Researchers Meeting Abstracts

Floods and hurricanes account for over 80% of all recorded losses in the United States. Eastern North Carolina, a predominantly rural region, has faced repeated flood-related losses following multiple storm events, including Fran in 1996, Floyd in 1999, Irene in 2011, Matthew in 2016, and Florence in 2018. Since the early 2000s, Federal Emergency Management Agency’s Hazard Mitigation Grant Program (HMGP) has funded buyouts considered to facilitate a managed retreat and reduce exposure to flood risks across the region. Funded by a 2019 National Oceanic and Atmospheric Administration’s Ocean Climate Applications Sectoral Applications Research Program (SARP) grant, a key objective of this research is to assess the perceived risks and needs of the hazard management and planning community in eastern North Carolina. This paper focuses on the challenges and impacts of HMGP buyout programs in the region and on rural counties in 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 24 planners and emergency managers during a flood workshop held at East Carolina University campus in February 2022. Findings show that while there were advantages to buyout programs in eastern North Carolina, such as preventing repetitive flooding and green space restoration, there was also local resistance to such programs because of attachment to property, disagreement with property appraisal value, and missing property-related documentation. Moreover, there were negative economic and health impacts on participants due to the long duration of the buyout process.

While the COVID-19 pandemic resulted in widespread economic disruption, the severity of the economic impacts varied substantially across locations, industries, and demographic categories. Experiencing a concurrent natural hazard could amplify the impacts of the COVID-19 pandemic on the local economy. The concurrence of acute shock interruptions, as well as chronic stressors that some communities faced prior to the COVID-19 pandemic, creates a complex event. While it is recognized that complex events tend to create an impact greater than the individual events, there has not been much research to investigate the impact of complex events. This paper investigates the impact of a natural hazard event during the COVID-19 pandemic on small businesses vital to community economy. The impact on small businesses is measured by the amount of loans given under the Paycheck Protection Program (PPP). Although PPP loans are meant to help businesses recover from the disruptions caused by one event (i.e., the COVID-19 pandemic), pre-existing socioeconomic stressors, and concurrent or previous natural disasters (e.g., hurricane events), may have changed small business utilization of PPP loans. This paper examines the correlation between PPP loan utilization and experiencing a concurrent natural hazard. Additionally, the paper explores the differential impacts of different size events, which is determined by the amount of public assistance given.

More than half of FEMA Region 7 was in moderate drought or worse in March 2023, with patches of exceptional drought, in Nebraska and Kansas. The last time the region was completely drought-free was in 2019. This study sought to understand how drought was being discussed on social media and by local stakeholders. Prior nationwide research covering a study period of 2017-18 found that people in the Plains are more likely to tweet about drought than people in other states. This study gathered pilot data about how drought information was conveyed, discussed, and interpreted by local stakeholders, as this health and safety crisis is managed in FEMA Region 7. The research question was, how do individuals use social media to manage this significant drought event? Initial social media data analyses suggested that conversations specifically about drought included perspectives on how drought affected individuals and communities. This research demonstrated the intensity of drought focus of social media users and how on-going drought conditions impacted livelihoods and quality of life. This work isolated risk communication dynamics between decision makers and the public during a significant drought event. These results will inform efficient and effective communication strategies designed to support governmental and other decision-making related to public risk communication surrounding drought preparedness and management. This research demonstrated the intensity of drought focus of social media users and how ongoing drought conditions are affecting livelihoods and quality of life directly. The results can inform official public communication strategies via social media for longitudinal events.

Disaster risk communication strategies are core elements of reducing risk exposure but can also be drivers of the risk creation process. Though often created in service of official narratives, analysis of these communication strategies as cultural artifacts shows the ways that these strategies exist independently of official narratives, persisting or evolving in ways outside state or institutional actors’ control. Among the most prominent of these risk communication strategies are the advertising campaigns that collectively form the character Smokey Bear. Originally formed in the 1940s in service of wartime goals around strategic resource protection, Smokey stands as one of the most recognizable symbols in American culture. However, despite attempts to introduce nuance to the character’s core message, various scholars point to the character as an element tying public attitudes around wildland fire management to outdated full-suppression strategies. To what extent do primary official actors, such as the United States Forest Service, still control the meaning behind Smokey Bear and to what extent is the symbol’s meaning tied to the stickiness of outdated policies? As a first step to answering these questions, this work explored the history of non-official, semi-official, and informal use, consumption, appropriation, and spread of the symbol, and the ways such evolution tied to (or broke from) Smokey’s original and evolving messages around wildland fire risk reduction. It employed archival analysis as an early stage of a larger project that will incorporate a series of interviews and direct observation of wildland fire management actors.

The Rhode Island Coastal Hazards Analysis, Modeling, and Prediction system (RI-CHAMP) is a high-resolution decision support tool designed to help state and local emergency managers (EMs) and facility operators anticipate the consequences of a major storm striking their community. Traditional tools available to EMs, such as vulnerability assessments and Sea, Lake, and Overland Surges from Hurricanes (SLOSH) models, do not provide actionable data regarding specific local concerns, such as emergency vehicle access and potential communication disruptions. This research approach captures critical infrastructure (CI) managers’ concerns about hurricanes and Nor’easters for use in an online dashboard viewer that integrates those concerns with high-resolution wind, wave, and storm surge model outputs derived from Advanced Circulation (ADCIRC) modeling. The usefulness of the system for emergency management relies on the quality of the data collected from CI Managers. Collecting, handling, and sharing potentially sensitive data across private, state, and federal entities presents new challenges for federally funded researchers. This presentation provides an overview of this eight-year project, with a focus on lessons learned through deep engagement with government agencies about collecting and handling sensitive data.

Effective disaster debris management requires reasonable predictions and estimates of debris for a community to get back to normal sooner. However, there is a lack of data related to post-disaster waste quantities that could validate and improve debris predictions. This knowledge gap can be addressed by establishing reconnaissance methods for safely quantifying disaster debris promptly following a disaster using remote sensing or imaging technology. This study aimed to demonstrate and compare multiple imaging tools available for quantifying disaster debris using post-disaster data collected following Hurricane Ida. The tools used in this study are satellite imagery, high-altitude emergency response imagery, unmanned aerial vehicles (UAV), and terrestrial laser scanning (TLS). The study found that satellite and high-altitude emergency response imagery are well-suited for debris management in the disaster response phase. They are useful for determining areas that need immediate resources, such as transportation networks that need debris cleared to facilitate the provision of emergency services, as well as detecting transported vegetation debris. UAVs and TLS are well-suited for debris management in the disaster recovery phase when debris is piled at curbs or staging sites and can provide precise volumes. Guidance is provided for selecting an imaging method based on the user’s desired application and available resources, which can assist decision making for disaster waste managers. 

The COVID-19 pandemic caused the shutdown of schools across the United States. As school closures were extended and virtual schooling became the norm, children with disabilities were at risk of falling behind their peers academically, developmentally, and socially. This study highlights how school closures instigated inequitable educational practices that impacted the social vulnerability of these children. This study examines these issues using in-depth interviews with parents of school-aged children in grades pre-kindergarten through eighth grade (n=39). Interviews focused on the process of at-home education, and the ensuing impacts on children’s physical, psychological, educational, and socio-emotional development. Parents reported that children with disabilities lacked access to education, therapeutic services, and mental health care. These changes to schooling, along with the social isolation and uncertainty experienced by families, were factors in the increased vulnerability of these children, as they fell behind their peers and suffered from a decline in mental health. This study highlights inequitable educational practices during the pandemic and recommends that educational and disaster management policy consider children with disabilities and plan for more equitable responses to disaster, including plans for continued learning, therapy, and mental health services.

Several cemeteries, including historical ones, are in coastal communities that are increasingly at risk from coastal hazards due to rising sea levels. To protect cultural resources, land use planners consider historical resources in long-term comprehensive plans as they do natural resources. However, land use plans and the environmental planning cultures in southeast Louisiana incentivize industries such as oil manufacturing companies to buy contracts with the state for political, economic, and governing factors that put cultural resources like cemeteries at risk. In this study, the research team assessed cemeteries in coastal parishes in Louisiana at risk of coastal inundation under three sea-level rise scenarios until 2100. In addition, the team quantified the number of cemeteries at risk of inundation and identified possible differences between the demographics of the cemeteries. Data come from the U.S. Census geodatabases, the National Flood Hazard Layer from FEMA, and Digital Elevation Models from the U.S. Geological Survey to spatially analyze the location of cemeteries and low-lying coastal areas of the Gulf of Mexico. The results show that of out of the 72 cemeteries in the study area, most are at risk of the 100-year flood already today. Considering the most conservative prediction of sea level rise of 1-meter by 2100, inundation could increase by more than 52% across the region. At least 55 cemeteries are at risk of destruction, which embodies more than 76% of the county’s indispensable and irreplaceable social assets.

Federal disaster aid has played an increasing role in assisting with housing recovery in post-disaster communities. This aid primarily includes emergency grants from the Federal Emergency Management Agency, loans from the Small Business Administration, and disaster recovery grants from the U.S. Department of Housing and Urban Development. However, federal disaster aid is facilitated and, in some cases, only available if a presidential major disaster declaration is in place. While many factors affect a presidential declaration, the key determinant is a function of the impacted state’s total uninsured losses and the total taxable resources. Consequently, smaller and more concentrated disasters in wealthy states tend not to receive presidential declarations. This study investigates the impact of a presidential disaster declaration on the total post-disaster unmet housing needs in East Palo Alto following a hypothetical coastal flooding event. Two scenarios are considered: (1) the impact of a 20-year flood event and (2) the impact of a 100-year flood event. Results indicate that the 20-year flood event is unlikely to receive a presidential disaster declaration causing this event to lead to similar unmet housing needs as the 100-year flood event, highlighting limitations in the current approaches to declaring a major disaster. 

Disasters result in a wide range of complex impacts. Some impacts are tangible and visible while others are more difficult to identify. Individuals who experience the same disaster event can be affected in different ways. After disasters “hierarchies of affectedness” are formed, where people who have been affected by disasters are categorized by the severity of their impacts in comparison to others. This is sometimes done in formal ways and can play out in the media, politics, and through emergency management agencies. These types of hierarchies have influence on practical things such as who gets access to financial aid, and what recovery supports are provided. Other times, hierarchies form in more covert ways, and are more difficult to define. Even though these hierarchies form after every disaster, little is known about the impacts of this process. Using data from a broader qualitative study, this study progressed a model that was originally developed in Denmark to better understand hierarchies of affectedness and the role they play for those who have been impacted by disasters. This presentation looks at the following questions: What is a hierarchy of affectedness? How do hierarchies of affectedness form? What real world implications do these hierarchies have on the support disaster affected people have access to? Can hierarchies of affectedness be useful? How can governments and emergency management agencies understand their role in hierarchies after disasters and reduce harm?

Developing ways to measure resilience and build resilience strategies that reduce risk, enhance economic resilience, and address the most pressing needs of our communities is a national priority. However, communities – especially rural ones – typically lack the capacity and expertise to build collaborative resilience strategies. Here we present a university-community partnership model to conduct a resilience assessment and strategy with a rural Idaho community. We adapted the City Resilience (CR) Index, designed for urban centers, for the City of Kamiah (population: 1200), which is situated within the Nez Perce Reservation. The CR index includes the dimensions of infrastructure and environment, economy and society, leadership and strategy, and health and well-being. Fifty-two indicators are evaluated through 156 indicator questions to determine the community’s strengths and areas of need beneath each dimension. We held virtual group interviews with 49 community partners across all sectors of the community to assess resilience metrics and facilitate conversations about next steps for resilience planning. We then led a solutions-focused community workshop with 50 attendees to develop priority resilience actions and shared materials and results with the community through this website. To help the community get started on their resilience actions we connected Boise State professors and students aligned with specific research projects: flood mitigation; economic resilience; infrastructure and asset mapping; and affordable housing. While this pilot was successful, we discovered the need to develop more relevant metrics for rural communities and explore how to scale up this model for continued university-community partnerships.

Atlantic Canada—New Brunswick, Nova Scotia, Prince Edward Island, Newfoundland, and Labrador—is prone to several climate-induced extreme events. These increasing crises have distinct, catastrophic effects on humans and animals. In Canada and around the globe, pet owners consider their pets, service animals, or other companion animals to be integral parts of their family structure. Indeed, because of companion animals’ importance to families, animal-inclusive disaster and emergency planning are needed to enhance risk reduction and emergency preparedness. This study employed a collaborative approach, bringing together social science researchers, veterinary practitioners, and frontline community-based veterinary organizations to explore best practice approaches for integrating companion animals into disaster planning. The objectives of this approach were twofold. First, conduct a survey in Atlantic Canada to understand the current knowledge on animal-inclusive disaster planning. Second, use the research-practitioner-community partnership to create recommendations and mobilize knowledge for pet owners. The authors collected survey data from 843 pet owners in Atlantic Canada using the veterinarian team members’ and community partners’ personal and professional networks. The results showed that over 80% of the respondents indicated that having an animal-inclusive disaster plan before a disaster would be impactful. However, only one-third of the participants felt “very” or “extremely knowledgeable” about creating that plan. These results highlight a critical need for targeted knowledge mobilization to provide the necessary resources for pet owners. Using the collaborative approach, the authors created recommendations using infographics and mobilized the knowledge at community workshops in Atlantic Canada.

This Faculty Early Career Development (CAREER) award will advance scientific knowledge on the response of soils to earthquake ground shaking at multiple scales and enable its incorporation into system-level probabilistic seismic hazard assessments for water distribution systems. The spatially variable geologic structure near the ground surface exerts a significant influence on the intensity of ground shaking (known as site effects) and can exacerbate damage to the built environment. Drinking water and wastewater utilities are critical lifelines because of the significant negative effects of earthquake damage on the ability to fight fires, on public health, and the economy. This CAREER grant will foster a paradigm shift in system-level seismic hazard assessments for water distribution systems to overcome current practices that oversimplify the effects of near-surface geologic conditions. Accounting for site effects and uncertainty can reduce damage and service losses and improve post-event rapid damage assessments. The educational impacts include reshaping students’ training in data analytics and uncertainty quantification, while the main outreach impact is the launch of the first Earthquake Engineering and Seismology Community Alliance in Latin America (E2SCALA).

In the United States, alerting authorities are authorized by the Federal Emergency Management Agency (FEMA) to notify the public of imminent hazards and threats by sending Wireless Emergency Alerts (WEAs) through the Integrated Public Alert and Warning System (IPAWS). Although recent efforts have been made to examine historical WEA compliance with frameworks such as Mileti and Sorenson’s warning response model, less attention has been paid to information included in WEAs that is not prescribed by message design frameworks from risk communication scholarship. This paper explores the presence of situational crisis communication theory’s (SCCT) “base ethical response”—instructing information, to help receivers physically cope with the event, and adjusting information, to help receivers psychologically cope with the event—in terse mobile alerts. The authors conducted a content analysis of 4,777 WEAs sent between 2019 and 2022 to determine how often and in which contexts (i.e., hazard types, 90- or 360-character messages) these strategies are used. This work found that the limited definition of adjusting information used in prior research (e.g., direction to mental health resources) is rarely included in WEAs. The authors propose that details of organizational response efforts can function as adjusting information by reducing uncertainty and reassuring receivers that the event is under control, thereby amending SCCT’s base response to more closely align with the objectives and goals of warning message design.

Despite the fact the small businesses make up almost 99% of the U.S. economy, little research has been focused on recovery experiences of businesses and the challenges they face. This is problematic for community recovery at large which depends on successful businesses recovery. Existing research indicates that businesses are less likely to take mitigative actions prior to a disaster, are generally low resourced and limited in their ability to develop alternative supply and revenue streams, and have fewer but also less effective programmatic options for post-disaster aid. This study examined the recovery experiences of local businesses in the two communities of Comerío and Loiza in Puerto Rico after Hurricane Maria in 2017. Based on a purposive survey of 150 businesses in both communities conducted in June 2022, the study examined the challenges faced in business recovery well as the relationship between business recovery and interaction with other community stakeholders. The findings of this study highlight the importance of broader community-level interdependencies to local business recovery and can result in better economic recovery policymaking as well as overall community recovery planning following disasters. 

Disaster resilience of local communities has become a critical issue since all humankind face more severe weather events related to global warming. Governments around the world often rely on street-level bureaucrats to implement measures, enforce regulations, and deliver public services in disaster management. However, these bureaucrats often struggle with heavy workloads, limited financial resources, and insufficient staff. Research on how street-level bureaucrats manage their duties psychologically and behaviorally is limited. To fill the gap, this qualitative work aims to answer the following research questions: How do street-level bureaucrats in disaster management perceive their job? How do they manage high workloads? Interview data from 17 informants from district offices, city governments, and the central government in Taiwan who are responsible to disaster preparedness and response showed that street-level bureaucrats generally perceived themselves as policy implementers and their job is meaningful to the citizens and society, although they suffer with high job stress. They are also confronted with problems like citizens who often ignore disaster risks and are reluctant to participate in disaster preparedness activities. The bureaucrats also identified positive ways to effectively manage their workloads, including receiving support from family, colleagues, supervisors, community leaders, local businesses, and collaborative professionals; learning useful knowledge, skills, and experiences; understanding local community attributes and needs; establishing intrinsic and extrinsic motivations; and obtaining assistance from social networks.

This study summarizes lessons learned and best practices from recent research on the COVID-19 pandemic response by local governments. Across the United States, local governments assumed a leading role during COVID-19 adopting various measures to protect their communities—health services expansion, public health guidelines enforcement, socio-economic relief, among others. Concurrently, they met significant challenges, including political opposition, lack of personnel/expertise, and fiscal stress. Using the Pandemic Management Survey, a survey instrument of 52 questions distributed to Pennsylvania and Florida municipal executives in the spring of 2021, the authors present lessons learned and best practices from a series of COVID-19-related projects they developed. The survey assesses local governments for their (a) risk perception and level of preparedness for natural and health-related disasters; (b) sources of information and channels of communication during disasters; and (c) response actions to the COVID-19 pandemic. As such, this study identifies determinant factors that facilitate effective and efficient local government response during public health crises. According to the findings, doing so requires a council-manager government form, preparedness driven by past organizational learning, information from expert organizations, citizen engagement in planning, and fiscal support from higher levels of government. Given that local governments are typically the first responders in the wake of public health crises, examining factors that facilitate or obstruct their response is necessary to improve their decision-making process and overall strategy.

This paper presents the use of tsunami evacuation drills within a coastal community in the Cascadia Subduction Zone (CSZ) to better understand evacuation behaviors and thus to improve tsunami evacuation preparedness and resilience. The authors of this study used Global Navigation Satellite System embedded in mobile devices to collect evacuees’ spatial trajectory data. Based on the empirical trajectory data, the authors employed probability functions to model people’s walking speed during the evacuation drills. An Evacuation Hiking Function was established to depict the speed-slope relationship and to inform evacuation modeling and planning. The regression analysis showed that evacuees’ speed was significantly negatively associated with slope, time spent during evacuation, rough terrain surface, walking at night, and distance to destination. The authors also demonstrated the impacts of milling time on mortality rate based on participants’ empirical evacuation behaviors and a state-of-the-art CSZ tsunami inundation model. Post-drill surveys revealed the importance of the drill as an educational and assessment tool. The results of this study can be used for public education, evacuation plan assessment, and evacuation simulation models. The drill procedures, designs, and the use of technology in data collection provided evidence-driven solutions to tsunami preparedness and inspire the use of drills in other types of natural disasters such as wildfires, hurricanes, volcanoes, and flooding.

