Researchers Meeting Abstracts

Every disaster reinforces the lessons of those that preceded it: that individuals, households, and communities with the greatest social and economic resources are able to resist the worst effects while those with the least are left to struggle. This talk focuses on the longitudinal Gulf Coast Child and Family Health Study (GCAFH) to consider why. In public health, social determinants of health are often associated with access to knowledge, prestige, power, and capital, all powerful levers for assuring health or predicting disparities. The same principle applies to disaster recovery. Those with limited access to these assets take longer to be re-housed, longer to regain their economic footing, are more likely to suffer mental health effects, are confronted with “trade-offs” of equally poor options, and have protracted periods of recovery. This presentation will explore the empirical evidence for these social determinants, and consider implications for both policy and future scholarship.

This study focused on the socio-economic activities driving changes in land use land cover (LULC) and its relationship with the flood disaster in the urban watershed using geospatial techniques, questionnaires, and field observation. Information on property ownership, flood occurrences, drivers of land-use change, and other relevant information in each of the catchments were gathered.

The study revealed that 28.60% started living in the flood-prone area of the watershed before 1990, 18.9% between 1995-2000, and 9.4% after 2005. However, 49.2% stayed in the area because they owned the properties, and 40.0% due to ancestral property inherited. The difference observed was significant (0.05). Significantly high number of respondents claimed that the highest flood water level occurred in 2011 (70.5%), followed by 2013 as reported by 19.2% of the respondents. Also, the result showed that heavy floods occurred in all the catchment areas in 2011 except in catchments 18, 20, and 24. In all the catchment areas, the respondent reported that both home and community were flooded during the flood. All the differences observed in the proportion of respondents was significant (0.05). It was also revealed that 82.2% have no consideration for relocation while 17.2% have plans to leave the area in the nearest future because of continuous flood occurrence.

The study established that the continuous increase in the utilization of urban resources both natural and artificial in an unsustainable way exposes the people and their properties to flood risk and reduces the land resources available.

The emergence and spread of a novel coronavirus (COVID-19) happened rapidly from March 2020, resulting in a strenuous global pandemic. The intensity of social media interactions increased as people started maintaining physical distancing and staying at home. But the spread of misinformation through social media has created another chaotic situation, which is often called an "Infodemic." The goal of this study is to understand how misinformation spread through social media and influenced the recovery process from the COVID-19 pandemic, where defining misinformation is crucial. The notion of influence was captured by quantifying the network effects on such communication behavior and characterizing how information is exchanged among people who are socially connected online. Twitter data was used to identify the spread of misinformation and was processed by applying several machine learning and natural language processing techniques. The spread of misinformation also depends on the social network characteristics, which were considered to track misinformation propagation. Additionally, identification of the good and bad actors in social media also contributed. These findings can be used by public health and emergency management agencies to tailor effective information dissemination policies for users based on their social network characteristics, activities, and interactions to recover from similar public health hazards.    

Despite a growing literature on crisis communication, there is a limited comparative understanding of public health organizations' online communication during an ongoing health crisis. To address this gap, the study's authors investigated how national public health agencies' online communication objectives and styles vary in times of COVID-19 pandemic? They first introduced a novel theoretical framework for public sector organizations' crisis communication that combined: (1) communication objectives, which included instructing the public on protective measures, helping the public cope with the uncertainties, and strengthening organizational reputation; and (2) communication styles, which consisted of agentic (assertive and task-oriented) and communal (supportive and promoting a sense of collectivity) communication. They then applied this newly developed framework to examine COVID-19 online communication of national-level public health agencies in three countries: the National Institute of Health in Italy, the Public Health Agency in Sweden, and the Centers for Disease Control and Prevention (CDC) in the United States. The study was based on content analysis of Twitter data (n= 856) collected during the pandemic's first wave and analyzed using the NVivo qualitative data software. The initial findings indicate that while there were similarities in these agencies' communication objectives, public health agencies' communication styles varied substantially. While the Italian agency adopted more of an agentic communication style, the Swedish agency mainly followed the communal communication style. CDC's communication was in between the two styles. The paper provides insights on how public organizations can rely on different communication styles and serve different communication purposes during public health crises.

Annually, disasters displace over 20 million people globally and they are often relocated to reduce hazard risk. However, in communities, lived experience, natural hazards represent just one long-term risk, while they address daily risks to health, well-being, livelihood, social capital and identity. Planners often propose relocation without addressing this risk trade-off, resulting in unsuccessful and disruptive relocation projects.  

Many planning studies undertake technical and spatial analyses of hazard risk to assess the need for relocation. Other studies critically examine social dimensions of disaster-induced relocation. However, little exists in the literature to inform planning practice on the trade-off between hazard risks and other risks. This research examines risk trade-off and proposes an analytical framework to inform planning practice.

This research carried out qualitative research in coastal Louisiana, where hurricanes, flooding, land subsidence and ecological destruction have forced thousands of households to relocate. To explore the lived experience of risk, displacement and relocation and to conceptualize risk trade-off, the author interviewed members of vulnerable communities including a Native American tribe. To explore the influence of planning and policy, they observed a planning process (LA SAFE) and interviewed planners, researchers and experts.  

This research reveals trajectories of social, economic and physical displacement, identifying a spectrum of risks. Using Risk Trade-off Analysis, a method developed by economists, the researcher sort these risks into coincident risks (mitigated simultaneously) and countervailing risks (undermines risk-reduction). In conclusion, this paper presents a framework that practitioners can use to frame risk-superior policies that minimize disruption.

The mission of Recover Hampton Roads is to prepare ahead of time, under blue skies, for the repair of damaged homes and hasten the transition of storm-displaced vulnerable households back into their primary residence. Low-to-moderate income (LMI) and medically fragile populations suffer enormously in the wake of catastrophic severe weather events, often experiencing lengthy displacements due to home damage and the inability to make necessary repairs. Lengthy displacements have a direct, deleterious impact upon the household, health and well-being, exacerbating existing disparities and evoking equity and justice issues. Recover Hampton Roads is focused on the repair of storm damage to the primary residence of vulnerable households that are anticipated to have lengthy displacements. The approach of Recover Hampton Roads is to assess and align repair needs of vulnerable households with converging donated material and volunteer labor flowing into the region. The Convergence, Inventory, Matching, and Assignment (CIMA) management platform, central to the Recover Hampton Roads approach, inventories and assigns resources relative to work packages generated in the field, and schedules repairs in a fashion that will maximize a global reduction in displacement times.

Big Data is revolutionizing disaster science, but these newly available or analyzable data sources also create big problems for researchers in the areas of privacy, replicability, accuracy, and accessibility.

There are obvious advantages to having detailed data at countrywide levels, but these datasets still benefit from validation from local experts, and merging multiple validated data sets is an even more difficult data management project. Analyses using massive data sets without local experts can lead to issues with replicability/reproducibility and accuracy. Additionally, data merging capabilities, machine learning, and artificial intelligence create an ethical minefield, making it likely that data use agreements and individual privacy standards will be violated, despite good intentions. 

This paper discusses both the tremendous new capabilities and the limitations of nationwide datasets such as the COVID-19 infection, hospitalization, and death data, compared with validated local data. The principle of mediocrity suggests that issues found compared to one local data source are probably similar across the area of measurement and should be considered before using high-level national data sources to make local-level policy recommendations. This paper gives examples of similar issues with property hazard data and differential privacy techniques proposed in the latest Decennial Census, which can further complicate otherwise promising new analytic techniques.

These large and consistently formatted data sets lead to significant new contributions to research and real-time response and allow us to tackle previously unanswerable problems; however, we must acknowledge these as additional tools and not replacements for traditional studies and data.

Preparedness in times of public health crises such as the COVID-19 pandemic depends on public leaders' effective communication with public members. Such communication is critical for people to understand the levels of risk and take the recommended protective actions to mitigate the negative effects of the crisis, including the number of lives lost, the number of confirmed cases, and the toll on businesses and the economy. Despite a growing literature on effective disaster communication, we have a limited understanding of how leader gender shapes the public's risk perception and compliance with protective measures. 

To study this overlooked question, the authors of this study conducted a COVID-19 focused natural experiment in eight states led by male and female governors from both parties in the United States in Alabama, Connecticut, Iowa, Nebraska, Rhode Island, South Carolina, Oregon, and Washington. The natural experiment involved implementing an online survey via Prime Turk (n = 1,263) during the height of the pandemic (from June to August 2020). Preliminary findings suggest that compared to female governors' COVID-19 policies, male governors' policies are perceived as more accepted by state residents and receive greater compliance in protective measures (e.g., wearing masks, social distancing) from the public in terms of both past practices and future intent. The study offers insights into how public authorities' leaders at the federal, state, and local levels can recognize the effects of gender bias in communication involving public health emergencies like the COVID-19 outbreak. 

The practice of physical distancing during the coronavirus (COVD-19) pandemic has impacted our ability to engage in face-to-face interactions, further challenging our physical, psychological, and economic well-being. As is common following crisis events, recovery from the pandemic will likely rely in part on social resources. Therefore, individuals who have maintained or expanded their social support networks during the pandemic may experience faster recovery. This research examined the resilience of social support networks of 18- to 34-year-olds in California, Texas, and Illinois using a longitudinal survey. The first wave was deployed in February 2021 to 313 individuals. Regression analysis is used to examine correlations between support network changes and individual sociodemographics. Initial findings indicate that women are significantly more likely than men to experience decreases in the size of their social support networks during the pandemic, which may imply slower long-term pandemic recovery for women. Initial findings also suggest that on average, during the pandemic, respondents have relied increasingly on social support from family members and romantic partners, those who have been known for more than six years or less than one year, those living within the same household or neighborhood, and those who are racially and politically similar to the respondent. Regression analysis is currently underway to further examine the resilience of social support networks to physical distancing of young people along intersections of race, class, gender, and sexual orientation.  

Promoting the adoption of household preparedness to natural hazards represents a critical step toward building resilient communities. However, the level of disaster preparedness across the nation remains low. The authors of this study hypothesized that education based in active learning was effective at motivating preparedness and mitigation actions. This study used tenets of the Protective Action Decision Model (PADM) to examine how knowledge, perceptions, and attitudes toward preparedness actions influenced behavior of Boise, Idaho's Wildland Urban Interface (WUI) residents. They implemented a PADM-based questionnaire before and after a 60-minute education workshop designed to help participants better understand WUI hazards, personalize their household risk, and develop positive attitudes toward taking mitigation and preparedness actions. The workshop, developed in collaboration with the Boise Fire Department and Idaho Firewise, used active-learning and goal setting strategies to help participants engage with the material and set achievable goals. Analyses show a positive shift in knowledge, perceptions, attitudes, and preparedness intentions after the workshop. The attendees reported feeling more able to protect their family and property from the threat of wildfire after our workshop. They also reported an increased intention to take action to reduce household risk. This research demonstrates the efficacy of active-learning and goal-setting strategies to engage WUI homeowners in a way that helps them personalize wildfire risk and develop positive attitudes toward preparing. This work also demonstrated how giving the audience a voice through active-learning allows stakeholders to both recognize and resolve inaccurate risk perceptions and negative attitudes toward preparedness actions.

Since the 2016 Louisiana Floods and Hurricane Harvey, civilian volunteer rescue organizations--often referred to as the "Cajun Navy"--have emerged and played a pivotal role in assisting with response and relief efforts. While the narrative in the media and on social media often focuses on boaters making rescues, there are roles and labor that operate unseen, "behind the timeline." Using longitudinal, ethnographic research on civilian volunteer rescue organizations this presentation details the division of labor and role experiences in these organizations through a gendered lens. Prior research has examined gender and gender roles in disasters. However, there is a gap with regard to how gender operates in civilian volunteer disaster organizations, both in terms of division of labor and how members understand the role(s) they occupy. Nearly three years of participant observation and over fifty interviews were conducted with eight organizations and volunteer members. Preliminary results indicate that the labor and roles occupied by men are often visible to the public--positioned at the forefront of traditional and social media outlets. Meanwhile, women take on much of the labor that is critical to organizational operation but not publicized--labor and roles that occur "behind the timeline." These findings provide a further understanding of gendered division of labor and gender roles in once-emergent groups, disaster volunteerism, and disaster nonprofits.

The emergency management community relies on both public and private sector supply chains in disaster response and recovery efforts. This study used a single universal resilience function model for predicting resilience in complex networks to anticipate the resilience and continuity of public and private sector supply chains in the United States in response to disaster events. The identification of supply chain attributes, based upon the EPIC framework that combines economic, political, infrastructure, and technical supply chain workforce competencies to evaluate regional supply chain network capabilities, was applied to major metropolitan statistical areas in the United States that are known to be vulnerable to natural disasters. Publicly available U.S. Bureau of Labor Statistics location quotient and employment data were used in the assessment of attributes, as well as location specific indexed political attributes and infrastructure assessments. Levers to enhance resilience were exposed and yield points were identified using the universal resilience function. Practical guidance is then proposed for state and local emergency management organizations, as well as economic and workforce development agencies to enhance both supply chain and community resilience to disaster events based on the manipulation of levers and yield points specific to each metropolitan statistical area.

The containment measures to reduce the spread of the COVID-19 virus, such as limitations to mass gatherings and social distancing requirements pose considerable challenges to the safe operation of family farms. Family farms are central to sourcing local food systems, supporting local economies, and food security, but they may be especially vulnerable to the impacts of the COVID-19 crisis. This is because family farms depend on direct-to-consumers retail outlets such as farmers, markets that may be closed as a result of the COVID-19 crisis. Yet, there is limited knowledge on COVID-19 pandemic’s impact on family farms. This study assessed the coping strategies of family farms in North Carolina. The authors recruited six family farms with agritourism offerings. Through a series of three interviews each, they asked farmers about the practices they implemented to cope with the COVID-19 crisis. From 18 interviews, they identified emerging themes that illustrated family farms' resilience during the COVID-19 crisis. Study results indicated that family farms' resilience was built upon diversifying their operation, reorganizing their businesses, peer-to-peer learning, and leveraging a support network. This enabled family farms to continue to be sources of employment and expand access to food in their communities. Family farms also offered safe spaces for outdoor recreation, a crucial service when many suffered isolation as a result of stay-at-home orders and travel bans. Understanding family farms’ resilience in response to the COVID-19 crisis reveals promising practices for building resilience when facing a future major health crisis.

Pre-disaster plans aim to increase the efficiency of post-event responses. Such plans typically consider multiple hazards at once to create common response structures to serve any disasters. But are such plans effective when two very different hazards occur simultaneously? To wit, post-earthquake sheltering typically involves placing all affected persons in centralized locations, but pandemic response requires maintaining social distance. To add to this complexity, disasters affect the elderly, low-income, disabled, and ethnic/racial/gender minority groups more severely than others. The post-disaster response is a high-stakes game for these vulnerable populations: poor response means poor recovery, which eventually means poorer resilience against future disasters. 

Using the case of the COVID-19 outbreak and the M 5.7 March 2020 earthquake in Magna, Utah, this paper examined how effective the pre-event multi-hazards response plans are when two very different hazards occur simultaneously, and especially in terms of meeting the needs of low-income communities. The researchers employed qualitative inquiry techniques using content analysis, semi-structured open-ended interviews with local disaster response administrators, and focus group interviews with residents. They applied thematic analyses to examine administrative and sociocultural challenges, opportunities of providing effective disaster response during concurrent disaster events, and actions to meet the needs of low-income communities. Preliminary findings show that disaster response to concurrent disasters is complicated by organizational capacity, political will, and socioeconomic status, all of which raise questions for long-term disaster recovery. This study calls for pre-event disaster response planning to carefully consider the possibility of concurrent events and to centralize equity within this framework. 

Death Attitudes are under-examined in disaster-related research. This study examined the association between several dimensions of death attitudes (namely, fear of death and death acceptance) and experiences during a tornado and after COVID-19. The sample was from the first wave of a planned longitudinal study of 792 residents in Dallas, who were exposed to the 2019 EF-3 Dallas tornado. A subset of questions was assigned to individuals using the planned missing method; thus the working sample includes 497 respondents who were selected to answer questions on death attitudes. Multiple regressions were conducted to examine the association between different dimensions of death attitudes and exposure to tornadoes and COVID-19. Results showed that subjective stressfulness of the tornado experiences was significantly associated with higher levels of fear of death, but not with death acceptance. Similarly, difficulties associated with COVID-19 such as financial difficulties and feeling isolated were significantly associated with higher levels of fear of death but not death acceptance. Future studies will examine the cascading impact of disasters on individuals’ death attitudes.

Managed retreat and relocation initiatives are frequently presented as potential adaptation strategies to use in mitigating the risks associated with living in hazardous areas. Despite the positive benefits of using these strategies to mitigate environmental threats, the relocation of communities can present additional issues. Studies note that risk exposure is not necessarily reduced by relocation and has at times been found to result in potential increases in community vulnerabilities. These dynamics are well highlighted by exploring the relocation and resettlement of informal communities in Puerto Rico. These activities have been motivated by a variety of interests, including urban renewal, hazard exposure, public health concerns, and new development investments. Furthermore, devastation brought about by Hurricane Maria and more recent hurricanes have continued to highlight how climate and hazard issues impact informal areas, leading to renewed interests in advancing mitigation efforts. This study assessed these dynamics through a qualitative assessment of the relocation and resettlement of informal communities in Puerto Rico. The data collection efforts centered on (1) in-depth semi-structured interviews with local public officials, planners, and other contextual experts; (2) focus groups with community leaders; and (3) hazard risk assessments. The findings highlight the critical role risk and safety framings play in the decision-making surrounding relocation and retreat activities across the island. The results of this research aim to contribute to discussions surrounding the equitable implementation of hazard mitigation and planning efforts in informal and vulnerable communities.

Post-disaster residential mobility has adverse consequences for long-term housing recovery. The more local homeowners decide to sell or abandon their homes, the longer it takes for some sense of normalcy to be reestablished. Nevertheless, existing housing recovery simulation frameworks often assume that all homeowners are willing to stay and eventually will repair their homes. Models that account for the willingness to repair often treat it as an exclusive function of access to financing. However, the literature on residential mobility indicates that pre-disaster housing and neighborhood satisfaction are important factors. Underestimating post-disaster residential mobility hides a significant problem with severe consequences for the recovery of communities. This study presents a conceptual model for voluntary and forced residential mobility to improve our ability to simulate residential mobility and its impact on long-term housing recovery. Logistic regression models built using data from the American Housing Survey are used to assess pre-disaster susceptibility to housing mobility across San Francisco neighborhoods. The conceptual and logistic models are integrated into a post-earthquake housing recovery simulation model to investigate residential mobility in San Francisco. The prevalence of voluntary and forced mobility is estimated for all neighborhoods in the city. The socio-economic groups most prone to residential mobility are identified, providing insights for disaster recovery planning. While the case study focuses on post-earthquake mobility in San Francisco, the conceptual and logistic models developed in this study are versatile and can be tailored to any community and adapted to other hazards.

This presentation is on the longitudinal study of Lumberton, North Carolina. The study is conducted by a multidisciplinary research team from the National Institute of Standards and Technology (NIST)-funded Center of Excellence for Risk-Based Community Resilience Planning and NIST Engineering Laboratory. The study began shortly after Hurricane Matthew caused catastrophic flooding in Lumberton in October 2016. It has continued, longitudinally, for five years, with six data collection efforts to date. 

This presentation will cover challenges encountered in observation and measurement of impacts to the community and recovery from Hurricane Matthew in 2016, Hurricane Florence in 2018, and the COVID-19 pandemic. The presentation will touch on epistemic uncertainties encountered in resilience modeling based on data collection in the longitudinal study of Lumberton. Uncertainty is grouped into temporal, modal, and human factors. Temporal factors include the timing of data collection with respect to a given disaster event, and emphasis is placed on the effects of repeat, compound, and cascading disaster events. Modal factors are grouped by data collection methodology, including remote and field-based observation and measurement of damages to physical infrastructure, remote and field-based surveys of residents and business owners, and remote and field-based interviews with community leaders and decision-makers. Human factors include surveyor bias, training, and familiarity with the affected area. This presentation will present the need for quantification of these uncertainties in field data collection to better inform resilience modeling.

