Researchers Meeting Abstracts

Climate change crises, predominantly caused by unsustainable environmental actions, contribute to issues in urban planning, natural resource management, water sanitation, and hygiene infrastructure. This allows for large-scale exposure to flooding that affects the wellbeing and livelihood of people. This work examined the association between different health issues, like cholera and psychological illness, and flooding, especially in post disaster events. It considered the frequency of flooding and incidents of health problems between 2000-2020 using a systematic literature review and secondary data sources. Displacement is quantified, and the implications of flood on both the physical and mental health of flood victims, are discussed in this study. This study revealed a strong association between cases of illness experienced directly and indirectly in post-flood events. The most recorded infectious disease (cholera) was used as a case study in the mapping procedure. Outbreaks of cholera in 2010 and 2020 were speculated to be directly related to sanitation, and water supply exposure, due to high rates of flooding and runoff which likely aided its rapid spread. The mental stress and challenges experienced by affected victims remain the most unrecognized post-flood illnesses, leading to short- and long-term effects on the people. Based on findings of this study, there is a need for a more accurate epidemiological study on infectious diseases and disasters, especially those associated with climate change-related disasters. Psychological resilience intervention should be recognized as a mitigation measure to reduce the mental exposure experienced by flood victims.

Various post-disaster recovery programs are designed to support disaster survivors at different levels: federal, state, and local. Community Development Block Grant and Disaster Recovery (CDBG-DR), a federal disaster assistance program, supports disaster survivors’ transitions from temporary shelters to permanent housing areas impacted by Major Disasters for which significant unmet needs still exist after Federal Emergency Management Agency (FEMA) programs conclude. CDBG-DR programs measure success as recovery process completion by the applicant. However, recovery for the many who drop out of the program remains incomplete. For all applicants, social vulnerability plays a significant role in differential disaster outcomes and influences the pace and progression of the post-disaster recovery process. However, very few studies have explored the links between social vulnerability variables such as race and ethnicity and the completion of the reconstruction process. Guided by the Social Vulnerability Index (SoVI) framework, the authors analyze CDBG-DR applicants data after the 2015 South Carolina Flood and 2016 Hurricane Matthew to explore (a) differences in who dropped out and who completed the program, (b) which social vulnerability variables are driving dropouts, and (c) whether those who dropped out and those who completed the recovery programs were served differently during the reconstruction process. Preliminary findings reveal that various social vulnerability variables such as age, education, gender, housing, race and ethnicity, and unemployment play a role in dropping out from the program. These results could be useful for developing interventions aimed at keeping people in these recovery programs through completion.

Green Infrastructure (GI) projects at the neighborhood-scale are nature-based solutions that can provide flood protection, clean air, diverse habitat, and beautiful green spaces to urban areas that provide additional co-benefits to residents. GI is thought to increase disaster resilience and is increasingly popular in cities across the USA. Despite many ecological, economic, and social benefits associated with GI, there is a gap in our understanding of inequities embedded in urban greening aspects of GI projects, particularly when accompanied by increased gentrification. Such GI interventions can have an ambiguous social impact if the vulnerable citizens (e.g. lower-income populations) are forced to pay higher housing costs or are displaced, a phenomenon referred to as “Green Gentrification”. Here,a quasi-experimental design is implemented to estimate the causal effect of GI interventions on Green Gentrification outcomes. Using the case study of Washington, D.C. and a difference-in-differences geospatial regression model we estimate the effects of GI projects implemented between 2013 and 2019 on the median housing value, median gross rents, median household income, and the proportion of black residents at the census block group level. Findings have implications for better addressing social inequities and climate injustice issues associated with disaster resilience planning in urban communities.

