Vol.:(0123456789)

Natural Hazards (2021) 109:1097–1118https://doi.org/10.1007/s11069-021-04870-y

1 3

ORIGINAL PAPER

The six capacities of community resilience: evidence from three small Texas communities impacted by Hurricane Harvey

Thomas W. Haase1  · Wen‑Jiun Wang1  · Ashley D. Ross2

Received: 19 November 2020 / Accepted: 10 June 2021 / Published online: 25 June 2021 © The Author(s), under exclusive licence to Springer Nature B.V. 2021

AbstractThis article builds upon disaster scholarship that suggests community resilience is driven by six capacities: social, economic, physical, human, institutional, and environmental. Together, these capacities constitute a conceptual framework that can be used to investi-gate and assess community resilience. While recent scholarship has provided insights into how resilience operates in large communities, there remain questions about whether this conceptual framework is appropriate for the study of resilience in small communities. To narrow this knowledge gap, we conducted interviews with twenty-six subjects from three small Texas communities affected by Hurricane Harvey: City of Dickinson; City of Port Aransas; and Town of Refugio. Analysis of the interview data confirms that the six capaci-ties of resilience provide an appropriate framework for the investigation of resilience in small communities. Given the complex and dynamic nature of community resilience, the findings also suggest that it is unlikely policymakers will be able to develop a unified pol-icy solution for hazard events that is appropriate for all communities. Rather, policymak-ers need to consider community-based resilience solutions, driven by local strengths and weaknesses, that facilitate the reduction of risks associated with hazard events.

Keywords Community resilience · Hurricane Harvey · Rural · Natural hazard · Disaster response

1 Introduction

To reduce the risk and consequences of catastrophic hazard events, disaster management scholars have sought to define and conceptualize resilience (Comfort et  al. 2010; Demiroz and Haase 2018), identify the capacities and processes related to resilience (Ross 2013), and specify how resilience can be operationalized and measured (Peacock et al. 2010; Cutter et al.

* Wen-Jiun Wang [email protected]

1 Department of Political Science, Sam Houston State University, Huntsville, TX 77340, USA2 Department of Marine and Coastal Environmental Science, Texas A&M University at Galveston,

Galveston, TX 77553, USA

1098 Natural Hazards (2021) 109:1097–1118

1 3

2010; Cox and Hamlen 2015). This scholarship, in conjunction with increases in the strength and severity of hazard events over the last several decades, has encouraged governments to pursue resilience as a public policy goal (United Nations 2015; Office of the President of the United States 2017). Parallel to these developments, scholars have developed frameworks that identify the universal drivers—or mechanisms—of resilience, especially for metropolitan communities and national infrastructure systems (Cutter et al. 2016).

Disaster scholars such as Cutter et  al. (2016, p. 1250), however, have argued that “the drivers of disaster resilience are different, [which suggests] that efforts to improve disaster resilience must be customized to the local context and not applied universally to all places.” Although often situated at the periphery of disaster management scholarship (Brennan and Flint 2007), the resilience of nonurban, typically rural, communities has become the focus of inquiry. In the American context, researchers have explored how administrators perceive and practice emergency management in rural counties in Florida (Kapucu 2016) and the Gulf Coast (Ross 2013), how public wealth levels affect the post-disaster fiscal health of rural com-munities (Fannin et al. 2012), how “inherent resilience” has helped natural resource-depend-ent communities in coastal Louisiana cope with disaster events (Colten et al. 2012), and how social capacities have emerged in rural communities in post-disaster contexts (Ross and Clay 2018). These findings confirm that the dynamics of resilience are similar across communi-ties of different population sizes. These findings also reveal questions about how community characteristics impact resilience, for example, whether smaller communities differ from their larger—often urban—counterparts in terms of their strengths and needs (Kapucu et al. 2013). To answer such questions, however, researchers must first consider whether, and the extent to which, the resilience frameworks used to assess the resilience of larger communities are also applicable to smaller communities.

This article explores the mechanisms that influenced the resilience of three small Texas communities affected by Hurricane Harvey: City of Dickinson, Galveston County; City of Port Aransas, Nueces County; and Town of Refugio, Refugio County. Like the Houston metropolitan region, Hurricane Harvey wrought substantial damage upon these three com-munities. Unlike the Houston metropolitan region, these communities did not receive substan-tial media or research attention after the largest catastrophic event, in terms of cost, to have struck the region since Hurricane Katrina (Blake and Zelinsky 2018). In response to the call to understand the mechanisms of resilience of all places (Cutter et al. 2016), this study was guided by two questions. First, to what extent were the six capacities of resilience—social, economic, physical, human, institutional, and environmental—present in these three commu-nities? Second, what factors, if any, facilitated or inhibited the ability of these three communi-ties to prepare for and respond to the Hurricane Harvey event? To address these questions, we conducted interviews with decision makers present in these communities during Hurricane Harvey. We then assessed whether the perceptions of resilience reported by these decision makers corresponded with the six capacities of resilience. The findings suggest that the six capacities of resilience provide an appropriate framework for the investigation of resilience in small communities. The findings also indicate, however, that the six capacities framework does not provide policymaker with a unified prescription for the development of resilience for all communities.

1099Natural Hazards (2021) 109:1097–1118

1 3

2 Literature review

Contemporary resilience discussions can be linked to Holling (1973, p. 17), who argued resilience represents a system’s ability to “to absorb changes of state variables, driving variables, and parameters, and still persist.” Disaster scholars interpret this definition of resilience in different ways. For example, Tierney and Bruneau (2007, p. 15) define resil-ience as “the ability of social units (e.g., organizations, communities) to mitigate hazards, contain the effects of disasters when they occur, and carry out recovery activities in ways that minimize social disruptions and mitigate the effects of future disasters.” Similarly, Ride and Bretherton (2011, p. 7) define resilience as “the capacity of a community to cope with the emergency, to rebuild, and learn from the experience, such that the new physical, social, and political structures are better adapted to the environment.” Despite the inherent similarities of these definitions, scholars have yet to reach a consensus on what resilience means (Manyena 2006, 2014). Nevertheless, governmental entities at both the national and international levels have indicated that resilience is a research imperative and identified resilience as an important public policy goal (National Research Council 2012; United Nations 2015; Office of the President of the United States 2017).

2.1 Conceptualizations of resilience

While resilience definitions suggest that individuals, communities, and systems can persist through disruptive events, there are inconsistencies within the literature about how resil-ience should be conceptualized (Haase and Demiroz 2019). According to Demiroz and Haase (2018), resilience scholarship can be organized into three separate lines of thinking, each of which adopt a different approach to how systems react to disruptive disturbances. Scholars that pursue the first line of thinking conceptualize resilience as a system’s ability to resist disturbances. This conceptualization considers the extent to which a system pos-sesses rigidity (Kendra and Wachtendorf 2003; Gerasimidis et al. 2017) or redundancies (Wildavsky 1988). The second line of thinking conceptualizes resilience as an ability to bounce back after a disruption. This conceptualization considers the extent to which a sys-tem can absorb the effects of a disturbance and the time it takes to reestablish normal oper-ations (Tierney and Bruneau 2007). The third line of thinking conceptualizes resilience as an ability to adapt or adjust to a disruption. This conceptualization suggests that a system, properly designed, can possess the flexibility to transition from one operational state to another, thereby responding to the disruption without experiencing a severe degradation in its operational capacity. Rather than bouncing back, the system takes on a new form (Ross 2013; Haase et al. 2017). Others frame this third conceptualization differently, suggesting that resilience represents a system’s ability to “bounce forward” or to strengthen its institu-tions and structures after a disaster event (Manyena et al. 2011).

