend’s place planned to do the same in the future [χ2 (1, N = 42) = 3.612; p = 0.057]. Most used their personal vehicle as evacuation transportation (83%). Just 12% of respondents were aware of FL 511.
Respondents demonstrated a dangerous lack of knowledge about storm-surge flooding. Overall, only 11% of respondents believed that their mobile home was in the surge zone (i.e., flood zone as defined by state or local emergency management), when in fact 60% of respondents resided in MHPs located in the surge zone. Conversely, 35% did not report residing in the surge zone, yet nearly 80% of these respondents actually did. Misperception of the surge zone was widespread; one-third of respondents who thought they were in the surge zone were actually not in the surge zone. Many respondents “did not know” whether their mobile home was in a surge zone (39%), and approximately one-half of these households were actually in the surge zone. This question was unanswered by 16% of respondents, which perhaps indicated further lack of awareness, confusion, or constraints in taking action on the risk. These findings on misperceptions are consistent with other studies that have observed public misperceptions of wind speed risk and storm proximity to home location (Senkbeil et al. 2019, 2020).
Most participants expressed concern over climate change and sea level rise, the potential for future flooding, and the effects on them or their children, and considered these issues as important concerns in their voting decisions. Almost one-half (46%) of respondents felt likely to resettle in the same location after hurricane damage to their home, and only 25% expressed interest in participating in stakeholder discussions.
We present the results of four multivariable logistic regression models fitted for select respondent perceptions of interest for policy makers in Tables S2–S5 of the online supplemental material. There were four key findings related to hurricane knowledge and attitudes:
-
being a retiree and a longtime resident were associated with low surge knowledge (i.e., incorrect knowledge about one’s flood zone);
-
being a longtime resident (i.e., >10 years) and a renter were associated with a lower likelihood of planned resettling or rebuilding after a storm-induced loss of housing;
-
being a retiree who is skeptical of climate change, Hispanic, and a renter were associated with low concern about future floods and storm surges; and
-
being a retiree who is skeptical of climate change, Hispanic, and a renter were associated with low concern about future storm impacts on self and children.
The return-by-mail survey was distributed just 3 months before Hurricane Matthew skirted Florida’s eastern coast in early October 2016 before making landfall in South Carolina on 8 October. Despite our low response rate and subsequent sample size, our series of four simple multivariable analyses still identified demographic characteristics and beliefs that were more associated with low knowledge or levels of concern than issues associated with higher vulnerability to hurricanes. These results generally warrant further inquiry into sources of mobile home resident vulnerability and show promise for more broadly (and with greater geographic precision) identifying the characteristics associated with higher hurricane vulnerability in coastal areas (Tanim and Tobin 2018). The focus group narratives from 2018 filled gaps by contextualizing resident experiences and constraints.
b. Focus groups: Qualitative findings
The results of our focus group discussions were consistent with the tenets of CCT in shaping mobile home residents’ hurricane experiences, particularly the themes of assistance, stress and anxiety, and recovery. We observed limitations at individual, family, community, and government scales, as well as at the intersections of these constraints. Residents often described having reduced decision-making power and control that they perceived as shaped by their socioeconomic condition and closely tied to their social status and structural limitations. Despite the multiple vulnerabilities of MHP communities, many residents expressed coping through the collective action of community help. The community help code under assistance illustrated a strong sense of community, which we interpret as an important expression of resilience.
Our focus group qualitative analysis generated themes organized and categorized by the most common and distinctive experiences reported by residents before, during, and after hurricane events. Figure 2 highlights connections between existing codes on hurricane-related vulnerability across all the focus group discussions.
Thematic map presenting the hierarchy of analytic themes and sample quotations.
Citation: Weather, Climate, and Society 15, 3; 10.1175/WCAS-D-22-0122.1
In Fig. 2, the most prominent and consistent themes throughout the six focus groups are organized into a thematic map with quotations drawn from residents’ responses. The analysis resulted in six overarching themes on hurricane experiences: 1) preparation, 2) evacuation, 3) assistance, 4) stress and anxiety, 5) tree concerns, and 6) recovery.
Residents discussed elements of hurricane preparation that were also related to attitudes and vulnerability of structures. Residents described prehurricane conditions as not only being stressful, but also politically shaped, such as residents not having access to hurricane preparation assistance from the government or fearing a lack of assistance from government agencies for potential damages after a storm: “Some prepare and others don’t prepare. Some don’t care.” Residents also discussed preparing their homes and expressed the lack of concern by other MHP residents. The preparation theme relates to evacuation in that evacuation can be seen as a form of preparation for a hurricane event.
The evacuation theme illustrates the importance of close, reliable social networks of family and friends for shaping evacuation behaviors: “You have to leave, and leave everything.” This theme also describes the realities of some residents who felt they had to evacuate abruptly from their homes, and others who did not (or felt they could not) evacuate because of pets that could not be accommodated in all designated shelters. Residents largely associated difficult hurricane experiences with disparities from lack of resources, citizenship status, or inability to vote and subsequent invisibility to politicians who were perceived as providing constituents with hurricane support in return for future campaign support.
The skirting underneath the house got blown off. FEMA kept screwing around for months and months and months. I didn’t get nothing. All I wanted was skirting around the house.
Disaster assistance challenges was a common topic of conversation that related to the posthurricane recovery theme since residents’ preoccupation was making repairs to their homes and restoring the MHP community.
Emotions of stress and anxiety became a salient theme for experiences before, during, and after a hurricane primarily due to vulnerability residents face. Residents commonly expressed these emotions as “worry” or “fear” on potential and actual tornado danger, flooding, and tree damage as well as the concern for their pets during a hurricane. “The loss of a human is what worries one the most.”