Natural hazards often generate widespread physical damage to the exposed community. Underserved people experience disproportionately higher societal impacts, and rising environmental justice and equity concerns. Despite the ubiquity of open data, and advancements in geostatistical analysis, data-driven assessments of such societal impacts lag behind policy needs for developing equitable actions on natural disasters. The current study seeks to fill this gap through an application of geospatial data and complex network analyses. Using the case of Hurricane Maria in Puerto Rico, the authors leveraged publicly available data to demonstrate patterns of social inequities in access to critical infrastructure (e.g., hospitals, schools, ports, etc.), post disaster. This model encompasses three components that were integrated into a geostatistical regression analysis: (1) the Center for Disease Control social vulnerability index (SVI) for every census tract, (2) the United States Geological Survey map data showing the distribution and density of landslides triggered by Hurricane Maria, and (3) the map of primary and secondary roads in Puerto Rico. Once overlaid, the map of landslides and the roads were used to estimate road damages. Through complex network analysis the authors of the study estimated relative road access to critical infrastructure lost in each census tract, post-Hurricane Maria. The authors then quantified the relationship between lost-access patterns and social-vulnerability patterns to assess social inequity associated with this natural disaster. Results can have important implications to advance data-driven assessments of environmental justice, and to address the disproportionate impacts that climate change has on vulnerable communities.

Disaster risk reduction policies effectively address individual hazard risks in isolation, but recently, policymakers have also had to deal with cascading disasters. As if that was not complex enough, the COVID-19 pandemic and climate change impacts have added new uncertainties to a world overshadowed by many hazards manifesting separately, together, and in cascades. The emerging world must deal with complex risk; a concept that policymakers need to understand quickly and address. This study focused on complex risk in the coastal communities of South Louisiana, which involved qualitative research on risk tradeoffs in disaster-induced relocation. Interviewees dealt with hurricanes, storm surges, floods, land subsidence, salinity ingress, and environmental pollution, which destroyed their homes and livelihoods and eroded their health and wealth. They watched neighbors, grocery stores, churches, clinics, schools, and gas stations leave the bayous. Their communities shrunk, and fishing camps moved in. Their incomes crashed, but their expenses increased. Property values sank with the subsiding land, and insurance premiums rose with the floods. When they attempted to adapt to environmental change, a restoration project killed their oyster beds and decimated shrimp populations. The people of the bayous live in this cloud of risks, dealing with them one at a time, to the best of their ability. This study highlighted the complex risk in coastal Louisiana, and compared how the Coastal Master Plan and the LA SAFE planning process have approached disaster risk reduction differently in this context, as well as the implications of complex risk for planning and policy.

Disasters generate tremendous amounts of debris that negatively impact communities and overwhelm waste management infrastructure. Reasonable debris forecasts and estimates are critical to anticipate the management and disposal needs of a community following a disaster. However, detailed post-disaster waste data is often unavailable. In this study, a novel set of post-disaster waste data collected in Beaumont following Hurricane Harvey that included debris tonnages and coordinate locations of each debris removal in residential areas was investigated. The dataset was utilized to examine the factors that influenced debris generation, the amount of debris generated, and how those amounts compared to estimates from standard debris forecasting and estimation methods. The study found that elevation and proximity to flooding hazards played a significant role in the type of flooding experienced, and, in turn, the amount of debris generated. Areas in Beaumont that experienced inundation from river flooding rather than strictly urban flooding had higher water depths in homes and generated greater amounts of debris. The debris generated was overpredicted by standard debris forecasting and estimation methods. However, current forecasting and estimation methods are not meant to estimate debris from urban flooding. Urban flooding is an increasingly prevalent issue in many areas following a natural disaster and forecasting and estimation methods that consider urban flooding should be developed.