Lifestyle recovery is the return to essential facilities (like grocery stores and medical centers) and non-essential facilities (like entertainment and recreation services). As a critical milestone of disaster recovery, lifestyle recovery captures the effects on population, the restoration of infrastructure, and the economic vitality of businesses. The research study asks: to what extent do disasters impact lifestyles and how do these vary across flooding and demographics? The study leverages privacy-enhanced location intelligence data to characterize distinctive lifestyle recovery patterns of the 2017 Hurricane Harvey in Harris County, Texas. The analysis integrates multiple data sources to record the number of visits from home census block groups (CBGs) to different lifestyle facilities. First, primary clustering using k-means characterized four distinct essential and non-essential lifestyle patterns. For each primary lifestyle cluster, the secondary clustering characterized the impact of the hurricane into three recovery trajectories based on the severity and duration of recovery. The findings imply differential rates of lifestyle recovery in demographic groups and show that the impact of lifestyle recovery extends beyond flooding. The analysis and findings provide novel data-driven insights for public officials and emergency managers to examine, measure, and monitor lifestyle recovery. First, lifestyle recovery is a critical milestone that can be examined, quantified, and monitored in the aftermath of disasters. Second, the spatial structures of cities formed by human mobility and facility distribution extends the spatial reach of flood impacts. Third, future studies could investigate the equity of lifestyle recovery among different demographics and regions of the community. 

This study addresses the gap between knowledge and action that was demonstrated in the Kahramanmaraş Earthquakes of 6 February 2023, Türkiye. The contrast between knowledge and action regarding seismic risk is extraordinary. Türkiye has designed and approved building codes that are equivalent to the rigorous codes implemented in seismically challenged California. There are approximately 120,000 structural engineers in Turkey who have the knowledge and skills to construct buildings, roads, and dams that may suffer strain from seismic events, but not fail. Yet, choices in the implementation of these codes in construction and inspection led to the collapse of over 250,000 buildings in 11 provinces, over 50,000 deaths in Türkiye alone, and over 8,000 deaths in neighboring Syria. The challenge is how to rebuild the devastated communities. Türkiye can opt for short-term solutions, rebuild the same style of flawed housing and infrastructure, hoping that it will last at least a generation. Or, a network of engineers, economists, policy analysts, planners, and responsible leaders may lead the intelligent redesign of the built environment to live with seismic risk and engage the whole population of Türkiye in an on-going experiment to create a society that recognizes seismic risk as a continuing threat. As population increases, but risk remains unattended, the consequences increase in cost and losses for the population. This study explores what collective actions can be taken to reduce risk in rebuilding today that enables informed, resilient communities to face the next earthquake that will surely come.

Relocation initiatives are frequently presented as useful and, at times, necessary adaptation strategies for use in mitigating the risks associated with living in hazardous areas. However, these formal efforts may not fully address community needs and interests. This can motivate the initiation of community-led adaptive practices, which center on individuals and groups implementing measures to reduce their exposure and vulnerability to environmental risks. In this study, we draw on data collected from interviews with local planners and focus groups with community members in Puerto Rico to present findings on how community-led adaptive activities work to develop community resilience to hazards and address gaps in formal relocation practices. Through this work, we highlight how community-led efforts can serve as a teaching tool for understanding the implementation of relocation and broader resilience-building activities in hazards research and practice.

The climate change crisis begs a few questions for those involved in planning processes. Firstly, what are our plans missing? In other words, is climate change integrated into our plans and to what extent is it integrated? Secondly, what are our missing plans? In other words, do jurisdictions have climate change-specific plans, and how do they compare to the adaptability standards Indigenous Nations’ plans have set. This project considered three major issues. Firstly, it analyzed if and to what extent climate change adaptation is integrated into comprehensive plans and hazard mitigation plans at the state and county level in the United States. It does so for 12 states and 24 counties—the least populous and most populous counties in each state. Secondly, this project analyzed whether these jurisdictions have created climate change-specific plans. Thirdly, this project analyzed common adaptation strategies across 47 Indigenous Nations’ plans that the United States could learn from as Indigenous Nations have been leading the climate change adaptation movement for years. This project’s results showed that every state studied is planning for climate change. However, the extent to which climate change is integrated into hazard mitigation and comprehensive plans varies based on wealth and the level of government that created the plan. Climate change is an interdisciplinary issue that cannot be resolved with only climate change plans that have weak legal enforceability and standardization. This project recommends updating hazard mitigation plans and comprehensive plans to require climate change as an element.

Economic resilience, or the ability to adapt or recover from a negative financial shock, depends on an individual’s or household’s adaptive capacity, which is the ability to access sufficient resources from one or more sources such as income, savings, loans, aid/assistance, or insurance. The purpose of this quantitative research study is to explore the social and economic factors that influence household mitigation behaviors to disasters to promote household resilience to more intense and frequent climate and disaster risks. What social and economic factors influence household behaviors to address climate/disaster risk? To better understand these relationships, we explore principles of social psychology and behavioral economics in the context of disasters using the 2021 Survey of Household Economic Decision Making, a nationally representative cross-sectional survey of the United States conducted by the Federal Reserve Board of Governors. This study introduces the use of additional aspects of household economic wellbeing including indices that measure a household’s balance sheet to include income, assets, and debts. The data included modules on a range of topics of current relevance to financial well-being including credit access and behaviors, savings, retirement, economic fragility, discrimination, education, student loans, disaster risk perceptions, and mitigation behaviors. Practical implications from this may interest private market actors that support financial inclusion and asset building efforts (e.g., Community Development Financial Institutions, Insurance providers), policymaking that promotes wealth building and debt reduction (e.g., home ownership, savings, and other assets that produce income and appreciate), and housing and planning stakeholders.

Research has shown that Federal Emergency Management Agency (FEMA) aid is not available to all people who need it, their approach to providing aid is inequitable, and previous studies after Hurricane Katrina have found that lack of social support and financial burden are mediators on the path from surviving a disaster to generalized depression. Our study sought to examine if there is an association between household financial loss and depression symptoms among hurricane survivors in North Carolina. Using data collected through a survey of randomly selected households in counties affected by hurricanes Florence and Dorian in Eastern North Carolina and the Outer Banks, this analysis is restricted to all adults who completed the survey, including questions on depressive symptoms. An exploratory factor analysis was used to construct a financial loss variable using information collected on job loss, property damage and out of pocket expenses. Data on depressive symptoms using the Center for Epidemiologic Studies Depression Scale (CES-D) validated instrument was used to determine if the respondent had symptoms of generalized depression. Binomial regression was used to estimate the association between household financial loss and individual depression score controlling for storm they experienced, county, age, income, education, job loss, illness/injury and race/ethnicity. Among the 505 respondents to the survey so far 248 (49.9%) reported depressive symptoms consistent with being at risk of clinical depression. The results of this analysis are useful to public health in informing the utility of financial services in reducing mental health effects of hurricanes in coastal regions.

This study took an integrated social-science and engineering approach to evaluate coastal community connections via local and highway transportation networks following a multi-hazard event. Present-day risks resulting from the ground shaking and tsunami inundation associated with the impending Cascadia Subduction Zone rupture were considered. The resilience of Oregon coastal communities was valued based on: 1) post-disaster regional and local scale transportation network performance, 2) access to general utilities and services, such as food, education, and healthcare, and 3) community-defined metrics such as homes and place of employment. Specifically, the resilience of these coastal communities was analyzed within the framework of accessibility considering both immediate damage and post-disaster recovery. The methods used publicly available data, participatory mapping, and an engineering model of transportation network damage and recovery. Results showed regional differences in the connectivity between critical facilities and community identified assets across the Oregon coast. This work can be used by stakeholders and practitioners to inform disaster policies with specific metrics that promote equitable resilience planning and engineering practices that might best reduce community islanding.

Climate change poses significant risk to the U.S. housing and mortgage finance ecosystem, with many effects shown to disproportionately affect economically vulnerable populations. Households exposed to natural disasters face risks of damaged property, loss of residence, and income disruptions, among others. For mortgage borrowers, disaster related financial shocks could lead to significant financial distress resulting in mortgage default. Focusing on the Federal Housing Administration’s (FHA) single family mortgage insurance (MI) program, this study characterized past disaster exposure on a sample of 897,00 FHA insured mortgages from 2004-2017 and estimated a relationship between past exposure and homebuyer outcomes. Additionally, this study estimated risks to the FHA Mutual Mortgage Insurance Fund through a two-stage process. Despite FHA safeguards against disaster-related costs through requirements for casualty insurance and “preservation and protection” requirements, results indicated a positive correlation. Using a standard claims-probability logit model with zip-code level disaster exposure data from the Federal Emergency Management Agency and an 18-year loan level panel of FHA borrower and mortgage characteristics, the authors found disaster exposure correlated with a 20% increase in the probability of mortgage foreclosure with an MI claim. These effects were heterogeneous across race, disaster type and credit groups, providing evidence that economically vulnerable groups are more susceptible to adverse outcomes within the FHA portfolio. A second stage simulation using estimated coefficients found that disaster exposure caused between $0.8 billion and $1.5 billion in additional FHA MI claims from 2004 to 2019. The role of hazard insurance and disaster aid is also examined.

Despite the significant hazard that tornadoes pose, much remains to be learned about the impacts of tornadoes on critical facilities in the United States. While post-storm investigation reports have documented the damage from individual tornadoes or tornado outbreaks, there is a major gap in the availability of data describing the cumulative national impacts of tornadoes on critical facilities. In response to this gap, this study created a database of tornado impacts on critical facilities (e.g., schools, fire stations, etc.) by mining National Weather Service (NWS) narratives from the National Centers for Environmental Information Storm Events Database. The first stage in the creation of the critical facility impacts database included documentation of preschool-12th grade schools hit by tornadoes recorded in the Storm Events Database between 1993 and 2020. To date, the authors have identified 669 schools struck by tornadoes, for an average of approximately 24 per year. The NWS narratives do not include an exhaustive list of facilities affected by each tornado; therefore, it is likely that other schools were hit by tornadoes during this period but not recorded, and the true number of schools struck by tornadoes is higher. Geographic Information System (GIS) analysis of tornado damage polygons in two states suggests the database captures only 50% of the schools hit by tornadoes during the study period. The results from this study helped support the adoption of the first tornado load requirements in the International Building Code (2024) and supported state adoptions of tornado shelter requirements.

From February of 2020, the United States suffered from COVID-19, which became a global pandemic. By 2023, 27,976 cases and 318 deaths occurred in Roanoke, Virginia. Roanoke Area Ministries (RAM) serves the homeless in the New River Valley, Virginia. Services include daily hot meals, donated clothing, short-term housing and storage, and financial assistance based on income and documented need. RAM housing and social services were strained by the pandemic, with increasing numbers of homeless, some newly homeless, and a steep rise in food insecurity. Moratoriums on the application of eviction notices led to many being forced to leave their residences and become part of the homeless population. In 2021, the authors assessed the impact of the pandemic on the RAM homeless client base. They conducted 50 oral history interviews with volunteers and clients, documenting the struggles and vulnerabilities of this population in crisis. The authors also reviewed five years of quantitative intake data (N= 500), and Blue Ridge point-in-time assessments for the same time period (2015-2020), and detailed overall impacts on recent service patterns during the COVID-19 pandemic. Related social disruptions of homelessness, food insecurity, and job loss were then compared in historical context. The final report outcomes identify areas of greatest need and justify increased funding from support agencies of $600,000 over pre-pandemic support. Report data was also used to initiate adaptive innovations in client services and identify strategies to increase fundraising efforts for novel service programs.

With the resilience agenda gaining traction in governmental policies, researchers, practitioners, and policymakers are seeking ways to put community resilience plans into practice. As communities are complex systems facing complex sets of risks, interdisciplinary and transdisciplinary approaches are needed to operationalize resilience. These approaches must still be simple to understand and yield easy to use tools that empower frontline communities and organizations to act in alignment with community interests. This research introduces a temporally-based accessibility metric that is scientifically principled while being practical and easily understood by community decision makers, where variation in the metric across the disaster timeline is a proxy for community recovery. Defining accessibility as the actual use of available resources with a reasonable effort and net cost opens up the need to evaluate six multidisciplinary dimensions of accessibility, including: proximity, availability, adequacy, acceptability, affordability, and awareness. The authors’ proposed metric combines at least four out of these six dimensions (i.e., proximity, acceptability, adequacy, and availability), considering uncertainty in measuring each. The metric conjoins with common engineering-oriented functionality-based community resilience frameworks as the functionality level of the service providers has been used in its development. Operating at the household level, the metric determines the ratio of post-disaster access time to the intended resources against its pre-disaster levels. The metrics are illustrated for schools and pharmacies using the Lumberton Testbed and data collected following the 2016 flood after Hurricane Matthew. Findings offer new insight into ways to prioritize recovery plans and can be used to trigger protective actions.

Evacuations are often needed during extreme weather events to ensure the public’s safety. Past research suggests that administrative burdens—compliance, learning, and psychological costs—can reduce the likelihood of shelter evacuations. These burdens disproportionately impact socially vulnerable populations and create social equity issues. While most residents can safely evacuate to shelters during emergencies, some individuals have specialized medical needs and require a more acute level of care during evacuations. In 2007, Florida began operating special needs shelters to provide a safe environment and basic assistance to individuals needing a higher level of care and supervision, to ensure that special medical needs are met during an evacuation. This study examined the following research questions: What administrative burdens did individuals face when evacuating to special needs shelters during Hurricane Irma? Did counties with high social vulnerability experience a greater number of administrative burdens? The researchers used qualitative document analysis to answer the research questions by analyzing after-action reports that counties created following Hurricane Irma and using social vulnerability data from the United States Centers for Disease Control and Prevention (CDC). The study revealed that although individuals faced all three types of administrative burdens, there were far more instances of psychological costs than compliance or learning costs. The researchers also found that administrative burdens were more common in counties with high social vulnerability, as determined by the CDC. Findings from this study can arm public administrators and emergency managers with insights on how to reduce such burdens during future emergencies.  

A growing trend over the past several years has been the formation and formalization of civilian volunteer rescue organizations. What once were emergent groups made up of spontaneous volunteers are now, in many cases, nonprofit organizations with a formalized structure in place. This has presented challenges to emergency management who are often unaware of how groups operate in the field and unsure how to best engage with them. Similarly, volunteer rescue organizations struggle with building lasting partnerships with emergency management. The authors of this study have conducted interviews with over 100 volunteers who do this type of rescue volunteering along with over 40 emergency managers or other official responders (e.g., fire, police) to understand what these groups do and the challenges of coordination with volunteers like these. The authors have also spent days in the field participating in rescue operations, training, and team building activities with some of these groups or online supporting their efforts. Through the Emergency Management Use Case program with the Texas Department of Emergency Management, the authors have partnered with emergency management officials to develop practical materials for both emergency management and volunteer rescue organizations relevant to how emergency management may better partner with volunteer rescue organizations in emergencies and disasters.

Human performance during compounding and cascading natural hazards is an important factor affecting the socio-economic impacts of these events. Fire following earthquakes (FFEs) is one type of cascading hazard which can occur when a single triggering hazard (an earthquake) results in failure of various ignition sources, such as a ruptured gas line, electric arcing, or overturned equipment. Research shows that human performance, including occupant response and first responders’ activities, can impact FFE risk. While several studies consider the risk-mitigating role of first responders’ activities and human evacuation following the occurrence of a fire in a building, limited research explicitly captures the impacts of structural and nonstructural damages on human performance. This study presents ongoing research by the authors focused on explicit modeling of human activities during compounding and cascading hazards. In this presentation, the authors report progress on identification and quantification of causal modeling of human activities during FFEs. A causal model is developed as a Bayesian Belief Network (BBN) model, where the conditional probabilities in the BBN model represent the strength of the links between the causes and effects. BBNs allow for the use of expert opinion and evidence-based reasoning for quantification of the causal model. To quantify the BBN causal model, information from existing literature as well as expert opinion are utilized. The developed causal model enables explicit integration of both occupant and first responders’ performance and associated uncertainties into risk models to support decision-making and inform resource allocation to improve human performance during these events. 

Extreme heat is the number one cause of weather-related fatalities in the United States. Community-based studies assessing local-level vulnerability and adaptive capacity are needed to understand how urban heat is distributed and provide more targeted mitigation and adaptation responses at a local level. This pilot study was conducted in Knoxville, Tennessee, to assess heat vulnerability and adaption. This study utilized mixed methods and collected (a) quantitative data from surveying 422 community residents about their experiences with heat, and (b) qualitative data from in-depth meetings with 16 community stakeholders to identify needs and solutions in extreme heat planning, preparedness, and response. Results from the community survey found that extreme heat risks were not distributed equally. Low-income and historically marginalized groups suffered the most from heat-related health issues and encountered barriers to cooling mechanisms. Results from the stakeholder discussions included four main themes related to challenges in extreme heat planning, preparedness, and response. These included (1) collaboration and coordination; (2) education and awareness; (3) access to resources; and (4) policies and procedures. Stakeholders also discussed several approaches to address these challenges, including creating a list of collaborating organizations and holding bi-monthly meetings, identifying trusted sources of information to provide targeted information and education, and mapping areas where cooling resources are located throughout the city. This case study highlighted the importance of including drivers of vulnerability at multiple scales and enhancing the ability of residents and stakeholders to work together to respond equitably to extreme heat. 

This study examined inequities in flood exposure and risk perception in Maricopa County, Arizona. Part One analyzed neighborhood-level data to explore the association between federally overlooked 100-year flood risk (flood risk not recognized by the federal government) and neighborhood indicators of social disadvantage, flood type, and flood zone delineation. Results showed that over 207,000 people in Maricopa County live in federally overlooked 100-year flood zones. Multivariable models indicated that neighborhoods with lower household incomes experienced disproportionate risk in federally overlooked flood zones. Part Two focused on parcel-level data from the Phoenix Area Social Survey and examined differences in flood risk perception between respondents in federally overlooked flood zones and Federal Emergency Management Agency (FEMA) flood zones. Results indicated that residents in federally overlooked flood zones had lower risk perception compared to residents in FEMA flood zones, even after adjusting for flood experiences. In addition, Hispanic/Latinx respondents had higher flood risk perception compared to White/Anglo respondents. This study highlighted the importance of analyzing flood risk perception between residents in federally overlooked flood zones and FEMA flood zones and underscored the need for targeted interventions to address inequities in flood exposure and risk perception in Maricopa County.

This mixed-method study delved into the lives of individuals affected by the devastation of Hurricane Maria, exploring their decisions to relocate and their experiences in the Relocation and Reconstruction Program (R3). With over 100 surveys and interviews, this study examined the impact of the hurricane on their homes and the factors that influenced their relocation choices, as well as their experiences within the R3 program. The survey also captured demographic information and investigated issues pertaining to equity. The results of this survey will not only inform decision-making to improve the relocation process for those affected by natural disasters but also to identify areas of inequity and ensure a more equitable process.

Providing actionable warnings and information to people working and living in areas prone to geohazards such as floods, debris flows, landslides, rockfalls, and avalanches presents a significant challenge. The ease of access to hydrometeorological data and current software technologies reduces the effort and costs associated with providing information on geohazards to people at risk. The challenge is to provide an appropriate level of warning and guidance to stakeholders given the uncertainty associated with predicting geohazard phenomena. In 2021, BGC developed a “Geohazard Situational Assessment Bulletin” to support construction work along a particularly hazardous region of British Columbia. The automated bulletin integrated weather, snowpack, and streamflow data to provide workers in the field with information on the recommended caution levels for the current day and 1-3 days in the future. Since then, BGC has developed bulletins for other construction projects as well as for communities including First Nations communities. This presentation will introduce lessons regarding the challenges and opportunities of providing and communicating actionable geohazard information to workers and communities through examination of several case studies. The need for a cross-disciplinary team of geoscientists, meteorologists, software engineers, emergency managers, and decision makers is highlighted.