While disasters are often understood as acute events, there is a growing recognition that they are slow-onset phenomena, resulting from long-term, socially constructed processes. In the North American Arctic, environmental degradation and change intersect with other processes, including colonization, marginalization, forced relocations from traditional homelands, residential schooling, and cultural assimilation. For Indigenous Peoples in the Arctic, climatic change is characterized by temperature increases, reduction in sea ice extent and thickness, permafrost thawing, and reduced snow cover. These experiences are constructed by, and inseparable from, these other long-term, socially constructed processes. This PhD research used a root cause analysis approach to investigate the longer sequences of events that give rise to shifting safeties and the construction of risk for people traveling on land, sea ice, and water trails in the Arctic North. While flexibly adapting to the pandemic and limits to fieldwork possibilities, the author focused on diverse sources of secondary literature to untangle both the immediate drivers of risk and wider root causes. Following a review of peer-reviewed and grey literature, the author analyzed media reports of travel events on the land and sea ice for discourses of safety and reference to drivers of risk. They also drew on analyses of the colonization of time to understand how these disasters unfold temporally in people's daily lives. This work adds to the conversation on longitudinal approaches to disaster studies and root cause analysis and the discussion around the value of using the qualitative sources already available to us.

Measurement and analysis of long-term housing recovery informs disaster research, and stakeholders and policymakers concerned with housing recovery outcomes. While the housing recovery literature has grown, few studies have modeled housing recovery longitudinally observing the same units with consistent steps through time. Drawing on the literature, this study proposed a standardized approach for assessing a critical dimension of housing recovery, termed restoration, that facilitates comparison of outcomes across and within disaster events. This measure utilizes tax assessor data that features beneficial attributes, including granular resolution at the parcel-level and consistent temporal coverage throughout much of the United States, with assessments typically conducted annually during the first half of the calendar year. Utilizing these data, restoration was conceptualized as a residential housing structure meeting or exceeding 95% of its pre-storm improvement value. The authors demonstrated this approach by comparing two case studies: Hurricanes Andrew and Ike . Using a panel logistic regression model, they regressed restoration on a suite of variables common to disaster vulnerability and recovery studies (i.e., time, damage, tenure, income, and minority race status). The analysis showed that damage, tenure and time were common and significant predictors of restoration between the two events. However, differences arose between the two cases with respect to speed of restoration, and the effects of income and minority status demonstrating differences in community and hazard contexts. The authors discuss plans for future model iterations, application to other disaster events, and integration into a community resilience modeling computational environment.

Tropical cyclones (TCs) can cause immense human and material damage. The long-term distribution of these impacts can be anticipated by systematic risk estimates. However, recurring climate patterns like the El Nino Southern Oscillation (ENSO) and Atlantic Multidecadal Oscillation (AMO) complicate long-term prediction. These patterns influence TC frequency, spatial distribution, and intensity, and thus landfall hazards. Because each combination of climate patterns comprises relatively few years of a reliable historical record, existing observations may misrepresent true risk profiles, leaving emergency preparedness "blind spots." This study aimed to mitigate this by generating many synthetic North Atlantic TC tracks, modeled from resampled historical data following varying ENSO-AMO interactions. They further explored public health applications of this approach. 

The authors developed an approach for TC health impact assessment that accounts for these challenges. They drew on a climatology- and statistics-based synthetic resampling algorithm to generate large datasets for analysis. They modeled specific climate pattern scenarios using combinations of ENSO (positive, neutral, or negative) and AMO (positive or negative) historical data. The authors used projected maximum county-level experienced wind speeds in a Bayesian exposure-response model for all-cause mortality among U.S. Gulf and Atlantic coast Medicare recipients.

The researchers demonstrated the applicability of synthetic TC tracks to exploring TC activity and public health risk under alternative climate pattern scenarios. They tested up to 250 years of synthetic tracks for multiple climate scenarios without obstacles to further scaling. This approach could, with minor modifications, be extended to other climate and public health outcomes.

As of May 2021, the public rollout of ShakeAlert, the West Coast Earthquake Early Warning system, will be completed in three states (Washington, Oregon, and California). Yet critical questions remain about what people understand about the system and if they know what to do when they receive an alert. To evaluate whether the alerting system has been successful in answering these key research questions, the U.S. Geological Survey, in collaboration with its university and state emergency management partners, developed a social science initiative focusing on three goals:

1. Develop an understanding of current earthquake risk perception, protective action knowledge, and basic earthquake preparedness across Washington, Oregon, and California populations.

2. Use social science research to inform the communication, education, and outreach (CEO) program for ShakeAlert, outlined in the 2018-2019 ShakeAlert CEO Plan (2018).

4. Develop a monitoring and evaluation plan for ShakeAlert.

This presentation focuses on the 15 active projects for Year 3 and plans for Year 4 and 5.  

This presentation is on the longitudinal study of Lumberton, North Carolina. The study is conducted by a multidisciplinary research team from the National Institute of Science and Technology (NIST)-funded Center of Excellence for Risk-Based Community Resilience Planning and NIST Engineering Laboratory. The study began shortly after Hurricane Matthew caused catastrophic flooding in Lumberton in October 2016. It has continued for five years, with six data collection field visits to date. 

This presentation will cover background on the study, including the research and sampling design, data collection timeline, and metrics used for tracking recovery through time. This field study is characterized by novel field study data collection methods. The methods include a focus on social dimensions (e.g., education) critical to community resilience; surveys and sampling plans ensure representation of physical damage, socio-demographics, housing types, and business sectors; and structured social science surveys and engineering damage assessments are used to integrate physical and social impacts and track recovery over time. Unique research objectives guided each wave of data collection, ranging from the assessment of initial physical and socio-economic impacts to tracking recovery of community sectors such as housing and business. We will describe the significant challenges encountered alongside the beneficial outcomes that arise when disciplines and perspectives are integrated. Finally, this presentation will explore the cross-institution Institutional Review Board procedure, team training and safety protocols, and implementation of technology in the field.

While research has contributed to the development of comprehensive measurements of disaster recovery, individual recovery self-assessments are likely equally important to individuals' commitment to rebuilding post-disaster. This study examines correlations between individual's comprehensive measures and self-assessments of recovery among Vietnamese Americans, who experienced Hurricane Katrina in 2005. 

Data came from the 2018 survey of a sample of 112 Vietnamese Americans. Correlations between self-assessments and Abramson's comprehensive measures of recovery were examined. Abramson's framework of disaster recovery proposed five measurement domains: stable housing, good mental health, good physical health, stable economic circumstances, and social role functioning. At this time, mental and physical health component scores have been calculated using the SF-12 module.

About 10% of individuals still thought of themselves as victims of Katrina; another ~10% characterized their situation as worse than it was before Katrina. The average score was 39.3 for physical health and 42.7 for mental health, out of 100. Those with higher scores generally were less likely to still believe that they were Katrina victims. However, individuals were less likely to report being in a worse situation compared to pre Katrina with an increased mental health score, but more likely to do so with an increased physical health score. 

Moving forward, the analysis will include other measures of recovery following Abramson's framework. The study will provide insight into measuring recovery among Vietnamese Americans. It is also important to identify which individual-level measures of recovery may better predict the stability and rebuilding of post-disaster populations. 

This study examined communication and organizing processes that constitute resilience. The authors of this study proposed a multi-level resilience framework using field observations, interviews, and document-based data collected from two Houston, Texas-based nonprofit networks impacted by Hurricane Harvey in 2017. The data illustrated how organizing processes build up across levels involving plans then improvisation after a disaster. As individuals work independently then connect to colleagues, they expand organizational level capacity, which then provides a basis to reconnect to interorganizational networks. This study adds to the science of resilience by showing how plans set the conditions for improvisation and proposes that interaction levels better represent time-based disaster response than task-based phases. Findings also amplify persistent dilemmas when it comes to the impact of disasters on society. Resource deserts get exacerbated by social isolation, which amplifies vulnerability. Paradoxically, organizational resilience is carried out by employees whose vulnerabilities become secondary to their work. 

Amid the global pandemic of COVID-19, those who find themselves homeless, or in precarious housing situations, are potentially among the most vulnerable to the spread of this virus. When individuals are being asked by federal and provincial governments under states of emergencies to stay home and self-isolate to stop the spread of this virus, what do individuals do when they do not have a home to go to? Throughout the crisis communication and emergency response to this global health disaster, consideration of those experiencing homelessness has been lacking. This begs the question, where do the homeless go during this disaster? These individuals face unique challenges and threats during disasters. Cities across Canada have responded in different ways to address this issue among this marginalized population. As systems move to disaster recovery, responses need to move forward by avoiding returning to the status quo for addressing issues of homelessness. This Social Sciences and Humanities Research Council-funded mixed-method research study explored the impacts of COVID-19 on individuals experiencing homelessness and the responses by both informal and formal systems to support the needs of these individuals in Halifax (urban) and Cape Breton (rural) Nova Scotia, and how these systems will move forward in disaster recovery.

Society is increasingly focused on resilience to catastrophic events. Firms generally observe two forms of economic disruption–property damage and business interruption (BI). In contrast to pre-event mitigation intended primarily to avoid property damage, firms can use a variety of resilience tactics once the disaster strikes to reduce BI losses by improving stability and continuity of operations. An econometric analysis was performed of large-area survey data collected from firms affected by two of the most devastating natural disasters in U.S. history. Results showed that BI losses exceeded property damage losses by over 900%. Additionally, firms avoided an average of $4.57 in BI for every dollar spent on resilience. Generalizable comparative static estimates were presented for the manufacturing sector to test formalized relationships, derived from an economic production theory foundation, that indicated tactics capable of improving factor productivity can significantly enhance the cost-effectiveness of resilience.

This presentation is on the longitudinal study of Lumberton, North Carolina. The study is conducted by a multidisciplinary research team from the National Institute of Science and Technology (NIST)-funded Center of Excellence for Risk-Based Community Resilience Planning and NIST Engineering Laboratory. The study began shortly after Hurricane Matthew caused catastrophic flooding in Lumberton in October 2016, and has continued, longitudinally, for five years, with six data collection efforts to date.

This presentation will show stress tests for Lumberton’s critical infrastructure networks, including power and water systems, and place past hurricane disruptions in context relative to many possible failure modes. Stress tests unravel infrastructure-level hidden failures, typical and rare, to determine interventions that improve resilience, including alternatives that go beyond tried and true options. In particular, we consider strategies that endow infrastructure with more pathways to handle contingencies, making them agile and capable of handling unforeseen disruptions, essentially steering infrastructure towards antifragility. We will show the benefits of system decentralization in water networks, where no coordination or resource sharing is strictly necessary for network partitions to achieve system-level performance goals. We will also show the merits of network topology reconfiguration to changing power grid operational conditions relative to systems without interventions or traditional interventions such as hardening. Our granular assessments also offer links to account for key population dislocation and business interruption effects, transcending infrastructure interventions into palpable community resilience.

Flood risks are rising across the United States, putting both the economic and social value of growing numbers of homes at risk. In response, the federal government is funding the purchase and demolition of housing in areas of greatest jeopardy, tacitly promoting residential resettlement as a strategy for climate adaptation, especially in urban centers. Despite these developments, little is known about where people move when they engage in such resettlement or how answers to that question vary by the racial and economic status of their flood-prone neighborhoods. The present study begins to fill that gap. First, the researchers introduced a new typology for classifying environmental resettlement along two socio-spatial dimensions of community attachment: (a) distance moved from one's flood-prone home; and (b) average distance resettled from similarly relocated neighbors. Next, they analyzed data from 1,572 homeowners who accepted government-funded buyouts across 39 neighborhoods in Harris County, Texas—the core of the Houston metropolitan area. Results indicate that homeowners from more privileged neighborhoods resettle closer to both their flood-prone homes and to one another, thus helping to preserve the social as well as the economic value of the home; whereas, homeowners from less privileged areas end up farther from both their departed home and each other. Implications for understanding social inequities in government-funded urban climate adaptation are discussed.  Jame

Exposures to extreme events are increasingly common in many parts of the globe as a function of changes in weather patterns combined with rising sea levels and population growth in risk areas. This presentation traces the implications of exposure to an extreme event for health, well-being, and population change over the long-term, drawing on data STAR, the Study of the Tsunami Aftermath and Recovery. STAR has interviewed respondents from 10 months before the December 2004 Indian Ocean earthquake and tsunami for 10 years, collecting data on mortality, morbidity, family formation and fertility, and migration. The data provide an unusual opportunity to examine resilience and the role of rebuilding assistance in the aftermath of a large-scale disaster.

Since Hurricane Harvey, various initiatives around equity in flood infrastructure planning have been piloted in and around Houston. This paper traces these initiatives ethnographically during the period following the passing of the flood bond in the summer of 2018 up until the fall of 2020. From long standing bureaucratic measures like how Harris County Flood Control measures project impact in structures, not property value when they aren't taking federal dollars to the prioritization framework built by County Judge Lina Hidalgo's staff for the flood bond to the formation of the Community Resilience Task Force, this paper explores different ways of evaluating life, loss, and damage that are being implemented at the local level. It examines how the federal benefit-cost ratio intersects with historical redlining and demonstrates the limits of an ahistorical approach to equity in flood risk assessment and project planning. This paper draws on the author's ethnographic and archival research at Harris County Flood Control and additional observations and in-depth interviews during the post-Harvey flood infrastructure planning process more broadly. Through long-term ethnographic attention to local infrastructure, agencies, and actors, this paper identifies social, material, and bureaucratic levers of change difficult to capture in more quantitative social science work in longitudinal disaster recovery research. This paper offers an in-depth case study to complement and inform related work to combat the historic inequities of benefit-cost ratio in an era of more and worse weather events related to climate change. 

In May 2020, researchers at the National Institute of Standards and Technology and the National Oceanic and Atmospheric Administration launched a longitudinal study to assess complex event resilience of small and medium-sized enterprises (SME). In particular, the study was designed to evaluate: (1) how the COVID-19 pandemic is affecting SME response to natural disasters; and (2) whether natural disaster resilience is helping SMEs cope with the impacts of the pandemic. Out of about 1300 respondents to the first wave of data collection during summer 2020, 705 opted into a follow-up second wave. Between December 2020 and February 2021, 247 or 35 % of the opt-in panel responded to the second wave of data collection. Initial findings demonstrate that while only about 25 % had experienced a natural hazard event between March and August 2020, about 40 % of the panel experienced a natural hazard event after August 2020. Moreover, 68.8 % had implemented some type of adaptation/coping strategies since August 1, 2020, with 38.2 % reporting that disaster actions in the past helped cope with impacts of the pandemic. Looking forward, 28.7 % of businesses plan to adopt practices used during the COVID-19 pandemic in anticipation of future natural hazards. This effort addresses the gap in research on the experiences of SMEs dealing with complex events generally and those that arise during a pandemic, specifically. This longitudinal study accounts for SME vulnerabilities, which may further amplify the impacts of a singular or complex event.

Disasters are associated with housing damage and destruction, and hence residential change, but scholars know little about the scale of this phenomenon or the processes underlying it. The American Housing Survey, 1997-2013, provides a unique opportunity to examine the prevalence of disaster-related residential mobility, selection into this type of mobility, and residents' satisfaction with their new housing and neighborhood compared to their previous ones. Householders, the unit of analysis, are grouped into four mover types (employment-related, housing- and family-related, disaster-related, and other forced moves) according to their self-reported reason for moving. Findings show that disaster-related movers are a small proportion of all recent movers, although this proportion is positively related to high damage hurricane seasons. The profile of disaster-related movers differs for homeowners and renters is an important distinction, given that renters are more mobile than homeowners in general. Among homeowners, disaster-related movers are more likely to be single heads of households and African-American but have lower odds of being Hispanic or Latino compared to non-movers. Among renters, disaster-related movers are very similar to non-movers; however, they are less likely to be Hispanics or to hold a college degree. In general, disaster-related movers are more similar to non-movers than they are to those moving for other reasons, consistent with the idea that a disaster-related move is forced. Finally, disaster-related movers are less satisfied with their current housing and neighborhoods than other movers, again, consistent with the idea that their choices were made under duress. 

Frontline workers, police officers, firefighters, and emergency medical technicians (EMTs) continued to keep us safe and cared for in times of the COVID-19 pandemic, often putting their lives at risk. They also have continued to respond to other crises as part of their jobs, including blizzards, floods, hurricanes, earthquakes, socio-political crises (e.g., the Black Lives Matter protests), and other public health crises (e.g., opioid epidemic). Despite a growing literature on the COVID-19 pandemic, our understanding of the impact of concurrent crises on frontline workers remains limited. To fill this gap in the literature, this paper asks: (1) How has focusing on multiple crises at once impacted frontline workers? (2) What needs to be done to mitigate the short and long-term negative impacts of the pandemic on frontline workers?

To fill this gap in the literature, the authors of this study conducted a nationwide survey (n=7,815), interviews (n=60), and a photovoice project with frontline workers. Preliminary findings indicate that the COVID-19 pandemic has impacted frontline workers and their work-life balance the most due to its unprecedented impact and longer duration compared to other crises. The growing anti-police sentiment's impact on the police and the Opioid Epidemic's impacts on firefighters and EMTs have also been noteworthy. While frontline workers have developed their individual-level coping mechanisms (e.g., reduced work hours) for concurrent crises, there is a need for organizational-level policies (e.g., job rotations) that can make frontline workers more resilient in the face of these crises. 

The importance of non-profits has been recognized by planners; however, we lack information on how non-profits navigate compound and consecutive disasters, which may reduce their capacity to perform. This study examined non-profit experiences in the context of Hurricanes Irma and Maria, the 2020 Puerto Rico Earthquake Sequence, and the COVID-19 pandemic. We asked: (1) How do the challenges brought about by the prior extreme weather and/or geohazard events or the recovery process compound the effect of a public health disaster (or vice versa) on health and non-health focused non-profits? (2) How does a non-profit’s experience with service provision in one type of disaster event help them to prepare or respond to a different type? The study included surveys with non-profit agencies (n=300). The surveys were supplemented with longitudinal data from Form 990 Returns filed by non-profits (available through GuideStar) as well as with in-depth interviews with non-profit agencies,leadership (n=40). Our research found that health and non-health focused non-profits in Puerto Rico were able to use their Hurricane Maria and earthquake experiences to become more efficient at facilitating information flow, fill resource gaps, and mediate between the government and residents in recovery planning efforts during the COVID-19 pandemic. During Hurricane Maria and the 2020 earthquakes, non-profit organizations changed their missions, programs, and target populations in order to adapt to the current public health crisis. Our research can help planners, emergency managers, governments, and other stakeholders to better engage with non-profits during recovery from compound and consecutive disaster events.

In recent years, geospatial analyses of emergency shelter locations have contributed to improved operations related to disaster evacuations, as well as the provision of resources and humanitarian aid for affected populations. However, these analyses are usually limited in many of the world's poorest countries due to the frequent underreporting of geographic information related to disaster management and risk reduction. Limited access to technological equipment, inconsistent categorization of shelter infrastructure, and the lack of a nationally standardized street addressing system contribute to this underreporting. Until now, few studies have examined how emergency shelter location and accessibility patterns intersect with the historical impact of disasters, transportation/connectivity infrastructure, and sociodemographic characteristics in countries exposed to multiple hazards with frequent underreporting of geographic hazard-response relevant information.  

This research aims to design a protocol to derive and leverage historical geolocation information about emergency shelters and other variables of relevance to hazard exposure and response between 2012 and 2018 in Guatemala. This study employs a web-scraping methodology as a tool to retrieve geographic location information (combining manual and automated processes) and then performs spatial and statistical analysis on collected data. The objective of the analyses is to understand: (i) whether there is a relationship between the accessibility to different types of shelters, the level of exposure to a particular natural hazard, and the potential number of evacuees, and (ii) the relationship between the total shelter size/capacity and the sociodemographic and road connectivity characteristics of the Guatemalan population at the provincial level. 

The rate of disaster recovery depends upon preparedness and hazard mitigation. Evidence shows that public hazard mitigation investments are not distributed equitably. This paper builds off prior longitudinal studies that examine structural biases in the U.S. disaster aid system, with a focus on the Federal Emergency Management Agency’s Hazard Mitigation Grant Program (HMGP)). These papers examined distributions of HMGP-funded property buyouts and their social justice implications. The authors of this study extended the critical analysis of the HMGP program beyond buyouts to consider the distribution of other mitigation investments, like property elevations. The use of longitudinal data helped resolve limitations associated with cross-sectional data used in the early environmental justice literature and allows for the consideration of causal relationships. The scope was limited to coastal shoreline counties. Given competing pressures like development interests and climate risk, the coastal zone is a dynamic area to study hazard mitigation policy implementation. The sample compared over 15,000 private residential properties that have been elevated or bought-out in the coastal zone in the past 30 years using HMGP funds. Preliminary findings indicate that there are statistically significant differences in the socio-demographic composition of ZIP Codes where property elevations occur compared to where buyouts occur. Compared to ZIP Codes with buyouts, ZIP Codes with property elevations have on average significantly more White people, wealthier populations, higher home values, increased rates of homeownership, and greater proximity to the shoreline. These findings deepen concerns that the HMGP may be exacerbating patterns of spatial inequality.