Evacuation orders impacted over eight million people in the United States in 2017 alone. Shelter for displaced persons is among the seven community lifelines essential to human health and safety in the National Response Framework. Stabilization of lifelines is a primary disaster response goal. Shelter location models often couple shelter locations with evacuation routing decisions, as shelter locations directly impact evacuation time. Many existing shelter location models assume displaced persons will go to the nearest shelter. This study examined the nearest-shelter assumption for actual response shelter configurations in three North Carolina cities following Hurricane Florence and two model-prescribed configurations for the same event. The focus was on shelter placement decisions for evacuees whose accommodation type is public shelter and whose destination choice is local. Case study results indicated that a displaced person’s second-nearest shelter often required less than one mile of incremental travel beyond the distance to the first-nearest shelter. Further, nearest-shelter behavior resulted in demand imbalance across open shelters. Finally, nearest-shelter predictions of the proportions of shelter seekers choosing each open shelter did not align with historical client data. Taken together, these observations did not support a one-size-fits-all assumption of nearest-shelter behavior. Empirical research to understand shelter destination choice is needed, along with new shelter location planning models capable of a more holistic treatment of destination choice. As communities are coping with multiple hazards and the ongoing COVID-19 pandemic, it is increasingly important that shelter location models account for complex community and household risk and behavior factors.

Climate change impacts and rapid development in the wildland-urban interface are increasing population exposure and vulnerability to the harmful effects of wildfire and wildfire smoke. The direct and indirect effects of these hazards may impact future mobility decisions among populations at risk. To better understand how perceptions and personal experience inform wildfire and wildfire smoke-associated migration intentions, the authors of this study surveyed a representative sample of 1,108 California residents following the 2020 wildfire season. The authors assessed the associations between threat appraisal, coping appraisal, personal experience, migration intentions, the impact of wildfire and wildfire smoke on migration intentions and place satisfaction, and future migration potential. Results indicated that roughly a third of the sample intended to move in the next five years, nearly a quarter of whom reported that wildfire and wildfire smoke impacted their migration decision at least a moderate amount. Perceived susceptibility and prior negative outcomes (e.g., evacuating, losing property) were associated with migration intentions. Prior negative outcomes were also associated with a greater impact of wildfire and wildfire smoke on migration intentions. For those intending to remain in place, prior negative outcomes were associated with a greater impact of wildfire and wildfire smoke on place satisfaction, which was associated with greater future migration potential. The findings of this study suggest that perceptions of and experiences with wildfire and wildfire smoke may impact individual mobility decisions. These insights may be leveraged to improve normative models of climate migration and to inform educational outreach campaigns to encourage wildfire and wildfire smoke risk mitigation behaviors.

Recent events such as the 2022 Tonga eruption and tsunami have renewed natural hazard concerns along broad swaths of coastline, namely the entirety of the United States West coast. Infrequent yet critical, these inundation hazards pose massive risk to coastal infrastructure. Impact and damming of debris carried (e.g., cars, concrete, trees), a secondary hazard, further amplifies risk but is inadequately characterized for robust design. Through an extensive, paired program of multi-scale experimental and advanced numerical trials we quantify stochastic hazard to coastal infrastructure. Hundreds of varied debris-field and inundation specifications are executed experimentally, matched by a novel multi-physics numerical tool (Material Point Method, Finite Element Method, and Affine-Separable Fluid-Implicit Particles) to faithfully simulate complex debris-fluid-structure interaction in high-performance computing systemsat the frontier of computational capability (e.g., on multiple Graphics Processing Units). Results are refined for coupling with modern, Performance-Based Tsunami Engineering (PBTE) design frameworks. Herein, we detail key findings and preliminary guidelines for improved engineering design regarding debris-field hazards.

Natural hazards annually cause significant structural, economic, and physical damage in both urban and rural areas throughout the United States, with the most common and recurring hazard being floods. Studies have been made in the past to investigate hazard risk to formulate risk reduction strategies, yet typically look at defining factors, such as exposure and vulnerability, separately. Hazard vulnerability and risk assessments are typically conducted using physical hazard and sociodemographic inputs, but few methods consider the impact of culture and compounding impacts on vulnerability. More than physical risk and demographic data need to be analyzed to fully understand the complexity of hazard vulnerability and recovery. The study aims to analyze the following: a) marginalized communities are more frequently and disproportionately affected by natural and man-made hazards within the Sarasota County in Florida, b) the damage from past and present hazards interacts and have lasting effects that are not currently accounted for in risk assessment methodologies, and c) risk perceptions vary spatially over time and space, which can lead to increased vulnerability to specific hazards. By reclassifying FEMA flood maps, the analysis spatially identifies areas that are within 100-year, 500-year, or minimal flood zones. By overlaying the reclassifying flood map of the study area with geocoded survey points, the perception of those within apparent flood risk can be analyzed. This study delivers the opportunity to observe flood risk patterns while identifying a trend of how said risk is perceived in areas that are more at risk than others.