2.2 The six capacities of community resilience

Drawing upon the definitions and conceptualizations identified above, scholars have also developed frameworks that can be used to operationalize and measure resilience in at-risk communities (Cutter et al. 2010; Peacock et al. 2010; Cox and Hamlen 2015; Ross 2013). Although an in-depth discussion of these frameworks is beyond the scope of this article, they suggest that community resilience is driven by six core capacities. These core capaci-ties include human capital, physical capital, social capital, economic capital, institutional

1100 Natural Hazards (2021) 109:1097–1118

1 3

capital, and environmental capital. For communities that experience a disruptive event, it is these six capacities that drive a community’s ability to resist, bounce back, or bounce for-ward. Depending upon a communities’ characteristics, these six capacities can also work—individually and/or collectively—to undermine or reduce community resilience. For exam-ple, social capital that is predominantly comprised of bonding ties may inhibit full and equitable disaster recovery in a community (Elliott et al. 2010; Aldrich 2012). While com-munity resilience frameworks label and describe these six core capacities in different ways, they have substantial conceptual overlap.

2.2.1 Human capital

Human capital is perhaps the most important of the six capacities of community resilience. For Cox and Hamlen (2015, p. 225), human capital represents the “knowledge and skills translatable to disaster preparedness, mitigation, response, and recovery.” Such knowledge and skills are represented by the education level, language capacity, and overall level of health of those who live in a community (Peacock et al. 2010, p. 17). Others characterize human capital through measures of communication capacity (e.g., access to telephones) and transportation capacity (e.g., access to a vehicle) (Cutter et  al. 2010; Ross 2013). Human capital plays an essential role in the development of resilience because it enables the members of a community to access the benefits derived from the five other capaci-ties of community resilience. However, access may differ from community to community. There may also be trade-offs between the various types of human capital. For example, while smaller communities are often considered to have lower overall levels of human cap-ital—at least in terms of recruiting and retaining qualified personnel—those who live in these communities may have extensive knowledge and skills of how to deal with localized hazards and disaster events (Colten et al. 2012; Cox and Hamlen 2015).

2.2.2 Physical capital

A community’s physical capital—or infrastructure—can also promote or constrain its resil-ience. Infrastructure can not only be used to mitigate risk, and it can also be used to pro-mote information and resource access, communication, and coordination during disaster events. According to Peacock et  al. (2010, p. 17), physical capital consists of “the basic infrastructure or built environmental features that can help people, households, and busi-nesses support their livelihood.” This infrastructure can include a variety of built systems, including transportation systems, communication systems, lifeline systems, as well as criti-cal emergency and healthcare systems. Others take a broader perspective on physical capi-tal. Cutter et  al. (2010) and Ross (2013), for example, also consider home type (mobile homes vs. nonmobile homes), home age, shelter capacity (vacant rental units and hotel density), and the number of schools to be important characteristics of physical capital. The development and maintenance of physical capital can be difficult for some communities, especially those with limited revenue streams, as long-term infrastructure investments can present significant economic costs.

2.2.3 Economic capital

Economic capital is comprised of the financial assets and “resources that people and households use to support their livelihoods and maintain their living standards” (Peacock

1101Natural Hazards (2021) 109:1097–1118

1 3

et  al. 2010, p. 18). These resources are reflected by home ownership rates, employment rates, economic diversity, income equality, and the robustness of the private sector (Cutter et al. 2010; Ross 2013). Tax dollars to support emergency management personnel, equip-ment, and operations most directly affect community resilience for disaster preparedness and response. In both large and small communities, economic capital can quickly become exhausted after a disaster. For example, job market disruptions can have a significant nega-tive effect on the overall tax revenue of the locality, which may inhibit the community’s ability undertake disaster management activities over the long term (Chang and Shinozuka 2004; Andrew and Ismayilov 2017).

2.2.4 Social capital

Community resilience is also facilitated by social capital or the “social networks and resources that people and households can draw upon to support their livelihoods and other needs” (Peacock et al. 2010, p. 17). During a disaster, social capital facilitates the collection and distribution of information and resources (Aldrich and Meyer 2015; Aldrich 2012). As such, the presence or absence of social capital can impact a community’s health and vul-nerability (Cox and Hamlen 2015). Although social capital is often thought of as the webs of interpersonal relationships that exist within a community, some disaster researchers con-sider place attachment and political engagement to also be social determinants of com-munity resilience (Cutter et al. 2010; Ross 2013). Together, these lines of thinking suggest that interpersonal connections help to build trust, which, in turn, can facilitate cooperation and collaboration within a community (Ross 2013). A community with especially strong social ties and a deep sense of attachment may still be resilient to disaster events (Colten et al. 2012; Simms 2017), even if the community has a deficit in one or more of the other resilience capitals.

2.2.5 Institutional capital

National governments are increasingly requiring that local communities undertake the institutional tasks and responsibilities associated with comprehensive disaster manage-ment. These institutional tasks and responsibilities include mitigation planning, response planning, code enforcement, community engagement, and training (Waugh 2010; Kapucu et al. 2013). In their exploration of institutional capital in the disaster management context, Cutter et al. (2010) and Ross (2013) suggest that institutional capital is represented by miti-gation plan coverage, flood insurance coverage, political fragmentation, municipal service expenditures, and previous disaster experience. Cox and Hamlen (2015, p. 225) also stress the importance of institutional capital, but they approach the concept from two perspec-tives: governance and disaster preparedness. Together, these two dimensions capture the essence of institutional capital: that government needs to have the ability to engage the public and the ability to take effective action across all four phases of the disaster manage-ment cycle, mitigation, preparation, response, and recovery.

2.2.6 Environmental capital

The environment can contribute to a communities’ ability to reduce disaster risk and vul-nerability. Ross (2013, p. 105), for example, notes that “the preservation and/or restoration or natural defense features such as wetlands” is an important aspect of coastal community

1102 Natural Hazards (2021) 109:1097–1118

1 3

resilience. Narrowly, environment capital is determined by the extent to which a commu-nity’s environment has been restored or preserved, the prevalence of impervious surfaces such as building and parking lots, and whether development occurs within designated floodplains. More broadly, Cox and Hamlen (2015, p. 225) note that environmental capital, what they refer to as natural capital, is reflected by community characteristics such as natu-ral resources, resource management, energy production and supply, geography, geographic isolation, environmental conditions, and the state of ecological and natural systems. Like the five other community resilience capacities, environmental capital can increase or decrease community resilience, depending on how the community has chosen to interact with the natural environment.

2.3 The impact of community population size on resilience

A community’s characteristics will impact its resilience, which will impact its demand for response services after a catastrophic event (Comfort et al. 2010). Two community charac-teristics that seem to have a substantial impact on resilience is population size and popula-tion density (Kapucu et al. 2013). Thus, when it comes to understanding resilience, it is important to draw theoretical and empirical distinctions between communities according to their population characteristics. Yet, researchers have only recently begun to investigate how the resilience of small communities differs from the resilience of large communities (Jerolleman 2020).

Research has revealed five obstacles that inhibit the resilience of small communities. The first obstacle is geographic isolation, as distance and limitations of transportation infrastructure can inhibit the movement of emergency management personnel during an emergency (Schneider et  al. 2015). The second obstacle reflects deficiencies in human and institutional capital, as small communities often do not have the emergency manage-ment personnel, practices, and policies needed to implement effective disaster manage-ment activities (Kapucu et  al. 2013; Cox and Hamlen 2015; Cutter et  al. 2016). Third, the resilience of small communities can also be impeded by economic homogeneity and resource deficiencies. As noted by Kapucu et al. (2013, p. 215, citing Jansen 2006) small communities “may have a less diversified economic base and fewer financial resources to support disaster mitigation practices or rebuilding efforts.” The lack of economic diversity can be especially problematic for communities dependent upon agriculture or other natural resources (Sabates-Wheeler et al. 2008). Relatedly, communities with lower income levels often have greater hazard exposure (Sabates-Wheeler et al. 2008). Finally, the resilience of small communities can be affected by the lack of media attention, which is associated with the availability and distribution of response and recovery resources (Brennan and Flint 2007).