Residents were worried about the potential loss of property, injury, and life as a result of a hurricane event. The narratives around stress and the emotional toll for conditions before, during, and after hurricanes in south Florida (La Greca et al. 2019) have been documented but had not been contextualized for south Florida MHP residents. “Everything flies but it’s the trees that do damage here . . . and the flooding also, but not like the trees.” Tree concerns emerged as an independent theme because they not only described feelings of concern, but trees were discussed as an urgent matter to be addressed during the process of preparation for hurricanes and during and after hurricane events. Residents who had trees near their homes were more worried about evacuation, damage, and recovery efforts. Figure 2 shows the connection between the recovery and subcodes (damage, cleanup, and repairs) as they relate to posthurricane experiences, as well as nuances between the themes. “It [the hurricane] left us without water, without electricity, and the tree fell apart.”
Residents expressed the challenges they faced as they attempted to restore their homes and community for months following the hurricane event. The damage subcodes addressed water and tree damage to the home or MHP as a result of the hurricane, while cleanup described experiences with personal, MHP community, and MHP management cleanup efforts of debris, and repairs surrounded discussions on the process of fixing homes and MHP damages while dealing with a lack of resources.
The circle graph in Fig. 3 highlights the codes with most connectivity having to do with hurricane vulnerability. There were 11 codes that display strong connectivity. The circle graph analysis shows three 4-way connections, six 3-way connections, and one 2-way connection. These connections reflect our analytical and iterative comparison process. For example, the subcode tree damage is related to four other codes: conditions during hurricane, conditions posthurricane, preparations, and potential risk and damage. These five codes form a closed interconnected relationship. These relationships underscore the threat of trees when preparing for a hurricane and the risk trees pose to parks and homes during and after hurricane events. The tree concern theme is supported by the established literature documenting that mobile homes are among the top locations for wind-related tree fatalities during and after hurricanes, thunderstorms, and tornadoes (Schmidlin 2009). Similarly, the codes stress and anxiety, evacuation, no FEMA help, and repairs form a strong relationship that supports the narratives about being worried about limited choices for evacuation due to less support networks, their pets and fear of losing their belongings. The residents discussed experiences they have had in getting help from FEMA and being worried about how they will repair damages because of financial limitations. The residents all expressed not having any real home security because they are unable to get insurance for mobile homes because of homes being too old or not being up to building codes, which revealed serious structural vulnerabilities of government level, community level, and personal socioeconomic constraints of residents. All residents across MHP had concerns and were worried about damage from the last storm caused by trees that the city has not cut down, high winds, and living in a flood zone.
Hurricane vulnerability codes clustered by coding similarity, with the most interconnected codes emphasized by larger font size and darker shade.
Citation: Weather, Climate, and Society 15, 3; 10.1175/WCAS-D-22-0122.1
The graph highlights the impact of trees on events and conditions before, during, and after hurricanes. The salience of this code provides insights into the vulnerability and risk of trees for residents, MHP management, and the local government: “but we have to tell the office to please cut the trees.” Across focus groups, residents voiced concerns about obtaining assistance or supplies to cut down trees, such as removal (or planned removal) of potentially dangerous trees or otherwise physically protecting their homes before future hurricanes. One participant explained, “It’s so dangerous. I am going to show you what fell after . . . A tree that weighed almost two tons on the roof.” Another participant described the damage trees can, and have, caused: “There’s a trailer that a tree and a palm tree fell on top of. It turned it into powder. And thankfully [the residents] had left.” Others explained how MHP management dealt with their tree issues, their desire for expert advice on how to address tree concerns, and how the clean up after the hurricane was primarily around tree debris and damage from fallen trees: “Neighbors had to figure out how to cut the trees.” These participant quotations further highlight the relationship in Fig. 3 between tree concerns and the other codes in the analysis related to vulnerability.
c. In-person survey following focus group
Table 1 presents individual and household characteristics of the 44 in-person survey participants who were permanent residents at the MHPs. Most respondents completed the survey in Spanish (79.5%), were female (59%), had a mean age of 54 (standard deviation 17.5), had an experience with a hurricane prior to Irma (80%) and were personally affected by Irma a lot or extremely (46%), and had dependents (55%) or pets (46%, mostly dogs) in the household.
Individual and household characteristics of 44 mobile home park residents surveyed after Hurricane Irma.
Most residents owned their mobile home (86%) and confirmed that they were living in a mandatory evacuation zone (82%). Most participants (67%) had large trees near their homes, yet 41% had shutters on windows, and 21% had homes constructed to code. No one claimed to have homeowners, renters, or flood insurance before Irma. Most respondents (80%) reported having good or extremely good physical health before Irma, and 21% reported suffering from a chronic health condition. The vast majority of participants sought out media coverage for Irma a lot or all the time (91%), and were in contact with family (91%).
Table 2 summarizes residents’ worry and anxiety before, during, and after Hurricane Irma. Many residents disclosed fear of Irma’s potential impact (75%), believed in the chance of the hurricane hitting their community (84%), and worried that their home would flood (61%) or that they would lose their home (66%). Participants reported that they were scared during Irma (73%), feared for their life (61%), worried about their home and property (77%), worried about close family and friends (87%), and feared for their community (89%). After Hurricane Irma, 46% of participants rated their general anxiety level as a lot or extreme. Many participants were worried about their home and property (77%), close family and friends (84%), their community (84%), their environment (86%), and climate change (86%).
Reported worry and anxiety of 44 mobile home park residents surveyed after Hurricane Irma about their experiences before, during, and after the storm.