Children and older adults are two populations typically deemed vulnerable during disasters and targeted for interventions. With COVID-19, their vulnerability is attributed to greater health risks with exposure to the virus for older adults (over 65) and getting behind in their education due to school closures for children (ages 5-17). There are also concerns about isolation, emotional health, and economic instability. Indeed, age vulnerability is central in the pandemic, and yet the voices of the young and old are rarely heard. This research project explored the lived experiences of children and older adults in several sites in the United States and Canada, examining vulnerability, mobilities, and capacities. This CONVERGE-sponsored, Quick Response study utilized multiple methods, such as daily journaling of school-aged children and of older adults during the pandemic, as well as interviews, online surveys, drawings and mapping, focus groups, podcasts, and workshops, with both age groups, parents, and key informants in the community. The results of this project will contribute to better disaster preparedness, responses, and policies and support systems that address the specific challenges and resilience of these potentially vulnerable and highly diverse groups. The project also explored the potential for digital communication between generations. This study, which brings together an international team from three social science backgrounds, will help us gain a broader understanding of how to keep potentially vulnerable populations safe and resilient as the pandemic continues and prepare us for future disasters. 

The CONVERGE Training Modules have inspired innovative lessons and learning assessments for teaching the next generation of hazards and disaster practitioners and scholars. In this panel, five college and university instructors, Reggie J. Ferreira, Christine Gibb, Hans Louis-Charles, Rejina Manandhar, and Olivia Vila, will present their contributions to the CONVERGE Training Module Assignment Bank that they initially created for undergraduate courses in different disciplines. After a short introduction and overview of the CONVERGE Training Modules, each panelist will briefly present their assignment and how it is integrated into their course. Most of the session will be dedicated to a moderated discussion exploring student responses to the module and assignment, accompanying rubrics, lessons learned, recommendations for improving upon the assignment, and adaptations for in-person and online teaching modalities. The audience is warmly welcomed to participate in the discussion!

Long-term individual recovery from a catastrophic event involves the restoration of social, economic, physical, and emotional well-being, particularly for displaced and highly traumatized populations. One strategy for measuring recovery over time involves establishing a longitudinal observational cohort. This analysis examines the cost, effort, and strategies involved in developing and maintaining a longitudinal cohort. 

Few disaster studies have methodically traced these restoration processes beyond a year or two. The Gulf Coast Child and Family Health Study recruited 1,079 randomly sampled individuals in Louisiana and Mississippi within months after Hurricane Katrina and followed them for thirteen years. Participants in the study were interviewed in person five times over the study period. Despite the challenges involved in maintaining a transient and economically vulnerable study cohort, 80.2% of all eligible participants were surveyed at the fifth time point, over a decade after the event; at each round of data collection, the refusal rate ranged from less than 1% to 3.3%. Reasons for non-participation included institutional inaccessibility for those who were incarcerated, in treatment programs, or in nursing homes; physical and mental health issues that preclude participation, including cognitive decline; loss to follow-up; and death. At the final round of face-to-face interviewing in 2018, the average field cost per case was $340 (not including a respondent incentive of $50) and required an average of 15.2 hours per case. This report describes the constitution and maintenance of such a long-term disaster cohort and considers the potential biases of cohort attrition and strategies to mitigate them. 

As a result of California's housing and wildfire crisis, an increasing number of people live in areas likely to be impacted by natural hazards such as flash floods and landslides. Some are pushed into these areas for economic reasons, while others choose to live in these areas. Across these groups, many are largely unaware of the risk, which entails a combination of social, economic, institutional, and environmental vulnerabilities at varying scales, from individuals and households to entire neighborhoods and cities. The vast majority of studies related to natural hazards and vulnerability focus on the dimension of social vulnerability, which is itself multifaceted and a crucial area of study. This study focused on institutional vulnerability, a neglected area of study, to demonstrate the importance of incorporating this dimension into disaster risk reduction agendas to help to mitigate future risk and protect local communities. The authors drew on a case study of the 2018 Montecito debris flows in Southern California, which significantly impacted the community, resulting in 23 deaths, hundreds of injuries, and extensive damage to homes and public infrastructure. While Montecito is a wealthy community in the Wildland-Urban Interface with low social vulnerability, the authors found instances of high institutional vulnerability. Based on mixed-methods research, including interviews with 25 members of the community, this study reveals that a more comprehensive assessment of natural hazard vulnerability, including institutional vulnerability, is critical.  

Sport and activity can be important outlets in North American culture, can be used to maintain and strengthen fitness levels, and to ensure mental and physical well-being. A well-defined body of literature shows that the release of endorphins that occur when engaging in sport and activity is correlated with improved mental health and the ability to cope with stressful situations easier. This project focused on the loss of access to participation in sport and activity during the COVID-19 pandemic. To ensure the health and safety of North Americans during the COVID-19 pandemic, governments have closed most, if not all, sport and recreational facilities. The impact of the removal of sport and activity to this degree in some people’s lives are significant. Additionally, because access to sport and fitness facilities is a class-based privilege, the researchers were interested in how the closure of both public and private gymnasiums and parks has impacted respondents’ perception of fitness and overall physical well-being during the pandemic quarantine. In this project, participants were asked to submit virtual recordings of innovations to sport and activity that they are creating inside of their homes. The researchers were interested in how the lack of access to sport and activity has forced individuals and groups to get creative, and whether the perceived effectiveness of these innovative ideas creates similar feelings of well-being as were experienced through sport prior to the pandemic. 

While Oklahoma is known as one of the most tornado-prone areas in the U.S., earthquakes are a relatively new hazard for the state. Most of the earthquakes are relatively small, causing cracks in brick facade and other minor damage, with a few larger earthquakes causing considerable structural damage to both homes and businesses. To add a wrinkle to the Oklahoma hazardscape, earthquakes in Oklahoma are a techna hazard, where most geological studies suggest that the earthquakes are likely a byproduct of oil and gas exploration. This study explores the factors that shape Oklahoman's attitudes toward tornado and earthquake hazards. A total of 4977 questionnaires were mailed and delivered using a disproportionate stratified random household sample from 27 counties across Oklahoma during the fall of 2019. In total, 866 households responded with a response rate of 17.40%. Findings suggest that demographic variables are better predictors of tornado hazard attitudes than earthquake hazard attitudes. In addition, tornado risk perceptions and hazard salience both positively predict tornado dreadfulness and negative emotion. As for earthquake hazard attitudes, while mineral rights and oil industry employment do not predict earthquake hazard attitudes, earthquake risk perceptions are a significant predictor in all the models. More importantly, the findings suggest that politically conservative respondents are less likely to believe earthquake hazards are a novel risk in Oklahoma. Likewise, conservative respondents are less likely to have stronger negative emotions toward earthquakes. These findings can help shape risk messaging across the state to encourage households to undertake hazard adjustments.

Post-disaster housing recovery is often expected and yet, after a major disaster event, many residential structures are abandoned or razed, resulting in substantial numbers of vacant parcels. Disaster recovery is spatially uneven, resulting in some neighborhoods lagging behind in rebuilding. Consequently, resultant vacant parcels may take longer to or never recover . Neighborhood change theory and broken window theory suggest that in normal urban processes, vacant parcels that remain vacant long-term can be contagious, degrading adjacent property values and potentially increasing uncertainty with respect to reinvestment, leading to deferred redevelopment and a downward spiral. To explore this phenomenon in the post-disaster context, this study employed 11-years of Galveston County annual property tax records from 2007 before Hurricane Ike through 2017 to examine single-family residential parcels prior to the hurricane that became vacant after the hurricane and did not redevelop. On Galveston Island and Bolivar Peninsula, more than half of the vacant parcels remained undeveloped through 2017, although a small proportion were acquired through the repetitive loss buyout processes. The authors of this study designed statistical models that predicted which kinds of formerly residential parcels in different kinds of neighborhoods would not recover. The model results highlight that vacant parcels may interrupt redevelopment efforts in adjacent areas. The findings highlight factors that can hinder or facilitate redevelopment outcomes and can help community decision makers and planners in long-term recovery planning by identifying neighborhoods prone to long-lasting vacancy and help allocate recovery resources for vacant parcel management and regeneration strategies.

EM-DAT recorded 9,167,302 disaster fatalities since 1900 in India. HSBS Global Research has ranked India most vulnerable to climate change in 2018. Ham in India go to the disaster site with their transceivers and antenna to restore emergency communication. Communication failures in disasters cause avoidable fatality, morbidity, and losses. To mitigate this vulnerability, the establishment of fail-proof frugal Ham Club Station (HCS) in District EOCs that provide a variety of voice, text, image, and data communication modes is recommended. HCS will have communication capability, when other modes of communication fail. 

In the United States, HCS exist in EOCs. In India, it is illegal to possess a transceiver without an amateur radio license. This research finds U.S. regulations that allow HCS in EOCs in order to persuade Wireless Planning and Coordination Wing (akin to FCC), Ministry of Communication, Government of India for policy changes that allow establishment of HCS in District EOCs. The district collector (administrator) or his nominee may be custodian of HCS, without communication rights, unless the person is a licensee. 

HCS will promote amateur radio hobby through amateur radio training, help with getting licensed, and disaster drills. Ham will maintain emergency communication on activation of EOC. Establishing HCS in 739 Indian district EOCs will require about Rs.22 Crores (~ $3 million) investment in equipment, that will give maximum social benefit return. A Coalition of Disaster Resilient Infrastructure fellowship is funding this one-year ongoing project. 

Changes in natural and built environments occur continuously over time, and they are accompanied by changes in populations that occupy them and their sociodemographics. This study develops a methodology to quantitatively assess changes in topography, development, and impervious cover over time and their association with flood vulnerabilities in Houston, Texas. A concomitant longitudinal analysis of population dynamics and demographics, damage to natural and built environments, and economic recovery measures for Houston provides a resilience assessment framework for natural hazards, integrating change metrics for multiple indicators. Relationships between the observed changes in natural and built environments and those in populations will be developed to begin to elucidate their association with recovery from disaster.

Several studies have demonstrated the aggregate impact of damage and disruption of schools and hospitals after disasters and their critical role in community recovery. However, there is limited research on the disparities of impacts across educational and healthcare services based on various levels of vulnerability. To address this gap, the authors of this study adopted the lens of social vulnerability to understand the patterns of education and healthcare services impact and recovery in Paradise, California following the 2018 Camp Fire. This paper presents the findings from in-depth qualitative analysis of 33 semi-structured interviews with representatives of 18 Butte County organizations, and archival research of secondary media documents, reports and newsletters. The authors hypothesized that disproportionate impacts from the loss of individual services compound for socially vulnerable groups due to the interdependencies across various services. The magnitude of interconnected losses with massive community-wide destruction raises critical questions about recovery decision-making. Restoring vulnerable communities similar to Paradise after such widespread impacts involves more than restoring individual components, such as housing or infrastructure. It requires a comprehensive look at the many dependencies each component has on other sectors. Therefore, a key local recovery priority should be to develop a shared expectation among stakeholders and community members before a disaster on decisions to restore certain sectors or the whole community given the magnitude and interdependencies of expected losses.

Disasters often exacerbate the pre-existing socio-economic inequalities in society. Previous research shows that minority and low-income populations are disproportionately impacted by disasters with respect to housing damages and economic losses. These disparities in impacts and recovery are further heightened in low-attention disasters due to the lack of assistance funds, whereby vulnerable households rely on their own limited resources to work on repairs at a slower pace. This paper presents an in-depth qualitative analysis of semi-structured interviews, media articles, local organizations' documents, and social media content from impacted households in Marshalltown, Iowa, after the July 2018 EF-3 Tornado. Findings show that lack of media attention, inaccessible government assistance, and pre-disaster socio-economic inequality exacerbates and prolongs severe disaster impacts in a community when the majority of affected households are vulnerable due to compounding factors. For example, low income, lack of home insurance, health fragility, and inadequate repair assistance left no option for several households other than living in a damaged, unsafe home during the cold winter of Iowa. Their alternate access to repairs is dependent on the availability of help from family members and friends. Further, households at the intersection of immigrant status and English proficiency avoid applying for local non-governmental assistance due to the fear of potential deportation for their family members or relatives. These add to their burdens of lingering repairs, exacerbation of unrepaired damages, and declining quality of life and mental health. These findings show the mechanisms through which post-disaster recovery policy exacerbates the pre-disaster inequalities in housing.   

The central question of this research is whether serious games can provide a suitable method for communicating tsunami risk to children. Child centred approaches to disaster risk reduction have gained traction over the last decade; Equipping children with actionable knowledge and decision-making skills would enable them to make appropriate, potentially life-saving decisions in extremis. An imbalance towards teaching non-European disasters can result in considering hazards as remote risks for children who may travel outside their curriculum focussed geography. In the Mediterranean region, tsunami risk is under recognized but potentially dangerous, where actionable key messages about natural warning signs (NWS) inform evacuation decision-making. Children are particularly vulnerable in hazard situations, sometimes away from parents, unsupervised or supervising siblings. They are vulnerable to centralised, hierarchical command structures, sometimes inadequate situational awareness and incompetent decision-making. As part of this study, 170 Moroccan school children were surveyed relating to tsunami risk in Morocco and their knowledge of tsunami NWS. Results showed understanding that earthquakes can cause tsunami, but significant uncertainty in recognizing NWS existed, despite assertions to the contrary. Although most respondents regularly visit the Moroccan coasts and recognise the lack of warning systems, significant uncertainty about tsunami risk in Morocco existed. The study concluded that children unfamiliar with NWS are ill-equipped to make tsunami evacuation decisions and educational policy makers should look to include credible actionable key messages in curriculum materials. The work will go on to assess the potential of serious games to communicate natural hazard risk.

Successfully coordinating a sustained research effort at national and regional levels can be challenging, and best practices and lessons learned are not often documented. The National Institute of Standards and Technology (NIST) and the National Oceanic and Atmospheric Administration’s (NOAA) Regional Integrated Sciences and Assessments (RISA) program are collaborating on a set of research projects exploring Small and Medium Enterprise (SME) disruption and resilience. Some of the most significant costs associated with the impact of natural hazards stem from disruptions experienced by SMEs. Understanding the extent of these disruptions, how SMEs learn about and implement resilience measures, and how this differs across communities and populations are key for effective and equitable recovery.

While the collaboration was initially grounded in regional case studies focusing on the impacts of natural hazards in three locations, research was disrupted by the onset of COVID-19 in March 2020. Knowledge of SME resilience became crucial as SMEs faced closures, among other challenges. As the projects shifted to consider the complexity of the events, NIST and NOAA launched a parallel survey effort on a national scale to capture ephemeral and longitudinal data. As a whole, the national and regional data collections inform each other and account for the disruption and resilience of SMEs from complex events springing from the combination of COVID-19, natural hazards, and social vulnerabilities. This paper presents the progress of this effort and discerns lessons for other national-regional partnerships that focus on applied research, over long timeframes, and requiring flexibility in mode.

This presentation is on the longitudinal study of Lumberton, North Carolina. The study is conducted by a multidisciplinary research team from the National Institute of Standards and Technology (NIST)-funded Center of Excellence for Risk-Based Community Resilience Planning and NIST Engineering Laboratory. The study began shortly after Hurricane Matthew caused catastrophic flooding in Lumberton in October 2016. It has continued, longitudinally, for five years, with six data collection efforts to date. This abstract is being submitted alongside five other presentation abstracts to form an entire session proposed to be moderated by John van de Lindt, Colorado State University.

This presentation will cover the multiple impacts and long-term recovery process for businesses through the three events, including sector-by-sector differences, and interactions with housing recovery. In particular, the recovery linkages between businesses and households will be discussed. Customer loss, in particular, had a higher effect magnitude than initial damage in terms of hindering the recovery of businesses. Labor disruption caused by transportation issues and childcare or school closure issues had a smaller relative effect, but it also significantly lowered business odds of full recovery. Furthermore, measurable mitigation and adaptation efforts taken by households, businesses, and the City will be presented. Discussion as to the significance of these steps towards helping the Lumberton community be better prepared for future disasters will be discussed. 

In the wake of the COVID-19 pandemic small businesses made headlines as hard hit by evolving supply and demand-side difficulties, such as limiting customer interactions, employee availability, as well as supply chain and inventory issues. There were 22 billion dollar natural disaster events in 2020; thus, interruption of small businesses in areas vulnerable to natural disasters are particularly noteworthy in the complex context of COVID-19. Our findings indicate natural hazards experienced during the pandemic create greater likelihood for stopped operation and decreased revenue across small business types and sectors. But importantly, minority and women-owned businesses see largely amplified negative impacts from COVID-19 that far outweigh the compound impact of both a natural hazard and COVID-19 felt by majority-owned businesses. When hit by a natural disaster in addition to COVID-19, a minority-owned business is 67 % more likely to stop operations altogether than their majority counterparts. This brief report presents findings from the first wave of a novel longitudinal survey of over 1300 U.S.-based small business operators which accounts for past and current experiences with natural disasters and extreme weather events. These results demonstrate the importance of considering COVID-19 impacts and associated recovery within the context of complex pre-existing social stressors and other community disruptions. These findings have important implications for processes, institutions, and policies that are typically external to the small business operator, direct control and drive income losses and future economic inequality.

The emergency food system had not faced a public health and economic disaster such as the coronavirus pandemic. The main goal of this project was to investigate how food pantries responded to the current crisis in the first 3-6 months of the pandemic. Specifically, the authors of this study sought to document the challenges faced and the strategies used to meet those challenges, and the success stories or accomplishments that shed light on efforts that sustain organizational and community resilience during this crisis. 

The authors conducted seven virtual focus groups with food pantry directors in Missouri from July-October 2020. The pantry directors (n=21) represent a range of pantries that differ by geography, the number of clients served, and resources. All focus groups were recorded and transcribed. They conducted thematic coding of transcripts. 

This study found that pantry directors were concerned for the health and safety of pantry clients and volunteers. Social distancing meant pantries had to alter their food distribution methods and use volunteers to limit interpersonal contact. They also had to manage fluctuations in the demand for food and the increased supply of food available. Stress and exhaustion also weighed on the directors. In the face of these challenges, food pantry directors noted several unexpected opportunities, including new resources (donations and volunteers) and strengthened community ties.

The emergency food system is an important element of the disaster network in a community. They will benefit from inclusion in disaster planning efforts in the future.   

FloodWise is a pilot project that proposes to develop innovative new approaches for flood disaster resilience by applying modified nature-based conservation practices in Eastern North Carolina. The authors of this project will discuss the estimated costs of the selected natural infrastructure practices developed via literature reviews, data collection from secondary or primary sources, and the use of economic-engineering approaches for cost and benefit estimation. Based on the costs to implement water farming practices, they have estimated the number of payments necessary for farmland owners to break even to consider participating in the FloodWise program. 

This presentation will include a summary of the preliminary estimates of natural infrastructure costs and the discounted cash flow and capital budgeting procedures. Small modifications of existing conservation practices, such as stream buffers, cover crops, silvopasture, and tile outlet terracing, had the least costs to install. Major natural infrastructure projects that required substantial earth moving and flood control structures were more expensive and would have more adverse impacts on existing farm and forest management. Larger flood water structure projects could store more water for longer periods, however, which may offset their greater establishment costs. Subsequent analyses will examine landowners' interest in adopting such practices, and with agencies and nongovernmental organizations about means to develop FloodWise programs for landowners and communities. 