Empirical research has studied the public perceptions of hurricane tracking forecasts by surveying residents as “end-users” about their perceptions and use of hurricane forecasts. However, there is little research on how business sectors, especially the critical infrastructure providers, interpret and use forecasts and warnings, and how they respond to forecasts and warnings. Meanwhile, the historical linear model of the hurricane warning system has evolved into a nonlinear, networked, multilayered, and complex information flow system for decision makers in the public, private, and other sectors, as well as individual residents and businesses, which has led to a collective approach to making the best use of hurricane forecasts to reduce vulnerabilities and damages. This project stemmed from the authors’ research examining cross-sectoral partnerships in enhancing community resilience. It fills a research gap of how science-practice partnerships between weather forecast researchers and critical infrastructure operators can advance collective decision-making to reduce community service disruptions during and after hurricanes, improve the emergency planning and business continuity of critical infrastructure sectors (e.g., electric power, fuel, and roads) before, during, and after hurricanes, and build the interdependent infrastructure system resilience to disasters. An ‘end-user’ survey of critical infrastructure operators was discussed for weather forecast scientists and operational meteorologists to better customize their forecast products and applications for varied industries providing power, fuel, and transportation infrastructures and services to the communities they work with in Florida and other coastal regions.

Phenomena such as climate change, COVID-19, long-term environmental pollutants, and urbanization increase the probability of hazards occurring simultaneously or sequentially— creating a compound hazard. How governments plan for compound hazards needs to differ from how they plan for independent hazards, but there is little research on whether or how this is done. This study examined the depth of formal and informal state planning for compound hazards across the United States and emergency manager experience during COVID-19 and coincidental disasters. State emergency management plans were evaluated based on whether and how they addressed simultaneous hazards, cascading hazards, and environmental justice. Results indicated that 45 of 51 state and territorial disaster plans mentioned compound hazards in some capacity in at least one sentence of their plan. Only 22 discussed the implications of compound hazards or established a plan to deal with them, and only seven included more than one sentence of actionable procedures. Interviews were then conducted with state emergency management planners from 21 states to confirm the plan analysis results and discuss their experience handling COVID-19 and other disasters concurrently. Key themes identified included the need for more cross-training in emergency management positions for personnel, clearer roles for political actors and plans for how to handle politicized disasters, mental health resources to combat trauma and burnout, sheltering plans for rural areas, increased coordination with all other units no matter their seeming relevance, and the need for plans for long-term hazards.

How can the tradeoffs of coastal adaptation strategies be better characterized to promote equitable outcomes? Coastal communities face unique challenges to manage the impacts from climate change, like sea level rise (SLR) and storm surge. There are multiple ways to adapt, which may broadly be characterized as protection (e.g., sea walls), accommodation (e.g., elevating homes), and retreat (e.g., buyouts and relocation). The choice of which strategy to implement is context specific, value-laden, and involves multiple stakeholders. Preferences toward adaptation strategies vary across groups. This study used mixed methods and a case study approach to evaluate adaptation strategies for the barrier island community of Ortley Beach, New Jersey, which has suffered severe repetitive losses. Following prior research, the authors conducted 25 interviews with households, local officials, and state and federal decision makers to explore how values relate to preferences for adaptation strategies in Ortley Beach. The outcomes from the interviews identified key values people associate with different adaptation strategies. Key values that emerged included economy, social equity, and effectiveness. Three policy pathways were explored for managing at-risk properties in Ortley Beach: (1) elevate all homes, (2) piecemeal buyouts and elevations, (3) buyout all homes. Using both qualitative and quantitative techniques, the study characterized the pathways according to the salient values elicited from the interviews. Findings highlighted points of conflict and consensus, supporting action-oriented research relevant to a broader audience of scholars and practitioners involved in coastal adaptation planning.

That relocation sites are not camps is inferred by their discursive framing as long-term solutions where residents are safe from natural hazards and integrate into the social, political, and economic fabric of their new communities. Yet, the socio-spatial restrictions imposed upon residents and the latter’s utilization of these spaces belie this claim. This paper conceptualizes post-disaster relocation sites as camps, arguing this framing allows for scrutinizing the governing practices that persist in shaping both socio-spatial and ecological relations and subjectivities of disaster survivors long after a “solution” to their initial displacement has supposedly been attained. This privileging of “incompleteness” over “finishedness” better explains the (in)tangible forces that shape recovery and the regulations, negotiations, and resistances that transpire within these spaces. Drawing on empirical studies of three post-disaster relocation sites in the Philippines and Japan, this paper examines various forms of camp governance utilized by state, faith-based, humanitarian, civil society actors, and by residents themselves. Deploying three lenses (politics of loss, spatial forms and practices, and resistance), the authors interrogate the assemblages of rules that shape resettlement site life, the unrecognized modes of meaning-making etched into post-disaster spaces, the affects, emotions, and aspirations that are entangled in socio-spatial practices and encounters with material forms, and the (subtle) forms of resistances to prescribed “ways to live.” Examining post-disaster relocation sites as camps thus contributes to surfacing the continuing struggles and acts of agency of disaster survivors done in pursuit of their aspirations—elements that would otherwise be neglected or worse, forgotten. 

In this study, we discuss the Chain of Consequences method and its applicability to assessing the cascading consequences of, and potential interventions to, post-fire hazards. This method, which was first developed and applied by the US Department of the Interior Strategic Sciences Group, involves bringing together a diverse team of multi-disciplinary experts to map cascading consequences of hazards and develop novel interventions. To evaluate and discuss the Chain of Consequences method, we begin by reviewing its prior applications. Then, we apply the method in a workshop convening stakeholders with context-specific, scientific expertise in Okanogan County, Washington in response to two federally declared Major Disaster Wildfires. These fires, the 2021 Cedar Creek and Muckamuck Fires, have demonstrated and continue to be at risk of the cascading consequences of fires and post-fire hazards, including debris flows and concerns related to water quality, for example. After applying this method to post-fire hazards, we investigate its usefulness outside of its traditional emergency “triage”-like applications. Overall, the Chain of Consequences method involves diverse teams of stakeholders identifying consequences and interventions that transcend disciplinary boundaries to reduce disaster losses and build resilience. Thus, this method had the potential of contributing to valuable interdisciplinary convergence research.  

In 2022, the longitudinal study in Lumberton, North Carolina, aimed to investigate the ongoing recovery and mitigation status after Hurricanes Matthew (2016) and Florence (2018), and the impact of the COVID-19 (2020) pandemic on business recovery processes, mitigation, and preparedness actions. This is the fourth survey wave to be administered to a random sample of businesses affected and unaffected by the floods. This recent survey focused on recovery status, operating capacity, recovery funding, insurance coverage, primary concerns, preparedness and mitigation strategies, and recovery financial preferences. The results from the fifth wave suggest that the COVID-19 pandemic had a significant impact on business recovery, with only 34% of affected businesses having fully recovered. The study also investigated the financial preferences for recovery funds of individual businesses. A discrete choice activity was designed to determine preferences for locally funded disaster assistance loans. Respondents generally preferred smaller loan amounts with faster disbursement periods, though in most cases they did not wish to apply for an assistance loan. The longitudinal findings show the complex nature of recovery from consecutive events, as 80% of the businesses have recovered from Hurricane Matthew, but only 36% have recovered from Matthew, Florence, and COVID-19. Future work should focus on businesses still recovering, as well as observational data on business entries and exits. This study provides insights into the ongoing recovery and mitigation efforts in Lumberton, North Carolina, and the importance of financial assistance tailored to businesses’ specific needs during and after disasters. 

Enabling resilience to natural hazards and disasters will require adapting or redirecting a variety of very complex global and local human systems. Scholars and practitioners need collaborative tools to understand the dynamics of these systems and the key drivers of their behavior, such as barriers to progress and leverage points for driving sustainable change. This talk developed an approach that tailors system dynamics tools to data-poor environments with fragmented knowledge, like those encountered in international development and in many frontline communities. Specifically, the causal loop diagram (CLD) was extended with a data layer that enables analysis of a system’s dynamics and behavioral drivers without the extensive data or broad assumptions required for a simulation. The approach is transparent and invites broad collaboration. It was developed through a 4-year partnership with USAID/Uganda, with participation from many local Ugandan organizations. This talk illustrates the approach through two applications spanning international development and disaster resilience. The first showed that a lack of demand for loans, rather than physical access to financial institutions, is the main barrier to broader loan usage by rural smallholder farmers. The second identified in real time the likely resilience to and propagation of COVID-19-driven shocks through the agricultural market system. Both analyses suggested high-leverage directions for future USAID investment. This talk will reflect on how the approach enabled collaborative analysis of complex systems in Uganda and consider how it may be further developed to support future research on community disaster resilience in a wider array of settings.

National and local Disaster Risk Reduction strategies struggle to include the voices of people with disabilities, despite the disproportionate exposure of this community and the requirements of international legal frameworks (starting from the Sendai Framework for Disaster Risk Reduction 2015–2030, or SFDRR). Multiple institutional and research papers have primarily focused on analyzing field initiatives and technical tools. Instead, this research examined the implementation gap through public policy lenses, trying to answer the questions of whether policymakers are adopting the necessary Disability-inclusive Disaster Risk Reduction (DiDRR) strategies and, if not, what framework would help bridge this gap. The research is based on quantitative-qualitative policy analysis (e.g., Stakeholder Analysis, Forcefield Analysis, Network Maps), supported by the latest official data available (e.g., World Risk Report) and global tracking tools (e.g., Sendai Monitor indicators, United Nations and Organization for Economic Cooperation and Development databases). Findings indicate an insufficient number of national and local policies and a limited level of inclusion along the formulation process. This paper suggests a policy framework for adopting DiDRR strategies based on engaging all the relevant stakeholders (starting with persons with disabilities and their organizations), supporting the process with a distributed governance model and inclusive budget, pursuing a cross-sector approach and leveraging the enabling factors. The essay provides viable suggestions related to four main aspects: policy adoption, policy formulation, budgeting, and governance. Finally, it outlines opportunities for the improvement of the SFDRR policy tracking tool and points out a possible blind spot in the Framework.

This Faculty Early Career Development (CAREER) grant will yield improved understanding of the mechanisms that cause disparities in vulnerability between affordable primary housing and second homes-vacation rentals in United States coastal areas. Growing population and development in many coastal areas, combined with rising sea levels, more intense and frequent storms and flooding under climate change, mounting economic inequalities, and aging infrastructure all pose increased risks to coastal housing stocks and, in turn, economies. In many coastal areas of the United States, affordable primary housing is shrinking, aging, and deteriorating while seasonal vacation housing is growing and improving. Affordable housing is disproportionately damaged by storms, displacing residents who are more likely to be poor, elderly, racial and ethnic minorities, and service industry workers. While disparate patterns of vulnerability are visible, the mechanisms underlying them are little understood, which hinders improvements to housing policies. This project will integrate theories and techniques from multiple fields to explain how development policies and funding mechanisms within the context of tourism economies create and exacerbate disparities in housing vulnerability. This project will engage students from underrepresented groups as well as students drawn from affected local communities in disaster research, outreach, and policy development. It will also assist planners and policy makers by exploring the potential effectiveness of alternative housing programs and policies.

Housing repair and reconstruction are among the strongest indicators of a community’s recovery after a disaster. The literature has shown that the rate and the completeness of housing recovery have consequences for economic, commercial, household, and emotional recovery after disasters. Housing restoration and recovery pose a large financial burden on households, and access to financial resources—in the form of insurance payouts, federal aid, loans, and assistance from Non-Governmental Organizations (NGOs) or family and friends—determine whose home and which neighborhoods recover more quickly. This paper provides an in-depth quantitative analysis of the impact of financial resources and assistance on housing recovery based on housing surveys conducted after Hurricane Matthew in Lumberton, North Carolina. This study attempted to answer two key questions: Does access to recovery resources (in terms of types and timing) correlate with damage levels and household social vulnerability? Do resources lead to homes recovering faster after controlling for different damage levels and social vulnerability? The analysis provides insights into disparities in access to financial resources, types of resources accessible to households, and how quickly they receive financial assistance. The findings show variations in the effect of financial resources—insurance, federal and non-federal assistance—on housing repair and re-occupancy times. Importantly, the results also show that housing recovery progress is determined by the socioeconomic status of residents and the level of damage to the house. The research offers policy suggestions to improve post-disaster housing recovery in a more timely and equitable manner.

Access is one dimension characterizing social equity in emergency management. Ensuring that all hazard-affected communities have access to disaster recovery information is not only an indication of how socially equitable a local emergency office’s practices are, but also how ethical they are. Moreover, the notion of equitable access to information has heightened importance to ethnic and racial minority groups, such as Latinos, because of their relative need in the aftermath of hazard events. As such, this case study investigated the use of Twitter by county-level offices of emergency management in Florida during and after Hurricane Irma and assessed the social equity of access to disaster recovery information related to the Floridian Latino population. Recommendations are made to enhance crisis communication practices and future research.

In the United States, the most widely used social vulnerability indices incorporate a set of indicators designed to measure social vulnerability to hazards in general. However, research has shown that because the social processes that drive vulnerability vary between hazard and place, using indicator sets that are tailored for specific hazards and places will better reflect social vulnerability. This study sought to better understand the relationship between commonly used social vulnerability indicators and flood exposure, and how that relationship changes across census tracts in the conterminous United States. Multiscale Geographically Weighted Regression is used to examine the relationship of social vulnerability indicators to flood exposed census tracts and then to identify the spatial variations in that relationship. Understanding the spatial variations in the relationship of indicators to flood exposure can help inform and refine the development of social vulnerability indices to better identify populations at risk of adverse flood outcomes.

The Federal Emergency Management Agency (FEMA) does not consider dam breach inundations when it creates its flood risk maps, potentially leaving additional flood-prone areas overlooked. FEMA has also been shown to unfairly allocate more payouts to claims filed in predominately White and wealthy communities. Furthermore, loss of life and damage costs tend to be the only indicators used for measuring a dam failure’s impact, leaving demographic disparities unexplored. Acknowledging these oversights, it is necessary to determine if FEMA assistance is equitable after a dam failure disaster and ensure that those impacted and socially marginalized can recover effectively. Using the 2015 dam failure disaster in Columbia, South Carolina, as a case study, this research assessed if the distribution of FEMA claim payouts was equitable when regarding race and class. The author obtained and filtered relevant FEMA claim data for the disaster using R and ArcGIS Pro. The author then categorized the claim data by census tract and linked it to corresponding race and median household income data. After conducting a comparative analysis, results revealed that predominantly Black and lower-income census tracts had higher percentages of unpaid claims. This disparity implies that FEMA’s assistance was inequitable and thus could have excluded those who would need assistance the most due to social barriers. This research not only introduced an alternative way to assess the impact of dam failures but also provided findings that support the need to consider additional and more detailed impacts that tend to be overlooked.

During a natural disaster, people post text messages via various channels, such as social media platforms and short message service, to seek help and share information. These text messages often contain critical descriptions about the locations of victims and accidents, and accurately extracting these location descriptions can help disaster responders reach victims more quickly and even save lives. Recent research has shown that many of these location descriptions go beyond simple place names, such as city names and state names, and cannot be extracted using typical named entity recognition tools. While machine learning models can be developed for extracting these location descriptions, they generally require training data with labeled location descriptions, which are often not available. This paper presents GALLOC, a GeoAnnotator for Labeling LOCation descriptions from disaster-related text messages. As a geo-annotation tool, GALLOC provides functions that enable the effective annotation of three components of location descriptions, namely description texts, location categories, and spatial footprints. This work demonstrates the functions of GALLOC and was evaluated by comparing it against other previous geo-annotation tools.

The science associated with assessing natural hazards and informing decision makers contains many layers of complex and interacting uncertainties. This is compounded by the evolving nature of response needs and the changing network of communication pathways. Further, varied understanding about what scientific uncertainty is, and where it comes from, affects people’s trust in and use of science advice. It is crucial to understand these diverse perspectives to improve effective communication.

Twenty-five mental model interviews were conducted in Aotearoa New Zealand with a range of participants including physical scientists, policy writers, emergency managers, and the public. Three-phase interviews included initial elicitation of free thoughts about uncertainty, a mental model mapping activity, and a closing phase of semi-structured questions to further explore participants’ thoughts on scientific processes and their philosophy of science. Qualitative analysis led to the construction of key themes, including: (a) the importance of recognizing the role of human behaviors, (b) understanding the “actors” involved as sources of uncertainty alongside data sources, (c) acknowledging influences such as governance and funding decisions, and (d) the difficulty that most participants had in defining what uncertainty actually is.

Participants highlighted the positive role of uncertainty for promoting debate and as a catalyst for further inquiry. They also demonstrated a level of comfort with uncertainty and advocated for “sitting with uncertainty” for transparent reporting in advice. These findings are presented to enhance hazard and risk communication, alongside the design of our interview methodology, which could be adapted for participatory and co-development research.

After Hurricane María devastated Puerto Rico in 2017, the SLIDES-PR (Storm-induced Landslide Impact Dynamics on Environment and Society in Puerto Rico) project was launched in the University of Puerto Rico Department of Geology to document the topographic effect of the tropical cyclone across the island’s mountainous terrain. This effort led to (a) the development of an event inventory of 70,000+ mass wasting sites, (b) the production of a modern high-resolution rainfall-induced landslide susceptibility map for PR, (c) the development, publishing, and distribution of a Spanish-language landslide guide for residents of the island, (d) analyses of the short- and long-term effects of landslide-driven sedimentation on the island’s fluvial network and vulnerable reservoir system, and (e) establishment of a monitoring network with the goal to develop landslide forecasting capabilities.

The monitoring network was made possible via a formal agreement between the USGS Landslide Hazards Program and the University of Puerto Rico and currently includes 15 monitoring sites. Each station measures volumetric soil water content, soil suction, soil groundwater pressure, rainfall, and other environmental conditions. Near real-time data for each station are transmitted to our data center and are updated and publicly available via derrumbe.net. Once empirical hydrological soil and meteorological thresholds are established for the stations, the network is upgraded to be an operational forecasting system. We envision products will include 1-, 2-, and 3-day landslide forecast outlooks that can be used by diverse local stakeholders that include the weather service forecast office, state and municipal emergency managers, transportation officials, and others.

This study developed a Tribe-focused wholistic resilience planning framework for climate-influenced hazards, and an integration model for making connections and capturing potential synergies with neighboring jurisdictions and other entities. A qualitative approach involving interviews and workshop-based focus groups are used to engage members of Tribal communities in co-producing a resilience assessment that is tailored to the needs and contexts of Tribes, developing a resilience plan based on this assessment, and building an integration model that leverages and captures synergies across various partners, resources, and other planning strategies. Findings showed that The Resilience Adaptation Feasibility Tool (RAFT) Scorecard and resilience planning framework meets the basic needs of Tribal communities but requires adjustments to account for the role of institutional knowledge, the specific governance structure, cultural assets, and environmental and health interactions of Tribes. The RAFT process showed that housing, food security, and mental health, are linked to Tribes’ efforts and capacities to raise awareness, conduct outreach, and access funding for resilience. While personnel, resources, and governance communication pathways varied amongst Tribes, sources of funding for resilience were similar. Further, proximity in location contributed to collective action over shared concerns. The public health contributions of the study take the form of: (a) offering a wholistic resilience planning framework that prioritizes Tribes’ health-hazard priorities; and (b) developing a model for engagement to support resilience through integrating Tribal and external resources in the context of previous or existing activities.