Despite the growth of research focused on small communities, disaster scholars have yet to consider whether the six capacities of resilience provide an appropriate framework for the analysis of resilience in nonurban contexts. This may be due to the challenges asso-ciated with disentangling the ambiguous meaning of the word “rural.” While the word “conjures widely shared images of farms, ranches, villages, small towns, and open spaces,” researchers and policymakers have yet to reach a consensus on how to distinguish rural communities from urban communities; consequently, they have adopted and employed a “dizzying array of definitions” (Cromartie and Bucholtz 2008, p. 29). Some definitions emphasize population size, although there is no accepted threshold of what makes a com-munity “rural.” The US Census Bureau defines “rural” based on land use and population

1103Natural Hazards (2021) 109:1097–1118

1 3

density parameters. In contrast, the US Department of Agriculture Economic Research Service uses definitions that focus on economic activity, designating communities as rural according to the extent to which they are integrated with primary labor and trade mar-kets (Cromartie and Bucholtz 2008). This article does not attempt to resolve this impor-tant definitional dilemma. Rather, we recognize that “rural” is a multidimensional concept. As such, our study explores the resilience capacities of three small communities that were affected by Hurricane Harvey, each of which have different population characteristics.

3 Hurricane Harvey and case selection

Hurricane Harvey, a Category 4 hurricane approximately 200 miles wide, made landfall on the southern Texas coast at Rockport on August 26, 2017. The storm then stalled over the state for 5 days, dumping an unprecedented amount of rainfall and causing widespread human and property losses (Commission to Rebuild Texas 2018). The federally designated Hurricane Harvey disaster area included 41 counties, shown in Fig. 1, that comprise about 41,500 square miles of land mass—larger than the states of Connecticut, Massachusetts, New Hampshire, Rhode Island, and Vermont combined (South Texas Economic Develop-ment Center 2018). Many of the communities within the Harvey-affected area are small, providing an opportunity to explore the prevalence of resilience capacities in this context.

Upon landfall, Hurricane Harvey’s wind gusts reached 132  mph (115  kt) and storm surge levels exceeded 10 feet in some areas (Blake and Zelinsky 2018). Hurricane Har-vey was also a severe precipitation event. The storm dumped more than 40 inches of rain in 48 h across the Houston metropolitan area (National Weather Service 2020). Rainfall totals peaked at 51.88 inches in parts of the Houston metropolitan area (Morast 2017) and at 60.58 inches in other regions of the state (Blake and Zelinsky 2018). This precipitation generated riverine flooding throughout Texas.

The hazardous conditions created by Hurricane Harvey caused 780,000 Texans to evac-uate their homes (Federal Emergency Management Agency 2018). As Hurricane Harvey began to dissipate, government agencies organized a massive response (Ross 2019). FEMA (2017), for example, noted that dozens of federal agencies, including the Coast Guard, National Guard, National Parks Service, US Fish and Wildlife Service, Customs and Bor-der Patrol, the Department of Defense, as well as dozens of state and local government agencies, participated in response activities. In addition, nonprofit organizations and vol-unteers made substantial contributions, as did the individuals who formed spontaneous and informal disaster response groups to distribute aid and assistance.

The flooding that occurred during the Hurricane Harvey event caused 68 deaths. The storm also caused wide-spread property damage, as flood waters impacted 300,000 struc-tures and 500,000 automobiles (Blake and Zelinsky 2018). Of the households affected by Hurricane Harvey, 80% did not have flood insurance (Federal Emergency Management Agency 2018). Payments made through FEMA’s Individual Assistance Program, Small Business Administration Loans, and claims filed with the National Flood Insurance Pro-gram (NFIP) and Texas Wind Insurance Association reached almost $15 billion dollars (Commission to Rebuild Texas 2018). Altogether, Hurricane Harvey caused an estimated $125 billion in damages, making it the second most costly disaster, behind Hurricane Kat-rina, in American history (FEMA 2018).

Despite the tremendous efforts of relief and response organizations, Hurricane Harvey revealed the vulnerability of Texas communities to disaster events. Almost immediately,

1104 Natural Hazards (2021) 109:1097–1118

1 3

questions were raised about why Texas’ communities were not more resilient (Parvin and Shaw 2011; Russell and Rubin 2017). This study responds to such questions through an investigation of the factors that promoted or inhibited resilience in three small Texas com-munities that were affected by Hurricane Harvey but remained outside the media spotlight.

3.1 Case selection

The three cases selected for this study are: (1) City of Dickinson (Galveston County); (2) City of Port Aransas (Nueces County); and (3) Town of Refugio (Refugio County). These cases were chosen for the variation they provide in population characteristics, as shown in Table  1. First, with populations that range from 2,866 (Refugio) to 19,583 (Dickinson), none of these cases are large urban areas (US Census Bureau 2016); thus, it is reasonable to consider them all “small” communities. Second, the population density of the counties in which the cases are located is varied, ranging from the sparsely populated Refugio County

Fig. 1 Hurricane Harvey storm track across Texas. Hurricane storm fields are illustrated using the National Hurricane Center’s (2017) best track estimate and maximum sustained wind swaths (knots) for Hurricane Harvey. Shaded counties are eligible for FEMA individual assistance. Map created using Esri ArcGIS

1105Natural Hazards (2021) 109:1097–1118

1 3

with 9.6 persons per square mile to the high population density of Galveston County with 769.9 persons per square mile. Third, these communities are varied in their connection to primary economic markets. According to rural–urban commuting codes (Economic Research Service 2020), Dickinson, located only 30 miles from the fourth most populous city in the nation—Houston, is considered a metropolitan area that has high commuting flows (30% or more) to an urban area. Port Aransas, located on Mustang Island along the Middle Texas Coast, is a small town with its own economic core. Refugio, located about 15 miles from the coast in a largely uninhabited area, is a rural community that has flows to an outside urban area.

The cases represent variation in small communities (population size less than 20,000) in terms of population density—low (Refugio), medium (Port Aransas), and high (Dick-inson)—and economic integration—a rural community with economic activities depend-ent on an outside urban center (Refugio), a small town sustaining its own economic core (Port Aransas), and a metropolitan community highly integrated with an urban area (Dick-inson). These two characteristics—population density and economic integration—are critical for assessment of resilience as they are connected to challenges associated with disaster response and recovery, including geographic isolation and the coordination of efforts within broader areas, often defined by economic markets. We recognize that these population characteristics are also likely connected to financial (i.e., household incomes) and social (i.e., civic involvement) aspects of the community; however, for the scope of this analysis, we focused our attention on population characteristics of the community and assessed related community attributes through the lens of resilience capacities.

Hurricane Harvey’s impact on the three communities selected for analysis warrants additional consideration. As Table 1 reveals, the Texas Division of Emergency Manage-ment (2017) estimated that Hurricane Harvey caused a total of 100 million dollars of dam-age upon these three communities. In part, it is the location of these communities that explains why they experienced so much damage. Port Aransas and Refugio are located about 20 miles from where Hurricane Harvey first made landfall, and Dickinson is located where Hurricane Harvey stalled as an unprecedented rainfall event. Port Aransas and Refu-gio were affected by hurricane hazards, namely wind and storm surge, while Dickinson was impacted by the flooding. Although damage estimates for Refugio were not available at the time the aforementioned estimates were published, news reports indicated that “more

Table 1 Population characteristics and Hurricane Harvey damage estimates of study cases

a US Census Bureau (2016)b US Census Bureau (2010)c Texas Division of Emergency Management (2017)

Populationa (Per mile2)b

Number of housing unitsa

Homes impactedc Total damagec

Affected Damaged Destroyed

Galveston County 314,485 (769.9) 140,048 5439 17,367 1865 $111,939,510City of Dickinson 19,583 8849 4124 4413 54 $62,212,500Nueces County 355,667 (405.8) 145,791 340 5280 1036 $40,236,000City of Port Aransas 3861 4327 340 5280 1036 $38,426,000Refugio County 7315 (9.6) 3721 310 1795 442 $25,994,373Town of Refugio 2866 1333 – – – –

1106 Natural Hazards (2021) 109:1097–1118

1 3

than 50% of structures in Refugio County were damaged or destroyed by Hurricane Har-vey” (Faircloth 2018).

4 Data collection and analytical methods

To explore the resilience mechanisms of small communities, interviews were conducted with governmental and nongovernmental officials involved in disaster preparation and response activities in the three communities under analysis. Given that this study sought to understand perceptions of resilience during the response phase, fourteen interviews were conducted with twenty-six subjects during the 3 months that followed Hurricane Harvey. The subjects were involved in activities related to emergency management, agriculture, education, healthcare, regional governance, housing, and charitable donations.