Since the onset of the COVID-19 pandemic, broad vaccination coverage has been widely viewed as the key to overcoming rising infection numbers. Vaccination coverage is particularly important for rural and low-income households, which are disproportionately vulnerable to limitations in the availability of medical services. Meanwhile, public concerns regarding vaccine safety and efficacy are escalating as more becomes known with respect to vaccination side effects and effectiveness against variants. To better understand the willingness of rural households to receive a COVID-19 vaccination, this study surveyed 492 households in nine rural neighborhoods of Yaan, Sichuan, following the framework of the Protective Action Decision Model (PADM). The survey collected respondents' information reliance, preparedness level, previous protective action adoptions, emotional responses, protective action assessments, and demographic characteristics. The survey showed that a household's willingness to receive a COVID-19 vaccination bypasses two effect routes: respondents may either carefully evaluate the protective action in terms of hazard-related concerns and vaccine availability or heuristically follow governmental guidance without additional risk or protective action assessments. Observed risk exposure (e.g., the percentage of others wearing facemasks in public) was identified as a key predictor in determining the psychological route that respondents would follow. Surprisingly, although research stresses the importance of behavioral consistency in predicting human behavior, previous commitment to protective actions (e.g., wearing a facemask) had limited or only indirect effects on the willingness of respondents to receive a vaccination. The findings of this study will benefit local authorities in reducing the inherent urban-rural divide in vaccination. 

This study examined the potential for the further integration of urban planning and hazard mitigation planning by exploring the role of the built environment and social infrastructure in connecting people with essential resources that might be needed in a disaster scenario. The research design used a mixed methods approach, comprising a sample survey administered in three characteristically different Washington State communities and semi-structured interviews with community hazards mitigation planners and volunteer organizers. Using survey data indicating where people anticipate turning for different kinds of essential resources in an earthquake scenario, the authors identified patterns of expected resource seeking as they relate to the built environment. They then spoke with local leaders and volunteers to explore ways in which local social infrastructure and other aspects of the built environment could be adapted to better connect communities with essential resources, both in a disaster scenario and on an everyday basis. 

Through a synthesis of the quantitative and qualitative data, including lessons learned during the COVID-19 pandemic, they developed recommendations for an expanded role of the built environment in hazard mitigation and resource provision. In particular, the authors identified potential means of intervention, including designing multipurpose facilities, leveraging new technology, and building strategic partnerships. Lessons learned can inform both community planning practice and hazard mitigation planning, as well as highlight ways in which the two disciplines can continue to become better integrated.

The recent increase in the number of disasters and complexities surrounding existing emergency management efforts necessitates the application of cross-sector collaboration and coordinated risk communication. While previous scholarly works have discussed emergency management efforts from a collaborative perspective, little consideration has been paid to the role public-private partnerships (PPPs) and information technology (IT) play in emergency management. The purpose of this study is to examine the relationship between PPPs and IT-enhanced risk communication in building community resilience. Using a web survey of the emergency managers and operators in public, private, and nonprofit sectors in the East Central Florida region, this research proposal aims to answer the following research questions. RQ1) To what extent does IT-enhanced risk communication affect emergency managers' and operators' perception of community resilience? RQ2) What factors affect emergency managers' and operators' perception of PPP effectiveness and institutionalization? The result of this study will help emergency managers across multiple sectors and policymakers to better understand the factors that contribute to successful PPPs in emergency management and assist them in planning for, managing, and utilizing their scarce emergency resources when collaborating with other organizations.

Hurricanes with gusty winds and catastrophic flooding can severely affect household well-being in affected areas. Hurricane Harvey stalled over the southern part of Texas in 2017 and made landfall resulting in unprecedented damages to community infrastructures and livelihoods. It triggered the highest ever recorded rainfall in the United States (with peak accumulation of 60.58 inches) and caused extensive flooding. This paper investigated the major driving factors related to the extent of damage, utility disruption and hurricane preparation that are likely to affect the household well-being. For this analysis, the authors of the study collected survey data from more than 700 households in Texas living in Hurricane Harvey affected areas. The survey data included information related to respondent’s socio-economic, demographic and hurricane impacts following Harvey. Through empirical analyses, they demonstrated how the extent of utility disruption (electricity, water etc.) and damages intensified the loss of household well-being. The findings suggest that hurricane preparation and access to essential resources after the event can significantly improve household well-being and expedite the recovery process. The finding has policy implications for promoting household level disaster preparation for mitigating future damage.

While longitudinal research is important to understanding long-term recovery, the actual mechanics (e.g., careful relationship building, required travel time, and limited research funding) of such studies make it difficult to complete. Consequently, research often terminates after a short period, resulting in a limited understanding of the potential contributions of longitudinal studies.

This paper summarizes the unique contributions that longitudinal research can provide. Two research experiences in Tohoku, Japan, affected by the 2011 GEJE, and Leyte, the Philippines, devastated by the 2013 Typhoon Haiyan, are revisited; both used a long-term ethnographic approach to follow almost a decade of rebuilding processes. Engagement in both places was beyond conducting research; the author has served on research committees, led on-site classes, and engaged in planning efforts to support recovery. 

Although the two case study settings are different, long-term research uncovered several commonalities useful for understanding the recovery framework. First, government-managed recovery proceeded in three segments, namely: i) plan development based on damage assessment; ii) rebuilding governed with project programs, funding allocation, and stakeholder identification; and iii) project implementation designed to meet the envisioned goal with stakeholder involvement. Second, the mindset and perception of the stakeholders, including government officials and residents, transformed with project progress as they adapted to the environment over time. Lastly, it is important to note that while this paper finds the recovery process as a non-linear, multi-faceted development in time and perceptions, this was only identifiable due to the nature of longitudinal research. 

What explains the absence of electoral incentives for preparedness spending? Previous research demonstrates that voters reward politicians for relief spending after a disaster, but not for preparedness spending despite it being significantly more cost-effective than relief. This absence of political incentives for preparedness spending could exacerbate vulnerability to disasters. This paper seeks to understand how moral foundations and attribution shape individuals; attitudes towards disaster preparedness. An online survey experiment found that morality is associated with attitudes towards federal preparedness spending, with harm/care and fairness/reciprocity positively related to support for preparedness spending, and ingroup/loyalty positively associated with support for greater amounts of federal preparedness spending. Against expectations, it was found that the attribution treatments had no meaningful effects on support for preparedness, suggesting that attribution of blame is not related to public attitudes about preparedness spending. The results have broad implications for the understanding of public opinion about disaster preparedness, demonstrating how moral values shape attitudes about preparedness initiatives, but the attribution of blame has little effect on public sentiment about these initiatives.  

Free choice learning environments (FCLEs), such as museums, national parks, and libraries, are trusted sources of information and support lifelong learning. Earthquake education in these spaces creates communal awareness of earthquake hazards and risk, and in turn, may increase engagement in preparedness behavior when an earthquake strikes. The ShakeAlert Earthquake Early Warning System aims to help communities prepare by warning in advance of shaking from significant earthquakes along the West Coast of the United States. ShakeAlert has the ability to minimize earthquake damage by prompting automated actions (e.g., stopping trains) and prompt personal protective actions, like Drop, Cover, and Hold On. As the testing of public alerting to wireless devices expands, raising awareness of ShakeAlert and earthquake education, is integral to reach the broadest audience possible, and to reduce damage, and loss of life. For these reasons, engaging people in environments where they choose to learn, such as FCLEs, better facilitates earthquake education, and contributes to awareness. Currently, there is a dearth in the literature on how FCLEs approach earthquake education through display themes and development. This study analyzed a national sample of earthquake exhibits and their themes in order to better understand exhibits across common informal learning environments, while also exploring how different display types can be uniquely engaging. The authors anticipate that this work will set a foundational framework, using a combination of pedagogical and communication theories, for how best to incorporate ShakeAlert and earthquake hazard education into FCLEs and support place-based learning.

Ecosystem-based disaster risk reduction (Eco-DRR) measures are nature-based solutions (NbS) that have the potential for DRR and climate change adaptation while delivering multiple co-benefits such as urban cooling, flood mitigation, and improved air quality, biodiversity, and social well-being. However, actionable information on NbS‚ risk reduction, adaptation, and improved social well-being benefits is required before cities decide to invest in them.In a collective effort to reduce disaster risk and limit sprawl, El Volante community stakeholders in peri-urban Lima, Peru, established a 14-hectare forest on the steep Andean mountainside bordering their burgeoning neighborhood. We designed a socio-economic analysis of the Eco-DRR measure to determine its economic viability and sustainability for replication. A household survey (n=100) revealed community members' high concern for rockfall risk (mean 4.25 ± 1.12). Results from our double-bounded dichotomous choice contingent valuation model indicated that households (n=91) are willing to pay $3.44/month (Std. Error 0.49), ~0.6% of their monthly income, for forest maintenance. A risk-based benefit cost analysis (BCA) using probabilistic risk assessment revealed that the eco-DRR project is economically inefficient with a benefit-cost ratio (BCR) of 0.2 ( ±0.02) when considering avoided losses alone. However, inclusion of place-based economic co-benefits and the social values of direct stakeholders (household WTP) increased project BCR to 1.37 (±0.39). Our equitable risk-based BCA demonstrates an appropriate valuation of the eco-DRR project which generates greater benefits for social-ecological well-being in comparison to grey infrastructure solutions alone, with the potential to increase the community resilience to shocks and stressors.

Disaster recovery research has long shown that social vulnerabilities and inequalities affect post-disaster recovery. However, the longitudinal effects of these vulnerabilities on individual/life recoveries are understudied. This study answered the following research question by employing large-scale panel data: How do vulnerabilities in terms of gender, disability, and family structure affect individual/life recoveries (up to ten years)? The data consisted of the 2014-2015-2016-2017-2020 Natori City Life Recovery Population Panel Survey results. In 2020, 1,989 (out of 4,270 individuals, 44.4% response rate) from 899 (out of 1,803, 49.9% response rate) households returned. Regression models tested the interaction effects of gender, disability, family structure, collective/individual social capital, health issues among family members, and household financial difficulties by year on individual life recoveries. The results showed that the vulnerability gaps exacerbated over time. For example, the disability dummy variable became negative predictors from the year 2017 onwards. The necessities of disaster case management were discussed in order to alleviate the lens of vulnerability effects.

Disaster research perhaps formally consolidated as a field in the 1970s, including notable publications on disaster recovery. Efforts have emerged to preserve and make accessible the knowledge and publications from this era, noting that, in some respects, little has changed in the baseline lessons emerging from disaster recovery research today. In examining the original material, three principal factors were identified for this observation: (1) A new generation of disaster recovery researchers and practitioners is not always aware of, or lacks easy access to, the early work. There is not always a realisation of what has been learned, documented, and applied before. (2. Many of the pioneers and founders of this field are today less active in research and practice, including writing about, presenting on, and disseminating their knowledge on disaster recovery.

3. Some of the early publications are still not available online or digitally, with hard copies becoming fewer.

This presentation uses these findings to move forward by documenting, preserving, and making accessible early material from disaster recovery research. The outputs presented are:

1. A table of milestones, giving an overview of key developments across sectors.

2. An overview of the main, identified influential publications from the period along with making them available electronically.

3. Video interviews with people from this era, ensuring a diversity of backgrounds, so that they can reflect on their early personal engagement and lessons learned, as well as what is similar and different today.

After a disaster, calls to build back better often focus on rebuilding physical infrastructure to a higher safety standard in order to reduce risk from future disasters. But what happens when such rebuilding happens at the expense of other aspects of well-being? Four years after the Mw 7.8 Gorkha earthquake in Nepal, this research consortium conducted a field survey of households in earthquake-affected districts in Nepal (N=815) to understand recovery processes and decisions. The researchers found that the initiative to build back better was indeed successful, in that more households rebuilt with safer materials and technical assistance, but that households also took on higher levels of debt in order to fund reconstruction in this manner. Moreover, earthquake impacts caused other disruptions in life plans and unmet livelihood needs that affect well-being, and such effects are not captured in building-focused disaster impact metrics. This project, Informatics for Equitable Recovery, highlights the need for both improved metrics for post-disaster impact and a deeper understanding of recovery trajectories. They make the case for further research into the longitudinal implications of such tradeoffs in recovery, especially in rural contexts in the global South. They also highlight the need for disaster impact metrics that incorporate such non-physical aspects of disaster losses and nuance commonly-used definitions of recovery. 

Small businesses are critical stakeholders of community recovery, but they are often less attended to in the aftermath of disasters, which leaves them vulnerable to the next event. Policymakers could use this recovery moment to promote mitigation, but do they? This study uses a case study approach to examine whether and how small business recovery programs promoted mitigation planning after five recent disasters (2012 Hurricane Sandy, 2016 Hurricane Matthew, 2017 Hurricane Harvey, 2020 COVID-19, and 2020 Magna earthquake). The findings of this study will help assess whether post-disaster recovery programs are promoting mitigation planning among small businesses as well as identify areas of policy improvement. 

Story-telling often shapes the ways that disaster survivors recall and re-experience a catastrophic event. One promising mental health intervention developed among Holocaust survivors is testimony therapy, a narrative-based counseling modality. This study investigated the effectiveness of testimony therapy in alleviating the mental distress of hurricane survivors. 

The Life Story Project was conducted among Katrina survivors two years after their hurricane exposure. One hundred subjects were randomly selected from among participants in the longitudinal cohort of the Gulf Coast Child and Family Health Study and randomly assigned to an intervention or control arm. The cases received a video-recorded structured testimony therapy session with a psychologist or social worker. Ten-year trajectories of mental health changes in the Mental Health Composite Summary score between the cases and the controls were evaluated using multilevel modeling. Two additional control groups were tested: a post-hoc control group assembled using propensity score matching, and a post-hoc control group composed of all other members of the longitudinal cohort.   

Although mental health scores among cases increased by 0.15 points per month with a deceleration over time, they were not statistically significantly different than any of the control groups. 

This study did not discern a statistical significance of testimony therapy as an intervention to promote mental distress recovery. Still, the use of experimental methods in a disaster setting warrant further investigation.

Disaster risk reduction (DRR) policies like building codes and construction regulations are crucial for saving lives and protecting property. The dynamics of public support for DRR policies and their enforcement thus present an important research puzzle. This three-year NSF-funded collaborative project brought together researchers from Florida International University and Vanderbilt University. The research team deployed an original survey module on disaster risk and attitudes towards DRR in 16 Latin American and Caribbean (LAC) countries and the United States. The collected data was aggregated to produce an open-access public opinion dataset (n=25,500). The authors tested hypotheses on the effects of three sets of factors on public support for DRR policies and practices: (1) Disaster risk (perception of the likelihood, severity, and type of hazard events; perception of risk as individual or general; national and local hazard risk profiles); (2) “Experience” with disasters (as an individual/family, community, or nation); and (3) Governance (trust in government; perceptions of effectiveness / integrity in DRR; personal experience with corruption; national levels of corruption). In the final phase of research, they will select one LAC country affected by a major hazard event and home in on the “experience” question: whether disasters as “focusing events” shift public opinion and help open “windows of opportunity” for improving DRR policies and their implementation. The team will compare pre- vs. post-event public opinion; assess the extent and duration of any changes; and analyze causes of change over time at the individual level.

Long-term disaster management struggles with noncompliance and behavioral apathy due to lowered risk perception across social groups. Existing studies capture these perceptions within the early stages of risk communication. However, they seldom examine the moderating effect of these perceptions-when they become local or group norms/contexts-within long-term disaster scenarios. Understanding the moderating effect of these norms can augment emergency managers' situational awareness of critical risk communication needs. Consequently, this augmented situational awareness can enhance risk communication by facilitating appropriate, consistent, frequent, and timely risk information, which can motivate residents to take action weeks or months after a disaster strikes.

The extended duration of COVID-19 allowed the authors to test their posited model over several months and across differing regions. Their preliminary analysis of Chinese and South Korean datasets suggested that a weak norm among residents calls for a much greater emphasis on the provision of risk information to promote residential stakeholder perceptions and protective actions. For example, when the threat perception of COVID-19 or its relative perceived severity (in comparison to other pandemics or epidemics) was low, it was extremely important to minimize conflicting information from different sources and to publish frequent but timely information about the risk. These actions can increase stakeholder trust and responsibility and solidify perceptions of expertise. Moreover, these characteristics have a significant indirect effect on actual protective actions, mediated by stakeholder perceptions. Therefore, this moderating effect can be used to increase the efficacy of risk communications.   

Successful disaster risk reduction (DRR) intervention programs save lives, but there are few opportunities to publicly celebrate these success stories. This is in large part due to the invisible nature of these program benefits: in the best case scenario, a successful intervention results in no negative consequences or impacts. In short, nothing happens. This presentation will introduce four types of situations where successful DRR intervention benefits are made invisible: (i) benefits made invisible in the midst of broader disaster, (ii) benefits made invisible by nature of the success, (iii) benefits made invisible due to yet unrealized events, (iv) benefits made invisible due to the randomness of the specific outcome. To make these benefits more visible, the authors have developed a counterfactual risk framework to estimate the mitigating effect of disaster risk reduction interventions. Focusing on the first type of invisible benefit (i), this framework is illustrated using a case study on the School Earthquake Safety Program success amidst the 2015 Nepal earthquake. For this case study, probabilistic realizations of casualties without the DRR intervention are modeled and compared with actual losses to provide a powerful representation of the success of a DRR intervention. In order to encourage future investment and political action, this study suggests that the value of risk reduction interventions should not be judged only on the basis of specific outcomes, but on the basis of broader probabilistic outcomes.

This paper describes the Interconnected Network Community Resilience Modeling Environment (IN-CORE) building functionality taxonomy developed as part of the National Institute of Standards and Technology-funded Center for Risk-Based Community Resilience Planning. IN-CORE is a tool that allows researchers to model a community's exposure to natural hazards and explore a variety of public policy options to improve the community's resilience to the impact of natural hazards. The proposed IN-CORE building functionality taxonomy is intended for classifying building inventory to assess the loss and recovery of function of building physical and socio-economic systems. It expands and modifies the Global Earthquake Model building structural taxonomy to include novel features that (1) address multiple hazards, such as seismic, wind, flooding, tornado, tsunami, (2) arrange attributes in the context of buildings systems (architectural, mechanical, electrical, socio-economic, etc.) and (3) weigh the attributes for their significance and knowledge levels based on relevance and quality of the information. 

The taxonomy was arranged into seven building systems and subdivided into 39 multi-level attributes. The attributes were selected based on their significance for all the considered hazards and were assigned "significance levels" that represent the relevance of the information for damage assessment and recovery for a specific hazard. In addition, the attribute level was assigned a "knowledge level" value that estimates the quality and completeness of the available information. 

The resulting information is summarized into a taxonomic string, used to provide quick access to the building inventory information and to select and verify the applicability of damage and recovery models.

In April 2015, the Gorkha Earthquake triggered a massive glacier avalanche that destroyed the village of Langtang, taking over 300 lives and causing unthinkable destruction. As the people of Langtangpas worked to rebuild their lives in the wake of this disaster, glaciologists and climatologists also returned to the Langtang Valley to continue the work of understanding climate change and forecasting uncertain futures. In the aftermath of the disaster, the Langtang Valley of Nepal has become a critical site for climate science in the Himalayan region. Because climate change is expected to intensify avalanche risks-both in the Himalayas (as the recent Chamoli disaster in Uttarakhand has highlighted) and across the world-the work of reckoning climatic uncertainties and the dimension of potential cryospheric hazards is increasingly vital. But what does it mean to live with these vital uncertainties?

Drawing from ethnographic research conducted over six years, this paper considered the ways in which the 2015 Langtang avalanche, understood as a diachronic disaster that speaks to shifting hazard regimes, has catalyzed a variety of epistemological, political, methodological, and ethical questions were critical within and beyond the Langtang. Valley. Building from recent scholarship focused on epistemological pluralism, political ontologies, moral ecologies, and climate justice in the Himalayas, the author examined the ways that interactions between different modes of reckoning, shifting Himalayan hazardscapes and climatic uncertainties shape life in the Langtang Valley. 

In the United States, coastal areas of the Southeastern states exemplify the new normal consisting of a perpetual process by states, counties, and municipalities of preparing for a major disaster, recovering from one, and preparing for another close on the heels of the first. The conventional Disaster Mitigation-Preparedness-Response-Recovery cycle was never as linear as it sounded. This dynamic raises myriad questions about the role of discrete planning processes and documents intended to have long-range horizons but, in reality, needing to be updated before they are completed. Here, the authors tackled questions including: 1) to what extent are hazard mitigation and recovery plan documents integrated? 2) to what extent are hazard mitigation and recovery planning processes integrated? And 3) what state and local factors explain variation in the integration of mitigation and recovery efforts? They answered these questions using a comparative case study research design as part of an ongoing National Science Funded study (#1760183). They applied a systematic plan quality content analysis methods to sample hazard mitigation and recovery plans for nine counties in Florida, Georgia, and North Carolina. They also conducted digital surveys and semi-structured interviews with key stakeholders in hazard planning in each jurisdiction. Specific themes they explored across each data set included 1) integration across the plan documents, with particular attention to the policies and actions in each plan; 2) the connections (or lack thereof) in the network of stakeholders in each planning process; 3) connections to long-term land use planning and other planning processes.