In recent years, the importance of Community-Based Disaster Response Organizations (CBDROs) has increased in the United States and Japan, with new roles and structures emerging in both countries. This research explores the roles of community-based organizations in disaster prevention and response via a case study comparison of two examples: the Seattle Emergency Hubs in Seattle, Washington, and Sendai Bosai Leaders in Sendai, Japan. Through interviews and document analysis, we seek to understand the following: (a) how CBDROs respond to the shift toward community-scale disaster preparedness and, (b) the similarities and differences between government response and voluntary organizations, with a focus on successes and challenges. In both cases, the roles of CBDROs and that of government officials are seen as complementary, with recognition that in a catastrophic event, residents cannot expect to be rescued by officials alone. In the case of Sendai, this is based on the experience of the 2011 Great East Japan Earthquake. In the U. S. Pacific Northwest, a similarly-scaled event—a magnitude 9.0 Cascadia Subduction Zone earthquake—is anticipated to occur in the foreseeable future. We find that while volunteer-run organizations play significant roles in both countries, their functions and relationships to official government agencies vary greatly, as do the ways in which they are networked with other neighborhood associations. Comparing the experiences of Seattle Emergency Hubs and the Sendai Bosai Leaders provides insight into the successes of, and challenges faced by, local CBDROs, highlighting opportunities for learning across experience.

The challenges facing Florida’s property insurers have been part of Florida’s political landscape since Hurricane Andrew in 1992, but they have grown more acute with increased cyclonal activity, sea level rise, risk exposure, and replacement costs. Moreover, the state’s legal environment, buttressed through “roof chasing” lawyers and public adjusters, has put insurers at risk for claims of scope and size unique to Florida. Florida’s political economy has not yielded solutions that reconcile actuarial necessity with affordable property insurance. Florida’s conservative policymakers seek market solutions to the state’s property insurance dilemmas, but the last quarter century shows that reforms are difficult to achieve. The insurance market mess is truly a “wicked problem” in lacking agreement on the cause of the issues and consensus on immediate solutions. Conflicting values and policy perspectives, uncertainty about the impact of potential policy options, and political considerations create an environment in which incremental changes are advanced as temporary stopgaps, without a long-term solution in sight. The authors of this study’s research described this challenge with the understanding that despite the motivation for fiscal sustainability through “rate adequacy” and “actuarial soundness”, consumer pressures and political pragmatism preclude major sustainable reform to that effect.

Existing research has made significant strides in studying many aspects of emergency managers’ decision-making processes, but there have been limited discussions about the various factors that emergency managers consider for issuing hurricane evacuation orders. This study aims to provide a more robust understanding of the community, institutional, and other elements involved in emergency managers’ decision-making processes when issuing hurricane evacuation orders. We used the principles of the theory of bounded rationality to formulate research questions for understanding decision-making during uncertain times (i.e., hurricane evacuation orders). We then conducted 20 semi-structured interviews with county emergency managers in Florida to understand how this decision-making process unfolds. Results showed that emergency managers consider two main factors (random and fixed) that would blend in with other elements (planning, collaboration, and information assessment) during the response phase of an emergency. The interplay among these elements can ultimately influence emergency managers’ hurricane evacuation decisions. The findings of this study confirm and contribute to the theory of bounded rationality by discussing the role of environmental and community factors in decision-making uncertainty.

The recent increase in catastrophic wildfires in the United States has brought a concern about managing this risk while preserving the socioeconomic and cultural fabric. Unfortunately, most communities at risk of experiencing wildfires in the United States do not have any wildfire protection plans to prevent disaster, even as the Wildland Urban Interface (WUI) rates grow. This combination of increasing wildfire risk and severe under-preparedness raises questions about viable options to mitigate against and adapt to wildfires in these communities. Existing wildfire management practices tend to emphasize mitigation actions such as fire suppression over adaptive land use planning. The lack of attention paid to community-scale, land use-based wildfire adaptation may be attributable to various socioeconomic, cultural, and institutional reasons that have not been deeply explored. The challenges to land use-based wildfire adaptation are even greater for unincorporated areas with poorer records for land use planning. Using the case of the 2021 Marshall Fire in Colorado, this study explores the extent to which unincorporated communities adopt land use-based wildfire adaptation strategies and the challenges and opportunities for adopting such strategies. Study data collection was conducted in two parallel phases: a review of secondary documents related to wildfire management in the unincorporated area of Colorado, and key informant interviews with 20-30 government officials, representatives of nonprofit organizations, and local community leaders within Boulder County. Findings from the study will improve understanding of the challenges and opportunities of adopting land use-based strategies for wildfire adaptation in unincorporated areas of the United States.

As disaster and environmental scholars work towards addressing the impacts of human-induced climate change, the need for research and researchers representative of communities most at-risk of the effects of disasters becomes imperative. While efforts have been placed into identifying gaps and diversifying the field, many studies have fallen short of exploring the compounded, intersectional factors which contribute to the barriers faced by historically marginalized groups from engaging with National Science Foundation-funded resources and support. Utilizing theories of social capital and intersectionality, this study was driven by the following research questions: (a) What are the experiences of historically underrepresented or marginalized scholars or institutions engaged in hazards and disaster research?, (b) What barriers do historically underrepresented minority or marginalized scholars face when accessing or using existing NSF-supported risks and disaster-related services and resources?, and (c) How can NSF-supported disaster and hazard-related services and resources better support or prioritize the needs of historically underrepresented or marginalized scholars or institutions engaged in hazards and disaster research? To address the research questions, the research team conducted an exploratory study utilizing one-on-one semi-structured interviews with 13 scholars representing 14 institutions. Results show that marginalized scholars and students face many compounding barriers, including institutional and financial obstacles, low administrative capacity, and an exclusive, gatekeeping disaster science culture. Participants provided recommendations ranging from acknowledging systemic barriers like racism and sexism and incorporating mentoring and training opportunities for early career professionals.

As the frequency, scale, and intensity of disasters continue to increase, building and enhancing resilience to disasters has become a critical planning, engineering, policy, and governance challenge. Of particular importance is infrastructure system resilience to disasters because critical infrastructure systems support public safety and the continuity of operations of governments and businesses. Infrastructure system resilience is also important for effective recovery. Infrastructure systems and services are increasingly becoming more interconnected and interdependent. The authors of this study presented and applied a network governance perspective to examine the interactions and coordinating processes, mechanisms, and structures among key stakeholder groups that manage and operate infrastructure systems. The authors examined their application and potential significance to flood disasters in a complex system of intergovernmental and interorganizational settings in Australia, especially as compared to the counterpart in the United States. The authors examined the New South Wales Flood Inquiry Report, Legislative Council Report, and various government response documents related to the role of organizational and governance networks to explore how these networks could contribute to a better understanding of how to govern interdependent infrastructure systems to enhance a community’s resilience to future floods. The research utilized network analysis tools to assess the effectiveness of coordination efforts and identify potential gaps, especially regarding the need to leverage collaborative leadership and organizational capacity to develop robust and connected networks and enhance stakeholder engagement.

Heat vulnerability and homelessness are central public health concerns, and their intersectionality results in preventable heat-related morbidity and mortality, undermining health equity. Those facing unsheltered homelessness in cities globally seek shelter in tents with shading features, used by local authorities and welfare organizations to decrease unsheltered heat exposure during the extreme summer months. This study examines the effectiveness of different tent cover materials to reduce air temperature (Tair) exposures during hot summer days. We monitored within-tent Tair using Kestrel Drop devices across three typical August days in Phoenix, Arizona. Air temperatures were statistically compared between six side-by-side identical tents with different cover materials: mylar, white bedsheet, tarp, Sunbrella fabric, and aluminum foil as well as ambient Tair. The control had no cover. Results indicate that ambient air temperature conditions were lower, on average, than the mean within-tent air temperatures during combined day-and-night, and daytime-only time classifications. Reasons for higher overall Tair include reduced ability for heat and vapor to escape the tent when the cover is on and reduced ventilation. Hence, adding cover materials over tents for shade can increase heat risk of an already vulnerable population. Local authorities and welfare associations need to investigate the use of tent shade at various times of day as a common heat exposure strategy to ensure that prevention efforts do not further marginalize the most vulnerable. Policies and guidelines around reducing heat vulnerability should be reviewed periodically to reflect research findings about the most effective passive personal heat reduction interventions.

Historically, tornadoes and earthquakes have been common in Oklahoma; however, in the case of heightened wildfire risks, the risk perceptions of the vulnerable population vary significantly in different geographic areas. Such variations in the risk perception of different stakeholders may result in communication and coordination failure while responding to major disasters, especially in marginalized communities. This study developed a decision-support framework for stakeholders that integrated household-level risk perception data collected from the Oklahoma Meso-Scale Integrated Socio-Geographic Network (M-SISNet) survey (N=3,449). The survey was designed to capture the Oklahomans’ perceptions of changing weather patterns in the state, as well as residents’ views on government policies and societal issues, to help understand how perceptions and views might shape infrastructure (i.e., water, energy, transportation) use and priorities. Risk perception was measured on a Likert scale from one (i.e., no risk) to five (i.e., extreme risk). The geo-tagged households in the survey allowed integrating community-specific factors into the risk perception modeling and analysis by using national databases such as American Community Survey (ACS). The study revealed key household and community level factors that govern the differences in public risk perception for a diverse set of hazard events (i.e., tornadoes, wildfires, flooding, drought) in Oklahoma. The study also provides new policy insights on community resilience and informs disaster literature regarding how marginalized communities can be better served in future disasters.

Throughout the COVID-19 pandemic, there has been a huge investment in—and deference to—various quantified ways of understanding the pandemic. In technical scientific advisory venues, computational modeling of disease spread, and prospective interventions, has played a powerful role in shaping policy. In broader public discourse and political arenas, this modeling has been complemented by significant public opinion polling that provides rapid and precise feedback on public acceptance of different measures. Both forms of quantified data are appealing to decision makers for easy and rapid accessibility, claims to objective certainty, and the utility of framings offered, regarding both the problem and solutions. This research explores ways that quantified framings have influenced lived experiences and pandemic outcomes. More specifically, the authors argue that these framings have often been deployed in ways that “short circuit” inclusive forms of deliberation about the underlying values, priorities, and solution spaces available to meet pandemic challenges. To reflect on these challenges, the authors draw on data from multiple sources, including public opinion polling throughout the pandemic, scientific modeling made public, and first-hand experience in a science advisory role. They argue that hyper-quantified ways of knowing have narrowed understanding of the pandemic, constrained prospective solutions, and continue to limit creative responses to solving systematic challenges.

A community with “near experience” refers to one with a high hazard risk but has only experienced low-to-moderate type disasters or indirect damages caused by a nearby affected community. There is still a lack of knowledge about pre-disaster housing planning, especially in communities with near experience. There is a lack of studies examining how and when public housing residents, a highly vulnerable social group, recover after disasters. Considering the high dependency on government assistance for recovery in the aftermath of disasters, this gap is critical. Salt Lake and Utah Counties, Utah, are used as case studies to examine policies designed to provide post-disaster housing to public housing residents. This study combined semi-structured, open-ended key informant interviews with 22 local emergency management, city planning, and housing officials of both counties and their constituent cities to examine how these plans were being implemented, what motivated them, what challenges they faced, and how they coordinated with other entities. Findings suggest cities were not prepared to house public housing residents after a disaster with an absence of disaster assessments, poor coordination and information-sharing, insufficient funding, and poor leadership. Disaster planning should be promoted for public housing recovery by creating joint plans, engaging diverse communities in planning processes, committing funding for such activities, and reviewing existing protocols.

The United States is currently experiencing an affordable housing crisis, compounded by an increase in climate change disasters. Disasters damage and destroy housing, and disproportionately affect marginalized groups, such as renters living in subsidized housing units. Many of these units are managed by public housing authorities with widely varying responsibilities, including roles in disaster risk management. However, there is limited information about if and how public housing authorities implement disaster risk management strategies to mitigate negative outcomes after disasters. This study describes barriers, facilitators, and opportunities for public housing authorities to implement disaster risk management strategies. Semi-structured key informant interviews were conducted with federal housing officials, public housing authority leaders, and non-profit staff, and the Framework Method for qualitative research was used to mathematically analyze transcribed interviews. A total of 20 key informants were interviewed during 15 interviews. We identified several barriers and facilitators to the engagement of public housing authorities in disaster risk management, as well as practical recommendations to improve public housing authority disaster preparedness. Funding, administrative hindrances, lack of federal government direction, and an absence of disaster expertise among staff were notable barriers to engagement in disaster risk management. However, past disaster experience, strong local relationships, and existing internal disaster plans were common facilitators for implementing disaster risk management strategies. Interviewees described an opportunity to create a dedicated emergency management position at public housing authorities to support engagement in disaster risk management. 

The National Institutes of Health (NIH) has a long history of supporting research in disaster settings facilitated through the Disaster Research Response (DR2) Program. The new NIH Climate Change and Health Initiative aims to expand activities around understanding the health impacts of climate change. To better understand the history of NIH investments and research at the nexus of these two fields, a portfolio analysis was conducted to examine the corpus of climate-related disaster research grants. Funded climate-related disaster grants from 2010-2022 were identified and tagged for several factors, including: funding level, funding institute, disaster type, geographic location, health outcome, exposure agent, special population, age group, occupation, and funding institution. Trends in funding and multiple study characteristics were analyzed. Over 150 projects identified as “climate change research” involved climate-related natural disasters. The most prevalent climate disasters studied were hurricanes, general disasters, and other extreme weather events. Much of the research used data from a specific disaster event to gather general post-disaster health effect findings. Additionally, the disaster event itself, rather than a specific chemical or biological exposure agent, was often the exposure of interest in studies of health impacts. Many opportunities exist to expand disaster research, for example: studying a broader range of events, especially compound disasters; investigating exposures and health outcomes among certain at-risk populations; and longer-term mental health outcomes and protective and adaptive measures.

This study focuses on the role of social capital in the resilience of Asian American-owned small businesses in New York City after the COVID-19 pandemic. Despite the significant contributions of Asian American-owned businesses to local economic growth, their disaster recovery experience has not been extensively studied, and recent studies suggest that they have faced racism and social stigma and received less financial assistance and policy attention than other groups. This study aims to examine the extent of, forms of, and uses of social capital in Asian American-owned business recovery and to explore the opportunities and challenges of cultural and racial factors that facilitate or hinder social networks and solidarity toward building business resilience. The study involved face-to-face, semi-structured in-depth interviews with 20-25 Asian American business owners through a snowball sampling method. Thematic content analysis was used to analyze the data and identify key themes such as trust and reciprocal relationships, the role of a sense of belonging in recovery decisions, networking activities, and use of group membership, and factors related to culture and heritage. The project focused on understanding the importance of cultural and ethnic contexts such as trust, norms of reciprocity, the sense of belonging, network structure, and cultural beliefs during the recovery process. The findings of this study can help disaster planners and policymakers design policies that consider all the resources Asian American-owned businesses have to facilitate recovery and promote diversity and inclusion in the disaster recovery process.

Surrogate models, also referenced as metamodels, have emerged as powerful data-driven models that can offer computationally efficient and accurate predictions for storm surge estimation. They are calibrated based on an existing database of synthetic storm simulations and can provide reliable approximations of the expected storm surge, replacing the computationally expensive numerical model used to originally establish this database. The metamodel input is the parametric features that can be used to uniquely describe each storm within the available database (wind field, storm size, speed, and directivity). The output of interest, in this case, is the storm surge (peak values or time-history evolution). The development of a Gaussian Process (GP) metamodel for the prediction of storm surge time-series is outlined here, comparing some existing approaches relying on dimensionality reduction techniques and the development of multiple metamodels to the development of a single metamodel to predict across the augmented space that includes spatio-temporal and storm features. To circumvent the emerging computational challenges in the GP formulation associated with factorizations for the large dimensional covariance matrix across the augmented feature space, a separable GP formulation is adopted, treating separately the kernels and associated covariance matrices across the storm feature, spatial, and temporal spaces. The two approaches are tested in a case study for North Atlantic where time series predictions are offered for over 12,000 locations of interest simultaneously.

This presentation describes the opening chapter of my (in progress) book manuscript which articulates the political production of socio-ecological disaster. In this presentation, I consider the history, symbolism, meaning, and magic of chicory, the small blue flower that gives my hometown, Bluefield, its name and which is also culturally significant to Louisiana as the bittersweet coffee concoction that was born of innovation amidst scarcity. Chicory, a plant that grows in disturbed places and wastelands reflects the qualities of the people who have made their homes in the extractive terrains of Appalachia and the Gulf Coast, both sites of my fieldwork, both sites of extractive harm, and both sites of resilient communities. Through this presentation, chicory helps us to consider the perennial power of people and place to root, resist, and remain interconnected amidst extractive logics, policies, and practices. The wisdom of chicory invites us to look carefully to the past to understand the diminishing potential of human over nature relations, to the radiant beauty of the present moment, and the budding future that remains yet determined in this time of anthesis.

Experiencing a tornado could profoundly impact a person’s mental health. Demographic factors, socioeconomic status, and pre-tornado trauma experience may be risk factors for psychological issues after a tornado. The mental health of persons with disabilities is disproportionately affected by extreme weather events such as tornadoes. This study examined correlations between tornado exposure, protective resources, depression, and post-traumatic stress disorder (PTSD) among people with disabilities. An online survey was completed by 209 people with disabilities in May 2022, about five months after the Midwest and Southeast tornado outbreak on December 10 and 11, 2021. Participants were 18 and older and lived in one of the 64 counties affected by the tornado outbreak. The results showed that the degree of tornado exposure and the level of protective resources were significantly related to post-tornado distress. Tornado exposure was found to have a significant and positive relationship with PTSD symptoms and depressive symptoms. People with disabilities who had higher levels of physical, social, and emotional protection resources had a significant and inverse relationship with PTSD symptoms and depressive symptoms. These findings suggest that tornado exposure may be associated with long-term mental health problems among people with disabilities. They also highlight the essential role of both internal (e.g., problem-solving, distress regulation, optimism) and external (e.g., physical and social) protective resources in contributing to lower risk for adverse post-disaster mental health outcomes among people with disabilities.

Recent urbanization around the globe shows a tendency toward high-rise buildings, and South Korea is no exception. In spite of socio-economical gains from these developments, the downside is so-called building wind, not only during typhoon seasons, but also during everyday life, which often threatens the structural integrity of the buildings and pedestrian safety. The wind environment changes not only through the overcrowding of cities and high-rise buildings, but also due to the reconstruction of old buildings. In order to reduce the potential hazard imposed by changes in the building complex, the wind environment should be assessed from the building design stage. The purpose of this study was to analyze the impacts of building winds on the pedestrian wind environment in the surrounding area that may occur as an existing low-rise residential complex was reconstructed into a middle- and high-rise apartment complex. The wind speed and directional characteristics were analyzed and pedestrian wind environments were generated using computational fluid dynamics (CFD). Using computed results, the spatial distributions of the probability of wind speeds exceeding certain thresholds were compared to show the impact of the reconstruction of the building complex. This research was supported by a grant (RS-2022-00155691) of Disaster-Safety Industry Technology Commercialization R&D Program, funded by Ministry of Interior and Safety (MOIS, Korea).