Many participants reported lifestyle changes during and after Hurricane Irma (Table 3). Participants reported a worse diet (31.8%), lower food (34.1%) and alcohol consumption (47.7%), less exercise (54.5%), and less sleep (61.4%) during the storm. After the hurricane, diet, exercise, and sleep rebounded somewhat. While no one reported an increase in alcohol use, some residents reported higher-than-usual levels of food consumption (18.2%), especially junk food (15.9%), and prolonged less exercise (38.6%) and sleep (31.8%).
Reported lifestyle characteristics of 44 mobile home park residents during and after Hurricane Irma.
There were important post-Irma impacts on school and work attendance (Table 3). 39% of participants reported that their children missed school (range 4–30 days), and one-half of the survey participants reported missed work (range 2–30 days). Participants attributed missed work to dealing with property damage (37%), evacuation (23%), lack of childcare (18%), and transportation issues (18%). Missed work was an important burden (i.e., reported as a lot or extreme) for 43% of respondents.
Table 4 summarizes the 44 in-person survey participants’ general experiences during and after Hurricane Irma. During Irma, many residents reported having access to hurricane news (75%), having a lot or extreme contact with friends and family (77%), and being supported by friends and family (82%). Most residents spent the hurricane with family (84%) and/or friends (29%). A few stayed home (14%), in a community shelter (5%), at a hotel or rental (5%), or in other places like businesses or a local casino (9%). Some residents encountered health-related challenges during Irma, such as not having access to needed medication (11%), lack of refrigeration for medication due to power loss (14%), no access to needed medical devices (11%) or medical care (18%), or being unable to see a health care provider when needed (21%).
Reported storm experiences of 44 mobile home park residents during and after Hurricane Irma.
After Hurricane Irma, residents experienced the loss of electricity (96%) and running water (39%), and advisories to boil water before consuming (55%). Flooding in the home was minimal, but over one-quarter of residents reported significant property or yard damage, and more than one-half reported having a lot or extreme financial burden from these storm damages. Nearly one-half of the residents who completed the in-person survey reported being displaced for 1–9 days (46%) after the storm. Some residents reported a lot or extreme physical illness (39%) or were unable to obtain prescription drug refills within 2 weeks of the storm (23%).
d. Discussion
Both surveys had questions that were very similar, and the focus groups served to triangulate the survey tools. There were notable differences in responses for some key questions about before, during, and after a hurricane event. For example, for the mail-in survey, MHP residents reported not living in a storm-surge zone (35%) or not knowing whether they lived in a storm-surge zone (39%), yet in the 2018 survey, having experienced Hurricane Irma, most residents reported living in an evacuation zone (82%). Experiencing Irma may have shaped evacuation awareness in these MHPs, as far fewer participants reported staying home for hurricanes in 2018 than in 2016 (14% vs 40%). Participants were also more likely to report concern over climate change after Irma (86% vs 56%).
The findings from the mail-in survey, focus groups narratives and in-person survey provide insights for policy changes surrounding the challenges MHP residents face related to hurricane events. Participants repeatedly expressed interest in affordable and economically sound mobile home insurance options that would improve housing security and alleviate stress and anxiety over storm-induced damages. Better government-provided hurricane preparation assistance geared toward MHPs would protect structures through help with wood to board up homes, reinforcement of mobile homes, addressing tree hazard issues, and could provide better food, water, and energy resources like batteries, flashlights, and solar powered appliances to aid recovery. The MHP managers at our study sites wanted to provide their residents with better hurricane-related assistance. MHP managers are vital gatekeepers and obvious partners for local government agencies that manage hurricane preparation, evacuation plans, and recovery efforts such as cleaning up debris. Unfortunately, our study communities did not receive such services, another indication of community-level constraints.
Participants reported unrealistic buy-out options from FEMA that do not reflect the moving and setup costs of resettling elsewhere. Focus group participants also highlighted significant financial obstacles to fixing their mobile home. Residents disclosed facing challenges in receiving FEMA or other government assistance for repairs like roof leaks. Residents often resorted to repairing and reusing damaged house materials and minimally patching up damages due to cost. Hurricane-prone cities should have better policies to protect MHP residents from this sort of housing insecurity. These policies may include subsidizing repairs and mobile home upgrades and helping communities secure resources like generators during recovery to minimize injury and health-related power use disruption, especially for children and older residents.
Hurricane Irma inadvertently presented an important opportunity to study how logistical factors of hurricane experience—preparation and evacuation challenges, by storm path uncertainties and local politics—shape future risk perceptions and preparation in MHP communities. CCT helped us examine how MHP residents’ choices were constrained by socioeconomic status, lack of access to insurance programs, certain identity status around age, lack of social networks, and immigration status. The findings are also consistent with studies on social connection and decision to evacuate, by Collins et al. (2017, 2018), that found that during Hurricanes Matthew and Irma there was a significant relationship between density and diversity in people’s social network and decision to evacuate or not during a hurricane. Participants were most concerned with being able to obtain supplies to secure their homes for a hurricane landfall, and their limited evacuation options that could accommodate all residents and pets. Residents struggled to get hurricane assistance from local government, and MHP management had limited capacity, but residents reported that the community collaborated to help each other during preparation and recovery. Overall, residents expressed emotional toll from the stress and anxiety associated with losing work during and after the hurricane, home and evacuation insecurity, perceived (and real) dangers from tornadoes, tree and water damage, and the loss of electricity and water.