Though significant effort has gone into increasing household preparedness over the last few decades, levels across the United States have remained stagnant. To understand and investigate avenues for motivating preparedness, we examine the role information seeking, an understudied aspect of the Protective Action Decision Model (PADM), plays in motivating household preparedness. We focus our efforts on the Portland metropolitan region (PDX) in Oregon, which is home to over two million people and at risk of a Cascadia Subduction Zone earthquake; this style of earthquake could cause billions of dollars in damage and thousands to tens of thousands of fatalities. We employ a quasi-experimental longitudinal research design. In 2019 and 2020 we sent a series of three questionnaires (T1, T2, T3) to a random sample of 2,400 PDX addresses. The questionnaires assess individual characteristics, core perceptions, information seeking, and preparedness. After completing the T1 survey, respondents could request additional risk and preparedness information be sent to them. We sent the T2 questionnaire to those who requested additional information to assess immediate changes following receipt of information. After six months, we sent the T3 questionnaire to all T1 respondents. This study measures how PADM variables change over time and how access to information influences household preparedness. Preliminary results show information seeking behavior to be strongly correlated with household preparedness. The results of this study help examine understudied aspects of the PADM and provide a better understanding of the role educational materials play in motivating household preparedness.

Progress on mitigating the impacts of coastal flooding requires that residents plan for flooding and adapt. Populations along densely populated coasts will face residual flood risks even where the public sector builds flood protections. At the same time, understanding of flood risks and adaptation options among residents remains low. The study focuses on the diverse Rockaway area of New York City devastated by Hurricane Sandy in 2012 and builds on prior research that documented the recovery from Hurricane Sandy. The first phase investigated how and why vulnerability and resilience to flooding differ in the population. The authors will present preliminary findings from an ongoing second phase that tests an approach to building capacity among homeowners to reduce the impacts of future floods and thereby avert potential disasters. An interdisciplinary team of scientists and educators is collaborating with community groups of residents and the City of New York to co-produce knowledge about current and future flood risks and benefits and costs of adaptation actions that homeowners can take with their own resources. The study employs program evaluation methods rarely applied to study disaster mitigation programs to assess whether and how co-production of knowledge affects adaptation behavior relative to less resource-intensive and frequently used means of communicating information such as online tools. The evaluation investigates who would benefit from investment in improving the understanding of risks and options. The study examines how the experience with the recovery from Hurricane Sandy affects willingness and ability to reduce future flood impacts.  

Post-disaster recovery and resilience are now a significant issue in public administration’s emergency management literature because of their intricate linkages with governance issues. Using semi-structured interviews (n= 81) with respondents from National Reconstruction Authority, local governments, and international and national non-government organizations, the researchers examined post-earthquake housing reconstruction of affected households in the Kathmandu Metropolitan City (KMC), Nepal. The paper drew on the network governance concept to describe and understand governance factors shaping the housing reconstruction of the affected population in KMC. The paper showed centralized governance structure and distrust between organizations involved in the recovery network and their impact on post-earthquake reconstruction in the KMC. In addition, the network of influential stakeholders that comprised ruling governments, government bureaucracy, and financial organizations deliberately ignored issues of urban housing reconstruction because it involves broader political-economic issues that are beyond their time, resources, and priorities. Furthermore, inflexible institutional and bureaucratic practices of the central government added numerous barriers to housing reconstruction. These issues have forced urban poor residents to live in risky houses, sell them or move out of the place, or construct non-compliance houses. The researchers conclude by suggesting that post-disaster recovery and resilience initiatives in urban cities need governance structure and capacity that match the complex problems of urban areas.

Disaster events result in substantial climate impacts over various spatiotemporal scales due to fugitive emissions of potent greenhouse gases (GHG) either from the event itself (i.e., wildfires) or from generated waste materials. Climate impacts of the generated waste materials represent an overlooked, yet potentially significant source of GHG emissions. Knowledge of these climate impacts improves a community's ability to prepare for, respond to, and recover from future disaster events. Therefore, a dynamic probabilistic materials flow analysis (dpMFA) was conducted to quantify, for the first time, GHG emissions from disaster waste materials over immediate- (hours-days), short- (weeks to months), and long-term (years) timescales. A range of disaster events was targeted including wildfires, floods, and hurricanes. Advanced debris estimation software (e.g., HAZUS-MH and I-WASTE) was applied to ascertain both the quantity and composition of waste materials generated from these events. Post-disaster waste management was simulated using three distinct scenarios: a baseline, refined, and informed analysis, where the management pathways (i.e., landfill, recycling, and reuse), the amount/type of materials allocated to these pathways, and mitigation efforts (i.e., temporary storage/controlled disposal) were varied. Waste-specific GHG emission factors, which represent the quantity of GHGs released per unit volume of waste material, were developed to delineate the time dependent release and climate forcing impacts of these materials throughout various stages of waste management. The results of the dpMFA indicate that disaster waste materials have important climate impacts, where the management practice highly influences the extent and magnitude of GHG emissions over long timescales.

Unlike the United States, Japan has a system in which local municipalities are responsible for the evacuation shelter operation. Since that, the quality of the evacuation shelter operation varies among municipalities. This study aims to explore the elements necessary to solve this challenge. The researchers identified twelve positive deviant good practice evacuation shelter management cases from four significant disasters over the past ten years. Interviews with twelve leaders were transcribed. This produced 231,208 letters transcript text. Three disaster researchers from sociology, public health, and architectural backgrounds as well as two crisis management practitioners independently extracted key terms from the same transcript. This process created 1,222 key term cards. Through the Affinity Diagram method, eight mutually exclusive super-conceptual clusters emerged. Five out of eight super-clusters corresponded with areas that were prescribed by the National government-issued Evacuation Shelter Management Guideline: 1) functions, 2) a managing organization, 3) resources, 4) bed, bathrooms, dining area layouts, and 5) practice tips. All twelve shelters turned out to be compliant with the Guideline. Three unique super-clusters also appeared to be characteristic to the competent shelter operation: 1) common concrete missions/visions, 2) awareness of and adherence to core humanitarian rights/values, 3) efforts to learn lessons from previous good/bad practices. These three super-clusters are unique and essential for the following reasons: 1) these were not taught in regular procedure-oriented training and drills, 2) these were not observed in bad practices based on the experience of the five research participants. A quantitative study is needed to generalize this finding.

The complexity of managing a disaster increases with the number of jurisdictions involved. Disaster management and recovery are rarely confined to a single jurisdiction, land owner, or stakeholder group. More commonly, disaster scenarios involve multiple jurisdictions and levels of governance. In multi-jurisdiction disaster scenarios, it is important to quantify and accurately map the jurisdictions involved. By framing the problem with geospatial boundaries, the authors examined whether there would be a significant increase in management complexity to cross jurisdictions. Previous research has shown that it is easier to coordinate with other jurisdictions of the same governance level than it is across levels. Currently, there is no consistent reporting structure to count and analyze jurisdictional complexity on wildfire events. The authors aimed to define jurisdictional complexity in the context of wildfires by providing a method to systematically "lump" or "split" jurisdictions and the impact this has on wildfire management. Their current research examines wildfires occurring in the United States from 1999-2018 that were managed by either type one or type two incident management teams. They used wildfires to explore the use of higher order grouping of jurisdictions impacted (acres burned) compared to those that were threatened. They used fire perimeters, land management and ownership data to quantify jurisdictional complexity for these incidents. Overall, it is crucial for practitioners to understand how and why these boundaries are constructed and the implications it has for managing these incidents.

This study evaluates the association between cumulative natural disaster exposure and psychological distress among Gulf Coast children and adolescents. 

The authors of this study used data from Katrina@10, three cohorts affected by Hurricane Katrina. The exposure of interest was the number of additional natural disasters the respondent had experienced. Child psychological distress was based on their caregiver endorsing one of the following in the past month: feeling sad or depressed; having problems sleeping; feeling nervous or afraid; or having problems getting along with other children. They conducted bivariate analyses followed by logistic regression, evaluating moderation by race; annual household income; parental educational attainment; household financial constraints; stable housing; family functioning; parental coping; and parental mental component score.

Among 648 respondents with children, about 17% reported their children having psychological distress in the past month. Each additional disaster experienced by a respondent was associated with a 43% increase in the odds that his/her child had psychological distress (OR 1.43; 95% CI 1.14, 1.80). Those whose caregivers reported coping somewhat or not well at all with parenting responsibilities were over twice as likely to have psychological distress (OR 2.44; 1.46, 4.08) than those whose caregivers were coping very well. Children and adolescents whose caregivers' mental component scores were below or well below the threshold were also nearly three times as likely to have psychological distress (OR 2.87; 95% CI 1.70, 4.85).  

These results demonstrate the persistent effects of exposure to natural disasters on child/adolescent psychological distress.

Slow-onset climate disasters, including drought and temperature rise, acutely impact Australians. Yet, they are often intangible and fail to elicit a political response. HeatWatch aims to bridge this gap by translating geo-spatial data of temperature measurements and projections into localized reports and community consultations.

How does temperature rise impact Australian towns and cities? The initiative demonstrates the dramatic projected increase in days where it is uncomfortable or dangerous to operate outside (over 35 degrees celsius) and how lowering emissions can curb these trends. In addition, the analyses of historical weather station data show that summers in the most populous areas of Australia are a month longer than in the 1950s and 1960s.

In 2021, the project focuses on electorate-level analyses and consultations with political representatives of the most impacted electorates. By communicating climate data and pursuing targeted advocacy, HeatWatch spurs political discussions to proactively address and prepare for worsening temperature rise in Australia.

More than two years have passed since Hurricane Maria made landfall in Puerto Rico, yet its impact remains largely unknown. This paper tries to understand household-level evacuation experiences during Hurricane Maria in Puerto Rico and investigates how their experiences and perceptions have influenced future evacuation decisions and planning. In particular, the authors of this study collected survey responses to understand public perceptions and preferences for future hurricane evacuation during a surging pandemic. They expect households to perceive COVID-19 as a new main obstacle and thus change their evacuation behaviors, including their decisions on the evacuation destination and the mode of transportation. Other socio-demographic information was also collected to understand the differences in risk perception among different groups. These results hope to provide useful information on refining preparedness and response for coastal hurricane states in the face of both hurricane and pandemic disaster threats.

The Katrina Diaspora is synonymous with the devastation, destruction, and displacement of the costliest storm to ever hit New Orleans-Hurricane Katrina (2005). The prevailing narrative has been that the Katrina Diaspora was durable and permanent, and initial migration data supported this perspective: return migration rates to New Orleans were initially high and peaked at around four months after Katrina. After four months, these rates declined dramatically, and the likelihood of return among those displaced nine months or longer was very low. Within 14 months after Katrina, only 50% of displaced residents had returned to New Orleans. More than 15 years after Katrina struck New Orleans, it is time to revisit the Katrina Diaspora narrative. 

This study reexamined the Katrina Diaspora by merging qualitative and quantitative data from the Displaced New Orleans Residents Survey and the Displaced New Orleans Residents Qualitative Study. In doing so, the authors of this study updated the findings from earlier studies and found that nearly 80% of respondents had returned to New Orleans. They used survival analysis to determine the rate and timing of return for participants and assess differences between key social characteristics like race, homeownership status, housing damage, and employment status. They then used qualitative data to uncover the drivers of return migration and factors that influenced permanent relocation. Their analyses provide the basis for the development of a conceptual framework that addresses the socio-ecological factors that shape delayed return migration and permanent relocation for displaced residents following a disaster. 

Disasters can cause considerable human and economic losses, which often impose a non-negligible financial burden on governments. This paper empirically examined the dynamic impacts of extreme weather events, including flooding, hurricanes, and severe storms, on county government finance. Specifically, the authors of this study constructed a panel dataset of about 800 counties across the United States for the period 1980-2015 and combined historical data on the presidential disaster declarations (PDD) and annual county-level fiscal data on tax revenues, government expenditures, intergovernmental transfers, and long-term debt issuance. They employed a novel event study approach that allows for heterogeneous treatment effects and multiple treatments to examine the dynamic fiscal impacts of extreme weather events by type. Results indicate that a flood or hurricane PDD causes a decline in the local per capita tax and leads to significant increases in total county government spending and intergovernmental revenues from the federal and state governments. A severe storm PDD seems to have little impact on a county's government spending, but causes an immediate increase in its intergovernmental revenues and an increase in local tax revenues eight years after the storm. Among the three types of natural hazards, hurricanes result in the largest fiscal repercussions on revenues and spending, and a significant increase in long-term debt issued. This study contributes to the literature on the economic impacts of natural disasters by examining the disaster-induced dynamics in local government fiscal conditions and taking a multi-hazard approach to allow for comparison among different types of natural hazards.

Since the 2015 Nepal earthquakes, The Asia Foundation has used five rounds of surveys to track recovery. The research, done in partnership with two local Nepali organizations (Interdisciplinary Analysts and Democracy Resource Center Nepal), used quantitative and qualitative methods and a longitudinal design to follow a sample of households over time. In the latest round, additional questions were incorporated through consultation with professionals in Nepal on urban recovery, vulnerable populations, coping strategies, and economic impacts. Questions included housing condition and damage; reconstruction; assessment of housing reconstruction grants; retrofitting; coping strategies; livelihoods; access to aid and other services; security and well-being; and future outlook in earthquake affected areas.

This study compiled data from 5,857 households, many of whom responded to as many as four rounds of surveys over nearly five years (n=3,554). This data allowed for an assessment of how long recovery took, how effective government reconstruction programs were, and what was the cost to these households. A qualitative assessment accompanied each round of quantitative data collection to better understand the roll out of government services and where people seemed to be falling through the cracks. This presentation will highlight findings from the survey design and major themes from each stage of recovery. We conclude with quantitative analysis jointly looking at what factors contribute to the stratification of household reconstruction over all five rounds of surveys and a predictive model that estimates the likelihood to reconstruct using geospatial covariates. This research offers substantial data and unique approaches to measuring long-term recovery.

This study focuses on self-reported household recovery outcomes following the destructive 2018 Hurricane Michael, which struck the Florida Panhandle as a Category 5 storm causing widespread devastation. A phone survey of 451 participants was conducted approximately five months after the hurricane made landfall. The study was administered in the 15 counties designated by the Federal Emergency Management Agency as eligible for both individual and public assistance. While 51.6% of the respondents indicated that they mostly recovered, over 45% reported that they did not fully recover. The majority of the participants reported suffering damages to their homes and experiencing prolonged loss of electric power, cell and internet service. Difficulties getting fuel, food, and access to healthcare were also reported. Respondents were recontacted with a follow-up survey 1.5 years after hurricane landfall. The panel data consisted of 156 respondents who were matched across the two phone surveys. The majority of the participants (64%) in the matched sample reported they recovered while 34% did not recover. The results from a fitted random-intercept logistic regression model indicated that severity of the damage, extended power outages, and difficulties getting access to healthcare had highly significant inverse associations with reported positive recovery outcomes. Homeowners’ insurance had a highly significant positive association with recovery. Concerns over shortages of food supplies and fuel showed inverse associations with recovery. Compared to those who received disaster assistance, the odds of recovery for those whose applications were denied or still pending were significantly lower.

The United States is frequently impacted by natural hazards, including wildfires, floods, tornados, tsunami, and earthquakes. Among these hazards, floods stand as the number one natural hazard in loss of lives and economic impacts. In particular, flood events are often the result of a combination of flood mechanisms, such as storm surge combined with intense rainfall or riverine flooding. Multi-mechanism floods (MMFs) can cause more severe impacts on societies and the built environment compared to single-mechanism floods. Conventional probabilistic flood hazard assessment (PFHA) usually focuses on a single mechanism, while a realistic and comprehensive estimation of flood hazards requires the consideration of more than one flood mechanism. Extending PFHA for MMFs will introduce new challenges from the need to model the dependence structure between variables and to run computationally expensive models to represent the interaction between flood mechanisms. This study used a Bayesian-motivated approach to assess flood hazards from the combination of hurricane-induced storm surge, precipitation-induced runoff, and river flow. To demonstrate the proposed approach, a Bayesian network (BN) was developed to analyze flood hazards for a coastal area located on the Delaware River at Trenton, New Jersey. The target random variable in this study was river discharge, which was affected by storm-induced surges and precipitation. To generate conditional probability tables needed for the BN, a series of predictive models were developed using numerical, statistical (surrogate), and analytical models. Preliminary results are presented to demonstrate the development of a hazard curve to capture contributions from multiple mechanisms.   

Natural disasters, hurricanes in particular, often cause damages to critical infrastructures and public utility systems as well as costs to people. Regardless of evacuation decisions and status, hurricane-induced recovery is challenging for most people. In order to recover in the aftermath of a major hurricane, people seek assistance from different agencies and social organizations. However, the empirical literature suffers from a lack of understanding of the relative effectiveness of such aids on recovery. This study utilizes survey data from 780 households from several locations affected by Hurricane Irma to identify how different forms of aids contributed to recovery including local, state, and federal government support as well as assistance received from neighbors and friends. Results indicate that the federal government provides more support for evacuation and rescue, power supply, healthcare-related aids compared to others. Neighbors are found to be more helpful for healthcare-related aids compared to friends (21.5% vs. 11.5%). The result also shows that people spend more on power consumption compared to others when faced with a utility disruption. Such disruptions include, on average, two to four days without power, water, waste management, phone, public transportation, educational institute, financial institute, hospital, pharmacy, medical test center, grocery interruption, among others. As part of the Irma recovery process, people living in the apartments showed more tendency to change their household type compared to others. The insights from this research will help disaster response personnel to develop more efficient recovery strategies in the aftermath of a major hurricane and enhance community resilience.

The growth of land development projects in hazard-prone areas and the increase of extreme events have intensified disaster-induced losses. As a planning tool, predictive models can help devise data-driven policies that can better recognize pre-disaster mitigation needs and post-disaster recovery priorities by simulating possible outcomes of different plans. Researchers have introduced several recovery models to address this need. However, there are still gaps in understanding the individual and collective decisions and their spatiotemporal effects on post-disaster recovery. This research proposed a spatial agent-based model to longitudinally simulate and predict homeowners' recovery decisions through incorporating recovery drivers that could capture interactions of individual, communal, and organizational decisions. The model was designed with an emphasis on using publicly available data to enhance its general applicability through optimizing monetary and time costs associated with data collection. Recovery of Staten Island, New York, after Hurricane Sandy was selected as the case study for the model presentation. The required data were obtained from various sources using different techniques, such as mining tax assessment data to estimate damage and restoration of properties, iterative proportional fitting of Census data to approximate household-level attributes, collecting macro-level data on recovery aids and distributing them among impacted households, and quantifying qualitative reports. The analysis outputs confirmed that a combination of internal, interactive, and external recovery drivers affected households' decisions and shaped the recovery progress.

Research suggests that rates of mental health disorders, such as anxiety, depression, and post-traumatic stress disorder (PTSD), rise after a disaster but return to normal within a year or two. Studies on Hurricane Katrina found mental health disorders remained elevated instead of returning to pre-storm levels. However, these studies did not offer a longitudinal perspective of mental and emotional health in the long-term recovery of New Orleans after Hurricane Katrina and the federal levee failure. This longitudinal study uses qualitative interviews, and a content analysis of local and national news articles to explore the impact of recovery policy on mental health as framed by African Americans displaced from New Orleans. The findings show both residents and the media framed post-Katrina policy change as compounding resident vulnerability by restricting access to resources. These policy decisions exacerbated the negative mental and emotional outcomes, with many participants continuing to report struggling with depression, anxiety, and PTSD nearly 15 years after Hurricane Katrina. Participants framed their mental health as caught in the intersection of policy changes that reshaped housing, education, and employment opportunities in post-Katrina New Orleans. Meeting their needs amidst the changing landscape increased their stress, which deteriorated their mental health. In order to improve health outcomes following future disasters, recovery policies should: keep families and social networks intact, offer mental healthcare services for several years after a disaster in locations where people typically gather and provide home healthcare assistance for those that experience extreme negative consequences.