Despite being a common strategy to reduce hazard exposure, shelters pose a public health dilemma during the global COVID-19 pandemic. Ideally, shelters offer a safe place for those with access and functional needs away from hazard risk. However, shelters can also facilitate the spread of the SARS-CoV-2 virus and infectious disease. Disaster managers have generated ad-hoc strategies for keeping people safe during the 2020 and 2021 hurricane seasons, but further study is needed to understand how strategies for multi-hazard shelter planning continue to develop in the United States island territories and Hawai’i, given their high exposure, limited alternatives, and concentrated levels of risk. This research investigated the following questions: (1) which traditional and non-traditional public health agencies are responsible for multi-hazard shelter planning?; (2) what plans and strategies are in place for reaching and accommodating those with access and functional needs?; (3) what resources exist to support these agencies in multi-hazard evacuation and shelter planning during the pandemic in the U.S. island territories and Hawai’i?; and (4) what education and training gaps exist for multi-hazard evacuation and shelter planning?

Flooding comprises the largest proportion of economic losses derived from disasters in the United States. and is poised to become the main driver of disaster-driven economic loss worldwide. Community-level flood planning has the potential to alleviate part of this problem. The United States Community Rating System (CRS) incentivizes community floodplain management practices in exchange for flood insurance premium rate discounts. The main goal of the CRS program is to mitigate flood damage to insurable property. Using multi-period difference-in-differences, this paper quantifies whether and to what extent this goal is achieved by the program. The study finds inconclusive evidence on whether losses on insured buildings do decrease as a result. These findings have important implications when considering the design of future programs from a climate justice lens, as previous research suggests that not all communities have the resources to adhere to this program and only members of the communities that do are able to benefit from flood insurance discounts. Further research is underway to explore whether the program contributes to increasing other dimensions of community resilience beyond infrastructure protection.

The information-seeking behavior of communities was highly influenced by the level of risk perception during the COVID-19 pandemic. This study explored the relationship between risk perception and information-seeking behavior during the COVID-19 pandemic with data from Bangladesh. The study used two different theories and a mixed-methods approach to finalize results. Through a survey, in-depth interviews, and key informant interviews, we accomplished the data collection process and recognized the relationship between risk perception and information-seeking behavior. The study found that perceptions about risk and emergencies change depending on people’s demographic and socio-economic features. For instance, perceiving the risk of COVID-19 was different between urban and rural communities. The study also found that people look for different information based on their risk perception. People who highly perceive the risk of COVID-19 mainly looked for safety and medical information. On the other hand, people who were economically vulnerable due to continuous lockdown and perceived the risk of COVID-19 at a low level mainly looked for relief and job-related information. Also, social media was identified as the most regular source to seek information during the pandemic. The study concluded that to manage any emergency effectively, access to accurate and need-based information must be ensured. If people do not perceive the risk of any emergency correctly they will not look for proper information and vice-versa. It will be challenging to manage an emergency like COVID-19 if risk perception and information-seeking behavior are not taken into account.