The interview protocol asked subjects to answer semi-structured questions about their professional backgrounds and official responsibilities before, during, and after Hurricane Harvey. The governmental officials were also asked to share their perceptions of how their government and community performed with respect to preparing for, and respond-ing to, catastrophic disaster events. Specific attention was directed toward identifying the strengths and weakness that may have existed in these domains. The nongovernmental sub-jects were asked a similar set of questions, which also focused on their perceptions about how their government, community, and organizations performed before, during, and after Hurricane Harvey. Collectively, these questions enabled the subjects to identify and discuss the capacities they believe impacted their community’s ability to prepare for, and respond to, Hurricane Harvey.

The interviews were audio recorded, transcribed, and analyzed using the qualita-tive coding software MaxQDA 2018 (VERBI Software 2018). This software was used to review the transcripts for themes connected to the resilience of the investigated communi-ties (Saldaña 2016). More specifically, the study employed a thematic coding approach to assess whether the content of the interviews corresponded with the six capacities of resil-ience. Throughout this process, the transcripts were reviewed for references that did not align with these criteria. Three steps were taken to complete data coding. First, we used an open-coding process to identify prominent themes (codes) contained in each of the tran-scripts. The coding process was inclusive, meaning that potentially relevant phrases were interpreted broadly. Second the codes associated with the six capacities of resilience were classified according to whether the theme referred to something that promoted or inhib-ited community resilience. Third, all other relevant codes were classified according to their respective themes and sub-themes.

5 Results

According to Ross (2013, p. 28), resilience is driven by a set of capacities, which “imbue a community with the strength needed to respond, cope, and recover from a disaster event.” Using data from interviews with officials involved in Hurricane Harvey preparedness and response activities in three small Texas communities, we explored whether the six capaci-ties of resilience—human, physical, economic, social, institutional, and environmen-tal—existed in the communities under analysis. Driven by the themes extracted from the

1107Natural Hazards (2021) 109:1097–1118

1 3

interview data, we also identified the adaptive processes that supported the resilience of these three communities.

5.1 Resilience capacities

The findings indicate the presence of all six capacities of resilience in the communities under analysis. More specifically, the findings reveal that the interview subjects most often referenced social, physical, human, institutional, and economic capacities as essential to their community’s ability to respond to Hurricane Harvey. Less predominant were refer-ences to environmental capacities. The prominent sub-themes identified for each of the six resilience capacities are considered in the following sections.

5.1.1 Social capital

Social capital, defined as the social networks that facilitate trust and collaboration, was the most prevalent of the six resilience capacities mentioned by the subjects. Although there were a few instances where the subjects reported that there was a lack of trust among some of the volunteer organizations or questioned the motivations of certain charity organiza-tions, most of the interviewees indicated that connections facilitated response and disaster relief activities.

The references to social capital centered around two factors. First, when it came to dis-aster events, the communities were able to draw upon the strength of the relationships that existed between citizens. The subjects often mentioned the benefits of living in a small town, for example, the fact that many of the citizens are, in one way or another, related to each other, and that such relationships give rise to a close-knit community (Interview 7). These “grassroot” relationships had developed over several decades, meaning that “eve-rybody knew everybody” or “everybody knew somebody who could get the job done” (Interview 2 and Interview 14). These relationships facilitated responsive information and resource exchanges during the response to Hurricane Harvey. Given that “everyone was aware of the needs that need to be addressed” (Interview 7), citizens that encountered prob-lems were able to jump “out in front, [and do] whatever they could do to help” (Inter-view 2). Referencing the strength of the social relationships during the initial days of the response, this same subject revealed that “it felt like Christmas, with everyone giving and helping each other” (Interview 2).

Second, the social capital benefits provided by the relationships among citizens were often leveraged by community organizations. For example, churches and schools often served as hubs for social interaction and response activities after Hurricane Harvey. Emphasizing the responsiveness of the churches, one individual reported that “every time we requested something from the churches, for example we needed an additional shelter. One phone call, one person, boom, shelter” (Interview 2). Such helpfulness and close com-munity ties were also present the schools, which were connected with local resources and officials, included in emergency response plans, and assigned as shelters and relief dis-tribution sites (Interview 3). Several subjects also noted that schools were an important bridge between the citizens and disaster relief actors. For instance, teachers not only knew their community, and they could identify the families that needed assistance though their interactions with students (Interview 7). Additionally, as schools resumed operations, they provided the community with a much-needed sense of routine, while teachers and adminis-trators continued to monitor and respond to emergent needs of their students (Interview 7).

1108 Natural Hazards (2021) 109:1097–1118

1 3

5.1.2 Physical capital

The second most prevalent capital mentioned during the interviews related to physical cap-ital, which refers to the quality of the infrastructure—or the built environment—within a community. Most of the references to physical capital were associated with communica-tion infrastructure systems, namely cellular phones (including internet), social media plat-forms, and lifeline systems. The importance of these systems was stressed by a local offi-cial who referenced their personal experiences during weeks after the storm. The official commented:

…communication was horrible, and we didn’t have cellphone service. We didn’t have landlines. We didn’t have electricity a lot of times. We didn’t have water the first probably about 10 days. And so that just really hampers everything, when you don’t have basic utilities, or in our case we would work here all day and then drive to someone’s house … at night and use the internet and work, and try to return phone calls or answer emails or send emails that we needed to send till all hours of the night. So that really made things tough at the beginning, was just not having access to the basic infrastructure utilities to make things happen (Interview 6).

In areas where electricity had not been disrupted, communication systems and social media platforms enabled information to be exchanged between government officials and the pub-lic. Social media platforms and cell phones also facilitated the exchange of information between citizens. During the Harvey event, these systems and platforms facilitated the communication of information about risk, community needs, and protective actions, for example, evacuations from flooded areas. One of the subjects (Interview 6) provided a con-cise summary of their experience with cellphone and social media during and after Hur-ricane Harvey:

[W]e use Facebook. Our Facebook page for everything. If you’ll see, I posted lots and lots of times in the last few months. And we were using it beforehand. It’s just the easiest way, because you can use your phone and update when you’re … whatever you’re doing around here. And even that’s what I was doing at night, was when I could get cellphone service or Internet somewhere, was trying to post that informa-tion out. Because people just want to know. There’s a comfort, even if it’s not good news. The unknown is worse than the known most of the time.

Additionally, subjects from all three communities also reported several limitations of phys-ical capital, which inhibited their ability to respond to the disaster. The lack of housing was the most often mentioned physical limitation, as both the displaced survivors of Hurricane Harvey and the first responders did not have enough places to stay after the disaster event. There simply were not enough housing units to address community needs. The other two categories were closely related: transportation and the difficulties associated with commu-nity access. The problems mentioned by the subjects ranged from it took a long time for responders to arrive, to being unable to access damaged areas due to flooding or the lack of transportation services—e.g., bus drivers or a ferry system to island areas (Interview 3; Interview 5; and Interview 8). Referencing the experience of external responders, one subject indicated: “[t]hey’re not familiar with the area. Then you’re dealing with flooded roads, washed out roads, power poles and power lines down over across highways. That makes it impossible to drive down” (Interview 8).

1109Natural Hazards (2021) 109:1097–1118

1 3

5.1.3 Human capital

Human capital refers to the knowledge and skills that can be used to manage natural haz-ards. In the context of disaster resilience, this often refers to an individual’s or organiza-tion’s knowledge and capacity to respond or recover from disaster events. The subjects revealed that many of communities relied on human capital to prepare for, and respond to, disaster events. As an illustrative example, officials stressed the readiness of their local Community Emergency Response Team (CERT) programs, which had trained sev-eral hundred people to prepare for disaster events. Additionally, the communities drew upon a wide range of other skills possessed by their citizens, for example boating and kayaking, which were used to help rescue people from their flooded neighborhoods. Other officials stressed their pre-Harvey training, which enabled them to understand the tasks that needed to be undertaken and who would be responsible for those tasks. Sev-eral subjects detailed the training experiences that had occurred in their community. For instance, one of the subjects noted: “[w]e had several meetings right after the begin-ning of hurricane season with the community mayors, our commissioners, all our law enforcement, our firefighters, our fire chiefs, our public works folks to [unintelligible] folks, and we go over evacuation. We [went] over storm planning, contingency plans within contingency plans to deal with issues like a hurricane” (Interview 8).