In 2005, Hurricane Katrina and subsequent policy changes reshaped the environmental, built, and social landscape of the New Orleans metropolitan area. As recovery progressed, competing narratives and disparities in recovery emerged that demonstrated the perseverance of structural inequalities in light of federal, state, and local policy measures implemented to address long-standing urban social issues, such as concentrated poverty and underperforming schools. In the sixteen years since the federal levee failure, the area has faced repeated environmental threats, including the BP Oil Spill, multiple hurricanes, frequent localized flooding events, and a catastrophic public health disaster—the COVID-19 pandemic. Disaster exposure can contribute to the onset of significant depression and post-traumatic stress disorder (PTSD), even for people with no previous history of mental illness. To understand the mental health impacts of cumulative disasters, the research team conducted a mixed-methods case study with frontline essential workers, focusing primarily on teachers and healthcare workers. The project sought to understand the extent to which essential workers were impacted by the COVID-19 pandemic personally and professionally, as well as the extent to which they exhibited signs and symptoms of anxiety, depression, and PTSD. The study found respondents reported elevated levels of each of these common mental health disorders, with nearly a quarter of all respondents scoring moderate or severe on standardized screening assessments. Using findings from the screener, the research team collaborated with local organizations to devise strategies and tailored interventions for supporting essential workers' mental and emotional wellness.  

Disasters are often viewed as natural, resulting in hazard-focused ways of dealing with them that involve experts and expert knowledge. Meanwhile, Indigenous peoples have been subject to various representations, such as being custodians of the environment and being especially vulnerable to environmental change. Recently, there have been calls for greater inclusion of Indigenous peoples in disaster risk reduction (DRR), alongside decentralization and localization of humanitarian aid. This signifies a shift in the ways that DRR has traditionally operated and is viewed by many to be a welcome change from authoritarian, top-down DRR. In theory, this also suggests a shift in power structures, where the role of experts is contested. As discourse is an important means of communicating meaning, the authors of this study delve into these contests by conducting a critical discourse analysis of the elite news media focusing on Indigenous peoples in disasters and humanitarianism in 2015-2020. The results show that disasters are principally viewed as natural phenomena, with discourses of humanitarianism and technocracy (including settler humanitarianism) depoliticizing disasters and rendering Indigenous peoples helpless. However, weaker discourses recognize disasters as political. These focus on systems of oppression and self-determination in DRR and often overtly discuss the role of colonialism in disaster creation. Although they find that shifts towards decentralization and localization in DRR have not led to fundamental shifts in power, they highlight many ways in which the news media can contest dominant discourses and thus be more politically engaged. 

Eastern North Carolina (ENC), a predominantly rural region, often experiences the intersection of fluvial, pluvial, and tidal flooding, which leads to complex and impactful outcomes, such as morbidity, mortality, economic disruption, and loss of livelihood. Managing these connected hazards is challenging, especially as climate change drivers such as sea-level rise and wetter storms will likely lead to a greater incidence of compound coastal water events (CCWE). Funded by a 2019 National Oceanic and Atmospheric Administration Coastal and Ocean Climate Applications/Sectoral Applications Research Program grant, a key objective of this research is to assess the perceived risks and needs of the hazard management and planning community in ENC. This paper is focused on the economic impacts of CCWEs in ENC. The project focused on rural counties in eastern North Carolina located along the coast and those adjacent to it that share estuarine environments or linked riverine systems. Data was obtained through focus group interviews conducted with 41 planners and emergency managers during a flood workshop held at the East Carolina University campus on February 26, 2020. Initial findings showed the greatest economic impacts, identified by the hazard management community, are to the local physical infrastructure and to individual households (i.e., damage to homes and businesses), which in turn has a compounding impact on the larger community and their quality of life. At the same time, communities are unable to systematically track and do not have access to quantifiable data on economic impacts, particularly the impact of long-term economic disruption. 

While extensive research exists on pandemic mitigation, preparedness, and planning, research on lived experiences, particularly of diasporic communities during one, is scarce. This paper aimed to explore the role of social capital as a coping strategy among Chinese diasporic communities living in Kenya. A ubiquitous reality people of Asian origin or features live with during the COVID-19 pandemic is outbreak-related stigma. Globally, targeted communities have had to develop coping mechanisms as a survival and resilience tool. Social capital is a requisite for resilience building in disasters and is achieved through creating, enhancing, and leveraging community capacity. This research design was based on a qualitative case study using semi-structured interviews of Chinese diaspora community members in Nairobi. The findings suggest that connections between social capital and coping strategies are by no means limited to local networks. There was evidence of the instrumental role of social cohesion as a support system through the mobilization of resources among Chinese communities; however, social capital was not a static or neutral process. There was a high degree of complexity among other social capital components, such as trust in institutions, civic participation, and resource help-seeking from emergency management and public health professionals, such as bonding, bridging, and linking, that impact coordination and integration with local government agency response. Thus, a framework that includes a combination of social, cognitive, and structural forms of social capital is presented, bringing new light to the study of social vulnerability and help-seeking among diaspora communities in a pandemic. 

There is limited knowledge on the relationship between neighborhood effects and mental health among displaced disaster survivors, particularly in understanding how this relationship impacts women. Hurricane Katrina was the largest internal displacement in the United States, which presented itself as a natural experiment. We examined the association between changes in neighborhood socioeconomic opportunity and self-rated mental health among women survivors up to 10 years following Katrina (N = 394). We also investigated whether this association was modified by move status, comparing women who were permanently displaced to those who had returned to their pre-Katrina neighborhood. We used cross-classified multilevel linear regression models with repeated measures of mental health as the outcome nested within individuals within neighborhoods. We analyzed U.S. Census block group-level data from the American Community Survey (2005-2009, 2006-2010, 2012-2016) and individual-level data from the Gulf Coast Child and Family Health study (Waves 1-5, 2005-2015). The exposure, neighborhood socioeconomic opportunity, was created as a single-factor index which represented 10 social and economic characteristics of participants’ neighborhood environments. Individual mental health was measured using mental component summary (MCS) scores. Increased neighborhood socioeconomic opportunity was statistically positively associated with MCS score (improved mental health) among women after controlling for age, race/ethnicity, economic positioning, months since Katrina, and move status (19.6, 95% Confidence Interval: 5.2, 34.3). Neighborhood socioeconomic opportunity and mental health were also modified by move status. These findings underscore the need to better understand the impacts of socioeconomic gains and health outcomes among women affected by natural disasters.  

Both the natural landscapes and human communities are important when understanding the outcomes of a disaster. To understand these interactions in the context of Hurricane Michael's Impacts on the Florida Panhandle, this study integrated household surveys with spatial modeling of land-use to study relationships among ecosystems and residents; recovery. The study's authors surveyed residents between December 2019 and February 2020, about 14-16 months after the storm made landfall. This survey allowed residents to self-assess their damage state and recovery from the storm, including physical home recovery and psychological recovery one year after the storm. Using ArcGIS, they mapped responses from the survey and linked them to landscape and exposure characteristics. They used Ordinal Logistic Regressions (OLR) to better understand the complex influences on household-level recovery following a major storm. Winds were the primary predictor for high home damages (Odds Ratio = 1.033, p-value = 0.004), while the surrounding ratio between Built and Greenspace were key predictors for both property (OR = 1.012, p-value = 0.058) and psychological (OR = 1.020, p-value = 0.010) recovery. Specifically, showing that having more green space directly surrounding a home can increase residents’ levels of both physical home recovery and individual mental recovery. This study highlights the critical importance of understanding both social and ecological landscapes for assessing storm impacts and developing recovery and restoration plans.  

This study discusses the underlying issue of the long-term disaster recovery on housing which is deeply related with the substratum concept upon the neo-liberalism or welfare state through a case study in Kesennuma City as an impacted municipality of the Great East Japan Earthquake and Tsunami in 2011. At most 470,000 evacuees were displaced due to the event and approximately 30,000 newly constructed public housing units have contributed to provide permanent dwelling for the disaster survivors, while the public housing stocks significantly increased. The issue is how the large volumes of the public stocks would be sustainably managed in the shrinking society of Japan. This preliminary study applies a Mixed Method Research in Kesennuma City to identify qualitative perspectives such as place attachment, life history for the residents, and using quantitative data for the volumes of public housing from pre- to post-disaster phases. A primary result with the quantitative perspective revealed that the four times units of public housing were newly constructed and most of the pre-disaster public houses were degraded due to the age of the buildings. The emergence of the new stocks and degradation of the old stocks might cause another uprooting migration as well as financial challenges for the municipality due to the potential integrating process. This study also examined the pathways in the different types of public housing such as detached, apartment, and mixed land use with private housing.

Disaster recovery is most often expressed in terms of reconstruction of the built environment, repatriation of displaced populations, or resumption of economic activity. From a population health perspective, the most common metric of long-term recovery involves dimensions of mental health. Rarely do assessments of individual recovery consider all of these facets. The authors previously developed a multi-dimensional measure of individual recovery that included economic and housing stability, mental and physical well-being, and social role adaptation. Using data from the longitudinal Gulf Coast Child and Family Health Study, an observational cohort study of 1,079 randomly sampled survivors from Louisiana and Mississippi who were followed up over five waves of face-to-face interviews across thirteen years, a latent growth curve analysis was conducted comparing two models of individual recovery. The first involved a subjective assessment of disaster recovery, in which respondents reported whether they had returned to a status similar or better than before the hurricane, or whether they were worse off. The second involved an objective assessment of economic and housing resources as well as standardized assessments of physical and mental health. The resulting recovery trajectories reflected two distinct expressions of resilience, recovery, and chronic dysfunction, and suggest that subjective and objective appraisals of recovery may yield quite distinct interpretations of disaster recovery.

Disaster researchers must give thoughtful attention to research methodology. Pre-disaster data is uncommon, and getting into the field in a timely manner can also present challenges. In addition, the safety of researchers and participants is of utmost concern. These considerations became even more strained during the COVID-19 pandemic. While primary data collection about the pandemic provides valuable insight into disaster dynamics, travel restrictions, budgetary concerns, and ethical considerations place limitations on what research can be conducted. However, secondary data analysis provides an opportunity to examine patterns across both space and time. The COVID-19 in the Deep South working group, supported by the National Science Foundation-funded Social Science Extreme Events Research Network and CONVERGE, was assembled to conduct interdisciplinary research on how the Deep South is uniquely vulnerable to the pandemic and its residual impacts. Working group members are trained in sociology, demography, epidemiology, geography, and ecology and include junior and senior scholars. This presentation will focus on secondary data analysis and index creation methods employed by the working group, paying special attention to an effort to assess ecological, social, health, and economic vulnerabilities of counties to the COVID-19 pandemic. Results from this study suggest that the southeast United States exhibits relatively high vulnerability scores across the four domains. These indices can be used as contextual data in a multilevel framework, and they can be updated and tested with new data as they are released. The utility and implications of using publicly available secondary data will be discussed. 

Small businesses have experienced significant revenue decreases, employee and customer losses, supply chain disruptions, and government-mandated restrictions and closures due to impacts from the COVID-19 pandemic. In addition to the pandemic, natural hazards, including wildfires, hurricanes, and floods, have further disrupted small businesses and communities across the country. As small businesses provide two-thirds of net new jobs and almost 50% of the United States workforce, the growing impact of COVID-19 and hazards amplifies the need for small business resilience building. 

This study of small businesses and community assistance during the COVID-19 pandemic sheds light on existing, underutilized mechanisms to support community resilience and provides insights into factors that increase horizontal support networks. This research explored the role of small businesses in community resilience, focusing on three central questions: 1) How are small businesses connected to vertical and horizontal hazard support networks? 2) What are the factors impacting small businesses' involvement in their communities during disaster impacts and short-term recovery phases? 3) How might the COVID-19 lessons inform small business owners' community support networks during future natural hazards? The authors improve upon previous research methodologies by studying small business support networks at both local and national scales. 

The results include insights into support networks of minority- and women-owned businesses, lessons learned about the importance of federal financial assistance, and policy implications to enhance the resilience of communities.

In response to increasing threats from natural disasters, social scientists and disaster managers have conceptualized social capital as a fundamental building block for community resilience. However, this line of research often overlooks the complex ways in which different forms of social capital (bonding, bridging, and linking) can operate during and after a disaster. Although bonding social ties to family, friends, and close neighbors can help households weather a natural disaster, these same connections can also extend a household's indirect exposure. Utilizing two restricted access datasets gathered in Houston, Texas during the aftermath of Hurricane Harvey, this study investigates the frequency with which households are exposed to the impacts of disaster not only directly but also indirectly, through their bonding social networks, and how that extended exposure can in turn influence trust in local, state, and federal government -- key sources of linking capital during disaster recovery. Results show that households experience indirect impacts pervasively through their close social ties and that these indirect impacts correlate significantly with lower trust in government at all levels, net of direct impacts and other statistical controls. Implications for a more nuanced approach to social capital in disaster research and planning are discussed.

This study sought to understand the experiences of those in the Latinx community in the Oklahoma City metropolitan area and Guymon, Oklahoma, during the COVID-19 pandemic through the lenses of labor and gender. Extreme events exacerbate issues that low-wage earners face even in the best circumstances, such as labor violations, wage theft, unsafe working conditions, and vulnerability due to citizenship status. Additionally, labor expectations are often gendered. Not just in Latinx communities, but especially so, labor, both in and outside of the home, is divided, with women bearing the brunt of care work and household duties, and men working physically demanding, often dangerous jobs. Vulnerability has been more evident during the COVID-19 pandemic. Essential workers-those in service industries, education, healthcare, and construction-are working in conditions where exposure to the virus is more likely, often with little to no insurance coverage. Moreover, the lack of formal childcare/schooling has made providing for the needs of children more difficult, especially when a parent cannot take time off work or relies on the school to provide meals. 

Qualitative interviews were conducted with approximately 20 participants over work and home life. Themes such as cultural/language considerations, gendered differences, access to mental health care, attitudes about the workplace, attitudes about government assistance, and necessity of work will be discussed. Additionally, a few notes on research ethics will be presented. Stories from our communities most at-risk can help to inform policy decisions for local, state, and federal governments, as well as non-profit and faith-based organizations.

Texas is one of 13 states in the United States that does not provide system wide temperature regulation (air conditioning and heating) in their state prison systems. Without temperature regulation, prison agencies develop policies to mitigate the impact of extreme heat on individual incarcerated persons. Policies typically include providing additional resources like water, ice, cold showers, cooled respite areas, commissary items for purchase, and water breaks for workers. Yet each year, the Texas Department of Criminal Justice reports that a number of incarcerated persons have fallen ill or even died from complications from extreme heat in Texas prisons. This study included an analysis of 315 surveys from incarcerated persons in Texas about their experience with heat mitigation policies as well as the impact of extreme heat on their lives and personal health between 2018 and 2020. Findings demonstrate how the structure of prisons and characteristics of imprisonment present significant challenges to incarcerated persons accessing resources to mitigate the impacts of heat. Additionally, surveys collected in 2020 reveal how the COVID19 pandemic, which disproportionately impacted incarcerated populations, created additional challenges to individually focused heat mitigation strategies in Texas prisons. These findings suggest that system-wide hazard mitigation efforts would be more effective and should be taken to decrease the vulnerability of incarcerated persons to the impacts of extreme heat, including death. The need for temperature regulation in prisons is expected to grow due to an increase in extreme temperatures with climate change and growing public health threats such as pandemics.

In California, over one million people lack access to clean, affordable drinking water, a crisis made worse by the region’s hydrology which leads to frequent droughts. As rising water bills threaten to make water inaccessible to those who can not afford it, a disproportionate burden is placed on low-income populations who already spend a higher percentage of their income on water and, in many cases, have already cut back water use to the bare minimums. As climate change further stresses water supply infrastructure, utilities are tasked with creating a drought-resilient water supply that ensures reliable access to affordable water for all communities. 

In this ongoing work, the authors evaluated how decisions made by utilities in California to prepare for and mitigate droughts impact the accessibility and affordability of drinking water in the state. They developed hydrological systems models for water providers across the state and model household water use during drought and non-drought conditions. Using these models, they calculated how the accessibility and affordability of water change as a result of different utility decisions and calculated the disparities in basic access to water across advantaged and disadvantaged groups to quantify the distributional equity of drought resilience interventions. 

This study describes the residential and health trajectories of a vulnerable sample of Hurricane Katrina survivors over a 15-year period (n=515, 2003-2018), as well as relationships between neighborhood characteristics and health over time. Some existing research examines residential and health trajectories after disasters, but few follow survivors beyond the impacted location and for an extended period of time. This article begins to fill that gap using data from the Resilience in Survivors of Katrina Project, a panel study of low-income mothers that includes four waves of data, one before Hurricane Katrina (1-year pre-Katrina) and three after (1-, 5-, and 12-years post-Katrina). Before Hurricane Katrina, all respondents lived in Louisiana, in 11 counties, with 90% of respondents in Orleans and Jefferson parishes. Respondents were dispersed to 19 states and 65 counties 1-y after the storm, and twelve years later, they remained spread across 20 states and 68 counties, reflecting the long-term nature of displacement and relocation. Using various longitudinal modeling strategies, the authors of this study examined how neighborhood poverty, racial composition, and segregation change across time for vulnerable survivors and how these residential characteristics correlate with physical and mental health outcomes (e.g., self-rated health and psychological distress). In particular, they compare relationships separately for those who returned to New Orleans and those who did not return to understand health and neighborhood trade-offs.

Facilitating the implementation of housing resilience strategies requires careful planning that engages multi-sector stakeholders, including multi-level governments, private industries, non-government organizations (NGOs), universities, and community leaders. Engaging multi-sector stakeholders is the key to establish mutual resilience goals that guide the plans and foster coordination, incorporate diverse ideas and extensive knowledge, and potentially lead to smoother and more efficient implementation. Despite the importance of inclusive collaboration in facilitating housing resilience, there is limited understanding of how multi-sector stakeholders collaborate in housing resilience planning. To address this knowledge gap, the authors of this study used social network analysis (SNA) to analyze how multi-sector stakeholders collaborate or contribute to the development of plans that are related to housing resilience. A two-mode social network model was built based on the data collected from 39 planning documents on housing resilience in three regions, including City of Miami, City of Miami Beach, and Miami-Dade County. Social network analysis measures, such as degree centrality, closeness centrality, betweenness centrality, and eigenvector centrality, were used to offer a quantitative understanding on the level of collaboration and involvement among the different sectors of stakeholders. The results of the study show that there are significant differences in the network measures across different sectors of stakeholders. The findings from this study can advance the knowledge on stakeholder collaboration on housing resilience planning. They could offer insights on how to facilitate more effective and collaborative housing resilience planning.

Ethnic and racial minority groups are among the populations at risk of suffering most the consequences of disasters and tend to be far less prepared. In this regard, emergency managers and other stakeholders have failed to understand these minority groups' relevant cultural processes. The relevance of this understanding is based on culture as an element that shapes thinking processes about hazards and disasters. Low levels of preparedness are generalized among groups; however, Latin American immigrants face significant barriers to preparedness. This qualitative study aimed to explore and understand the process of disaster preparedness among Latinx immigrants of recent arrival at Omaha, Nebraska. Through a series of in-depth interviews, this study conveys participants' perspectives to shed light on the disaster preparedness process once arrived. Findings suggest that Latin American immigrants of recent arrival in Omaha, like the general population, have low levels of preparedness. Nevertheless, countries of origin influenced how they currently prepare, either due to previous experiences with natural hazards or through emergency response public education or training. The interviewees acknowledged natural hazards related to extreme cold weather and tornadoes in Omaha but admitted that experience and education to face such threats was lacking. Preparedness information is obtained from family or friends that have more time living in Omaha, not from other sources. Latin American immigrants of recent arrival tend to trust public authorities' effectiveness, mainly first responders; however, interviewees acknowledged that immigration status could represent a barrier to preparedness and response to potential disasters.

In order to create more effective weather messages, it is imperative to understand how people perceive, and respond to, weather alerts. This study examined how perceptions of a tornado alert text message, and communicative concerns, shape shelter-seeking intentions. An online survey was conducted comprising 189 participants. The survey included items concerning secondary communication (motivation to share information with others), health belief model variables (perceived vulnerability, barriers, and benefits to action), message perceptions (message cues, message directness), past experience, and shelter-seeking intentions. In regression analyses, the health belief model variables, along with past experience, accounted for 36.5% of the variance in shelter-seeking intentions, with past experience, severity and susceptibility being significant predictors. Additionally, severity, susceptibility, and benefits predicted messages-related variables. In mediation analyses, message cues mediated the relationships of HBM variables on shelter-seeking intentions, and message directness mediated susceptibility and benefits on intentions. Regarding secondary communication, messages cues mediated the relationship between HBM variables and seeking out others upon receiving the alert. The discussion highlights implications for education and information campaigns.