Protecting people who live in subsidized housing from natural and industrial hazards is important because their subsidy limits their mobility and the government has an obligation to ensure housing that is decent, safe, sanitary, and in good repair. Using proprietary data from First Street Foundation and publicly available data from the Federal Emergency Management Agency’s (FEMA) recently updated National Risk Index, the author examined flood hazard risks for residents of subsidized housing. This analysis builds on work by the Public and Affordable Housing Research Corporation (PAHRC) and the National Low Income Housing Coalition (NLIHC), as well as Climate Central. Both studies mapped risks for subsidized housing using the National Housing Preservation Database. PAHRC and NLIHC examined risks using a previous release of FEMA’s National Risk Index, while Climate Central developed their own coastal flooding model. This analysis adds an additional lens by using First Street Foundation’s proprietary Flood Factor model. This study estimated areas of greatest concern, where there is a combination of high flooding probability, high depth of projected flooding, and low housing quality. This work also examined the demographics of residents in areas of risk to understand how they compare to those in areas of less risk. The author concludes with recommendations for further research, with the goal of shaping policy recommendations for the federal government to better protect those they serve.

Traditional behavioral models have not fully examined spatial and temporal movements of evacuees from section to section on evacuation routes. They also perform poorly in predicting household-level evacuation decisions. On the other hand, big data analysis often trains models during non-emergency situations, based on limited behavioral factors (e.g., awareness of congestion or threat), and relies on a very large dataset that is unfeasible to many social science research projects. This study evaluates an existing model and its parameters predicting large-scale evacuation route choices spatially and temporally. It also develops algorithms to predict selected evacuation decisions (i.e., whether evacuated, daytime or nighttime evacuation, preparation/mobilization time, and evacuation time) using machine learning and deep learning. We trained a model using XGBoost and data from a hurricane evacuation (N ~ 1,300). The results are more accurate than statistical methods noticeably (i.e., from ~60% to 75% and R2 from three percent to more than 25%).

Celebration of successful disaster risk reduction (DRR) interventions can be a key component to sustaining and amplifying efforts that already exist, such as earthquake early warning (EEW) systems. By focusing on positive benefits rather than fixating on negative impacts or technical shortcomings which dominate the news and research literature, this research can shed light on otherwise invisible benefits and build support for the necessary resources to continue or expand successful DRR programs. This study focused on understanding the invisible benefits specific to technical end-users of the ShakeAlert EEW System in the water utility sector. Through the lens of downward counterfactual thinking (i.e., reimagining past events in ways that would result in worse outcomes than actually transpired), this research identified the currently invisible benefits and synergies to the water utility sector. This work was informed through semi-structured interviews with ShakeAlert technical end-users in the water utility sector, media content, and existing literature. By examining successes of technical end-users, we can identify broader implications and benefits of ShakeAlert-powered alerts.

Disaster risk analysis in low- and middle-income countries is expensive and challenging, particularly for vulnerable schools. This study streamlines vulnerability identification by developing transfer learning models, clustering similar vulnerability curves, and conducting economic assessments. The simplified process aids in timely decision-making and reduces the need for time-consuming data collection and expert on-site evaluations. The study identified the typology of specific buildings using input of eight valid images, four from exterior and four on diaphragm. These images were then fed into numerous feature models, each determining the existence of its trained feature. Given a predetermined logic tree, a list of existence and non-existence of building features decided the typology of individual buildings and thus provided vulnerability assessments. Significant effort was invested into identifying differentiating features and screening photos for ground truth values. Among all potential features, Reinforced Concrete Columns, Timber Diaphragm, and Multiple Stories were significantly influential to the typology decision and considerably easier to identify. The study implemented multiple transfer learning models on the selected features, and the base model ResNet50 yielded the most satisfactory results of a 0.97 accuracy and a 0.98 f1 score, essentially proving the potential of the framework when deployed in a complex setting. As the model prioritizes accessibility, reduces cost and time, and provides detailed data for informed decision-making by school administrators and policy makers, it promotes community resilience and child safety while simultaneously mitigating inequality and corruption by prioritizing schools based on need.

Quantitative modeling of potential impacts from a natural hazard largely employs the disaster risk modeling framework, which focuses on immediate losses irrespective of who might be affected. While countless post-event studies have shown that marginalized groups can be disproportionately impacted by disasters, current modeling frameworks do not explicitly consider such disparities. This can lead to interventions that are poorly targeted, or worse, deepen existing inequalities. This research examined emerging approaches toward the development of equity-aware risk models—or risk models that acknowledge and illuminate disparities in risk between gender, racial groups, age, or other sociodemographic groups. The research team convened a working group of risk modelers and managers and conducted case study evaluations of equity-aware risk models for application in low and middle-income countries. The initial workshop held in November 2022 identified key themes required for developing equity-aware risk models—applications, methods, and data availability—that were explored further through a series of working group meetings. Simultaneously, case study evaluations compared multiple methods ranging from risk disaggregation, social vulnerability indices, and probabilistic approaches. Results show that several risk modelers and managers are prepared to account for equity in quantitative assessments of disaster impacts and that a number of promising methods are available. Applications range from comparing disaster mitigation projects to integrating social protection programs with disaster insurance. This study explored the suitability of various approaches for equity-aware risk assessments in low and middle-income countries, though additional research is required to understand how these methodological approaches vary across contexts and applications.

Global dependence on satellite technologies, along with expanding space exploration, and commercial space travel have made rocket launches routine modern-day events. As a result, there is an increasing probability of civilian casualty from rocket launch anomalies, or from falling space debris already in orbit. This study provides an empirical analysis of local emergency management publicly available guidelines for the risks associated with human-made space hazards, natural space hazards, and unknown space hazards. A multi-case approach was utilized with a document analysis of 391 emergency management documents provided by 512 local jurisdictions across the states of California, Florida, Texas, and Virginia. Descriptive statistics and a QGIS vector spatial analysis are conducted to identify high risk counties in close proximity to rocket launch sites. Results from a logit regression model were used to examine the relationship between a county containing a rocket launch site or being in close proximity to a launch site and their likelihood of including falling space hazards in their emergency management documents. A significant number of documents mention falling aircrafts, missiles from tornadoes, and weapons of mass destruction. Rocket launch anomalies and falling space debris remain outside the scope or imaginability of local emergency management plans.

With the rise in frequency and intensity of hurricanes, geographically vulnerable regions like Puerto Rico face unforeseen threats. Hurricane Maria (2017) devastated Puerto Rico’s physical infrastructure, induced extended utility disruptions, and displaced a large segment of the population. With preexisting social vulnerabilities, the widespread hurricane-induced damages, disruptions, and displacements left a lingering impact on household recovery and well-being. This study assesses the long-term impact of Hurricane Maria on Puerto Rican households to analyze the role of storm-induced damages and existing socio-economic and environmental factors. The authors surveyed 502 Puerto Rican households in 2021 and collected data on hurricane exposure, physical losses, protective actions, disaster relief, and household characteristics. The ordered logit regression was used to analyze factors’ association with household impacts. The results suggest that households who experienced extended utility disruption (electricity, cell phone network, internet, and public transport) and higher property damages were more likely to have an impact for an extended period. Additionally, households encountering damages from hurricane-force wind were more affected than those encountering flood damages. Prolonged evacuation, income loss, and insufficient disaster aid also increased household suffering. While some preparedness measures (e.g., owning an alternative power source before the storm, hurricane-proofing the house) lessened the impact, having insurance (wind, flood, health) also contributed positively to reducing the long-run effects. While households with elderly members suffered more adverse impacts, larger households tended to be less impacted. These research findings may be useful for improving disaster management and reducing long-term impacts in vulnerable regions.

Wildfires pose a significant ongoing global threat, causing widespread damage to the environment and communities each year. Obtaining real-time situational awareness is crucial for timely and effective responses to mitigate their impact. However, traditional survey-based methods of investigating situational awareness can be time-consuming and labor-intensive, which can delay response times and hinder effective decision-making. The increasing use of social media provides an opportunity to attain and improve situational awareness during wildfire response. In this study, the authors explore the potential of using topic modeling techniques to analyze public opinions and behaviors during wildfires. The researchers focused on the 2020 western United States wildfire season and collected relevant Twitter™ data from September 2 to October 4, 2020. Using Named Entity Recognition (NER) techniques, we first identified geo-information in the tweets. Next, the researchers employed BERTopic, a transformer-based topic modeling technique, to identify the most common topics present in the tweets. Results demonstrate that topic modeling techniques can effectively extract important situational information related to wildfires from social media data. The researchers identified several key topics, including “forecasting and preparation,” “impacts,” “casualties,” “rescue and response,” “cause,” “sympathy,” and “recovery”. With the identified location and topic information, this analysis of the topic distribution and evolution in tweets can provide valuable insights for wildfire managers to make informed decisions and prioritize operational needs. Findings suggest that social media can potentially aid in decision-making, optimizing resources, and building a more resilient community.

Communities exposed to risk from a Cascadia earthquake are diverse and have varied information needs. However, information about earthquake risk is often developed for “general” audiences, and rarely tested for its effectiveness at improving risk understanding or preparedness intentions. Additionally, though advances have been made in translating earthquake risk and preparedness information into other languages, many resources are only available in English. Best practices in risk communication suggest that providing personalized, clear, and culturally appropriate information will improve people’s willingness to engage with the material, and their learning outcomes. In this study, the authors work closely with four Spanish-speaking community liaisons to provide input on and co-facilitate six, two-hour earthquake preparedness workshops. Workshops were carried out in Spanish online and in-person during the summer of 2022. The workshop included topics related to Cascadia earthquake impacts, preparedness actions, and goal setting activities that help participants think about how to get prepared. The authors used a pre- and post-test experimental design to survey the 48 participants who attended the workshops. We are in the process of analyzing the pre- and post-workshop data to understand which variables are the strongest predictors of people’s intentions to prepare and whether the workshop changed risk perceptions, hazard adjustment perceptions, earthquake knowledge, and intentions to prepare. The authors anticipate these findings will help advance strategies for developing earthquake information that meets underserved community needs.

With the climate warming every year, the trend of destructive wildfires is expected to continue. To coexist with wildfires and minimize casualties and losses to properties, an understanding of the interaction between wildfires and the built environment is required. In this study, a graph-based approach is presented and used to predict the spatial distribution of damage to individual buildings during the 2022 Marshall Fire. The validity of the graph model is first shown by spatially comparing the predicted and the observed damaged patterns. The model was utilized to investigate the sensitivity of the damage patterns to wind conditions to understand better the correlation between wind characteristics and the resulting damage. The results show the variation in damage distribution among all wind directions to be significant, highlighting that wind conditions need to be carefully considered for wildfire risk assessment and that tailored and community-specific solutions should be developed to minimize potential losses.

As climate change intensifies the frequency and severity of hurricanes, managing evacuation traffic becomes increasingly challenging. Based on interviews with emergency managers from Florida counties, this study identified specific challenges faced in managing evacuation traffic after rapidly intensifying hurricanes. Findings indicate that inadequate awareness of hurricane risks is one of the key factors that contributes to people’s reluctance to evacuate on time. This may be due to lack of recent hurricane experience, migration with no prior knowledge of hurricanes, or forgetting past hurricane impacts. Moreover, people may not follow evacuation instructions or rely on unofficial sources of information. The paper argues that effective management of evacuation traffic and its impact requires clear communication and coordination with diverse stakeholders at multiple scales, especially community-based stakeholders, to increase awareness and provide accurate information. By addressing these challenges and implementing the proposed solutions, emergency managers in Florida counties can improve the effectiveness of their evacuation efforts and ensure public safety during hurricane events.

Homelessness creates physical and psychological disadvantages for students, and often disrupts school access. Despite efforts to end homelessness in the United States, student homelessness has gradually increased over the past decade. Research suggests that students who experience prolonged dislocation and school accessibility loss after a disaster are mainly from low-income communities. Dislocated student recovery is a complex process driven by household social vulnerability. This study develops a two-tier framework to capture recovery and long-term homelessness of kindergarten through high school (K-12) students after a natural disaster. Using the National Center for Education Statistics (NCES) school attendance boundaries, residential housing inventory, and U.S. Census data, the first tier identifies students within school boundaries and links schools to students to housing. The second tier tracks the stage of housing for each dislocated student after a disaster, based on predicted building damage. The recovery of dislocated students is predicted using an existing multi-state Markov Chain Model, which captures sequences of households transitioning through the four stages of post-disaster housing (i.e., emergency shelter, temporary shelter, temporary housing, and permanent housing) based on household social vulnerability. Finally, the two-tiered approach predicts the number of long-term homeless students by school. The framework is exemplified for the case of Hurricane Matthew-induced flooding of Lumberton, North Carolina. The example demonstrates the ability of the framework to capture long-term K-12 student homelessness and can aid decision-making to reduce future disaster impacts on students.

Hurricanes historically present the most significant threat to coastal communities and frequently cause the most substantial damages when compared to other natural disasters in the United States. While flooding and winds are inherent dangers associated with hurricanes, debris can have cascading acute and long-term effects on infrastructure and be very expensive to manage. Furthermore, hurricane-induced debris behavior is a complex phenomenon due to the interconnected nature of human-built-natural systems embedded within coastal communities. As population growth, urban development, and societal reliance on fossil fuels continue to affect hazard exposure and natural disaster intensities, there is a need to create future scenario-based models to fortify community resilience and reduce the negative impacts of debris. Therefore, this study leverages multiple, convergent modeling efforts of future land cover projections, storm surge scenarios, and hurricane-induced debris impacts to estimate debris volumes and locations following future development and hurricane conditions. In our methodology, we leverage human, built, and natural environment parameters, scenario-based climate projections, and storm intensity measures within a predictive model of debris and its impacts on communities. Results from our modeling efforts include future development patterns in Galveston County for years 2025 to 2040, 45 total storm scenario simulations, and debris volume and location estimates from storm scenarios. Ultimately, our results can inform urban planners, emergency managers, and other decision makers with future, policy-based hazard scenarios and offer support for proactive policies aimed at reducing impacts from future natural hazards.

To address the uneven impacts of climate change and disasters on marginalized and oppressed communities, disaster risk reduction and resilience program efforts must be associated with broader justice and empowerment initiatives. Coastal flooding poses a significant risk to personal and public properties in Jacksonville, Florida. The impact of climate change on marginalized communities in Jacksonville has been exacerbated by historical injustices such as redlining policies. To promote equity and uphold individual rights, we are conducting research on the unjust distribution of resources and services in Jacksonville. Participatory action research (PAR) was employed for this study, which was led by a team of academic investigators in partnership with local community researchers from Edward Waters College and local community organizations. This study aims at empowering underserved and overburdened communities to change their current circumstances and status quo. Although PAR is widely recognized as one of the most powerful tools for empowering communities and improving their capacity to handle risks, disaster research has often neglected its theoretical foundations in participatory democracy, social justice, and critical thinking. We hope to enrich the disaster research discourse by using a Freirean perspective and applying it to our case study. It is intended to address research ethics and its impacts on the community in relation to the PAR approach in the context of resilience. It will also be discussed how to balance local engagement with broader involvement of organizations.

This Faculty Early Career Development (CAREER) grant will further the understanding of how communities can effectively leverage philanthropic resources to meet housing-recovery needs after disasters. As disaster costs and disaster displacement increase, governmental assistance to individuals and private insurance often are inadequate to ensure full recovery for all affected people. Philanthropic resources can address unmet needs of disaster survivors if used effectively and efficiently. Locally led nonprofit long-term recovery groups are often charged with distributing these resources, but little is known about these organizations’ efforts or what makes their operations more or less effective in promoting community recovery and resilience. This project will assess how philanthropic housing-recovery practices affect individual unmet needs, post-disaster equity, and the overarching philanthropic ecosystem of affected communities. This project will use the research results to inform and test a training program for locally based nonprofits, government officials, and foundations that will improve their effectiveness in managing philanthropic resources for disaster recovery. Educational outcomes also include undergraduate research experiences to foster under-represented student engagement in STEM and graduate student internships coordinated with disaster recovery nonprofits to further their data management skills.

Timely research is critical to furthering the collective understanding of how disasters and other public health emergencies impact human health, but many challenges remain to conducting research in disaster settings. The Rapid Acquisition of Pre- and Post-Incident Disaster Data (RAPIDD) human subjects research protocol was developed by the National Institutes of Health (NIH) Disaster Research Response (DR2) program in 2015 to address the gaps that impeded the rapid collection of data and samples from disaster response personnel and to allow for studies of a wide range of potential adverse exposure scenarios and health-related effects. To expand the scope beyond these worker populations and to help research teams with faster investigation of the impacts of disasters on communities, the DR2 program launched a free online tool for managing disaster-related research in March of 2023. Also available on mobile devices, the RAPIDD Protocol Designer is a customizable template that researchers can adapt to fit the needs of their research question and facilitate timely data collection following disasters and public health emergencies. This new tool contains pre-approved NIH Institutional Review Board (IRB) language and connects to hundreds of publicly available, curated resources—surveys, procedures, templates, and other tools—submitted by NIH grantee institutions and interagency partners to the DR2 Resources Portal. As disaster events increase in volatility and frequency, the RAPIDD Protocol Designer is a useful resource for research teams to get into the field more quickly while respecting the time and experiences of impacted communities.

Natural hazard-related disasters have had increasingly damaging impacts around the world. Due to intersecting vulnerabilities and a rise in climate change-induced hazards, Southeast Asia is particularly vulnerable. Such locations are often important tourist destinations, so tourism is a key aspect of disaster recovery and may play a positive role in assisting broader community recovery. As much post-disaster recovery literature focuses on tourism’s recovery from disaster damages or tourism as a livelihood diversification strategy, there is a research gap concerning how the sector can play an active role in disaster recovery. This study examined how different forms of tourism played a role in post-disaster recovery from Typhoon Haiyan in the City of Tacloban, Philippines. The role played by domestic tourists and international aid workers (defined here as tourists) during the recovery period was explored using 29 semi-structured interviews with key stakeholder groups including tourism sector employees, government officials, and indigenous groups. Preliminary qualitative analysis shows that: (a) aid workers filled to some extent the hiatus in domestic tourism caused by the typhoon, and facilitated an emerging demand for international tourism through social capital developed in the years post-disaster; (b) Haiyan acted as a catalyst for development of the current tourism strategy—Meetings Incentives Conventions Exhibits (MICE); and (c) disaster tourists directly supported the hospitality industry in the post-disaster period. The Tacloban experience demonstrates the importance of developing a broader post-disaster tourism strategy to leverage the networks and resources present within the tourism sector that can help facilitate recovery.

Risk communication in times of disasters is complex, involving rapid and diverse communication in social networks (e.g., public and/or private agencies; local residents) coupled with limited mobilization capacity and operational constraints of physical infrastructure networks. Despite a growing literature on infrastructure interdependencies and co-dependent social-physical systems, an in-depth understanding of how risk communication in online social networks weighs into physical infrastructure networks during a major disaster remains limited. Building on this gap, this study offers a holistic network diagnostic framework that takes into account complex interdependencies in coupled social-physical infrastructure systems. The research considered the case of several extreme weather events in Oklahoma for the year 2020 to co-locate social risk communication networks, captured through Twitter, with physical infrastructure networks in vulnerable communities as they responded to these major crisis events. The risk narratives on Twitter (i.e., tweets) were analyzed using novel natural language processing and text classification techniques, while the networks (both social and physical) were analyzed using metrics available in network science literature. The proposed approach and study findings provide new insights to improve disaster response strategies.