Additionally, an individual capacity and attitude toward preparing for and responding to disasters constitutes another type of human capital. In the present study, this individ-ual dimension of human capital seemed to have an impact on whether the affected com-munities needed to wait for external assistance. In fact, many of the subjects indicated that a sense of “self-resilience” within their communities helped to facilitate the recov-ery process. Recognizing that government assistance would be slow to arrive, the sub-jects stressed that their citizens acted and began to coordinate their activities prior to the arrival of state and federal officials and resources. Focusing on readiness to take action that exists in smaller communities, as opposed to waiting for assistance, one official noted, “[w]e really are self-sufficient in many ways that people in a bigger city are not. I think they had lots of, as I understand it with the flooding, there was lots of not neces-sarily fighting but arguing and people weren’t responding quick enough. We honestly did not have that here. People pulled together” (Interview 13). Another official reported the self-directed nature of activities that spontaneously unfolded in their community:

Our community didn’t wait. Our folks didn’t wait on any government interven-tion. We personally witnessed, when we actually left this room, we personally wit-nessed people pulling sheet rock out, and carpet, and other belongings, and the piles of debris in front of their houses were bigger than their houses. And our community folks didn’t wait. They didn’t wait for their insurance companies or FEMA or any kind of aid, and for the most part our folks, hundreds and hundreds of people that we encountered just driving the streets to do assessments, wind-shield assessments (Interview 2).

Notwithstanding the importance of these successes, subjects also reported that their communities suffered from human capital deficiencies. For example, subjects noted they suffered from high stress levels and were overwhelmed by the scope of the event (Inter-view 8; Interview 13). Focusing on the exhaustion experienced by response personal after Hurricane Harvey, one subject explained (Interview 8):

1110 Natural Hazards (2021) 109:1097–1118

1 3

From a personal standpoint, I went the first nine days 24 hours a day because dur-ing that period of time the phone calls never stopped. Once the internet service was up, that never stopped, the continual contacts that you got during a 24-hour period was just incredible. I don’t know how or why our bodies do what they do, but when you can maybe pass out for 15 or 20 minutes and then wake up and go for hours, it’s exhausting. Somehow, we did it. Not only me, but everybody…

Other subjects noted that the lack of community knowledge about hurricane dangers was problematic, as memories fade about how bad things can get. Referencing what they called “storm amnesia,” the subjects indicated that the general public is not prepared when they “go 5, 10 years without a disaster” (Interview 2). They also noted that the community did not understand the nature of the risk, as this was a different type of event. For instance, as media referenced the strength of the storm, which weakened as it hovered over the Houston region, the public began to think: “So I don’t have to worry about it because it’s only a tropical storm for us. Well, it’s the tropical storm that put [out] more rain than any other storm, in history” (Interview 4). Specifically, the community was not prepared for how the storm surge would interact with the flooding generated by the record levels of rainfall.

5.1.4 Economic capital

Economic capital, or the financial strength of individuals as well as public and private insti-tutions, also contributed to the resilience of the communities under analysis. The references to economic capital shared by the subjects focused on the community as well as individu-als. At the community level, the subjects indicated that their public agencies and organi-zation often did not have the financial resources they needed to prepare for, and respond to, events such as Hurricane Harvey. This concern extended to the long-term recovery, as smaller communities would not have the same access to recovery funds as larger communi-ties. Central to this point, the subjects noted the disaster destroyed valuable properties and caused local businesses to shutdown, which would undermine their communities’ long-term ability to generate tax revenue. A subject expressed how changes in tax-rolls might impact smaller communities:

So I’m looking at these small towns, and recovery to me is probably the biggest issue on the business and the individual. Because government will or won’t recover based on its occupants. They’re so dependent on the occupants because that’s where they derive their taxes from, so that’s their tax-based liens. It’s like using buyout for recovery, so everyone’s clamped screaming for buyout. People don’t understand, you permanently remove that piece of property off the tax rolls. So if I remove, 80% or 50% or even 20% off my tax roll, that’s a huge, huge hit to the city. And then whether or not the citizens understand that… They don’t care [be]cause they care about them[selves], "It’s about me, cause my house got damaged. And I want my money [be]cause I see that’s going around in Houston." You know they had all these buyouts – we had 600 buyouts over in [Redacted]. And we’ll look at that and, some of it was probably very good and some of probably could’ve been done a little bit differently (Interview 4).

There were also economic challenges at the individual level. While the subjects indicated they were grateful for the donations that arrived during weeks that followed the disaster, they also indicated that the recovery of their communities would be constrained by low incomes levels and the lack of insurance. Referring to the storm’s impact to the economy,

1111Natural Hazards (2021) 109:1097–1118

1 3

one subject indicated that “there’s nowhere for people to work. They have no income; you know the two-week cycle…[a]nd you’re always 2 weeks behind or you’re living 7 days to 7 days. Well now they’re 2 weeks behind their 2 weeks because they have had no paycheck, they can’t catch up. I mean we’re still; everyone is still catching up” (Interview 13). A dif-ferent subject echoed this problem, discussing how the lack of insurance would constrain the community’s long-term recovery. In the subject’s own words: “if you didn’t have insur-ance, and I understand about 80% of people didn’t have insurance, then this is not going to be over this year, or the next year, or the next. This is ongoing, we will be dealing with this for many years, families will be dealing with it, employees will be dealing with it” (Inter-view 3).

5.1.5 Institutional capital

Institutional capital, or the processes and activities associated with disaster management, were also identified by the subjects. Most often mentioned was the collection and dissemi-nation of information, especially from the National Weather Service, which supported situ-ational awareness and decision making. Also mentioned was the institutional capacity to “cut the red tape,” meaning officials figured out how to streamline the approval processes to get businesses opened (Interview 5), or figured how to undertake tasks before other juris-dictions, for example, with mosquito spraying (Interview 8). The design of the emergency management system also seemed to contribute to community resilience. This appeared to be especially important for one community, where government officials reported that the emergency management processes and activities were easy to understand (Interview 2). The subjects were not suggesting their jobs were easy, but rather, that they did not need to reinvent the wheel when they made decisions about how to proceed after Hurricane Har-vey. The steps to be taken were specified in policy documents.

Reflective of this point is that subjects stressed that their communities had spent time on activities such as preparation, training, and communication. With respect to preparation, one community official pointed out the tasks that had been undertaken before Hurricane Harvey, which included the development of preposition contracts, the adoption of strin-gent building codes, the identification of important stakeholders and community partners, and the execution of monthly meetings (Interview 5 and Interview 4). Pre-event training was also deemed for the development of community resilience. As an illustrative exam-ple, a subject with emergency management experience noted: “we’ve practiced transferring resources, we’ve practiced working together and communicating during times like that. So, I think that helps out a lot. We have what … 8 years of [practice since] the hurricane con-ference” (Interview 6). A separate subject indicated that they would work with members of the community to ensure that they were prepared, for example, “people who are day-to-day work human resources or day-to-day work in engineering or finance, they have a role in the management of a large incident, so those are the people that we reach out to, those are the people we try to give them a day or a day and a 1/2 to 2 days of exposure to hurricanes or to large responses and stuff like that” (Interview 1). This same individual also com-mented on the importance of communication, where community members built up their institutional capacity by holding regular “discussions about … mitigation strategies and looking to partner on opportunities that otherwise would not be available to the community because we were impacted by such a severe disaster” (Interview 1).