Housing is a fundamental component of the physical environment of human communities, and in places dominated by rental housing of any kind, post-disaster stability requires that landlords remain solvent. Situated within the context of the COVID-19 pandemic, this study examines the decisions that landlords made during the pandemic and considers how those decisions shape opportunities for rental housing stability during recovery. Unlike disasters where the hazard event physically damages the housing stock, the COVID-19 disaster left housing units untouched while limiting tenants' ability to pay rent. As the pandemic progressed, many landlords had to manage revenue shortfalls while trying to fulfill a range of financial obligations such as mortgage payments, insurance, taxes, and building maintenance. Even landlords sympathetic to the plight of their tenants found themselves caught between a regulatory requirement to keep tenants in place and their need to ensure their own long-term financial stability. Using qualitative data from 80 interviews in four cities with landlords who self-manage their residential rental properties, this study asks the question, how are landlords responding to the tension between rental restrictions and their financial obligations in the context of rental moratoria? Initial analysis identifies key outcomes, most notably a desire to leave the market altogether. The findings point to the necessity of post-disaster recovery policies that look beyond tenant displacement and develop a more nuanced understanding of how disasters affect urban rental housing systems.

In the last five years, people living in Puerto Rico have been experiencing consecutive extreme events, including hurricanes, landslides, an earthquake sequence, and the COVID-19 pandemic. The outcome of these events has been catastrophic, adding layers of complexity which would confound any social system not designed for multiple concurrent disasters. Understanding how different stakeholders perceive risks, what their informational needs are during crises, and how they make protective action decisions are questions of the utmost importance for disaster risk reduction. Prior research on earthquake aftershock communication suggests that people have a wide range of information needs, evolving over time and varying by location, that could be largely fulfilled before a major event. This presentation shares ongoing research with three communities in Puerto Rico to understand how risks are perceived, how information needs have and may evolve, and how risk communication products may be co-designed considering contextual cues and prior experiences with disasters to inform protective action decision-making. Further, it offers recommendations that governmental and non-governmental organizations could adopt to meet the information needs of different stakeholders. 

This study focused on understanding individual decision-making under uncertainty during the COVID-19 pandemic. The researchers explored these issues by employing the theoretical lens of subjective resilience, intersectional vulnerabilities, adaptive, and coping capacities to understand individual cognition of risks and perceptions of recovery from the pandemic for those living in hurricane-prone regions. They examined several factors affecting subjective perceptions of resilience and recovery such as capacities to adapt and cope, risk perceptions, communication sources, intersectional and cumulative physical and social vulnerabilities, the ability to prepare for hurricanes among the pandemic, perceptions of government policies, social capital, and partisanship. To analyze these factors and to understand changing perceptions, they used individual-level data collected from two cross-sectional surveys administered in August (1,392 respondents) and November 2020 (1,911 respondents) using both random (via telephone landlines) and nonrandom (via the Internet) sampling in Florida. Using logit regression modeling, their preliminary analysis found that expectations of the length of recovery were positively associated with several factors, including adaptive capacities; risk perceptions about the pandemic; higher levels of stress; lack of preparation for a potential hurricane during the pandemic; the perceived effectiveness of pandemic management policies; less social connectedness; and political party affiliation. The results thus far also indicate differential recovery trajectories, particularly when intersectional vulnerabilities are analyzed. The findings contribute to better the understanding of resilience, vulnerabilities, and factors affecting recovery in areas susceptible to multiple hazards during a pandemic. The results help inform policies and practices to improve pandemic preparedness and management.   

The growing impact of climate-driven coastal disasters necessitates a more comprehensive understanding of how coastal populations engage with adaptation decisions. While many studies explore factors that may drive residential adaptation in a general sense, few evaluate how residents decide between pursuing in situ, protective adaptation vs. out-migration or retreat from at-risk areas, and fewer still evaluate how residents' intent changes over time.  

To address this, they distributed two surveys to residents of North Carolina’s Albemarle-Pamlico Peninsula (APP), a low-lying area at the frontier of sea-level rise. The first wave, distributed in 2017 (n = 227), asked about respondents' property, knowledge, beliefs, opinions, experiences with flooding, storms, saltwater intrusion, and aspects of their community engagement. The second wave, distributed in 2020 (n = 87), repeated much of the original survey with added questions about residents' experiences with Hurricanes Florence (2018) and Dorian (2019), which impacted the APP during the intervening period.  

The researcher's initial results found that, at either point in time, there was no significant relationship between residents' willingness to protect property or structures (i.e., in situ adaptation) and their willingness to migrate (i.e., retreat), indicating that individuals may be open to consider either in situ or retreat responses, but rarely both. However, these positions shifted in the intervening years, suggesting (1) increased disaster exposure could lead to greater openness for more proactive adaptation measures; and, (2) while residents are unlikely to consider in situ protection and retreat simultaneously, the shift from preferring protection to retreat could occur progressively over the long-term. 

Research on post-disaster economics has shown that a brief period of disruption for the impacted area is expected. However, prolonged economic changes in locations with stagnant recovery periods are less understood. 

The region affected by Hurricane Michael provides a unique opportunity to study long-term employment and workforce disruptions because of its higher Social Vulnerability Index (SVI) score. This study compared the extent of these disruptions to other Florida locations with higher SVI that also recently experienced a major hurricane.

Florida counties impacted by Hurricanes Michael and Irma were examined for employment and unemployment rates, workforce employee count by industry, and wages. This data has been collected, estimated, and reported by the U.S. Bureau of Labor Statistics and Florida Bureau of Workforce Economics and Statistics Research. Trend analysis was used to compare changes from two years pre-storm to one year post-storm across locations. 

Preliminary results showed that post-Hurricane Michael counties, characterized with moderate to high SVI (>= 0.5), experienced low employment, high unemployment, and decreases in hospitality, retail, education, health care, and social services workforces, as well as an increased construction workforce. Findings showed that they experienced this period of economic disruption in employment and unemployment for a longer time compared to post-Hurricane Irma counties with low to moderate SVI (< 0.5).

These results highlight the need for increased disaster resilience against economic disruptions in communities with higher social vulnerability factors. Longitudinal studies of post-disaster employment measures contribute to the understanding of how to mitigate prolonged disruptions to an area's workforce.

The Federal Emergency Management Agency-funded buyout program in the United States is the longest-running buyout program worldwide. As climate change intensifies flood risks in many places, the demands on floodplain buyouts have increased in recent decades to expedite residents' relocation from hazardous areas to safer ground and minimize future disaster risks. Despite the potential future needs, limited research has focused on the long-term effects of neighborhood change on communities left behind after buyout implementation. 

Focusing on changes from 1990 to 2015 in 2,144 block groups in Harris County, Texas, the authors sought to examine how buyouts implemented between 2001 and 2010 have altered neighborhood characteristics compared to the normal trend of neighborhood change over time, specifically in racial composition. This study used a mixed-methods approach, including longitudinal multilevel analysis for the quantitative study and a phenomenological qualitative approach. A piecewise discontinuity model was developed in the quantitative research to compare trajectories of neighborhood change between the pre-buyout and the post-buyout phases. Questions derived from the statistical results were posed to residents who participated in the buyouts and opted out. The findings suggest shifting racial compositions in buyout neighborhoods from white to more Hispanic. Also, its impact has been greater in the wealthier buyout neighborhoods. 

Reviewing the local buyout practice over the long-term and analyzing its consequences in neighborhoods, this study sheds light on the neglected long-term impacts of floodplain buyouts on the remaining communities. It also suggests a better-designed floodplain buyout program to promote hazard mitigation and community resilience. 

Flash floods are multifaceted and complicated hazards that claim the lives of thousands globally every year. This study aims to assess whether a convergent parallel mixed-methods study design can capture the physical and human complexity of a flash flood event, focusing on a case study in Kerala, India. It is projected that both extreme climate events and urbanization will continue to influence flooding in the region. This project draws on research from socio-political geography to examine the relationship between government institutions and people's perceptions of flash floods through semi-structured interviews with government officials and analysis of policy documents and newspapers. In parallel, the authors of this study analyzed rainfall remote sensing data from the Climate Hazards Group’s Infrared Precipitation with Stations dataset between 1986 and 2020; using results from the interviews to guide the analysis and suggest avenues for further qualitative research. 

The following three questions inform this research: 1) what factors cause impactful flash floods in Kerala; 2) what are the government and public perceptions of flash flooding; and 3) what spatial-temporal rainfall statistics capture impactful flash flood events? We found that beyond simple urbanization, the cause of flash flooding, the public's perception is that drainage systems, coastal construction, and dam management practices have exacerbated flash floods in Kerala. This was reflected in the fact that the maximum 5-day summed rainfall was the most effective statistic in capturing flash floods in the public's consciousness. Together these suggest avenues to better guide flash flood prediction and policy. 

Disaster researchers within the vulnerability paradigm have recognized that a disaster's impact on a population is negotiated by complexity and unequal power relations. The recovery, response, adaptation, and vulnerabilities during disasters are, in turn, shaped by these power relations. This understanding, however, is inadequately integrated within Disaster Management practice. Depoliticization of disaster in policy-making process and practice often discount power relations existing at different levels. The top-down epistemic exclusivity and the abject obliviousness of power structures within disaster management practice often result in the reproduction of processes resulting in exclusion and marginalization. The study hypothesizes that the current disaster governance processes, deeply enmeshed within power structures, (re)create privileges and oppressions. This is done through the entrenched politics of exclusion-inclusion within society not independent of the governance process. The thematic engagement of the ongoing research is on exclusion and day-to-day disaster risk governance. Both formal and informal governance actors are the participants of the study. Using Social Construction of Target Population theory and Intersectionality frameworks, the study will investigate how the intersection of socio-economic (namely gendering, caste, and class) and governance processes shape social exclusion in the Kosi floodplain. Located in the Interpretivist research paradigm, the study will deploy focused group discussions, semi-structured interviews, and observation as methods of data generation. The study would address both the knowledge and population gap while studying exclusion in the Kosi floodplain, which in spite of being chronically suffused with embankment policy disasters since the 1950s, is still meagerly represented in the academic literature. 

The National Institute of Standards and Technology (NIST) Center of Excellence's (CoE) creation of IN-CORE incorporates hazard and engineering sciences with social sciences to provide models and predictions that can inform a community's resilience decisions and policies in various areas such as the local economy, housing, natural environment, and public health. The CoE is conducting stakeholder engagement with a sample of partner communities across the nation to pilot the use of IN-CORE for resilience planning to enhance the IN-CORE modules and interface to better serve local resilience planning needs. This community engagement approach links the work of modelers-researchers and practitioners. One of the key steps towards this community engagement involves developing an evaluation of resilience practices, policies, and funding in the partner communities and examining the most effective approaches and policies in each local community for influencing long-term resilience outcomes. After a full review of what resilience plans, policies, and practices are in place, enables informed and targeted use of various hazard, damage, impact, and recovery models in IN-CORE tailored to the resilience needs, capacities, priorities, and progress of the partner communities. This active and grounded outreach and engagement process has allowed NIST CoE to consider what direct policies can lead to a more speedy and equitable recovery. The uniqueness of this project is that it combines the modeling work of researchers at the NIST CoE with the on-the-ground experience of resilience practitioners at our partner communities to determine applicable and feasible policies and actions that support a successful recovery. 

How climate change will impact social inequalities is a critical topic of social research, especially with the predicted more frequent and more severe extreme weather events. Parallel to this work is scientific research on climate change attribution; it seeks to explicitly quantify how the occurrence of extreme weather events is linked to climatic change. Little work, however, has explicitly analyzed this intersection, namely who experiences the brunt of impacts of climate change from extreme weather events. To analyze this question, the authors carried y out a novel analysis using climate change attribution science to assess if and to what extent socio-spatial inequalities were found in flooding during Hurricane Harvey in Harris County, Texas (of which Houston is the central city) that could be explicitly tied to climate change. To do so, they synthesized data on flooding during Hurricane Harvey that was attributed to climate change with social data on individual parcels from Harris County Appraisal District parcel data and from the U.S. Census. The results show that a majority of parcels had climate change-related flooding including a subset of properties that experienced greater than one foot of additional flooding from climate change alone. These impacts were unequal: they were most acutely felt in racially diverse neighborhoods and on multi-family properties. The conclusions point to how climate change is already accentuating social inequalities.

Advancing the concept of more resilient communities requires a better understanding of the role played by those tasked with administering hazard mitigation grant programs in the United States. To address this issue and to apply the findings to the emerging Building Resilient Infrastructure and Communities program, the Federal Emergency Management Agency (FEMA) asked researchers from North Carolina State University to explore the roles that states and territories play in assisting local governments in developing and implementing Hazard Mitigation Assistance (HMA) programs. A national survey of State Hazard Mitigation Officers (SHMOs) in U.S. states and territories was conducted to help answer this question. The survey addressed the following topical areas: 1) state and territory engagement in HMA programs, 2) assistance provided by state and territory officials to local governments, 3) state and territory commitment to hazard mitigation, and 4) state and territory’s perceived local government needs, including low-capacity communities. The results highlight low to modest participation and understanding of supplemental the FEMA capacity-building funding programs (program administration by states and advance assistance), significant variation across states and territories in their ability to provide necessary technical assistance, insufficient state- and territory-level leadership (including among emergency management officials and governors), and poor strategies to identify and reach out to underserved communities. While the results are concerning, they provide direction to FEMA and state and territory officials who seek to improve their capacity to help local governments become better equipped to apply for and implement FEMA HMA grants. Gavin Smith, North Carolina State University

The colleges of engineering at California State University, Los Angeles (Cal State LA) (a Hispanic-serving institution) and California Polytechnic State University (a primarily white institution) are investigating the impact to the educational mission of the disruption caused by the COVID-19 pandemic. This study is temporal and inclusive. Performance is measured over time through a longitudinal survey that includes students, staff, and faculty and tracks demographic categories within each group. The study considers two institutions in the same system but with different student bodies, which allows for comparisons between populations with different ethnic and socioeconomic demographic distributions. The presentation will focus on the research design and data analysis completed to date. Data collection (numerical and open-ended questionnaires) occurred at two-week intervals at both campuses during spring and summer 2020, immediately following campus closures. Data collection continues at Cal State LA twice per semester. The quantitative analysis of complete or nearing completion concerns (a) the immediate impact of campus closures and (b) the temporal variation during the spring semester at both campuses and the significant differences among them. The former, corresponding to the initial drop in the resilience triangle, considers the impact on different demographic groups. The latter, corresponding to the early phase of recovery, uses general estimating equations as the analytical framework. Future analysis will entail (c) longitudinal, thematic analysis of the qualitative data (open-ended questions), (d) analysis of the long-term recovery at Cal State LA, and (e) exploration of the structural relationship between variables using structural equation modeling. 

This presentation is on the longitudinal study of Lumberton, North Carolina. The study is conducted by a multidisciplinary research team from the National Institute of Science and Technology (NIST)-funded Center of Excellence for Risk-Based Community Resilience Planning and NIST Engineering Laboratory. The study began shortly after Hurricane Matthew caused catastrophic flooding in Lumberton in October 2016, and has continued, longitudinally, for five years, with six data collection efforts to date. This abstract is being submitted alongside five other presentation abstracts to form an entire session proposed to be moderated by John van de Lindt, Colorado State University.

This presentation will present findings from surveys administered to households throughout the past five years. Housing units are categorized based on whether the same household has resided in them through the past five years, or if a new household has moved in at any point since 2016 Hurricane Matthew. For the former, four housing recovery metrics are tracked through time and multiple disasters. For the latter, two recovery metrics are tracked. Analyses reveal differential impacts and recovery trajectories attributed to housing type and ownership status, household social vulnerability characteristics, and differential access to recovery resources. The presentation will then integrate the recovery metrics from the surveys for public housing with interview findings from public housing residents and local-level decision-making regarding repair/replace decisions of the public housing units. The presentation will conclude with the major takeaways from the on-going study that can be applied more broadly to housing recovery in the United States and beyond.

Disaster events and recovery periods bring about change in access to essential services through the closure of facilities, transportation, evacuation orders, power outages, and other barriers. Understanding changes in access to essential services following a disruption is critical to ensure just and equitable recovery and more resilient communities. Location-based services (LBS) data collected from cell phones offer new opportunities for identifying patterns of unique visits to facilities and, subsequently, deviations from those patterns that may indicate unavailability of those facilities, i.e., functional closures. In this analysis, the authors of this study introduce a data-driven approach to identify functional closures at the facility level by assessing patterns in unique user appearances over time from LBS data. They show the application of this approach applied to samples of supermarkets, schools, health care facilities, and home improvement stores in Southwest Florida leading up to and following the landfall of Hurricane Irma in 2017. This paper promotes the adoption of LBS data for hazards and risk analysis research for understanding behavioral responses to disruptions and the factors contributing to those behaviors.   

From earthquakes to extreme temperatures and variable levels of precipitation, many communities are exposed to multiple hazards. This has never been more apparent than preparing for and responding to natural hazard threats during the coronavirus disease 2019 (COVID-19) pandemic. Using semi-structured interview data conducted with twenty-nine emergency managers and hazard specialists working in British Columbia, Canada, between August and November 2020, this research examines how emergency and risk managers view the threat of multi-hazards and how emergency management decisions are made in the face of multiple threats. This research presents findings on capacity differences across communities, community member and politician buy-in, and the ever-important task of understanding and responding to the needs of vulnerable populations. A framework is presented that combines the needs of short-term actions and long-term preparedness in the face of multi-hazards and incorporates the changing threats from climate and population changes. Concurrent, compounding, and cascading threats are discussed at the individual and society levels, as well as from the perspective of local, regional, and federal government representatives. This research can further our understanding of how best to support communities adapting to multi-hazard threats and how to support the individuals living within at-risk communities.

What determines recovery? Longitudinal studies can answer this question by providing evidence about pre-existing inequalities as well as post-event social environmental processes. Community-based participatory research with the 1995 Kobe Earthquake survivors led to the Seven Critical Element Model of Life Recovery. The 2001-2003-2005 Hyogo Life Recovery Panel Study highlighted individual/collective social ties as one of the most critical elements for life-recovery-bouncing-back within the most disadvantaged group. The five Natori Life Recovery Panel Surveys on Great East Japan Earthquake (GEJE) survivors between 2014 and 2021 cross-validated the significant effects of housing, livelihood, health, and individual/collective social ties on life recovery. Those with limited access to these elements during pre- and/or post-disaster periods recovered the least. This presentation focuses on empirical evidence that 1) those who suffered the most were also struggling prior to disaster and that 2) post-disaster personalized measures aiming at social tie enrichment can enhance resilience over time among even the most disadvantaged.

Relief and recovery from natural disasters require a coordinated approach across several federal and state government agencies. In order to achieve optimal resource allocation and first responders, accurate and timely information of the impact as well as the assessment of the destruction is the cornerstone to any recovery effort. In addition, this knowledge should be gathered within 24-48 hours and shared for informed decision-making. To this end, this research illustrates two real-world experiences in addressing post-disaster recovery efforts for the Bahamas due to Hurricane Dorian and Puerto Rico due to earthquake and subsequent flooding. The research illustrated that a coordinated effort is needed for planning as well as executing for improved human security.

Existing flood models do not fully consider expanding development in flood-prone regions, rapid rain accumulation, construction methods and materials, climate change, or population growth. The underlying process of creating these maps is heavily centralized (i.e., authority-oriented) and disproportionately influenced to benefit wealthier, more privileged communities. A 2017 report by the Department of Homeland Security’s Office of Inspector General found that only 42% of the total flood map miles in the Federal Emergency Management Agency’s inventory were updated and valid. A major limiting factor in covering neighborhoods and communities using the current flood sensing capabilities is the high cost of sensor acquisition, installation, maintenance, and a lack of skilled operators. 

This Texas Sea Grant-funded project augmented flood management practices in Texas coastal communities through citizen science, artificial intelligence (AI), decision science, and cyberinfrastructure. Part of this project included a survey component that asked respondents to 1) share their experiences with flooding and rescue operations in the past and 2) what types of information would have helped them while making evacuation decisions. Through integrating the findings of this survey into technology development, the author’s long-term goal is to design a user-inspired flood risk mapping tool for coastal communities. This submission will present the survey results, which will be used to incorporate preferred information into a CyberGIS spatial decision support system to assist in both evacuation and rescue operations decision-making. 