Responding to a crisis event involves interactions that are top-down (e.g., government agencies) and bottom-up (e.g., social networks). Such interactions are sometimes inefficiently organized, which leads to coordination challenges hindering optimal response. The challenge is to enable innovative coordination mechanisms that can allocate critical resources equitably without undermining each other. During major disasters, different stakeholders (i.e., public and/or private agencies, emergency management officials, and first responders) experience complex decision-making scenarios while lacking critical resources and efficient response strategies in place. Natural hazards may occur at any time, leaving key stakeholders with a narrow lead time, especially in short-notice risk events. Such decision-making processes are complex due to the vast amount of information they are exposed to, combined with cognitive limitations. This research considered emergency management agencies and city officials from different rural communities in Oklahoma. Data was collected by conducting systematic one-on-one ethnographic interviews, a day-long participatory workshop, and a follow-up network survey using an ego-centric network design approach. The study then developed critical decision trees of the key stakeholders and analyzed their ego-centric networks. The findings from this research can potentially save lives by enabling public authorities to use more effective and efficient ways of communicating risk information to a target and aligning critical resources under acute budget and resource constraints.

To reduce risk from natural hazards, mitigations should be informed by the complex, multi-hazard environment in which they will be implemented. Coastal communities face earthquake, liquefaction, and flooding hazard, all of which can damage housing and dislocate households. In order to compare mitigation actions to reduce risks, a unified approach is necessary to consider damages from each hazard to the housing stock and to compare the effects of these damages on the households. This presentation describes a model for housing impacts and recovery that is compatible with engineering assessments of physical damage. This agent-based model simulates building-level recovery times, including the need for financing, which is dependent on owner income. In using time to recover in place of initial damage, income-based inequities are included in the comparison. This model is applied to a San Francisco Bay Area case study to demonstrate how impacts to residential housing come from earthquake and liquefaction hazards. This work is being applied to a flood mitigation alternatives analysis to compare impacts under various future scenarios. Working with a community partner, this engineering model is used to simulate potential future changes to reduce flood risk and quantify the reduction in risk to residential housing in the case study area.

The Hunga-Tonga-Hunga-Ha’apai (Tonga) volcano erupted on January 15, 2022, and triggered a tsunami predicted to reach North America, providing a unique opportunity to investigate risk perception and communication among coastal emergency managers (EMs). In response, this research explored: (a) how risk might be most effectively communicated where alerts may come from a combination of physical signals and explicit messaging and (b) how risk perceptions associated with “distant” tsunami alerts and warnings affect EMs’ willingness to issue emergency alerts. A purposive sample of coastal EMs (n=21) in the Pacific Northwest (PNW) participated in semi-structured interviews that probed these topics regarding the Tonga tsunami. Participants represented Tribal, county, state, and federal jurisdictions. Transcripts were deductively coded and thematically analyzed using NVivo qualitative analysis software. Coded text was synthesized to develop salient themes. Participants described tsunamis as rare but devastating events. While they perceived low risk from the Tonga distant tsunami, many had to strike a balance between notifying the public and avoiding the negative impacts of their messaging (e.g., “crying wolf,” panic, curiosity). The tsunami’s distance from the PNW created a window of uncertainty for EMs but also provided time to evaluate response options. Participants used a multi-pronged alert dissemination strategy to reach their community (e.g., alert systems, social media, and door-to-door). This study highlighted key elements that inform EM risk perception and communications in the context of uncertainty due to the absence of local ground shaking from an earthquake. Findings can inform tsunami preparedness among PNW communities.

Improving the resilience of a community is a complex and costly process with many stakeholders reaching for limited potential sources of funding. Communities are showing interest in private market financial tools to fund such efforts, including resilience bonds. Charleston, South Carolina, is among the communities pursuing funding sources for mitigation projects, addressing disasters like flooding, sea level rise, king tides, and hurricanes. The City of Charleston, a historical city on the Atlantic coast, has mostly directed funds to near-term disaster mitigation projects such as strengthening buildings, land use regulations, freeboard requirements, green infrastructure, building retrofits, and holistic approaches promoting community resilience. Thus far, Charleston has pursued funding sources from taxes, federal and state grants, and private philanthropies for these mitigation projects. But the city is now facing limited funding available for the implementation of large-scale infrastructural mitigation projects. This research used an in-depth qualitative analysis of semi-structured interviews with local resilience stakeholders and financial experts, and review of media articles, and local organizations’ documents in Charleston to understand various dimensions of resilience investment priorities, opportunities, and constraints within a historic context of development and disasters in this area, centered around equity consideration.

Given the rising exposures to coastal hazards in highly vulnerable areas like Puerto Rico, there is a growing urge to integrate natural hazard risks into public health preparedness for enhancing community resilience. How public health care providers, hospital management staff, nonprofit and community organizations, emergency management personnel, and the public interact in responding to natural hazards is highly critical for building a resilient healthcare system in affected communities. Against this backdrop, we conducted interviews to analyze the perspectives from patients, community members, public health professionals, and emergency management personnel to gain deeper insights into the nature and extent of key public health management challenges following recent natural disaster events in Puerto Rico. The two overarching research questions are: (1) To what extent are these key challenges due to resource constraints, and to what extent are they due to management practices and policy constraints? (2) What are the actionable steps that can be specified to overcome these constraints for reducing mortality and morbidity risks during the next disaster? We tried to address these issues in this project. The preliminary findings indicated that high mortality and morbidity risks in the aftermath of Hurricane Maria in Puerto Rico are primarily driven by resource constraints which are further aggravated by management practices and policy constraints. Our findings have significant implications, as it can potentially help to build a conduit between the community, public health stakeholders, and emergency management agencies for laying down a foundation of a resilient health system in Puerto Rico.  

Rural and disadvantaged communities are often the most susceptible to climate disasters that exacerbate the socioeconomic disparities. However, it is unclear to what extent disasters and their cascading failures of critical infrastructures disproportionately impact vulnerable groups. This research proposes a human-centered uncertainty-informed decision framework to advance fundamental knowledge, build capacity, and forge research-to-practice partnerships that can strengthen and transform the post-disaster recovery of communities in an equitable way. This decision framework would also help to identify the dynamic processes and complex interactions in the nexus of disasters, humans, and interdependent critical infrastructure under deep uncertainties. To establish the framework, we use Hurricane Fiona and its impact on socially vulnerable communities in Puerto Rico as a case study. Through a deep convergence process, the research team collected data on cascading critical infrastructure failures, impacts, and recovery challenges from local stakeholders and communities using focus groups, surveys, and in-depth interviews. Data were analyzed using a mixed-methods approach where the qualitative findings from interviews helped validate secondary data analysis. This research presents insights from two perspectives: First, it reveals the community perspective which focuses on how communities were impacted by the disaster and delayed restoration of critical infrastructure. Second, it describes the stakeholder perspective which sheds light on the challenges involved in reconstructing critical infrastructure and how those challenges delayed community recovery. Novel solutions are provided in collaboration with community and infrastructure manager experts that will advance post-disaster recovery and reconstruction.

Following major disasters, media reports emerge presenting a picture of mass chaos, mass panic, and a significant increase in crimes, specifically looting behaviors. While research has shown that individual instances of panic and looting occur after disasters, it is not widespread. In fact, survivors are more likely to engage in prosocial behaviors, such as searching for and rescuing neighbors and friends. Disaster science is rooted in empirical evidence that mass looting does not occur following disasters; however, the myth still thrives and is presented as fact following major events today. This research looks back at the rhetoric presented in the media following the 2005 landfall of Hurricane Katrina and the 2023 landfall of Hurricane Ian. Preliminary findings show that populations blamed for the looting behaviors were primarily ethnic/racial minorities. While after Hurricane Katrina the blame was placed on survivors inside the disaster zone, the rhetoric following Hurricane Ian focused on immigrants coming from outside the community to loot the impacted area. This research adds to the more than 50 years of scholarly works on disaster myths.

With the increasing occurrence of wildfires threatening and affecting society, the National Oceanic and Atmospheric Administration (NOAA) has increased its strategic investments and planning to address wildfire-related hazards. One important aspect of this is the establishment of a new NOAA Fire Weather Testbed (FWT). This FWT was designed to bring together the wildfire communities including decision makers, researchers, and operational fire weather forecasters to improve and tailor tools, applications, products, and information, with a heavy focus on understanding the information needs of the wildfire communities. The FWT will work closely with other NOAA Testbeds and Proving Grounds to facilitate various evaluation and transitional fire weather-related activities; and, it is a joint effort between three of NOAA’s Line Offices: NOAA Research, National Weather Service, and National Environmental Satellite, Data, and Information Service. This presentation will provide an update on the status of the infrastructure, staffing, and planned execution of the FWT. Housed within NOAA’s Global Systems Lab in Boulder, Colorado, the FWT will consist of physical and virtual facilities for conducting evaluations and experiments, with a heavy reliance on cloud computing resources and methods.

Dealing with the internet’s superabundance of disaster preparedness information is challenging. To understand how young adults (ages 18-29) in Portland, Oregon, respond to preparedness content online, the authors conducted an experiment (n = 125) comparing self-reported self-efficacy, intent to act, and steps taken in eight categories of earthquake preparedness action among participants who spent up to 45 minutes either web searching or playing video games. Surveys were administered before and after the experiment and three months later. Those in the web search condition could browse at will or use three starter links to emergency management websites. All web search participants (n = 59) began their explorations with the curated sites, but 32 also did free searches for additional information. Compared to the additional search group, the group which viewed only curated sites spent significantly less time on the task (16 vs. 23 minutes) but rated their sources as more trustworthy (p = 0.001) and deemed the information more reliable. Of 27 web searchers who responded to the follow-up survey, those in the group which only viewed curated sites (n = 12) showed greater increases in self-efficacy, intent to act, and steps taken than those who viewed the curated sites but also spent time searching other sites (n = 15). The differences were significant for four and two out of eight categories for intent to act and self-efficacy, respectively, and not significant for steps taken. Emergency management website managers should prioritize discoverability and content that will retain young adults’ attention.

Site response in sedimentary basins is generally governed by mechanisms such as resonance, surface wave propagation, and three-dimensional (3D) seismic wave effects due to basin geometry. Combinations or subsets of these processes can lead to the amplification and extended duration of observed ground motions from earthquakes. For this reason, current ground motion models (GMMs) incorporate basin effects using the time-averaged shear wave velocity in the upper 30m (VS30), and the depths to a particular shear wave velocity horizon. This approach captures an average representation of site response features associated with basin sediment deposits but is limited in its description of basin and non-basin conditions, as well as lateral and other 3D contributing effects. Recent research has addressed the issue of within versus outside basin conditions in the GMMs, however, this study investigated alternative means for describing and using geometric basin properties in the development of region-specific non-ergodic models to improve the characterization of site response in sedimentary basins. This was accomplished by developing basin geometric parameters derived from basin shape, with predictive power assessed using ground motion residual analysis. The basin shape was established and validated using community velocity models, gravimetric data, chronostratigraphic models, and available subsurface investigation data. Southern California has a variety of sedimentary basins and was selected as the region of study due to the immediate availability of requisite data and resources. The performance of the developed models will be evaluated by assessing their ability to reduce bias and uncertainty of ground motion models.

Category 5 Typhoon Hagibis hit Japan in 2019, causing a record-breaking flood and more than 100 deaths, the second largest death toll caused by a typhoon in more than 50 years in the country. Two years after the historical disaster, another heavy rain caused flooding, with rivers reaching overflowing levels in Nagano City. Many residents along the river, however, quickly reacted and evacuated. Previous tangible flood impacts might have changed their behavior. Our research investigates whether 24-hour weather prediction technology that detects flooding risks could affect pre-evacuation activities. The authors administered consecutive workshops designed with neighborhood leaders, government officials, and experts in hydrology, socio-hydrology, and urban engineering to compare pre-evacuation activities during the 2019 Typhoon and the 2021 heavy rain. The workshops were held from January to March in 2023, with a total of 60 disaster survivors, asking what they were doing 24, 12, and 6 hours prior to the highest water level in the two events. The results indicated that evacuation preparations such as moving expensive agricultural equipment to the levee had been underway for 24 hours before the heavy rain in 2021, whereas activities during the day prior to 2019 flooding were full of regular daily activities without uncertainty about the levee breaking. Furthermore, the participants were questioned about the potential benefits and challenges of the 24-hour weather prediction technology. They identified early evacuation of the disabled and elderly as a benefit and mentioned that neighborhood security post-evacuation could be a dilemma.

Storm surge is a major natural hazard in coastal regions, responsible both for significant property damage and loss of life. While high-fidelity regional- and global-ocean circulation models such as the ADvanced CIRCulation (ADCIRC) model can accurately predict storm surge, they are very computationally expensive. Here we developed a novel surrogate model for peak storm surge prediction based on a multi-stage approach. In the first stage, points are classified as inundated or not. In the second, the level of inundation is predicted for each point. We demonstrate our modeling framework on two study areas: the Texas coast and the northern portion of the Alaskan coast. For Texas, the model is trained with a database of 446 synthetic hurricanes. The accurately matched ADCIRC predictions on a test set of synthetic storms. We further present a test of the model on Hurricanes Ike (2008) and Harvey (2017). For both storms, we find that the model predictions have comparable accuracy to ADCIRC hindcasts when compared to actual observational data. For Alaska, the model is trained on a dataset of 109 historical surge events. We tested the surrogate model on actual surge events including the recent Typhoon Merbok (2022) that took place after the events in the training data. As with the Texas dataset, the surrogate model achieved similar performance to ADCIRC on real events when compared to observational data. In both cases, the surrogate models are many orders of magnitude faster than ADCIRC.

Building codes are often touted as the solution to high earthquake mortality. Yet implementing building codes is often difficult in low-income countries. One aspect of this challenge can be seen in the experience of Nepali masons trained in constructing earthquake-resistant houses according to code. Based on interviews and focus group discussions with 49 trained masons who took part in seismic-resistant construction training through the National Society for Earthquake Technology (NSET), it was clear that masons wanted others to understand them as professionals, with the prestige and jurisdiction capture this would bring. Trained masons formed professional groups, felt heightened pride in their work, discussed it using concepts of altruism and public good, and believed they gained more societal recognition. Yet, they continued to experience competition from untrained masons who constructed seismically fragile structures. This research shows both the promise and limitation of building code adoption processes as a tool for risk reduction, especially in contexts of low government capacity.

The outbreak of large-scale infection of COVID-19 around the world has not only caused a large number of human infections and deaths, but also severely damaged the urban life that humans are accustomed to. This study explores the influencing factors based on the literature of the diseases that have caused the impact, and then designs a questionnaire based on the changes and influencing factors of urban life summarized by various countries. The results of the study showed that the frequency of use of public transportation and indoor public spaces decreased significantly during the severe epidemic period, and risk perception was the main factor. In addition, the factors affecting the frequency of urban space and transportation use are also different in time periods of different severity. For example, factors affecting the frequency of workplaces and leisure venues have shifted from risk perception to income and urban environment characteristics during the raging epidemic. By analyzing the usage cognition and perception of urban space and public transportation during the epidemic, the critical urban characteristics which were valued and worried by residents during the epidemic can be understood. In the era of coexistence with the virus, urban planners can apply relevant experience to reduce urban risks and eliminate doubts in the direction of designing urban environments under the epidemic.

Essential service critical infrastructures are impacted by individual and organizational resilience. Yet, how can systems be improved to maintain resilience? When examining childcare as an invisible, yet critical infrastructure, a precarious relationship is revealed: Providers served the community in a time of tremendous need, while placing themselves in significant harm for low wages. Even as the authors conceptualize resilience as never-ending, the role strain language that permeates open-ended data suggests that childcare providers rose to the challenge, but at deep and unsustainable costs, professionally and personally. This raises important theoretical considerations: How can this simultaneous experience of resilience and role strain among childcare providers expand the conceptualization of such a core analytic construct? What parameters might be put on resilience to make the concept more analytically meaningful, particularly for service infrastructures? Based on a survey of 141 childcare providers in the five-county metropolitan region around Rochester, NY, this study examines the intermediate stage of an ongoing disaster landscape, with data collected between late September 2021 and March 2022, and reflects on the impacts of all waves through the Omicron variant on childcare providers. The sample is a census of all childcare centers identified by the New York state online database, and the professional networks of a local Child Care Resource and Referral agency (CCRR) who helped circulate the recruitment email.

In September 2017, Puerto Rico was hit by the destructive Hurricane Maria. The hurricane severely damaged critical infrastructure, leaving residents without power, water, communication, and other utility services. The devastation led to a massive displacement of the people of Puerto Rico. This study aimed to assess whether those who were displaced due to Hurricane Maria had more mental health issues compared to their counterparts. To conduct this study, the authors surveyed 507 Puerto Rican adults in the aftermath of Hurricane Maria in 2019. Using logistic regression analyses, the authors assessed the relationship between displacement and mental health outcomes (depression, anxiety, post-traumatic stress disorder, among others) and controlled for socioeconomic factors, pre-disaster characteristics, and perceived risk. These findings suggest that mental health challenges were prominent among Hurricane Maria survivors, especially those who were displaced permanently. Based on these results, the authors recommend that local, state, and federal governments invest in intervention programs that facilitate access to mental health services in the affected communities and communities where displaced Puerto Ricans have resettled.

How do disasters influence the local perceptions of risk and disaster preparedness? Studies show that experiencing a disaster shapes the perception of risk, but do disasters still affect the personal imagination even after all those who directly experienced it are long gone? To answer this question, the author considered the compound earthquake-tsunami disaster of 1908 in the Strait of Messina and Reggio in Italy. The disaster caused the destruction of the city of Messina and took approximately 80,000 lives. The destructive event shaped permanently not only the new urban landscape but also the residents’ outlook on disasters and prevention. This project focused on the topics of disaster storytelling and the long-term impacts of disasters on risk perception. In the case of Messina, disaster stories are passed down through family stories, yearly memorial events, municipal signs of “Messina before the disaster” around the city, and historical tours for tourists and local high schools. The personal and collective disaster memory of the 1908 disaster led to the adoption of notions of disaster preparedness but also a sense of inevitability and a critical outlook on the efficacy of local authorities and disaster preparedness. The project data included early century journal articles, interview transcripts, and field notes. The analytical framework was based on the literature on disaster storytelling, human geography, and critical discourse data analysis. This project’s results have practical implications for local policy makers interested in improving existing disaster preparedness policies to increase local awareness and participation.

Urbanization is increasing day by day and data shows that 55% of the world’s population lives in urban areas, a proportion that is expected to increase to 68% by 2050. This research focused on urban-rural connectivity from a disaster resilience perspective where the existing status of urban-rural linkage was analyzed by choosing a case study from a developing country, Nepal, particularly converging on road and water connectivity. How past disasters wedged on urban-rural settings, and in what way conflict over common resources generated hurdles between urban-rural connectivity was analyzed from available primary and secondary data. A qualitative approach with figures and tables was applied to analyze the received data from the purposive and semi-structured interview. Poor connectivity between urban and adjacent rural administration for disaster management and lack of uniformity in data management were observed as major problems. In contrast, integrated data management systems for disaster management, appropriate integrated policy by addressing road/water problems, and tax systems for water issues were suggested in the urban-rural setting to create disaster-resilient communities.

In this study, we explored determinants of disaster preparedness and vulnerability among historically marginalized households. We followed two lines of inquiry. First, “what characteristics drive preparedness and vulnerability?” Second, we turn towards application and ask, “what can this study do to inform disaster preparedness practice for historically marginalized households?” We attempt to answer these questions by marshaling a unique data source for disaster studies research. A short preparedness questionnaire was included with the 2020 Collaborative Multiracial Post-Election Survey (CMPS). A popular data source in political science, the CMPS has 17,000 respondents and oversamples historically marginalized populations. Additional variables on the CMPS include targeted questions about COVID-19 response along with traditional variables measuring household demographics, social trust, information sources, and similar areas of interest. Responses were analyzed using cross-sectional statistical analysis. We believe that disasters are, in part, political phenomena. As such, we draw on data sources used by this field to explore disparities in preparedness and vulnerability. Integrating with other theories, we also consider the various intersections of vulnerability and how they challenge definitions of preparedness. Our results suggest that African American households show strong levels of disaster planning. However, vulnerabilities remain around issues of recommended response and recovery resources. Comparison with other benchmark studies reveals similar results, which offers support for further use of the CMPS in disaster studies. Turning theory to action, the survey results provide indirect support for the efficacy of planning activities while raising unanswered questions about resources and vulnerability.