In contrast, the subjects indicated that the lack of institutional capital was one of the largest obstacles for community resilience. The smaller the community, the more the

1112 Natural Hazards (2021) 109:1097–1118

1 3

subjects would make references to deficiencies in institutional capacity. For instance, most of the subjects expressed concerns about the size of their respective emergency manage-ment budgets, the need for emergency management equipment and knowledgeable per-sonnel, the need for full-time emergency management personnel, and the need for robust recovery plans. Other deficiencies were associated with the management of the response activities. Especially problematic was donations management, which included communi-ties being inundated with donations, having to manage donations—including the sorting of 4,000 pairs of shoes—and matching donors with recipients (Interview 3; Interview 11; Interview 13). The subjects also reported they had difficulties navigating the federal sup-port agencies, labeling them as “bloated,” documentation-driven, and “slow to follow up” on requests (Interview 5). Finally, the subjects reported that people were impatient with the speed of the recovery, especially in the smaller and more remote communities, where dam-aged homes had yet to be assessed 2 months after Hurricane Harvey (Interview 10).

5.1.6 Environmental capital

The least prevalent capital discussed by the subjects related to the environment, which can have an impact on both the nature of hazard events and the extent to which hazard events can disrupt a community. As expressed in the literature review, environmental capital can reflect a range of community characteristics, for example, a community’s geographical or ecological features. In the present study, there were very few instances when the subjects specifically mentioned the environment. In one instance, the subjects mentioned their geo-graphic location, for example, its distance from neighboring urban areas or the fact that it was surrounded by flood water, meant that external relief organizations had a difficult time delivering assistance to affected populations (Interview 3). Additionally, another subject indicated that the natural environment, specifically the humidity and mosquitos that perme-ated the areas affected by Hurricane Harvey, created difficulties for first responders once they arrived (Interview 8). The subjects also indicated that the environment had positive impacts on community resilience. To this end, a subject indicated the environment was an attractor for tourism, suggesting that the environment was a natural resource that could impact economic capital (Interview 5). While the conclusions that can be drawn from these findings are limited, in part because of the interview protocol and the number of subjects, the findings suggest that environmental conditions can have both positive and negative impacts on a community’s resilience.

5.2 Resilience processes

Beyond providing insight into the prevalence of adaptive capacities in the three communi-ties under analysis, the interview data also provided insight into what Ross (2013) refers to as “adaptation processes” or the mechanisms that enable a community to adapt in response to risks and disruptive events. The two processes that were most prevalent were “response flexibility and creativity” and “collaboration and coordination.” One of the most prominent examples of response flexibility and creativity was the ad hoc establishment of a barbeque pit that enabled cooking without electricity in one of the affected communities, which was used to feed the responders and volunteers (Interview 7). Another example related to the previously mentioned mosquito problem, which inhibited power crews from completing their tasks. In this instance, the solution to this problem was identified after a local offi-cial indicated he knew someone who was permitted to use aircraft to spray for mosquitos

1113Natural Hazards (2021) 109:1097–1118

1 3

(Interview 8). These examples suggest that aspects of the response to Hurricane Harvey were not scripted, but rather, involved a series of decisions about how to respond to unex-pected circumstances and problems.

The subjects also stressed the role of collaboration and coordination among the actors involved in the Hurricane Harvey response. For instance, one official stressed that the foun-dation for post-disaster collaboration is planning at various jurisdictional levels. In the words of the interview subject: “[you] plan for this, plan your work, work your plan, every piece of it. Because like I said, if it’s well thought out and it’s discussed and it’s a collabo-rative effort … not just within the community but with your regional partners” (Interview 5). Similarly, one of the subjects referenced the need to engage in a learning process, as they did not have a response system in place when Hurricane Harvey struck (Interview 13). Other than the occasional reference to the processes of resilience, the subjects did not elab-orate on the nature of these processes or how they contributed to community resilience. These findings, however, do indicate that improvisation, creativity, collaboration, and coor-dination warrant further exploration, as they appear to be mechanisms through which a community can unlock the benefits of the six capacities of community resilience (Comfort 2007; Comfort et al. 2010).

6 Discussion

The results lend support to many of the conclusions expressed in the disaster management and resilience literatures, namely that the factors that promote and inhibit community resil-ience are driven by local context (Cutter et  al. 2016). Specifically, the findings confirm that the six capacities of resilience, derived from the community resilience frameworks discussed in the literature, have utility in that they provide a lens through which disas-ter resilience of small communities can be analyzed. The findings also indicate that adap-tive processes, namely flexibility and collaboration, contribute to the resilience of small communities. Finally, although this research did not undertake a comparative analysis, the interview subjects referenced physical, economic, and institutional constraints that inhib-ited their capacity for resilience. These constraints are consistent with the disadvantages of small communities, as compared to larger communities, mentioned in the disaster resil-ience literature (Cross 2001; Cutter et al. 2010; Kapucu et al. 2013, Ross and Clay 2018). Together, these findings provide insight into the capacities and processes that drive resil-ience, underscore the complexity of resilience, and suggest there are limitations to the approaches that are used to measure and assess resilience.

Given the lack of consensus on how to define and conceptualize resilience (Demiroz and Haase 2018; Haase and Demiroz 2019), quantitative measures alone may not capture the nuances of capacities and the dynamics of adaptive processes particular to small com-munities. For example, with regard to human capital, the interview data confirmed that individuals within small communities accumulate the knowledge and skills they need to respond to local disaster events (Colten et al. 2012; Cox and Hamlen 2015). Thus, quanti-tative assessments of human capital, typically measured as aggregate levels of education, may not adequately reflect the extent to which individuals in a small community possess the tacit knowledge and skills needed to prepare for and respond to disaster events. The same can be said of the self-reliant attitude noted by the subjects. Although self-reliance may not be captured by measures of income or education, this trait may be a foundation upon which the human capital needed to promote community resilience is built.

1114 Natural Hazards (2021) 109:1097–1118

1 3

Collectively, these findings suggest that qualitative methods—when used in conjunc-tion with quantitative methods—may provide researchers with a richer understanding of the capacities of resilience and the dynamic intersection between context, processes, and capaci-ties. There are several ways that qualitative method can contribute to understandings of resil-ience. In terms of theory, qualitative studies of how communities—large and small—manage risk can help researchers to consider the adequacy of contemporary resilience frameworks, for example, by revealing whether the quantitative measures of resilience are appropriate or whether new quantitative measures should be developed. Peering beyond the numbers, qualitative research can also reveal how the presence of one resilience capacity might help a community overcome a deficiency in another resilience capacity. In the present study, indi-viduals from small Texas communities suggested that they were able to leverage high levels of social capital to overcome perceived deficiencies in economic capital, at least during the initial response period. Furthermore, quantitative data cannot reveal the competing interests that exist within a community or provide in-depth insights into the reasons why certain capac-ity levels increase or decrease.

To this end, researchers can use qualitative data to reveal the multifaceted—and intercon-nected—nature of the six resilience capacities and where interventions in one capacity area may boost capacities in other areas. As the interviews revealed, schools constitute a physical capacity in that they are buildings, but when it comes to resilience, schools are more than just a physical asset—they can also facilitate social capital and bridge gaps that exists between emergency management officials and the members of the community who have response needs. Finally, as a complement to quantitative data, qualitative data can also help research-ers to unpack the complexity and dynamics of the interactions that occur between resilience capacities and resilience process, which together facilitate a communities’ ability to adapt and adjust in response to disruptive events (Comfort 1999, 2019; Comfort et al. 2019).

The experiences of the small communities that were captured in this analysis revealed the appropriateness and utility of employing the six capacities of resilience to observe and describe the resilience in at-risk communities. The findings also suggest, however, that the way these capacities will be expressed in individual communities, as well as the prevalence of these capacities, will likely differ from community to community. As such, there remain important questions about whether quantitative measures can be used to fully assess the six capacities of resilience, and by extension, whether it is appropriate to use a single index assess resilience within and across communities. Given that resilience is the result of the interaction between a set of capacities and processes derived from, and influenced by, local contexts, our resilience frameworks could benefit from continued refinement. Additionally, policymakers should be cautious in their adoption of resilience as a public policy goal and risk-reduction metric (Chandler 2014). At the very least, there is a need to recognize that there is no such thing as a “one-size-fits-all” resilience intervention. This is because the resilience capacities and processes that undergird a community’s ability to adapt to a disaster event are complex, and the mechanisms that connect resilience capacities and processes are context-specific and rooted in legacies of the community.