First, the researchers investigated how poverty rates, hurricane damages, and disaster aid were correlated. Next, case studies were conducted in two municipalities selected based on quantitative findings. Then the team examined correlations between property damages, hurricane fatalities, and the way economic inequality and COVID-19 infection rates were affected by disaster aid distribution patterns. After accounting for changes in population, the researchers found that municipal poverty rates increased at a faster rate after the hurricanes. Moreover, poverty accelerated faster in areas that received more disaster aid. Case studies also revealed unequal distribution of disaster aid within municipalities, compounding poverty for already vulnerable populations. We found similarities between the two municipalities, such as a sense of violence from bureaucracy and governmental neglect, and differences, such as weaker social support networks where poverty was less extreme. Lastly, findings indicate that the cumulative number of COVID-19 cases were positively correlated with each of the following ordered from strongest correlation to weakest: disbursed disaster aid, hurricane fatalities, economic inequality, and hurricane property damages. The top policy recommendation is to immediately adopt an equity framework in all disaster responses and policy making for disaster recovery. Public health experts, social workers, and community representatives should collaborate on policy design.

Blight contributes to increased health risks and is unevenly distributed, exacerbating geographic health inequities. Increased blight is a common consequence of natural hazards, and transforming blight is an effective resilience strategy. What are the economic, social and health costs of abandonment? What are the benefits for public health emergency preparedness and disaster recovery of eliminating abandonment? Does engaging in collective action to transform abandonment into community assets improve social capital and/or community health?

This study used survey data from abandonment inventories, local health data, US Census data, and Geographic Information Systems (GIS) to illustrate spatial patterns of abandonment across Puerto Rico and identify associations with health risks, economic costs and social vulnerabilities. The authors conducted two case studies to elucidate the health, economic, and social costs of abandonment, and the benefits of community-led abandonment transformation. Additionally, to identify potential impacts on intermediate measures associated with multiple health outcomes, including agency, connectedness and social capital, among people who engage in actions to transform abandonment, and protective health benefits of community assets produced from the transformation of abandoned buildings, the authors conducted a multi-site survey. Qualitative, quantitative and geographic findings describe how Puerto Rico is fighting blight, the health, social and economic costs of blight, and the benefits of community-led blight transformation processes. Policy and programmatic recommendations will support improved blight abatement across Puerto Rico and contribute to defining blight abatement efforts as a public health strategy in disaster mitigation, readiness, response and recovery.

The International Development team of the Institute for Sustainable Futures (ISF), UTS (Australia) works extensively within the development practice, climate change adaptation (CCA) and resilience niche. While ISF’s CCA and resilience projects fundamentally involve fieldwork and travel to the Asia-Pacific region, the team made a pragmatic change in the research approach due to the Covid-19 pandemic and switched from face-to-face to remote research.

Given that ISF has been successful to continue ongoing projects and commence new projects with international stakeholders, the team’s researchers were asked for a retrospective analysis as a year-ending team activity in 2021, to identify the tools and techniques of remote research that the researchers applied in the past years. The analysis involved Donald Schon’s ‘theory of action’, to reflect on an experience to determine the actions of the experience, consider the spaces for improvement, and recognize the positives from the interaction. The reflective practice underpinned that a strong and effective partnership with local stakeholders is fundamental for a remote research project. Maintaining good relationships through regular check-in meetings, online celebrations to mark milestones of the projects, participatory methods in sensemaking workshops for local researchers to strengthen local research skills and enable primary data collection, and creating conversations on critical inquiry and research impact with local research stakeholders, were identified as most efficient tools and techniques for remote research. Availing flexible planning and introducing the mentoring process to ensure quality skill development, were surfaced as two potential aspects for consideration in designing future research methods.

This study proposes a method to deduce minimum household income for single-family homeowners using publicly available data via the tax assessor and current population survey. The approach is versatile and can be tailored to any community in the United State. Income and the cost of repairing damages play a central role in determining the capacity of impacted households to cope with and recover from disasters. While the simulation of damages and corresponding repair costs can be conducted at individual building resolution using publicly available data, household income data is only available aggregated across specified brackets at larger spatial units such as census tracts. Researchers typically employ random sampling to generate synthetic household income for individual households. This approach leads to unrealistically low income being attributed to some households in highly valued homes that tend to have costly repairs after a disaster. Thus, the coping capacity of select low-income households will be substantially underestimated because their assigned income often cannot finance the repairs. Results for a case study featuring 95,387 single-family detached residences in San Francisco show better agreement between the estimated homeowner income and home value compared to distributions drawn from conventional methods. The proposed approach leads to more reasonable income estimates that are crucial for adequate pre-disaster planning.