Mobile home residents face mandatory evacuation orders during hurricanes of any category because of the structural weaknesses of their homes. The propensity for sheltering in place (as indicated by 23% of the return-by-mail survey respondents) may point to disdain for or ignorance about this norm, obliviousness to one’s vulnerability, inability to leave home, or combinations of these factors. Regardless, staying in place is a worrisome choice from an emergency management perspective. Our findings also suggest that for some residents, past evacuation behavior informs future evacuation behavior, as 14% of post-Irma survey participants sheltered at home during Irma. Evacuating out of town was, and continues to be, the least preferred practice, while the preference for local (shorter distance) evacuations is consistent with evacuation literature (Cheng et al. 2008), as is the high preference for staying with friends and family, and the very low preference for evacuating to public shelters (Lindell et al. 2011). However, our survey question regarding previous hurricane experience does not capture important details about the nature of the hurricane experience, such as hurricane severity, and extent of damage, factors that can influence future evacuation behavior (Bowser and Cutter 2015; Tobin 1999). Studies indicate that the relationship between previous hurricane experience and evacuation decision is not consistent, and is complicated by the longstanding difficulty in defining and measuring experience (Baker 1991) as well as in understanding the interactions between various weather experiences in the past (Demuth et al. 2016; Meyer et al. 2018).
Other than evacuation destination choice, there were some other findings that have implications for emergency management. An overwhelming 82.5% of respondents said they would evacuate using personal vehicles. This is consistent with studies on hurricane evacuation vehicle usage, which have found that not only do a vast majority of evacuees use personal vehicles (Baker 2000; Lindell et al. 2011; Siebeneck and Cova 2008), but also that many evacuating households use more than one vehicle (Lindell et al. 2011; Siebeneck et al. 2013; Wu et al. 2012). South Florida is a commuter community with poor public transportation during the best of times. Although special evacuation transportation arrangements are made during hurricanes, our small survey sample was unambiguous across respondent characteristics in not utilizing it. Although our survey did not ask for the reason for this choice, we speculate that it could result from the convenience of using one’s own vehicles or about the inadequacy of public transportation. Regardless of cause, the choice of personal vehicles is likely to complicate evacuation planning challenges with roadblocks, traffic jams, etc. The challenge is further compounded by low awareness of FL 511 across all demographic groups, which is consistent with other studies (Alawadi et al. 2020).
The general perception is that only narrow coastal stretches are at risk of flooding from surges. However, in very low-lying coastal locations with a network of inland water bodies that connect to the ocean, the surge can travel and inundate locations inland from the coast (Fig. 1). Insufficient or incorrect knowledge about surge zone risk reported in our analysis is consistent with the findings in the broader hurricane hazards literature (Frazier et al. 2010; Morrow et al. 2015; South Florida Regional Planning Council et al. 2012). Our survey respondents were more worried about wind damage than flooding. Given the structural vulnerability of mobile homes, this response is unsurprising. However, it could also indicate low perception of flood risks, which may be linked to low knowledge of storm-surge locations.
Our study was limited by a small sample size from the return-by-mail survey, which was compounded by missing responses to several survey items. This prompted the second qualitative wave of our study, but we were able to convene six focus groups, which is the recommended minimum for qualitative data analysis. Also, we only captured the experiences of permanent mobile home residents in south Florida. Seasonal residents—an important demographic group occupying mobile homes in south Florida—generally only visit during the winter months and are unlikely to experience a hurricane, so the vulnerability and risk perception of seasonal mobile home residents may be very different from (and potentially interact with) permanent residents’ risk perception. Although we focused on one of the most vulnerable subpopulations in the region, our analysis was cross-sectional and only provides a partial portrait of risk perception in many MHPs.
4. Conclusions
Our study of mobile home residents in Broward and Miami-Dade Counties provides important insights about the drivers of hurricane vulnerability in MHPs. The near-miss of Hurricane Matthew in 2016, and the experience of Hurricane Irma in 2017, shifted this unawareness and disengagement but revealed several social and structural issues that limit the hurricane response options of mobile home residents. The devastation in Puerto Rico of Hurricane Maria in 2017, the impact of Hurricane Michael in the Florida Panhandle, and the catastrophic damage of Hurricane Florence in the Carolinas in 2018 resulted in increased focus on preparedness improvements and vulnerability reduction (Fucile-Sanchez and Davlasheridze 2020; Ma and Smith 2019). There are roughly 35 000 mobile housing units in the study area, and most are extremely vulnerable and economically marginalized communities. MHPs would benefit from disaster education outreach programs and the development of financial instruments to help households insure their property and recover faster, thus potentially mitigating dependence on social services after a severe storm.
Our study goals were to assess hurricane-related knowledge, attitudes, and practices and to understand the most salient social and structural determinants that shape hurricane preparedness and decision-making. This mixed-method approach provided the opportunity to examine nuance in hurricane vulnerability and some of the constraints experienced by MHP residents. CCT provided a framework for understanding the influence of various social and structural limitations during hurricane events for MHP residents. The examination of social constraints using qualitative and quantitative approaches enhances existing scholarship on hurricane vulnerability by demonstrating how structural factors constrain MHP residents from enacting certain adaptation strategies. Understanding these constraints could help disaster preparedness officials understand barriers to relocation and resilience that lead to unnecessary disparities in stress, injuries, and the perpetuation of poverty through these social and structural disadvantages.
The results underscore the importance of improving hurricane preparedness, planning, risk communication, and engagement of mobile home dwellers, as well as mobile home park owners and managers. The ongoing assessment of hurricane risk perceptions over time, particularly before and after the hurricane season, would help inform the optimum timing and communication channels for hurricane planning and relief by health and safety officials for our most vulnerable neighbors. Future research should aim to increase engagement with MHP management, representatives from community organizations who provide hurricane-related resources, and residents in disaster resilience policy and planning. The role and perspectives of local residents, MHP management staff, government officials, and stakeholders who provide hurricane-related assistance are all vital for limiting the vulnerability of MHP residents, improving their options and dignity, and mitigating their reliance on social safety net programs after hurricane events.