7 Conclusion

This article builds upon research that suggests resilience is driven by six community capacities: social, economic, physical, human, institutional, and environmental (Cutter et  al. 2010; Peacock et  al. 2010; Cox and Hamlen 2015; Ross 2013). Although these

1115Natural Hazards (2021) 109:1097–1118

1 3

capacities can be organized as a conceptual framework, questions remain as to whether this framework can be used to investigate the presence of resilience in small commu-nities. Through an analysis of data collected from interviews with governmental and nongovernmental officials in three small Texas communities—Dickinson, Port Aransas, and Refugio—this study used the six capacities of resilience framework to investigate perceptions of how these communities prepared for and responded to Hurricane Har-vey. The results support and extend previous research that found that the resilience of small communities is driven by six interrelated capitals (Ross and Clay 2018), includ-ing human capital, physical capital, social capital, economic capital, institutional capi-tal, and environmental capital. When it comes to resilience capitals, the findings also reveal that communities are not homogenous: communities have different strengths and weaknesses.

These findings suggest that policymakers in smaller communities should consider how to leverage the resilience capacities present in their communities to overcome, or at least manage, the challenges that stem from deficits in other capacities, whether economic, physical, or institutional. Particularly, policymakers should consider how social and human capacities can support response flexibility and creativity and collaboration and coordina-tion, all of which help a community to adapt to disruptive events. Given the contextual nature of community resilience, however, it is unlikely that state and national legislatures will ever be able to advance a singular policy solution that will bring the resilience of all communities to the same level. Rather, with the support of state and national legislatures, local policymakers will need to develop community-based solutions, driven by their com-munities’ strengths and weaknesses, that facilitate the reduction of the risk associated with disaster events (Kapucu et al. 2013). In short, there is no unified recipe for the development and maintenance of community resilience (Manyena 2014).

The findings discussed above are not without their methodological limitations, which restrict the generalizability of this article’s conclusions. The principal limitation is that the interview subjects who participated in this study were limited in number and experience. The inclusion of additional public and nonprofit officials, as well as individual citizens, would have enriched the data collected for analysis. Additionally, the interviews revealed the presence of non-English speaking sub-communities, namely Hispanic communities, whose Hurricane Harvey experiences were not directly captured by this study. Future research should expand the scope of individuals who can provide qualitative data about both the capitals—including environmental capital—and the processes of community resil-ience. Finally, research efforts should expand the number and diversity of the communities analyzed as case studies. In doing so, attention should be directed toward the completion of comparative analyses that explore how the six capacities of resilience are affected by fac-tors such as community size, location, wealth, and demographics. If community resilience efforts are to be localized (Cutter et  al. 2016), then the resilience frameworks that drive these efforts must be derived from data ascertained from a variety of contexts.

Acknowledgements The authors thank Abbey Hotard, PhD student in the Marine and Coastal Manage-ment and Science program at Texas A&M University at Galveston, for creation of the map figure. The research leading to these results received funding from the National Academy of Sciences, Engineering, and Medicine Gulf Research Program under an Early Career Fellowship awarded to Ashley Ross under Grant Agreement Number 2000008414. The National Academy of Sciences, Engineering, and Medicine Gulf Research Program had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. The authors are also grateful for the nonfinancial support provided by Texas A&M University at Galveston, Department of Marine and Coastal Environmental Science and Sam Houston State University, Department of Political Science, for this research.

1116 Natural Hazards (2021) 109:1097–1118

1 3

Funding The research leading to these results received funding from the National Academy of Sciences, Engineering, and Medicine Gulf Research Program under an Early Career Fellowship awarded to Ashley Ross under Grant Agreement Number 2000008414.

Availability of data and materials The datasets generated during and/or analyzed during the current study are not publicly available due to the need to protect the anonymity and confidentially of the interview sub-jects but are available from the corresponding author on reasonable request. The interview protocol for local and regional public officials is available via Harvard Dataverse: https:// doi. org/ 10. 7910/ DVN/ YB0OS9.

Declarations

Conflict of interest The authors have no conflicts of interests to declare that are relevant to the content of this article.

Ethical approval and informed consent The Institutional Review Board (IRB) at Texas A&M University reviewed and approved this project (IRB number: IRB2017-0631D). The researchers obtained informed consent from all participants, who indicated they understood the nature, demands, benefits, and risks of participation.

References

Aldrich DP (2012) Building resilience: social capital in post-disaster recovery. The University of Chicago Press, Chicago

Aldrich DP, Meyer MA (2015) Social capital and community resilience. Am Behav Sci 59(2):254–269. https:// doi. org/ 10. 1177/ 00027 64214 550299

Andrew SA, Ismayilov O (2017) The effects of natural disasters on local government finance. In: LePore A (ed) The future of disaster management in the U.S: rethinking legislation, policy, and finance. Rout-ledge, New York

Blake ES, Zelinsky DA (2018) Tropical cyclone report Hurricane Harvey: 17 August–1 September 2017. National Hurricane Center, Miami

Brennan MA, Flint CG (2007) Uncovering the hidden dimensions of rural disaster mitigation: capacity building through community emergency response teams. South Rural Sociol 22:16

Chandler D (2014) Resilience: the governance of complexity. Routledge, New YorkChang SE, Shinozuka M (2004) Measuring improvements in the disaster resilience of communities. Earthq

Spectra 20:739–755. https:// doi. org/ 10. 1193/1. 17757 96Colten CE, Hay J, Giancarlo A (2012) Community resilience and oil spills in Coastal Louisiana. Ecol Soc

17:5. https:// doi. org/ 10. 5751/ ES- 05047- 170305Comfort LK (1999) Shared risk: complex systems in seismic response. PergamonComfort LK (2007) Crisis management in hindsight: cognition, communication, coordination, and control.

Public Adm Rev 67:189–197Comfort LK (2019) The dynamics of risk: changing technologies and collective action in seismic events.

Princeton University Press, PrincetonComfort LK, Boin CC, Demchak CC (2010) Designing resilience: preparing for extreme events. University

of Pittsburgh Press, PittsburghComfort LK, Haase TW, Ertan G, Scheinert SR (2019) The dynamics of change following extreme events:

transition, scale, and adaptation in systems under stress. Adm Soc. https:// doi. org/ 10. 1177/ 00953 99719 869991

Commission to Rebuild Texas (2018) Eye of the storm: report of the Governor’s Commission to Rebuild Texas. Governor of the State of Texas, Austin

Corbin JM, Strauss AL (2015) Basics of qualitative research: techniques and procedures for developing grounded theory, 4th edn. SAGE, Los Angeles

Cox RS, Hamlen M (2015) Community disaster resilience and the rural resilience index. Am Behav Sci 59:220–237. https:// doi. org/ 10. 1177/ 00027 64214 550297

Cromartie J, Bucholtz S (2008) Defining the “rural” in rural America. Amber Waves, June 1. https:// www. ers. usda. gov/ amber- waves/ 2008/ june/ defin ing- the- rural- in- rural- ameri ca/

Cross JA (2001) Megacities and small towns: different perspectives on hazard vulnerability. Glob Environ Change Part B Environ Hazards 3(2):63–80. https:// doi. org/ 10. 1016/ S1464- 2867(01) 00020-1

1117Natural Hazards (2021) 109:1097–1118

1 3

Cutter SL, Burton CG, Emrich CT (2010) Disaster resilience indicators for benchmarking baseline condi-tions. J Homel Secur Emerg Manag. https:// doi. org/ 10. 2202/ 1547- 7355. 1732

Cutter SL, Ash KD, Emrich CT (2016) Urban–rural differences in disaster resilience. Ann Am As Geogr 106:1236–1252. https:// doi. org/ 10. 1080/ 24694 452. 2016. 11947 40

Demiroz F, Haase TW (2018) The concept of resilience: a bibliometric analysis of the emergency and disas-ter management literature. Local Gov Stud. https:// doi. org/ 10. 1080/ 03003 930. 2018. 15417 96