This research has focused specifically on the vulnerability and gap in research of rural communities. Given that 19.3% of the U.S. population lives in rural areas, and 97% of the total U.S. land is classified as rural, and differences in governance, social networks, access to resources, and other factors, it is important to investigate the specific needs of these communities and their members. In May of 2019, an EF4 tornado moved through the heart of Northeast Kansas, causing 18 injuries and no fatalities. The tornado path stretched for over 30 miles and narrowly missed several cities, reaching a width of one mile. The field data described in this presentation are from three waves of collection, occurring three days, six months, and roughly one year after the event. The data collected consists of structural assessments of physical damage and repair, and surveys administered to households on disruption and recovery. This presentation will present factors that contributed to household and housing-unit recovery in rural areas after tornadoes, and plot trajectories for housing unit repair, functionality restoration, and recovery. Taking this holistic approach to tracking recovery will deepen understanding of the recovery process after disasters, and can make skillful interventions more applicable to rural communities in the future. 

Oregon coastal communities face challenges related to both chronic (storms, sea-level rise (SLR)) and acute (earthquake/tsunamis) natural hazards. Furthermore, these coastal communities are isolated, "geographically, economically, and culturally" from urban inland areas. Coastal intra- and inter-community economic disparities are extreme, leading to heterogeneity in hazard preparedness. Thus, successful adaptation and mitigation measures must address local concerns and diverse perspectives while being cognizant of limited financial resources and community capacity. This paper presentation discusses current research efforts to model feasible hazard mitigation and adaptation policy pathways that Oregon coastal communities may implement to increase their resiliency to chronic and acute hazards. The researchers hypothesized that significant gains in developing mitigation and adaptation pathways could be realized by quantitatively examining potential policies and their associated trade-offs across scales ranging from the community to county to statewide. Utilizing the agent-based spatially explicit Envision model, they evaluated the efficacy of different hazard mitigation and adaptation policies for a variety of resilience metrics such as direct losses (life safety, capital), indirect losses (recovery), and social equity (unequal loss). This paper presentation will provide an overview of the Envisioning Oregon's Coastal Futures Project and showcase preliminary results. Additionally, the researchers will highlight the community engagement efforts used to identify mitigation/adaptation policies and resilience metrics, including the incorporation of qualitative place-based values into the Envision framework. They will also discuss our efforts to train a cohort of transdisciplinary students in the co-production of actionable knowledge for hazard resilience. 

Hurricane Katrina did not affect every part of Biloxi, Mississippi, the same way, devastating the oldest part of the city the hardest, filled with shotgun houses, social clubs, and churches rooted in ethnic and national heritage, the seafood industry, and casino barges. The recovery process since has been shaped not only by what had physically been there before and the path of the hurricane, but also by pre-disaster social, cultural, political, and economic forces and the city's history and people's memories of it. Neighborhood, ethnic, racial, and socioeconomic class identities are interwoven with memories of earlier hurricanes to shape post-Katrina responses and recoveries. Understanding this context is vital to our understanding of long-term recoveries from disasters in general, including Katrina’s lasting impact on Biloxi, and of the ways in which cycles of disasters can influence recovery, in turn setting up what and who will be hit by the next disaster. Following Katrina, decisions were made to relocate casinos onto land, evolving into questions for some local residents about why they too could not have a plan for recovering. Long-term recoveries in Biloxi, then, have been shaped not only by ideas about recovery but by a complex web of risk perception, economic need, and cultural identities and values.  

In the Great East Japan Earthquake, the mortality rate of people with disabilities (PWD) was more than doubled in Miyagi Prefecture, where normalization was advanced and widely practiced. Once the disaster happened, however, these services stopped, and PWDs were left behind. The root cause of their high mortality lies in the siloization of normal time social services and emergency time operations. The fundamental solution is to seamlessly connect those two silos by involving professional social workers who would plan normal time formal social services as well as emergency informal evacuation/sheltering resources. Beppu City, Oita, Japan initiated this silo buster practice. The key person of the Beppu Project is Junko Murano, who acted as a city inclusion manager. Interviewed with the positive deviant, 87,990 letters transcript was produced. Two researchers conducted content analysis and key phrase findings until the reliability of Cohen’s coefficient reached 0.75. This produced 306 key phrases. To categorize these, the affinity diagram method was employed. Cycles of diagram making, sharing with stakeholders, including citizen groups, local communities, and the local government officials; and revising the diagram were repeated six times over the past five years. As a result, three mutually exclusive super-conceptual clusters (boundary crossing, boundary spanning, and empowerment/strength-building) emerged. One of the major findings is that 1) boundary spanning or coordination among stakeholders in usual terms is an outcome of boundary crossing, 2) mission and evidence bases enable the act of boundary crossing, and 3) empowerment/strength-building was brought out as a result of boundary spanning.

The KATIVA NOLA project began in 2003 as a study of how Vietnamese immigrants in New Orleans fare at their new home vis-à-vis their counterparts who remained behind in Vietnam after the collapse of the South Vietnamese government in 1975. An initial wave of the U.S.-based sample took place during the summer of 2005, just weeks before Hurricane Katrina flooded a principal enclave in eastern New Orleans. Subsequent survey waves in 2006, 2007, 2010, and 2018 and extensive qualitative work follow and document their recovery from that historical disaster. Early results indicate a much more robust recovery than similarly-affected groups, and more recent work explores the reasons why. Implications for current frameworks of resilience and recovery will be discussed.

Yearly, the Federal Emergency Management Agency (FEMA) provides hundreds of millions of dollars for hazard mitigation through their Hazard Mitigation Assistance (HMA) grant programs. Unfortunately, this funding is most accessible to already resource-rich communities that have the means necessary to complete complex applications and implementation processes, leaving communities that are known to be more vulnerable to disasters the least able to leverage federal mitigation funding. This distributional inequity is consistent with decades of environmental justice research. While distributional equity is the most recognizable dimension of environmental justice,  two other important dimensions contribute to environmental justice: recognition and procedural equity. These dimensions are interrelated, and if the equity in FEMA mitigation funding is to be nurtured, special attention must be paid to addressing inequities associated with distribution, procedure, and recognition in FEMA programs and processes. This presentation highlights the results of a national survey conducted with 43 State Hazard Mitigation Officers (SHMOs) that explored the role of states and territories in facilitating distributional, procedural, and recognition equity in their jurisdictions in relation to FEMA HMA programs. The results highlight state-level shortcomings, including limited understanding of underserved communities, poor procedures for identifying and engaging with underserved communities, and limited local engagement in the state- or territory-sponsored conferences, trainings, meetings, and policy discussions. However, more importantly, the results yield insight that can guide the improvement of state-level policies and practices that govern the state’s interaction with local communities that are least equipped to effectively apply for and implement FEMA HMA grant programs. 

Much of the current disaster literature adopts a social vulnerability perspective, which considers how political, social, and economic factors influence pre-disaster preparation and post-disaster recovery. Even with this focus, however, there remains a lack of literature on immigrant populations and their long-term recovery trajectories. As such, this project involves a long-term, intersectional, multi-level analysis of Mexican immigrants and their disproportionate vulnerability in post-disaster recovery. This research builds on previous fieldwork conducted in 2019 in Houston, Texas after Hurricane Harvey, which focused on the role of community-based organizations in assisting Mexican immigrant community members to move forward in post-Harvey housing recovery despite anti-immigrant policies. Future data collection will include: a) content analysis of disaster policy in the United States; b) follow-up semi-structured interviews with community-based organizations to assess the progression of Mexican immigrants‚recovery, particularly in light of the subsequent tropical storms and the COVID-19 pandemic that have impacted the Mexican immigrant community in Houston since Harvey’s landfall in 2017; and c) semi-structured interviews with Mexican immigrant households to understand their recovery from their point of view. The long-term nature of this study will provide a unique opportunity to elucidate the lived reality of compound disasters for Mexican immigrants and those who provide service on their behalf. Overall, this project addressed the challenges for and the needs of Mexican immigrants in post-disaster recovery. Findings will be of value to stakeholders involved in post-disaster recovery work. 

This presentation introduces an empirical predictive modeling approach to quantify flood vulnerability at the community scale. The researchers consider the effects of hazard strength, built environment, and social landscape. To showcase the utility of the proposed methodology, they applied it to census tracts in North Carolina. They empirically model flood vulnerability as the expectation of fraction of uninsured property loss given water depth. They derive flood risk as the expected annual uninsured property loss and loss ratio. They applied an Akaike information criterion-based two-stage model selection algorithm to choose the optimal input variables. For model calibration, they adopted modeled flood loss data from flood events induced by Hurricanes Matthew in 2016 and Florence in 2018 and insurance claims data from the Federal Insurance and Mitigation Administration’s National Flood Insurance Program. To calculate annual expected losses, they adopted the Federal Emergency Management Agency’s National Flood Hazard Layer and the empirical probability distribution of water depth of flood events to determine geographic distributions of flood hazard frequency and exposure. Results indicate that social indicators can be effectively used along with water depth and environmental variables to model community vulnerability to flooding. Although both environmental and social vulnerability factors were statistically significant in contributing to overall flood losses, flood exposure was the strongest predictor of flood risk. The resulting risk maps indicated pronounced spatial patterns of flood risk across North Carolina, suggesting a low risk for the Piedmont plateau relative to the Appalachian Mountains and Atlantic coastal plain regions. 

The National Flood Insurance Program (NFIP), which supplies the majority of flood insurance policies in the United States, plans to move to risk-based rates to ensure the financial stability of the program. This move to risk-based rates will cause premiums to rise for some homeowners. Under the program's current structure, homeowners are eligible for subsidized rates if their home was constructed before NFIP flood insurance rate maps were created for their community; however, these subsidies would be phased out in the move to risk-based rates. This loss of subsidies combined with the rate increases may increase the financial burden of flood insurance for households. This study looked at the effect that income-based subsidies have on communities facing flood hazards. The authors of the study used an agent-based model to evaluate the effect that subsidized, and unsubsidized flood insurance premiums have on communities in high-risk flood areas and the flood mitigation strategies employed. They applied this model to the community of Fargo, North Dakota, which is bordered by the Red River and experiences seasonal floods. They compared the effects of insurance subsidies on mitigation decisions, damages incurred, move-out rates, and the number of active flood insurance policies held. Damages and mitigation decisions are evaluated from the perspective of both individual homeowners and the community at large. These results may help us to evaluate the value of flood insurance and insurance subsidies for communities.    

The RISK project began in 2004 as a study of poor single parents enrolled in community colleges in the New Orleans area. Baseline data were collected on 1,019 participants age 18-34, in 2004, including physical and mental health, social support, and neighborhoods. A second wave of data collection reached about half of respondents in the summer of 2005 when it was interrupted by Hurricane Katrina. Since then, respondents have been followed to 38 different states, and surveyed three more times, most recently in 2018. In addition, 90 respondents did qualitative life history interviews. In this talk I will survey what we have learned about physical and mental health, residential and socioeconomic trajectories, as well as subjective well being of respondents over the last 15 years.

Business interruption is a common metric used to understand the economic impact of a disaster event. However, business interruption takes on new meaning for businesses that provide access to essential services, like fresh food. With this in mind, traditional metrics such as length of closure may not fully capture the impact on the quantity and quality of fresh food available to the community. This research explored the results from a survey of food retailers after Hurricane Harvey in Harris, Jefferson, and Orange County, Texas. Over 450 retailers were randomly sampled based on whether they accept Supplemental Nutrition Assistance Program (SNAP) benefits, their location relative to the floodplain, and their location relative to food deserts. In-person surveys were conducted five to seven months after Hurricane Harvey. Collected data included 210 complete surveys and are publicly available on DesignSafe-CI. This paper discusses alternative metrics for food retail interruption focusing on food access. The analysis considered the importance of factors that affect business interruption for food retailers, specifically the role of critical infrastructure (utility service, road network, and communications), and the key components needed for the retailer's functioning (supply chain, personnel, and the physical storefront). The findings of this research can inform food retailers about how to prioritize their critical inputs to reduce interruption of their operations and how interruption might ultimately lead to greater demand in food aid after disasters.

Community-based resilience programs are a key part of the United Nations Disaster Risk Reduction program. In general, the cost-benefit and economic analyses of such programs are conducted after the fact. Additionally, these analyses typically focus on more easily estimable costs and benefits related to infrastructure and less oN quantifying the costs and benefits of soft resilience‚ like individual training or preparedness. To address these limitations, a survey-based contingent valuation (CV) method was used, which focused on individual willingness to community-based resilience training. Specifically, a single-bound, dichotomous choice design was administered through an online survey to a sample of 667 residents of Tennessee, North Carolina and Virginia. Preliminary estimates suggest a mean willingness to pay (WTP) for a five-week community-based resilience training between $110 and $340 per individual in a one-off payment. Individual factors such as experience with extreme weather are associated with the likelihood of being willing to pay. When aggregating to the adult population of these states, the estimated value of the described program ranges between $488 million and $2.26 billion. Under the assumption that trained individuals provide value to their personal and professional networks, a more conservative estimated value of the program ranges between $560,000 and $28 million. Given the low cost of implementing such a program, even these lower total values suggest a high potential benefit-cost ratio. Policymakers in charge of improving disaster resilience and risk reduction may consider collaborative, community-based resilience training as both a publicly desired and cost-effective option.

A pandemic differs from most other disasters because, rather than bringing people together in recovery, it tends to isolate them in anxiety or divide them in disagreement about risks. Against the backdrop of their previous, quantitative research on recovery from Hurricane Katrina in New Orleans, the authors of this study present qualitative evidence on COVID-19, mostly in southern Louisiana. They have completed about 250 hour-long, in-depth qualitative interviews with a diverse sample of residents since March 2020 and are continuing data collection at present writing. Unlike the aftermath of Katrina, residents seem more isolated, and rather than dealing with shared damage, they seem divided by divergent fears of virus exposure, concerns about the economy, desires to return to normal, and rejection of cautionary advice. For instance, older people are sometimes fearful of younger people who might not be careful and spread the virus, even within a family. Residents do support each other within households and within closer contacts. Yet, while some organizations and residents provide organized support, this seems rarer than after Katrina and seems to require special efforts and strategy, especially in social distancing. As one respondent said, "it’s different from a Hurricane because there are no actual physical contacts." As the vaccine rollout proceeds at present writing, we are watching for an easing of these concerns. Finally, as one respondent said, "if one does nothing about COVID-19 (e.g., not masking), it makes it worse. With hurricanes, doing nothing doesn’t make it worse."   

Landslides are frequent, damaging natural hazards that threaten the people and the natural and built environments of Puerto Rico. In 2017, more than 70,000 landslides were triggered across the island by heavy rainfall from Hurricane María, prompting requests by local professionals for landslide education and outreach materials. A novel Collaborative Risk Communication framework was developed to meet those requests, shaping the creation of a Spanish- and English-language Landslide Guide for Residents of Puerto Rico. Collaborative risk communication is defined as an iterative process guided by a set of principles for the interdisciplinary coproduction of hazards information by local and external stakeholders. The process behind this form of risk communication involves mapping out the risk communication stakeholders in the at-risk location—in this case Puerto Rico—and collaborating over time to address a shared challenge, such as landslide hazards. A core team of government and university partners was formed to conduct collaborative work with an informal network of hazards professionals from diverse sectors in Puerto Rico. The following principles guided this process: cultural competence, ethical engagement, listening, inclusive decisions, empathy, convergence research, nested mentoring, adaptability, and reciprocity. We describe these principles and the associated process in order to motivate collaborative risk communication efforts in different geographic and cultural contexts. While our work focused on addressing landslides, the principles and processes are transferable to other hazards. They may also serve as a roadmap for future partnerships among government agencies and university researchers to inform the co-creation of risk communication tools.

Information on past flash flood events can be useful for improving understanding of this hazard and planning effective mitigation strategies for it. The internet is potentially a rich repository of this information. However, conventional internet search engines (such as Google or Bing) are becoming increasingly less precise for domain-specific search due to the growing size of the internet and the commercial aspects of the search technology. This drawback poses a challenge to researchers, policymakers, and emergency management officials interested in studying past flash flood events. For example, trying different search queries and separating relevant web pages from non-relevant web pages can be time-consuming. This presentation introduces a novel domain-specific search engine for flash flood events. This search engine is analogous to the Google Scholar search engine for scholarly literature. The user enters the date and location (county and state) of the flash flood event(s) of interest. The algorithm uses the National Oceanic and Atmospheric Administration Storm Events database to compose customized search queries on the flash flood event(s) of interest to guide the search process. Then, it uses machine learning techniques to retrieve web pages containing information relevant to these events. The study shows that the information is retrieved using this domain-specific engine at a better accuracy than conventional search engines. The presentation outlines the steps involved in developing this search engine and demonstrates its utility for assembling micro-data on past flash flood events. The presentation concludes by discussing how developing this search engine is generalizable to other disaster types. 

International post-disaster short-term reconstruction and long-term recovery initiatives have been arranged in such a way that they give priority predominantly to the emergency response and intensive built environment reconstruction (e.g., housing and infrastructure systems), without enough or equal attention given to long-term social recovery. A more intense focus on long-term social recovery and rehabilitation can promote holistic approaches to building resilience and sustainability at individual, family, and community levels. Based on a 10-year longitudinal study of post-Wenchuan earthquake reconstruction and recovery efforts, this research examined the unique background of its government-led intensive built environment reconstruction, unveiled the dynamic interplay between physical reconstruction and social recovery, and addressed how the physical reconstruction and social recovery did or did not reinforce each other in the promotion of individual, family, and community resilience and sustainable development. The project illustrated that physical reconstruction should not only offer its inhabitants solid and safe housing to stay, but also should provide a foundation to strengthen ongoing long-term social recovery. Meanwhile, long-term social recovery improves the quality of the built environment by facilitating the participation of earthquake-affected residents in community-based place-making activities, and it also promotes community-level long-term sustainable development through rebuilding residents' lives and livelihoods.

Natural disasters adversely impact children's mental health, with increased parent or child exposure and subsequent parental distress predicting poorer outcomes. It remains unknown, however, whether the psychological consequences of disasters for children persist long-term, and if so, why and for whom. The authors of this study examined the effects of mothers’ exposure to Hurricane Katrina on adolescent children's mental health 12 years later, distinguishing between direct effects of disaster exposure and effects mediated by maternal distress. They also evaluated moderation by child age and gender.

Data were from a 2003-2018 study of young, low-income, primarily African American mothers living in New Orleans when Hurricane Katrina occurred in 2005 (n=328). Mothers rated their mental health about one year pre-Katrina and one, four, and 12 years afterwards. They reported on an adolescent child’s (ages 10-17, mean=14.46) internalizing and externalizing symptoms 12 years post-Katrina using the Strengths and Difficulties Questionnaire. 

Path analytic models adjusting for mothers’s pre-disaster distress showed that, whereas the direct effects of maternal disaster exposure on child mental health were not significant, the indirect effects were. Specifically, mothers who experienced more Katrina-related stressors had higher distress thereafter, which predicted poorer child outcomes. Results did not differ by child age. Gender differences are discussed.

Disasters may affect children's mental health for many years, even for those who were very young or not yet born at the time, due to parents' ongoing disaster-related distress. Addressing parents' mental health needs in the aftermath of disasters may improve child well-being long-term.

Wildfires are a growing threat to communities across the United States. Research has shown that the intensity and frequency of and the social harm due to wildfires have increased in recent years, largely as a result of climate change. Meanwhile, urban and suburban growth has led to the expansion of the wildland-urban interface (WUI), making more communities vulnerable to wildfire risks. For example, the 2019 Kincade Fire burned 77,758 acres in Sonoma County, California within two weeks, forcing nearly 200,000 residents to flee. To reduce the life safety risks of wildfires and to enhance the resilience of WUI communities, it is important to understand household behavior and movement during wildfires. This talk presents new evacuation behavioral findings for the Kincade Fire. Specifically, the authors of this study conducted a survey study (see here for study information: https://hazards.colorado.edu/news/award-news/upcoming-research-2019-california-wildfires) to investigate how individual, social and environmental factors (e.g., evacuation orders and other information, previous experience, and preparedness activities) affected people's decisions whether or not to evacuate in the Kincade Fire. In particular, they applied both logistic regression and machine learning techniques to model evacuation decisions and compare their performance in terms of predictive accuracy and behavioral interpretability. They will conclude the presentation by discussing the potential of leveraging artificial intelligence-enhanced choice modeling to generate richer insights to inform emergency planning for WUI communities.