Data availability statement.
Because of privacy and ethical concerns, focus group data cannot be made available. Survey data are available from the corresponding author upon reasonable request.
REFERENCES
Alawadi, R., P. Murray-Tuite, D. Marasco, S. Ukkusuri, and Y. Ge, 2020: Determinants of full and partial household evacuation decision making in Hurricane Matthew. Transp. Res., 83D, 102313, https://doi.org/10.1016/j.trd.2020.102313.
Aldrich, N., and W. F. Benson, 2008: Disaster preparedness and the chronic disease needs of vulnerable older adults. Prev. Chronic Dis., 5, A27.
Ash, K. D., 2017: A qualitative study of mobile home resident perspectives on tornadoes and tornado protective actions in South Carolina, USA. GeoJournal, 82, 533–552, https://doi.org/10.1007/s10708-016-9700-8.
Ash, K. D., M. J. Egnoto, S. M. Strader, W. S. Ashley, D. B. Roueche, K. E. Klockow-McClain, D. Caplen, and M. Dickerson, 2020: Structural forces: Perception and vulnerability factors for tornado sheltering within mobile and manufactured housing in Alabama and Mississippi. Wea. Climate Soc., 12, 453–472, https://doi.org/10.1175/WCAS-D-19-0088.1.
Baker, E. J., 1991: Hurricane evacuation behavior. Int. J. Mass Emerg. Disasters, 9, 287–310, https://doi.org/10.1177/028072709100900210.
Baker, E. J., 2000: Hurricane evacuation in the United States. Storms, Vol. 1, Routledge Press, 306–319.
Balaguru, K., D. R. Judi, and L. R. Leung, 2016: Future hurricane storm surge risk for the U.S. Gulf and Florida coasts based on projections of thermodynamic potential intensity. Climatic Change, 138, 99–110, https://doi.org/10.1007/s10584-016-1728-8.
Barbour, R. S., 2008: Introducing Qualitative Research: A Student Guide to the Craft of Doing Qualitative Research. Sage, 319 pp.
Bird, C. E., and P. P. Rieker, 2008: Gender and Health: The Effects of Constrained Choices and Social Policies. Cambridge University Press, 256 pp.
Bowser, G. C., and S. L. Cutter, 2015: Stay or go? Examining decision making and behavior in hurricane evacuations. Environment, 57, 28–41, https://doi.org/10.1080/00139157.2015.1089145.
Brodar, K. E., A. M. La Greca, N. Tarlow, and J. S. Comer, 2020: “My kids are my priority”: Mothers’ decisions to evacuate for Hurricane Irma and evacuation intentions for future hurricanes. J. Fam. Issues, 41, 2251–2274, https://doi.org/10.1177/0192513X20933931.
Burton, C. G., S. Rufat, and E. Tate, 2018: Social vulnerability: Conceptual foundations and geospatial modeling. Vulnerability and Resilience to Natural Hazards, Cambridge University Press, 53–81.
Chakalian, P., L. Kurtz, and D. Hondula, 2019: After the lights go out: Household resilience to electrical grid failure following Hurricane Irma. Nat. Hazards Rev., 20, 05019001, https://doi.org/10.1061/(ASCE)NH.1527-6996.0000335.
Chatterjee, C., and P. Mozumder, 2014: Understanding household preferences for hurricane risk mitigation information: Evidence from survey responses. Risk Anal., 34, 984–996, https://doi.org/10.1111/risa.12196.
Cheng, G., C. G. Wilmot, and E. J. Baker, 2008: A destination choice model for hurricane evacuation. Proc. 87th Transportation Research Board Annual Meeting, Washington, DC, Transportation Research Board, 13–17, https://www.ltrc.lsu.edu/pdf/2008/08-0840.PDF.
Collins, J., R. Ersing, and A. Polen, 2017: Evacuation decision-making during Hurricane Matthew: An assessment of the effects of social connections. Wea. Climate Soc., 9, 769–776, https://doi.org/10.1175/WCAS-D-17-0047.1.
Collins, J., R. Ersing, A. Polen, M. Saunders, and J. Senkbeil, 2018: The effects of social connections on evacuation decision making during Hurricane Irma. Wea. Climate Soc., 10, 459–469, https://doi.org/10.1175/WCAS-D-17-0119.1.
Corbin, J., and A. Strauss, 2008: Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. 3rd ed. Sage, 28 pp.
Cutter, S. L., and C. T. Emrich, 2006: Moral hazard, social catastrophe: The changing face of vulnerability along the hurricane coasts. Ann. Amer. Acad. Pol. Soc. Sci., 604, 102–112, https://doi.org/10.1177/0002716205285515.
Danielle, P. 2017: Florida has 828,000 mobile homes. Less than a third were built to survive a hurricane. Washington Post, 12 September, https://www.washingtonpost.com/news/wonk/wp/2017/09/12/florida-has-828000-mobile-homes-only-half-are-insured/.
Demuth, J. L., R. E. Morss, J. K. Lazo, and C. Trumbo, 2016: The effects of past hurricane experiences on evacuation intentions through risk perception and efficacy beliefs: A mediation analysis. Wea. Climate Soc., 8, 327–344, https://doi.org/10.1175/WCAS-D-15-0074.1.
Desbarats, J., 1983: Spatial choice and constraints on behavior. Ann. Assoc. Amer. Geogr., 73, 340–357, https://doi.org/10.1111/j.1467-8306.1983.tb01421.x.
Edwards, J. N., D. L. Klemmack, and L. Hatos, 1973: Social participation patterns among mobile-home and single-family dwellers. Soc. Forces, 51, 485–489, https://doi.org/10.2307/2576695.