Economic Research Service (2020) 2010 Rural–urban commuting area codes. United States Department of Agriculture, Hermon

Elliott JR, Haney TJ, Sams-Abiodun P (2010) Limits to social capital: comparing network assistance in two New Orleans neighborhoods devastated by Hurricane Katrina. Sociol Q 51:624–648. https:// doi. org/ 10. 1111/j. 1533- 8525. 2010. 01186.x

Faircloth R (2018) Refugio will never be the same: county slow to rebuild from Harvey. StatesmanFannin JM, Barreca JD, Detre JD (2012) The role of public wealth in recovery and resiliency to natural dis-

asters in rural communities. Am J Agric Econ 94:549–555. https:// doi. org/ 10. 1093/ ajae/ aar068Federal Emergency Management Agency (2017) webpage resource: https:// www. fema. gov/ press- relea se/

20210 318/ histo ric- disas ter- respo nse- hurri cane- harvey- texas. Accessed 15 Sep 2019Federal Emergency Management Agency (2018) 2017 Hurricane season FEMA after-action report. Federal

Emergency Management Agency, WashingtonGerasimidis S, Khorasani NE, Garlock M et al (2017) Resilience of tall steel moment resisting frame build-

ings with multi-hazard post-event fire. J Constr Steel Res 139:202–219. https:// doi. org/ 10. 1016/j. jcsr. 2017. 09. 026

Glaser BG, Strauss AL (1967) The discovery of grounded theory: strategies for qualitative research. Aldine de Gruyter, New York

Haase TW, Demiroz F (2019) Considerations of resilience in the homeland security literature: towards con-ceptual convergence? J Homel Secur Emerg Manag. https:// doi. org/ 10. 1515/ jhsem- 2018- 0078

Haase TW, Ertan G, Comfort LK (2017) The roots of community resilience: a comparative analysis of structural change in four Gulf Coast Hurricane Response Networks. Homel Secur Aff 13:26

Holling CS (1973) Resilience and stability of ecological systems. Annu Rev Ecol Syst 4:1–23Jerolleman A (2020) Challenges of post-disaster recovery in rural areas. In: Laska S (ed) Louisiana’s

response to extreme weather. Springer International Publishing, Cham, pp 285–310Kapucu N (2016) Community-based research in generating usable knowledge for public policy and adminis-

tration. Adm Soc 48:683–710. https:// doi. org/ 10. 1177/ 00953 99713 519095Kapucu N, Hawkins CV, Rivera FI (2013) Disaster preparedness and resilience for rural communities. Risk

Hazards Crisis Public Policy 4:215–233. https:// doi. org/ 10. 1002/ rhc3. 12043Kendra JM, Wachtendorf T (2003) Elements of resilience after the world trade center disaster: reconstituting

New York city’s emergency operations centre. Disasters 27:37–53. https:// doi. org/ 10. 1111/ 1467- 7717. 00218

Manyena SB (2006) The concept of resilience revisited. Disasters 30:434–450. https:// doi. org/ 10. 1111/j. 0361- 3666. 2006. 00331.x

Manyena SB (2014) Disaster resilience: a question of ‘multiple faces’ and ‘multiple spaces’? Int J Disaster Risk Reduct 8:1–9. https:// doi. org/ 10. 1016/j. ijdrr. 2013. 12. 010

Manyena SB, O’Brien G, O’Keefe P, Rose J (2011) Disaster resilience: a bounce back or bounce forward ability? Local Environ 16:417–424. https:// doi. org/ 10. 1080/ 13549 839. 2011. 583049

Miles MB, Huberman AM, Saldaña J (2018) Qualitative data analysis: a methods sourcebook. SAGE, Los Angeles

Morast R (2017) Hurricane Harvey by the numbers. Houston Chronicle. https:// www. houst onchr onicle. com/ life/ artic le/ Hurri cane- Harvey- by- the- numbe rs- 12172 287. php. Accessed 19 Sep 2019

National Hurricane Center (2017) webpage resource: https:// www. nhc. noaa. gov/ data/# hurdat. Accessed 15 Sep 2019

National Research Council (2012) Disaster resilience: a national imperative. National Academies Press, Washington

National Weather Service (2020) Major Hurricane Harvey—August 25–29, 2017. Weather Forecast Office, Corpus Christi

Office of the President of the United States (2017) National security strategy 2017. In: White House. https:// www. white house. gov/ wp- conte nt/ uploa ds/ 2017/ 12/ NSS- Final- 12- 18- 2017- 0905. pdf. Accessed 22 Sep 2017

Parvin GA, Shaw R (2011) Climate disaster resilience of Dhaka City corporation: an empirical assessment at zone level. Risk Hazards Crisis Public Policy 2:1–30. https:// doi. org/ 10. 2202/ 1944- 4079. 1069

1118 Natural Hazards (2021) 109:1097–1118

1 3

Peacock WG, Brody SD, Seitz WA et  al (2010) Advancing resilience of coastal localities: developing, implementing, and sustaining the use of coastal resilience indicators: a final report. Hazard Reduction and Recovery Center, College Station

Ride A, Bretherton D (2011) Community resilience in natural disasters. Palgrave Macmillan, New YorkRoss AD (2013) Local disaster resilience: administrative and political perspectives. Routledge, New YorkRoss AD (2019) Public sector agencies and their formal legal and administrative responsibilities. Oxford

University Press, OxfordRoss AD, Clay LA (2018) Capital Assets And Rural Resilience: An Analysis of Texas Communities

impacted by Hurricane Harvey. J Nat Resour Policy Res 8:154–186Russell P, Rubin D (2017) Hurricanes propel forward thinking on risk, resilience. https:// www. enr. com/ artic

les/ 42831- hurri canes- propel- forwa rd- think ing- on- risk- resil ience. Accessed 26 Feb 2019Sabates-Wheeler R, Devereux S, Mitchell T et al (2008) Rural disaster risk—poverty interface, Brighton:

IDS. https:// opend ocs. ids. ac. uk/ opend ocs/ handle/ 20. 500. 12413/ 2550. Accessed 26 Feb 2019Saldaña J (2016) The coding manual for qualitative researchers, 3rd edn. SAGE, Los AngelesSchneider AH, Mort A, Kindness P et al (2015) Using technology to enhance rural resilience in pre-hospital

emergencies. Scott Geogr J 131:194–200. https:// doi. org/ 10. 1080/ 14702 541. 2014. 978810Simms JRZ (2017) “Why would i live anyplace else?”: Resilience, sense of place, and possibili-

ties of migration in Coastal Louisiana. J Coast Res 332:408–420. https:// doi. org/ 10. 2112/ JCOAS TRES-D- 15- 00193.1

South Texas Economic Development Center (2018) The economic aftermath of Harvey. Texas A&M Uni-versit – Corpus Christi, Corpus Christi

Texas Division of Emergency Management (2017) No change: DSO spreadsheet 17-0021 Harvey 2017 112917. Texas Department of Public Safety, Austin

Tierney K, Bruneau M (2007) Conceptualizing and measuring resilience. TR News Transp Res Board Natl Acad 250:14–18. http:// www. trb. org/ Publi catio ns/ Blurbs/ 158982. aspx

United Nations (2015) Sendai framework for disaster risk reduction 2015–2030. United Nations Office for Disaster Risk Reduction, Sendai

US Census Bureau (2010) Census of Population And Housing: Population Per Square Mile. Census of Pop-ulation and Housing, Washington

United States Census Bureau (2016) webpage resource: https:// www. census. gov/ acs/ www/ data/ data- tables- and- tools/ data- profi les/ 2016/. Accessed 15 Sep 2019

VERBI Software (2018) MAXQDA Analytics Pro [Computer Program]. VERBI, BerlinWaugh WL (2010) Looking for the FEMA guy, mayor, local planning guy, governor, and others. Adm Soc

41:1004–1008. https:// doi. org/ 10. 1177/ 00953 99709 348781Wildavsky A (1988) Searching for safety. Transaction Press, New Brunswick

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.