Alternative fuel vehicles (e.g., battery, plug-in hybrid, and fuel cell electric vehicles) may pose challenges to evacuation planning due to their short driving range and sparse refueling or charging infrastructure on transportation networks. Deployment and strategic siting of emergency and portable refueling stations are needed to support the evacuation of alternative fuel vehicles to reach a shelter or a safe zone. This study proposes a location-routing problem with hop constraints that optimize the placement of emergency refueling stations of each alternative fuel vehicle type to support evacuation routing. A decomposition method with a transformed network is developed to solve the location problem and a matheuristic method is proposed to design evacuation routes that minimize travel and refueling time while satisfying the refueling needs of each vehicle type. A series of numerical experiments focus on the South Florida transportation network, which is susceptible to preemptive hurricane evacuations. The experiments provide recommendations to decision-makers on how to comprehensively plan preemptive evacuations for a heterogeneous set of alternative fuel vehicles and reinforce the resilience of vulnerable-to-extreme hazards communities.

This study compared the resilience and performance of critical infrastructure systems in the United States, focusing on the impact of Hurricane Harvey and Hurricane Irma in 2017. The aim was to identify any differences in the ability of critical infrastructure systems to withstand and recover from such disasters in different regions of the country. The study gathered data through household surveys in Florida (Hurricane Irma) and Texas (Hurricane Harvey) during the landfall of both hurricanes, with a sample size of 780 respondents for each location. The study investigated the number of households that experienced various types of utility service disruptions during the hurricanes, including electricity, water, phone/cell phone, internet, transportation, workforce, and grocery stores. In Florida and Texas, 74% and 66% of respondents reported electricity disruptions, and 41% and 50% lacked water supply. Phone reception was unavailable to about half of the surveyed households, and transportation was disrupted for over 40%. Accessing groceries and work was difficult for more than half of the respondents in both states. The study incorporated the survey responses into the Dynamic Inoperable Input-Output Model to understand the economic consequences of utility service disruptions that households faced during Hurricane Harvey compared to Hurricane Irma. The results may provide valuable insights into the strengths and weaknesses of critical infrastructure systems in the United States and inform future efforts to improve resilience and performance in the face of natural disasters. Identifying vulnerable sectors can inform targeted interventions to mitigate the impact of future disasters.

Many crops and agricultural regions in the United States are vulnerable to flood loss due to their proximity to flood-prone areas. However, estimates of such flood losses are expensive, labor intensive, and typically have limited geographic coverage. Much of the crop loss occurs in the country’s rural regions, and if the communities are socially vulnerable, they can be at greater flood risk. This study first estimates the agricultural flood loss for the contiguous United States and the geographic distribution of the loss. Using a predictive flood loss-assessment model, this study quantified the magnitude and geographic distribution of average annualized loss for all major crops in each state. Second, the researchers used spatial regression to determine which social vulnerability variables explained flood loss and to identify population characteristics of high and lower-loss regions. The loss for a June 1 simulated flood event is approximately $550 million, with higher losses distributed in the U.S. Midwest and Mississippi River Basin. Results of spatial regression show employment and income characteristics significantly relate to higher loss, while mobile homes, language, renters, and female-headed households are associated with lower loss. The results identified locations where flood mitigation strategies can reduce crop loss and target communities most in need. Our findings also indicate that methodological modification is needed to improve the quantification of agricultural flood loss and its relationship with vulnerable populations.

Compared to other members of society, people experiencing homelessness during natural disasters and extreme weather events face unique challenges, such as a lack of shelter, transportation, access to the news, and the need to carry all their belongings. When faced with a natural disaster, this population can become even more vulnerable because they lack physical protection and, therefore, have greater exposure to hazards and adverse weather effects. In addition, the unsafe living conditions coupled with the uncertainty of the COVID-19 virus, presented unique challenges for homeless communities, such as an increased risk of virus transmission to the exacerbation of social disparities. This study explored the direct experiences and perspectives of people experiencing homelessness when they were making evacuation decisions during disasters utilizing an adapted Protective Action Decision Model (PADM) framework. The findings presented are the result of 41 in-depth semi-structured field interviews with people experiencing homelessness in Fort Lauderdale, Broward County. The direct experiences that they communicated have essential contributions to theory and practice. To the theory, it contributes to the understanding of human behavior using an adapted evacuation decision framework for the homeless community. To practice, it acknowledges the voice of a marginalized population, providing an opportunity to enhance the current evacuation emergency plans by local governments and community partners that assist the homeless community. In addition, it provides human agency to a marginalized population by opening the conversation for disaster research questions on shelter perceptions and utilization in homelessness research.

Small Island Developing States (SIDS) are amongst the most vulnerable to hazards such as tropical cyclones, earthquakes, tsunamis, volcanism, and flooding. With the changing climate, the intensity and frequency of hazards have increased leading to near constant disaster events. Communities are not only exposed to an individual hazard but to a host of multiple hazards occurring simultaneously or in short succession. COVID-19 impacted Fiji in March 2020, eight months later Fiji experienced Tropical Cyclones Yasa and Harold. As many as 23,000 people went to evacuation centers in the midst of COVID-19. A multi-hazard risk assessment was crucial to investigate the relevance of concepts that see specific vulnerabilities. The study focused on social science assessments of multi-hazard exposure and related them to community engagement, government policies, and resource allocations. This introduced a multi-level perspective from local, regional, to national perspectives and actions. The study found that multiple hazards have great relevance to people living in Fiji. However, the concept of multi-hazard was never part of national disaster management legislation. It recently has been included in policy frameworks, where it mainly focuses on early warning systems. There are institutional, financial, technical, and capacity challenges to the current disaster management. To enhance multi-hazard perspectives in disaster risk reduction the study recommends engaging communities to a greater extend in disaster management policy formulation. There is a need to have a distinct orientation to the protection from multiple hazards to ensure effective disaster management and to provide better security to individuals and communities.

Ethnographic research on earthquake early warning technology in Mexico have utility that stretches far beyond that nation. Research on how Mexican engineers and scientists develop, maintain, and advocate for earthquake early warning technologies shows that their work entails understanding and attempting to transform relationships between environment, society, and technology. This talk presents high level findings from the book ¡Alerta! Engineering on Shaky Ground. It offers a set of provocations related to the ways risk mitigation is understood and enacted that are relevant for researchers, practitioners, students, and perhaps others.

The interactions between communities and their built environment, while somewhat studied, is not yet fully understood particularly for extreme, disruptive events such as floods. In this project, the authors examined how perceptions of flood risk may differ from indicators of actual hazards. The perceived risk of flood, defined as an individual’s subjective assessment of the likelihood and severity of a flood event, can influence their preparedness, response to flooding, and engagement in flood mitigation. However, perceived risk may not always align with the actual risk of flooding, which is determined by factors such as topography, land use/land cover, weather patterns, and institutional contexts for flood management. The authors presented data from a survey of Houston area residents and overlayed their responses in a geospatially coded hazard indicators flood risk mapping system. The survey questions covered topics such as perceptions of flood risk, experience with flooding, flood preparedness behavior, and demographic information. The mapping of hazards included Federal Emergency Management Agency flood zones (e.g., 100-year flood) in addition to historical flooding data and inundation data associated with compound flooding. The results supported the main hypothesis of the research that in order to integrate community input, decision makers must understand community perception of their built environment and how it relates to their perceived quality of life.

Floods have a devastating impact on developing nations. The focus of current resilience models is predominantly on developed countries, and less is known regarding how they might perform in other contexts. This paper presents a comparative case study in Bangladesh using two established hazard resilience models: the Disaster Resilience of Place (DROP) Model and the Spatially Explicit Resilience-Vulnerability (SERV) model. Using publicly available data, we applied the two respective models at the district level in the context of flooding, the most extensive hazard in Bangladesh. The results of our model implementations were evaluated by Bangladeshi emergency managers and researchers regarding their potential utility. While emergency managers preferred the SERV model results, researchers thought the DROP model results were preferable. This suggests an opportunity to create a model that would serve the flood resilience needs of managers and researchers.

In an era where media coverage of governmental issues focuses almost entirely on national-level actors (e.g., presidents, the Federal Emergency Management Agency), it is still the case that local organizations are key actors in emergency planning and response. In the case of a hazard, it is local organizations that will be the first responders on the site of the event. While studies of trust in government have revealed connections to household preparedness, few of these studies have focused on local level organizations. This presentation used Boolean approaches to measuring trust in local organizations involved in emergency management to investigate the connections between this trust and the extent to which households report preparing for hazards in their community. The authors reviewed the results of a national survey of 3,870 respondents to build a model of the connections between trust in local first responder organizations, other key controls for preparedness behavior (e.g., risk perception, disaster experience, demographics), and an activity-based measure of household preparedness for disasters. Results suggest that the minimum, mean, and maximum level of trust are all positively related to greater degrees of household preparedness, even controlling for general social trust. The presentation will discuss how local organizations can use this trust to overcome the problems related to historic lows in national and federal-level trust in government organizations.

The Marshall Fire (December 2021) was the most destructive wildfire in Colorado’s history. The fire burned 1,084 homes in the City of Louisville, the Town of Superior, and unincorporated Boulder County, all upper-middle class communities with progressive climate change goals. After 18 months, what is the progress of rebuilding, and how has the fire shaped the public’s support for long-term resilience strategies? This paper draws on a unique, multi-wave household survey (n=824 and 576) of the residents of Louisville, Superior, and Boulder County and semi-structured interviews with 22 recovery decision makers from local and state government, community-based organizations, and non-profit entities. We focus on two key findings. First, that rebuilding progress is strongly influenced by income and insurance resources, even in relatively prosperous communities, and second, that public support for climate resiliency policies decline in the face of a disaster recovery. We discuss these findings and how these dynamics shape local government decision-making about resilience policies. These findings are relevant and transferable to other communities that will be newly under threat as climate change increasingly fuels extreme weather events like wildfires.

While there are more than six decades of research on how people respond to risk information, less is known about information seeking in the context of competing official and non-official sources of risk information available through social media. There is also a dearth of social science research on disasters in Puerto Rico, especially on risk communication. Moreover, Puerto Ricans have been coping with consecutive extreme events in the last five years, including hurricanes, landslides, an earthquake sequence, and the COVID-19 pandemic, challenging more traditional event-based definitions and studies of disasters. For the last two years, our team has been conducting research to understand how in the context of navigating multiple disaster cycles simultaneously, emergency management-related personnel, as well as residents, make decisions about what actions they should take to protect themselves and others against the possibility of a significant earthquake in the future. Using data collected through in-depth interviews with key informants (n=51), a semi-structured survey (n=428), and focus groups (8 facilitated) in three communities in the municipalities of Guayanilla, Guánica, and San Juan, we explore how emergency responders and residents' access, interpret and use official and non-official earthquake risk information available through social media and other sources to inform immediate and long-term protective-action decisions. This research is vital for disaster reduction as it may assist in effectively planning and promoting situational awareness, hazard adjustments, and preparedness for future earthquakes.

During disasters, households may experience lost communication and roads may become impassable; thus, a common recommendation is to prepare an emergency supply kit. The decision to prepare for disasters becomes more challenging during pandemics because individuals may be hesitant to evacuate to shelters. Therefore, many estimates before 2020 may no longer reflect current levels of preparedness. This study aimed to provide information on what actions people take to prepare for disasters, determine whether households have supply kits, and help guide next steps. The authors conducted descriptive statistics, chi-square analyses, and backward-step regressions on 10 questions collected through Porter Novelli’s Consumer Styles surveys in Fall 2020 (n=3,625), Spring 2021 (n=6,455), and Fall 2021 (n=3,553) to examine factors that contribute to household preparedness levels and supply kit ownership. Being married, having children in the home, and having a household income of $150,000 are all associated with increased preparedness levels. Persons living in mobile homes, RVs, boats, or vans are half as likely to have preparedness plans compared to those living in single family homes. While there was a significant association between preparedness level and supply kit ownership, overall kit ownership is lacking. While most believed a kit would help their chance of survival, only a third had one. Age, sex, education level, and region of the country were significant predictors of kit ownership. This study helped close existing gaps surrounding preparedness and emergency supply kits to guide public health research and prevention strategies to help reduce adverse health impacts during disasters.

Lack of household preparedness is a public health concern because it consumes first responders’ time and can deplete community health resources. A common recommendation is to prepare an emergency supply kit that can be used to shelter-in-place for approximately three days. This study’s goal was to determine the efficacy and public health impact of emergency supply kits among Hurricane Ian’s impacted population. The authors conducted a cross-sectional survey shortly after Ian’s landfall (October-December 2022). Sampling included areas within the hurricane’s force wind swath, excluding mandatory Evacuation Zone A. Using Address-Based Sampling (ABS), the authors mailed a survey to 5,000 households with an option to complete the survey by web. The authors received 1,342 (27%) unique responses. Preliminary results showed most respondents (74%) indicated they did not evacuate. Among these households, 66% reported minor damage, 34% major, and 1% destroyed. While most (88%) knew what an emergency supply kit was, only 61% had prepared one prior to Ian’s landfall. Most (73%) believed their kit was extremely or very helpful. A radio, generator, and cash were the top items that respondents reported not having in their kits and wishing they had included. Overall, emergency supply kit ownership can be improved. These data are essential in characterizing emergency supply kit ownership and documenting how kits were used immediately after the impact of a major hurricane. Data can help understand why and how households prepare for emergencies, improve emergency supply kit item lists, detail effectiveness of their use, and tailor messaging for future storms. 

Hurricane Michael was a Category 5 storm that hit the Florida Panhandle in October 2018, causing massive destruction and a prolonged recovery. This study aimed to understand whether impacted coastal counties experienced excess mortality post-storm, and if so, which demographic groups and causes of death experienced that excess. Vital statistics death records from 2013 to 2019 for Bay, Gulf, and Franklin counties were analyzed using seasonally adjusted auto-regressive integrative moving average to forecast mortality rates with 90% and 95% confidence intervals for one-year post-storm based on the trend of the previous five years. All county resident mortality as well as by age, sex, race, and cause of death was analyzed by annual quarter to determine if any observed mortality rates were beyond the upper bound of the forecasts. For quarter 2 of 2019, the forecasted mortality rate for Gulf and Franklin counties was 227.8 per 100,000 (95% CI, 159.8), 316.0 vs. 322.5 observed mortality rate. Observed mortality rate for those 55 and older (87.9% of records) was 771.2 vs. forecasted 575.4 (90% CI, 426.0, 755.2). The observed mortality rate was 365.0 for Whites (88.2% of records) vs. forecasted 253.9 (95% CI, 177.9, 352.3). During quarter 3 of 2019, the observed mortality rate for cancer was 80.6, vs. forecasted 48.7 (90% CI, 28.9, 76.2). Bay County showed no significant difference between observed and forecasts. These results provide the opportunity to enhance hurricane mitigation, preparedness, response, and recovery by identifying those populations that are vulnerable to prolonged natural disaster recovery.

This research explores the experiences of vulnerable households in the aftermath of the EF3 2018 Tornado that hit Marshalltown, Iowa (United States). The study focuses on low-attention disasters, which are often overlooked in the media and by policymakers, but which can have significant impacts on vulnerable populations. Discussions on consequences of low-attention disasters are mostly confined to developing countries in the literature. Few studies address the manifestations and consequences of low-attention disaster in the global north. Our case study of Marshalltown adds to the literature by highlighting an example of neglected disasters in the developed countries, and its consequences for the vulnerable population and low-income households. Using qualitative data from interviews with low-income, minority, elderly, and disabled households in Marshalltown, the study examined the challenges faced by vulnerable households in the aftermath of the tornado. The results indicate that vulnerable households faced unique challenges, including limited access to resources and communication barriers. When there is less aid available to distribute, those in need and vulnerable are more likely to have even more difficulty accessing aid compared to times when more aid is available. These challenges were compounded by pre-existing inequalities and systemic barriers that exacerbated the impacts of the disaster. This shows that while there are major differences between countries in terms of the amount of available housing recovery resources, neglected needs and lack of access to resources in low-attention disasters leads to comparable consequences in different places.

Despite the increase in frequency and intensity of wildfires around the world, little research has examined evacuation logistics and household preparation time in such a rapid-onset scenario. To address this gap, this study analyzes 152 household residents’ responses to the devastating 2018 wildfire in Mati, Greece, where the second-deadliest wildfire of the 21st century took place. The survey instrument measured residents’ awareness of wildfire risks and pre-disaster evacuation preparation, as well as decision about route choices, number of vehicles to take, preparation time, and travel time. Analyses show that disaster preparedness practices (e.g., having a fire-resistant house, managing vegetation) as well as exposure to previous wildfire events influence the residents’ wildfire logistics-related decisions. We also compare households’ evacuation preparation time during the Mati wildfire event with past tsunami studies in four Cascadia Subduction Zone (CSZ) communities to highlight the differences in evacuation preparation time estimates between different disaster scenarios. These analyses can help emergency managers to develop effective wildfire evacuation plans that reduce the loss of lives.

The effects of Hurricane Katrina and other natural disasters have increased the focus on nursing homes, particularly nursing home emergency preparedness. Due to numerous emergency preparedness insufficiencies, requirements were implemented to protect the vulnerable population of older adults who reside in nursing homes. In 2016, the Centers for Medicare and Medicaid (CMS) added specific emergency preparedness requirements to the existing fire safety requirements for regulated nursing homes. These deficiency evaluations are the only routinely collected source of information detailing compliance with the Emergency Preparedness Rule for regulated facilities. The purpose of this study was to provide a baseline assessment of nursing home facility characteristics associated with having an Emergency Preparedness Rule deficiency citation. This study tested the Organization-Based Conceptual Framework that relates internal and organizational factors to Emergency Preparedness Rule deficiencies for CMS regulated nursing homes. This research is significant in several ways. First, examining CMS Emergency Preparedness Rule deficiencies as one measure of resident safety may provide further understanding about how complex resident safety issues are in nursing homes. Second, no research to date has presented a nationally representative picture of CMS Emergency Preparedness Rule deficiency citations in nursing homes. The study found that there was support for four of the five variables of interest identified in the study. The resulting data can be used to evaluate facility compliance with the new federal guidelines and consider the overall effectiveness of the CMS Emergency Preparedness Rule on the emergency preparedness process in U.S. nursing homes.

This longitudinal study in Lumberton, North Carolina, is investigating ongoing recovery and mitigation processes after Hurricanes Matthew (2016) and Florence (2018) and the impact of the COVID-19 (2020) pandemic on recovery and resilience. Five survey waves were administered to a random sample of households between November 2016 and June 2022. Information was collected across several dimensions, including occupancy status and tenure, damage, dislocation and recovery status, recovery funding, insurance coverage, mitigation and preparedness strategies, and the effects of the COVID-19 pandemic on housing repair and household recovery. In the fifth wave, 38% percent of respondents reported housing unit damage from both hurricanes, 18% from Matthew only, and 5% from Florence only. Of those reporting damages, only 58% had completed repairs. Given COVID-19 and its economic impact, this wave highlighted the compounding effects of sequential disasters on households’ recovery processes. Newly added survey questions subjectively measured perceived recovery which provides a different perspective than typical objective measures such as return to permanent housing or housing repair completion; this supports the notion of separate and heterogeneous household and housing recovery outcomes. Furthermore, respondents tended to evaluate their own recovery more positively than the recovery of their neighborhoods and city overall. The longitudinal results show the complex nature of recovery from consecutive crises. Households reporting damage from both hurricanes are less likely to have completed all needed repairs. These households reported the most use of recovery aid from government and nongovernmental sources. Approximately 25% of households reported that COVID-19 disruptions majorly impacted repairs.