Elliott, J. R., and J. Pais, 2006: Race, class, and Hurricane Katrina: Social differences in human responses to disaster. Soc. Sci. Res., 35, 295–321, https://doi.org/10.1016/j.ssresearch.2006.02.003.
Ersing, R. L., C. Pearce, J. Collins, M. E. Saunders, and A. Polen, 2020: Geophysical and social influences on evacuation decision-making: The case of Hurricane Irma. Atmosphere, 11, 851, https://doi.org/10.3390/atmos11080851.
Esnard, A.-M., A. Sapat, and D. Mitsova, 2011: An index of relative displacement risk to hurricanes. Nat. Hazards, 59, 833–859, https://doi.org/10.1007/s11069-011-9799-3.
FEMA, 2018: Hurricane Irma in Florida: Building performance observations, recommendations, and technical guidance. FEMA Mitigation Assessment Team Rep. FEMA P-2023, 181 pp., https://www.fema.gov/sites/default/files/2020-07/mat-report_hurricane-irma_florida.pdf.
Finch, C., C. T. Emrich, and S. L. Cutter, 2010: Disaster disparities and differential recovery in New Orleans. Popul. Environ., 31, 179–202, https://doi.org/10.1007/s11111-009-0099-8.
Fitzpatrick, K., 2002: Unhealthy Places: The Ecology of Risk in the Urban Landscape. Taylor and Francis, 288 pp.
Frazier, T. G., N. Wood, B. Yarnal, and D. H. Bauer, 2010: Influence of potential sea level rise on societal vulnerability to hurricane storm-surge hazards, Sarasota County, Florida. Appl. Geogr., 30, 490–505, https://doi.org/10.1016/j.apgeog.2010.05.005.
Fucile-Sanchez, E., and M. Davlasheridze, 2020: Adjustments of socially vulnerable populations in Galveston County, Texas USA following Hurricane Ike. Sustainability, 12, 7097, https://doi.org/10.3390/su12177097.
GeoPlan Center, University of Florida, 2017: Florida parcel data statewide—2018. Florida Geographic Data Layer, accessed 1 June 2018, https://fgdl.org/meta/PARCELS.xml.
Hager, M. A., S. Wilson, T. H. Pollak, and P. M. Rooney, 2003: Response rates for mail surveys of nonprofit organizations: A review and empirical test. Nonprofit Voluntary Sect. Quart., 32, 252–267, https://doi.org/10.1177/0899764003032002005.
Haney, T. J., J. R. Elliott, and E. Fussell, 2007: Families and hurricane response: Evacuation, separation, and the emotional toll of Hurricane Katrina. The Sociology of Katrina: Perspectives on a Modern Catastrophe, Rowman & Littlefield, 71–90.
Howell, J., and J. Elliott, 2019: Damages done: The longitudinal impacts of natural hazards on wealth inequality in the United States. Soc. Probl., 66, 448–467, https://doi.org/10.1093/socpro/spy016.
Huang, S.-K., M. K. Lindell, and C. S. Prater, 2016: Who leaves and who stays? A review and statistical meta-analysis of hurricane evacuation studies. Environ. Behav., 48, 991–1029, https://doi.org/10.1177/0013916515578485.
Hudak, S., 2017: Hurricane Irma chases some trailer dwellers out, but others hunker down. Orlando Sentinel, 12 June, https://www.orlandosentinel.com/2017/09/09/hurricane-irma-chases-some-trailer-dwellers-out-but-others-hunker-down/.
Irish, J. L., D. T. Resio, and J. J. Ratcliff, 2008: The influence of storm size on Hurricane Surge. J. Phys. Oceanogr., 38, 2003–2013, https://doi.org/10.1175/2008JPO3727.1.
John, S., 2015: U.S. coastline vulnerability to hurricanes is growing to unprecedented levels. Washington Post, 26 August, https://www.washingtonpost.com/news/capital-weather-gang/wp/2015/08/26/u-s-coastline-vulnerability-to-hurricanes-is-growing-to-unprecedented-levels/.
Karaye, I. M., A. D. Ross, M. Perez-Patron, C. Thompson, N. Taylor, and J. A. Horney, 2019: Factors associated with self-reported mental health of residents exposed to Hurricane Harvey. Prog. Disaster Sci., 2, 100016, https://doi.org/10.1016/j.pdisas.2019.100016.
Kusenbach, M., 2017a: “Look at my house!” Home and mobile home ownership among Latino/a immigrants in Florida. J. Housing Built Environ., 32, 29–47, https://doi.org/10.1007/s10901-015-9488-8.
Kusenbach, M., 2017b: “It’s not where I’d be running like an idiot for a small one:” Hurricane perceptions and evacuation decision making among Florida mobile home residents. Int. J. Mass Emerg. Disasters, 35, 91–119, https://doi.org/10.1177/028072701703500206.
Kusenbach, M., J. Simms, and G. Tobin, 2010: Disaster vulnerability and evacuation readiness: Coastal mobile home residents in Florida. Nat. Hazards, 52, 79–95, https://doi.org/10.1007/s11069-009-9358-3.
LaGory, M., 1982: Toward a sociology of space: The constrained choice model. Symbolic Interact., 5, 65–78, https://doi.org/10.1525/si.1982.5.1.65.
La Greca, A. M., K. E. Brodar, B. A. Danzi, N. Tarlow, K. Silva, and J. S. Comer, 2019: Before the storm: Stressors associated with the Hurricane Irma evacuation process for families. Disaster Med. Public Health Prep., 13, 63–73, https://doi.org/10.1017/dmp.2019.9.
Lindell, M. K., J. E. Kang, and C. S. Prater, 2011: The logistics of household hurricane evacuation. Nat. Hazards, 58, 1093–1109, https://doi.org/10.1007/s11069-011-9715-x.
Litman, T., 2006: Lessons from Katrina and Rita: What major disasters can teach transportation planners. J. Transp. Eng., 132, 11–18, https://doi.org/10.1061/(ASCE)0733-947X(2006)132:1(11).
Lovekamp, W. E., 2008: The sociology of Katrina: Perspectives on a modern catastrophe. Contemp. Sociol., 37, 260–261, https://doi.org/10.1177/009430610803700336.
Ma, C., and T. Smith, 2019: Vulnerability of renters and low-income households to storm damage: Evidence from Hurricane Maria in Puerto Rico. Amer. J. Public Health, 110, 196–202, https://doi.org/10.2105/AJPH.2019.305438.
MacTavish, K. A., 2007: The wrong side of the tracks: Social inequality and mobile home park residence. Community Dev., 38, 74–91, https://doi.org/10.1080/15575330709490186.
Masozera, M., M. Bailey, and C. Kerchner, 2007: Distribution of impacts of natural disasters across income groups: A case study of New Orleans. Ecol. Econ., 63, 299–306, https://doi.org/10.1016/j.ecolecon.2006.06.013.
Meyer, M. A., B. Mitchell, J. C. Purdum, K. Breen, and R. L. Iles, 2018: Previous hurricane evacuation decisions and future evacuation intentions among residents of southeast Louisiana. Int. J. Disaster Risk Reduct., 31, 1231–1244, https://doi.org/10.1016/j.ijdrr.2018.01.003.
Mitsova, D., A.-M. Esnard, A. Sapat, and B. Lai, 2018: Socioeconomic vulnerability and electric power restoration timelines in Florida: The case of Hurricane Irma. Nat. Hazards, 94, 689–709, https://doi.org/10.1007/s11069-018-3413-x.
Morgan, D. L., 1997: The Focus Group Guidebook. Vol. 1, Sage, 120 pp.
Morrow, B. H., J. K. Lazo, J. Rhome, and J. Feyen, 2015: Improving storm surge risk communication: Stakeholder perspectives. Bull. Amer. Meteor. Soc., 96, 35–48, https://doi.org/10.1175/BAMS-D-13-00197.1.
Murphy, B., 2017: Mobile home residents without insurance in FL flood zones fall through the cracks. Naples Daily News, 30 September, https://www.naplesnews.com/story/weather/hurricanes/2017/09/30/mobile-home-residents-florida-flood-zones-without-insurance-fall-through-cracks/709388001/.
Myers, C., T. Slack, and J. Singelmann, 2008: Social vulnerability and migration in the wake of disaster: The case of Hurricanes Katrina and Rita. Popul. Environ., 29, 271–291, https://doi.org/10.1007/s11111-008-0072-y.
Pais, J. F., and J. R. Elliott, 2008: Places as recovery machines: Vulnerability and neighborhood change after major hurricanes. Soc. Forces, 86, 1415–1453, https://doi.org/10.1353/sof.0.0047.
Peacock, W. G., B. H. Morrow, and H. Gladwin, 1997: Hurricane Andrew: Ethnicity, Gender, and the Sociology of Disasters. Routledge, 277 pp.
Pescosolido, B. A., 1992: Beyond rational choice: The social dynamics of how people seek help. Amer. J. Sociol., 97, 1096–1138, https://doi.org/10.1086/229863.
Pielke, R. A., J. Gratz, C. W. Landsea, D. Collins, M. A. Saunders, and R. Musulin, 2008: Normalized hurricane damage in the United States: 1900–2005. Nat. Hazards Rev., 9, 29–42, https://doi.org/10.1061/(ASCE)1527-6988(2008)9:1(29).
Prasad, S., 2016: Assessing the need for evacuation assistance in the 100 year floodplain of south Florida. Appl. Geogr., 67, 67–76, https://doi.org/10.1016/j.apgeog.2015.12.005.
Prasad, S., and J. Stoler, 2016: Mobile home residents and hurricane vulnerability in south Florida: Research gaps and challenges. Int. J. Disaster Risk Sci., 7, 436–439, https://doi.org/10.1007/s13753-016-0101-x.
Reid, M., 2013: Social policy, “deservingness,” and sociotemporal marginalization: Katrina survivors and FEMA. Sociol. Forum, 28, 742–763, https://doi.org/10.1111/socf.12051.
Rickless, D. S., G. E. Wilt, J. D. Sharpe, N. Molinari, W. Stephens, and T. T. LeBlanc, 2021: Social vulnerability and access of local medical care during Hurricane Harvey: A spatial analysis. Disaster Med. Public Health Prep., 17, e12, https://doi.org/10.1017/dmp.2020.421.
Rieker, P. P., and C. E. Bird, 2005: Rethinking gender differences in health: Why we need to integrate social and biological perspectives. J. Gerontol., 60B, S40–S47.
Rufat, S., E. Tate, C. T. Emrich, and F. Antolini, 2019: How valid are social vulnerability models? Ann. Amer. Assoc. Geogr., 109, 1131–1153, https://doi.org/10.1080/24694452.2018.1535887.
Rumbach, A., E. Sullivan, and C. Makarewicz, 2020: Mobile home parks and disasters: Understanding risk to the third housing type in the United States. Nat. Hazards Rev., 21, 05020001, https://doi.org/10.1061/(ASCE)NH.1527-6996.0000357.
Saldaña, J., 2009: The Coding Manual for Qualitative Researchers. Sage, 440 pp.