Floods are among the most destructive natural hazards, with devastating socio-economic consequences that disproportionately affect marginalized communities. Floods have become more frequent and intense recently, requiring collaboration among various stakeholders with different authorities and interests for effective flood management. However, the lack of coordination and collaboration among stakeholders has often resulted in counterproductive efforts, exacerbating the flood problem. This study examines the flood management challenges on the Ciliwung river in Indonesia by analyzing the perspectives of various stakeholders including public, private, and civil society actors. Network analysis techniques are used to identify the key actors involved in flood management, revealing insights into the relationships between stakeholders. We also employed a heatmap analysis to determine program priorities and budget allocations within the government’s flood management program. Using interviews and qualitative analysis, the findings reveal some misalignments of perspectives that impede collaboration among actors in flood management. The government-led efforts to build consensus across stakeholder groups through coordinating teams have failed due to unfulfilled stakeholder expectations. To address these challenges, this study maps the emerging perspectives based on the Sendai Framework guiding principles to develop recommendations for action planning. The three dominant guiding principles (i.e., shared responsibility, coherence policies, and local community empowerment) outline the priority steps in improving flood management to enhance community resilience. This study emphasizes the importance of aligning stakeholders’ perspectives to support effective flood management and discusses ways to help the stakeholders develop strategies to enhance community flood resilience and promote sustainable development.

Over the last decade, there has been a noticeable increase in the recognition of the value of cross disciplinary teams to address complex and system problems, particularly in the arena of climate adaptation science (inclusive of hazard mitigation planning/implementation, disaster risk reduction, sustainable land use, and watershed management). As a result, a new interdisciplinary field has emerged—the Science of Team Science. This presentation offers an exploration of the benefits, justifications, and challenges of integrating team science in post disaster and hazard-prone landscapes. A critical aspect of this type of work is the consideration of justice and equity. Through a case study assessment of a multi-institutional NSF funded project analyzing the impact of Hurricane Maria on critical infrastructure and livelihoods in Puerto Rico, the presenter examines the application of interdisciplinary and transdisciplinary science to a complex social, cultural, and geophysical setting. 

Wildfires, which are increasing in frequency, duration, and intensity, are measurably affecting vulnerable populations, labor, housing, and education. Wildfire smoke is a growing problem for groups that face greater economic barriers than the general population, such as low-income families, housing-vulnerable communities, and frontline workers. Smoke—in contrast to the wildfires from which it originates—presents hazards hundreds of miles from the locations of property losses connected to wildfires, still the damages from smoke are just becoming better understood. For this work, an 11-year record (2011-2021) of wildfire smoke plumes maintained by the National Oceanic and Atmospheric Administration (NOAA) was combined with U.S. Census block group centers of population and corresponding social characteristics to center populations, communities, and economic sectors disproportionately burdened by climate hazards. The results show that the average community in the United States went from experiencing about 29 days per year where any wildfire smoke might be overhead from 2011-2015, to experiencing about 45 days per year from 2017-2021. Over 85% of the U.S. population lives in counties that experienced a significant increase in the number of days of the densest “heavy” wildfire smoke between the two periods. Dramatic increases in disruptive smoke overlap with eligibility for existing financing programs that could help build resilience to smoke-related damages, particularly for frontline communities. Programs that target low- and moderate-income communities and communities of color may have outsized importance in building broad economic resilience to climate risks. 

This research project is an intersectional, multi-level qualitative analysis of Mexican immigrant women that aims to reveal how housing tenure and legal status affect access to disaster recovery resources. This research focuses on Hurricane Harvey, which made landfall in Houston, Texas in 2017. For Part I, 172 U.S. disaster recovery planning and policy documents at the federal, state, and local levels were collected to conduct an intersectional and critical document analysis. This improved understanding of the legislative possibilities and constraints that shape the disaster recovery process for Mexican immigrant women. For Part II, preliminary interviews with service providers (n=19) from community-based organizations (n=7) in Houston were conducted in 2019. Follow-up interviews with providers (n=8) were conducted in 2021 in order to understand how subsequent disasters, such as Tropical Storm Imelda (2019), the start of the COVID-19 pandemic (2020), and Winter Storm Uri (2021) have affected their efforts to continue assisting Mexican immigrants in the four years since Harvey—with a specific focus on women’s experiences. Almost 40 hours of ethnographic observations with the organizations were also compiled. Finally, for Part III, qualitative, semi-structured interviews with Mexican immigrant women (n=56) were conducted to understand their recovery from their own point of view, with a focus on how housing tenure and legal status affect access to disaster recovery resources. This presentation discusses the strategies employed when conducting qualitative research with a hard-to-reach population, such as sharing research plans with community leaders and partnering with a community member for recruitment.

Many communities are at risk from multiple hazards. However, understanding of the effects of multi-hazard contexts on risk judgments and behaviors is limited. Past studies tended to test which of multiple hazards people choose to think about and act on. A novel approach was used to test these effects, using an experimental survey where groups of participants from Wellington, Aotearoa New Zealand (NZ) were asked to consider a different combination of the same two hazards which both pose a risk to the city: earthquake and tsunami. Asking people about one hazard had significant and meaningful effects on their judgments for a second hazard. Generally, these findings reflect a decrease in risk perception of the second hazard, suggesting that people’s risk perception can be overloaded if they are asked to consider multiple hazards simultaneously, which is frequently the case in reality. Further, there were differences in the pattern of effects depending on which hazard was presented first, with important implications for public education efforts. In particular, many communities in NZ are at risk of local source tsunami where strong earthquake shaking is a key warning. However, the findings suggest that communication around tsunami evacuation needs to bear in mind the impact that experiencing a strong earthquake will have on tsunami risk perception and ability to act quickly.

Communities in the United States are striving to improve their hazard resilience, climate change preparedness, environmental sustainability, and social equity, among other crucial goals. One of the key methods to achieve these goals is by including them in community plans, often using terms such as “resilience”, “adaptation”, “sustainability” and/or “equity”. These terms encompass a variety of somewhat complementary goals and approaches to community planning and in themselves have diverse and evolving meanings in the academic and practitioner communities. Yet, little is known about how these goals can be best integrated to produce more effective and/or just planning outcomes, or how public perceptions of these terms may influence support for such efforts. As part of a larger body of work informing integrated planning approaches, the goals of the present research were to better understand perceptions of resilience, climate adaptation, sustainability, and social equity among a representative sample of the U.S. population. Data were collected by employing a 1 x 4 experimental survey design, to understand what concepts and goals are associated with these four terms as they relate to community planning, as well as concerns about, engagement in, and support for such efforts. Analysis identified areas of goal and definitional overlap and distinction among the public, and examined which associated factors (e.g., geographic region, trust in planning institutions, issues, and hazards of concern) were most predictive of support. These results contributed to guidance for integrating and communicating emerging, convergent planning strategies, with the goal of enhancing community resilience and disaster recovery.

Social vulnerability refers to the social aspect of a societal entity’s inability to withstand adverse impacts of a hazardous event. It is usually quantified as a function of pre-event indicators of societal characteristics. To facilitate disaster management, many scholars have proposed indices to measure and map social vulnerability. However, most of these index-based works do not consider historical information on event losses. The lack of empirical association with loss data throws into question the veracity and validity of the index-based methods. Alternatively, in this research presentation, the researcher introduces a new methodological paradigm of empirical predictive modeling (EPM) for quantifying social vulnerability. The EPM approach leverages data on hazard strengths, exposed values, vulnerability indicators, and losses of events to empirically establish mathematical models to indicate social vulnerability. The produced metrics of social vulnerability correspond to the expected probability or expectation of loss of a societal entity given its possible experience of hazard strengths of a potential or future event. Coupled with the estimation of exceedance frequencies of hazard strengths of events, the EPM social vulnerability metrics can be applied to assess risks in terms of expected future losses in specific units such as lives and U.S. dollars. The practicality of the EPM paradigm is presented via its applications to census tracts of North Carolina, townships of Taiwan, districts of Nepal, and countries around the world within the context of the hazards of flood and earthquake, respectively.

As the financial costs of disasters continue to rise, many communities struggle to secure enough funding to implement hazard management strategies. Identifying federal disaster programs can be challenging because programs can be added and removed over time, eligible activities can change across events, and some programs may not be available until years after a major event has occurred. To address this issue, this research established a systematic approach to identify disaster programs through the digitization and analysis of the annual Federal assistance listings (formerly known as the Catalog of Federal Domestic Assistance) for years 2003-2019. Whereas previous research on federal spending may focus on singular programs, this analysis takes an alternative approach by aggregating disaster spending from the individual program level, allowing for spending analyses by program characteristics. Therefore, this presentation addresses the following research questions: (a) What programs are available for disaster management and how consistent are these programs through time? (b) How are these programs distributed across the mitigation, preparedness, response, and recovery phases of emergency management? (c) Who is eligible (e.g., for-profit businesses and households, nonprofits, local governments, and other entities)? And (d) How much funding is available? The findings presented will include the substantive results of the disaster spending analysis, a discussion of the reliability of the data and how other researchers might apply to methodology to other topics, as well as guidance for how communities might use these results to link programs with their disaster planning and policy initiatives.

Stated preference contingent valuation methods allow researchers and policymakers to estimate economic value for non-market or otherwise difficult-to-value goods. Critics are concerned with overly optimistic estimates and other biases that result from the hypothetical nature of reported preferences. Researchers have a number of methods to address these concerns, and many are widely used. One less common method is spike models, despite being introduced in the late 1990s. These models allow for the probability, thus possibility, of zero willingness-to-pay (WTP). This paper demonstrated the utility of spike models in estimating WTP and economics value using a case of community-based resilience, a United Nations Disaster Risk Reduction priority, in three states in the United States. Estimates not incorporating spike models suggested mean WTP for a five-week community-based resilience training between $42 and $135 per individual in a one-off payment. However, spike models resulted in much more modest estimates of individual WTP of approximately $5 in a one-off payment. These differences suggest modelling choices that allow for zero or even negative WTP estimates are essential to informing policy and decision-making and greatly affect aggregation exercises. Results also suggested that key economic demographics such as income level and home location are associated with the likelihood of being willing to pay. However, given the low cost of implementing most community-based resilience programs, even these lower total values suggest a potentially positive benefit-cost ratio. Thus, policymakers interested disaster resilience and risk reduction may consider collaborative, community-based training as both a publicly desired and cost-effective option.

Increasingly, Pre-K-12 public school teachers are expected to assume responsibilities related to emergency preparedness and response associated with threats of violence, disease outbreaks, and natural hazards. Such expectations coupled with a lack of teacher involvement in disaster planning can result in limited knowledge of and decreased motivation to comply with safety protocols, including disaster drills. Furthermore, teachers may experience anxiety about their responsibilities in a crisis related to keeping themselves and their students safe. This study proposes that the key to enhancing emergency preparedness in the school workforce is to provide opportunities for engagement with district safety and security staff through town hall sessions, and to incorporate “psychological preparedness” and peer support within an emergency management training framework emphasizing shared leadership. Initially, the authors consulted 64 teachers across six school districts in Colorado about emergency management and mental health concerns. After developing a half-day group intervention in response to identified needs, the authors implemented the training in six schools, training 434 members of the public school workforce. Using a matched comparison waitlist control design, the authors administered surveys at two time points for a total 232 participants completing both surveys. This work explores how the intervention influenced the following domains: emergency preparedness, psychological preparedness, mental health, peer support, and shared leadership. A brief cost-effective school-based mental health integrated emergency preparedness intervention has the potential to enhance preparedness while also providing a mechanism for continued workforce engagement in the process of making related improvements throughout the school district. 

Landslide hazards are of special concern for public health during hurricane disaster response in Puerto Rico. After Hurricane Maria in 2017, more than 70,000 landslides occurred across communities in the mountainous interior of the main island, and hundreds more occurred after Hurricane Fiona in 2022. Despite documented impacts, it was previously unknown how many people lived with landslide risk. This study used a mixed-methods approach to assess the social vulnerability of landslide-prone communities in Puerto Rico, guided by the following research questions: What is the total population living in high-risk areas for landslides in Puerto Rico? What is the relationship between social vulnerability and landslide susceptibility? How do communities experience and self-define vulnerability to landslides? The authors first analyzed the population exposed to landslides at the 100-meter scale using a landslide susceptibility model and gridded population data, finding that about 33% of the population (approximately one million people) live in areas of high to extreme landslide susceptibility. The authors also conducted two focus groups in the municipality of Utuado (n=22) and semi-structured interviews with agronomists, emergency managers, community leaders, and local researchers (n=11) to assess the social dimensions of landslides. Initial findings indicate that landslides were highly disruptive to the road network, which limited residents’ mobility and access to water, food, and healthcare. Landslides also blocked access to farms, leading to lost harvests and income. Age, disability, and remoteness were highlighted as top indicators of vulnerability. Finally, an interactive map was developed to visualize and share synthesized results.

The advent of artificial intelligence (AI) tools for internet search (like ChatGPT) promotes calls for enhanced internet search and information retrieval capabilities. In response to these calls within the natural hazards and disasters domain, this study focuses on introducing Flood Finder, an information retrieval system that leverages AI to extract information on past flash flood events from the internet. The system operates in three stages. First, it integrates publicly available data, and machine learning to retrieve web pages containing relevant information about past flash flood events from the internet. This stage eliminates current trial-and-error searches in conventional search engines. Second, it uses a domain-specific multi-label classification model to classify paragraphs in relevant web pages into specific topics based on their content. Finally, it employs a custom named entity recognition model to extract key information from classified paragraphs. The study also includes use-case examples of Flood Finder to demonstrate how AI can be used to make internet searches for information on past flood events more accurate, and efficient. Practitioners and researchers in the natural hazard and disaster community can use this system to obtain detailed data on past flood events to inform the development of more effective preparedness and mitigation plans for future events.

The COVID-19 pandemic dramatically changed the way society interacts with spaces and places. This is especially true for shelter residents who were perceived to be at a greater risk for contracting the virus due to several factors, including their communal living environment. To understand how shelter residents defined their risk and resilience, the authors implemented a PhotoVoice project in collaboration with five artist-participants living in a shelter in Miami, Florida. Although the shelter was at times perceived to be a safe place, it was often at the expense of self-determination. Despite a restrictive environment, residents found ways to adapt during an unprecedented and challenging time.

The negative impacts of climate change and climate-related incidents call for building the resilience capacity of entrepreneurial teams as it enables them to better cope with these disasters, further promoting their community contributions. Focusing on entrepreneurs in Canada, this project aimed to build individual and collective resilience capacity in community settings in two ways. First, by building their capacity to navigate community-based resources to support, maintain, and enhance their and their teams’ health and well-being. Secondly, by generating their capacity to facilitate community-based resources to serve them and their teams in culturally meaningful ways. This project used in-depth interviews with 60 entrepreneurs from 13 provinces and territories across Canada. Preliminary results detail the current knowledge base and explore barriers surrounding community-based, diverse resources that support entrepreneurs’ health and well-being post-disaster. Themes included discussions of what resources and supports are currently available, barriers to those resources and supports, and what supports are needed in entrepreneurial spaces. From these initial results, the authors provided evaluations and best practices in entrepreneurial resilience for those at the incubator/accelerator level to enhance the entrepreneurial ecosystem holistically. These outcomes support entrepreneurs in facilitating different community-based resources to serve them and their teams in diverse community contexts post-disaster. The Canadian findings shed light on valuable references for international entrepreneurial teams to build their resilience capacity to arrange community-based resources to deal with climate change, disaster, and other crises.

Land use and population in major river basins have drastically changed since World War II in Japan, including urban development in flood-prone areas. Although there are several studies describing the transformation of agricultural land into urbanized land use, few studies have compared the dynamic interrelations between land use and flood-exposed population. This research examined two questions; (1) How have land use and population transformed in each river basin over the last 50 years in Japan? (2) What types of land use have been sacrificed for urban development in flood-prone areas? These questions can inform urban planning, especially policy options with nature-based solutions for flood mitigation. The authors conducted tempo-spatial and quantified mesh-unit analysis with Geographic Information System (GIS) and R software approximately every ten years since 1976 to typologize the urbanized river basins in Japan. For the mesh-unit analysis, the authors used the segment concept, which categorizes river basins based on the four steps of gradients. The tentative results showed that river basins in category 1 maintained paddy fields in flood-prone areas despite the high population growth, while natural land use in the other river basins declined and was replaced by urban development. As a result, the population exposed to floods increased. Moreover, the ratio of land use in relation to buildings in flood-prone areas relative to whole river basin does not necessarily correspond to that of the population. The significance of this study suggests comparison of river basin land use and population in tempo-spatial dynamic manner.

Globally, disaster displacements disproportionately affect Indigenous communities due to historical and ongoing racist policies that dispossessed land, attempted cultural genocide, and forcibly relocated people to hazardous locations. Indigenous communities are simultaneously vulnerable to disasters—through shared histories of colonization and systemic exclusion from power—and resilient, through localized knowledge systems and cultural practices. Drawing on decolonial disaster risk reduction, this research focused on disaster evacuations: a well-intentioned emergency management practice that often results in compounding trauma within Indigenous communities. The authors adopted a cultural safety lens for mitigating the unintended consequences of technical fixes for fundamentally human problems. The research journey was guided by the convergence framework: (1) Researchers: Guided by the Indigenous Circle of Matriarchs, this project brought together Indigenous community members to conduct research in Lil’wat Nation (British Columbia), Siksika Nation (Alberta) and Mattagami First Nation (Ontario). (2) Community-led education/training: The research training for community leaders was emergent, intergenerational, and peer-led. It created safe space for mutual learning and recognized lived experience as expertise. (3) Solution: A space was created for emergent solutions and collaboration such as guidance for culturally safe evacuations, design for hosting sites, and Indigenous-led training. (4) Connecting researchers/coordinate diverse teams: Within an Indigenous team, diverse geographies, ages, genders, community sizes, and varied lived experience with disasters were connected. (5) How can Indigenous-led research, data collection/sharing, and solutions implementation be supported and funded? The authors offer this question as an opportunity to discuss and collaborate.

Some disasters with a historically low probability of occurrence in an area have become more frequent in different parts of the world. According to the Protective Action Decision Model (PADM), the pre-decision process (risk information exposure, attention, and comprehension) affects risk perceptions. However, we know less about how risk information regarding a rare disaster event affects people’s risk perceptions. Specifically, tornado warnings have a lead time of 10 to 15 minutes before a tornado strikes. People have a limited time to perceive the risk and take protective actions. Therefore, it is important to understand what risk information people utilize to inform their risk perceptions if they are unfamiliar with the hazard. This experimental study used tornado scenarios to examine this issue. The study recruited 136 participants from the state of Washington. A machine learning (random forest regressor) approach was used to analyze the data. Leave-one-out cross validation was employed to evaluate the model (average absolute error 0.56). The results showed that text messages that describe potential hazards, location impacted, and storm distance to the hypothetical home location are the most distinctive factors in shaping participants’ risk perception, not the map-based images such as tornado warning polygons or doppler radar images. Further, findings suggested map-based images negatively impact people’s risk perceptions when they face a rare tornado event. The findings of this study could help meteorologists, emergency managers, and local officials better communicate risk information to the public when a rare tornado event happens.