Schmidlin, T. W., 2009: Human fatalities from wind-related tree failures in the United States, 1995–2007. Nat. Hazards, 50, 13–25, https://doi.org/10.1007/s11069-008-9314-7.
Schmidlin, T. W., B. O. Hammer, Y. Ono, and P. S. King, 2009: Tornado shelter-seeking behavior and tornado shelter options among mobile home residents in the United States. Nat. Hazards, 48, 191–201, https://doi.org/10.1007/s11069-008-9257-z.
Schreiber, S., 2005: Mobile homes and hurricanes: The crisis in Florida. Encounters/Encuentros/Recontres: Proceedings of the 2005 ACSA International Conference, D. Covo and G. Mérigo, Eds., ACSA, 39–47, https://www.acsa-arch.org/proceedings/International%20Proceedings/ACSA.Intl.2005/ACSA.Intl.2005.5.pdf.
Senkbeil, J., J. Collins, and J. Reed, 2019: Evacuee perception of geophysical hazards for Hurricane Irma. Wea. Climate Soc., 11, 217–227, https://doi.org/10.1175/WCAS-D-18-0019.1.
Senkbeil, J., and Coauthors, 2020: Perceptions of hurricane-track forecasts in the United States. Wea. Climate Soc., 12, 15–29, https://doi.org/10.1175/WCAS-D-19-0031.1.
Siebeneck, L. K., and T. J. Cova, 2008: An assessment of the return-entry process of Hurricane Rita 2005. Int. J. Mass Emerg. Disasters, 26, 91–111, https://doi.org/10.1177/028072700802600202.
Siebeneck, L. K., M. Lindell, C. Prater, H.-C. Wu, and S.-K. Huang, 2013: Evacuees’ reentry concerns and experiences in the aftermath of Hurricane Ike. Nat. Hazards, 65, 2267–2286, https://doi.org/10.1007/s11069-012-0474-0.
Simmons, K., and D. Sutter, 2008: Manufactured home building regulations and the February 2, 2007 Florida tornadoes. Nat. Hazards, 46, 415–425, https://doi.org/10.1007/s11069-007-9192-4.
South Florida Regional Planning Council, Florida State Emergency Response Team, and the East Central Florida Regional Planning Council, 2012: Regional behavioral analysis summary. Statewide Regional Evacuation Study Program, South Florida Region Technical Data Rep, Vol. 1-11, 36 pp., https://sfregionalcouncil.org/wp-content/uploads/2019/08/Vol1-11_ChIII.pdf.
Strader, S. M., and W. S. Ashley, 2018: Finescale assessment of mobile home tornado vulnerability in the central and southeast United States. Wea. Climate Soc., 10, 797–812, https://doi.org/10.1175/WCAS-D-18-0060.1.
Strader, S. M., K. Ash, E. Wagner, and C. Sherrod, 2019: Mobile home resident evacuation vulnerability and emergency medical service access during tornado events in the southeast United States. Int. J. Disaster Risk Reduct., 38, 101210, https://doi.org/10.1016/j.ijdrr.2019.101210.
Stuart, D., and R. Schewe, 2016: Constrained choice and climate change mitigation in US agriculture: Structural barriers to a climate change ethic. J. Agric. Environ. Ethics, 29, 369–385, https://doi.org/10.1007/s10806-016-9605-z.
Sullivan, E., 2017: Displaced in place: Manufactured housing, mass eviction, and the paradox of state intervention. Amer. Sociol. Rev., 82, 243–269, https://doi.org/10.1177/0003122416688667.
Sullivan, E., 2018: Manufactured Insecurity: Mobile Home Parks and Americans’ Tenuous Right to Place. University of California Press, 250 pp.
Tanim, S., and G. A. Tobin, 2018: Social factors and evacuation vulnerability: An application in Pinellas County, Florida. Pap. Appl. Geogr., 4, 123–136, https://doi.org/10.1080/23754931.2016.1266020.
Tobin, G. A., 1999: Sustainability and community resilience: The Holy Grail of hazards planning? Global Environ. Change, 1B, 13–25, https://doi.org/10.1016/S1464-2867(99)00002-9.
Tobin, G. A., and J. Burton, 2009: Building codes and residential structures: Spatial and temporal patterns of hurricane wind damage. Applied Geography Conf., Baton Rouge, LA, AGC, 134–143, https://digitalcommons.usf.edu/geo_facpub/81/.
Vuolo, M., J. Kadowaki, and B. C. Kelly, 2016: A multilevel test of constrained choices theory: The case of tobacco clean air restrictions. J. Health Soc. Behav., 57, 351–372, https://doi.org/10.1177/0022146516653790.
Wu, H.-C., M. K. Lindell, and C. S. Prater, 2012: Logistics of hurricane evacuation in Hurricanes Katrina and Rita. Transp. Res., 15F, 445–461, https://doi.org/10.1016/j.trf.2012.03.005.
Wynter-Minott, S., 2017: Hurricane Irma and Florida: Learning from our disaster experiences; dialysis facilities. Nephrol. Nurs. J., 44, 500.
Zandbergen, P., 2009: Exposure of US counties to Atlantic tropical storms and hurricanes, 1851–2003. Nat. Hazards, 48, 83–99, https://doi.org/10.1007/s11069-008-9250-6.
Zhang, K., 2011: Analysis of non-linear inundation from sea-level rise using lidar data: A case study for south Florida. Climatic Change, 106, 537–565, https://doi.org/10.1007/s10584-010-9987-2.
Zolnikov, T. R., 2018: A humanitarian crisis: Lessons learned from Hurricane Irma. Amer. J. Public Health, 108, 27–28, https://doi.org/10.2105/AJPH.2017.304192.