Following Hurricane Maria, death certificate quality limited accurate determinations of the number of people who died directly or indirectly from the hurricane. Previous studies have attempted to measure post-hurricane mortality. However, little has been done to identify which communities experienced more mortality. The objectives of this study included identifying how many excess deaths occurred in Puerto Rico following Hurricane Maria, which community-level factors are associated with higher risk of death following Hurricane Maria, most common causes of death after the storm, and geographic clusters of excess mortality. Researchers obtained death certificates from Puerto Rico’s Department of Health Demographic registry for 09/13/2016 - 03/31/2017 and 09/13/2016 - 03/31/2018. Overall, the study found 1,650 more deaths in the six months following Hurricane Maria than the same period the year before. An increase of more than 100 deaths due to Alzheimer's disease, diabetes mellitus, sepsis, chronic respiratory disease, and hypertension occurred after Hurricane Maria. The study found minimal differences for asthma-related deaths while there were 126 fewer cancer deaths in the post-Maria period. A significant increase in the mortality rate post-Maria compared with pre-Maria was seen for all disease categories examined excluding Chronic obstructive pulmonary disease (COPD), Mental Health Conditions, Asthma and Cancer. The study found that the rate of death per 100,000 was slightly higher post-Maria compared to before Hurricane Maria (risk ratio=1.14, 95% Confidence Interval:1.00 - 1.17). Researchers geocoded death certificates in order to conduct a spatial analysis at the census tract level to identify clusters of excess mortality and social vulnerability using the Getis-Ird (Gi)* statistics.

After the emergency declaration of the COVID-19 pandemic, small businesses quickly became a focal point for the drastic changes occurring throughout the world. Essential designations, loan and grant access, and employee losses led to critical discussions on equity, community resilience, and disaster response. Studying small businesses also furthered understanding of adaptation and resilience across geographies and during complex disasters. Although there was suddenly an opportunity to answer burning research questions, methodologies had to be reconsidered or entirely rewritten. The pandemic highlighted the need to develop resilient research proposals where recovery and adaptation are just as important for the project as for the research participants. This presentation discusses the findings from three years of small business research in the coastal Carolinas and draws parallels with lessons learned while completing a dissertation during a pandemic.

The authors of this study investigated the extent to which federal disaster funds are distributed equitably. As federal agencies have been tasked to address environmental justice and equity, lessons can be learned from federal Emergency Management Agency (FEMA) allocations. The authors implemented social vulnerability and social capital theories to understand: (1) the effect that bonding social capital and bridging social capital have on allocations of disaster assistance, and (2) whether these types of social capital have an interaction effect with social vulnerability. In other words, do any protective measures of social capital equally benefit more and less vulnerable communities? Or does the interaction exacerbate inequities in access to FEMA assistance? This national case study implemented a temporal econometric model to analyze FEMA Public Assistance allocations from 2005 to 2017 for climate-related disasters at the county scale. There were three findings. First, federal funds underserve socially vulnerable groups, especially communities with greater proportions of elderly, disabled, low-income, and less educated individuals. Second, without distinguishing for social vulnerability, this study found federal aid allocations have a negative relationship with bonding social capital and a positive relationship with bridging social capital. Third, this study revealed that there is a significant interaction effect between social vulnerability and social capital which suggests bridging social capital has a negative relationship with disaster assistance for populations that are more socially vulnerable. The authors of this study advise scholars to interpret with great caution any conclusions made from models that assess social vulnerability and social capital theories separately. The task to allocate federal funds equitably is also complicated by these findings.

This presentation highlights strategies, practical lessons learned, and an in-depth discussion of critical issues in collecting data through Systematic Social Media Data Collection (SSMDC) as a primary recruitment strategy. While unique methods beyond face-to-face data collection are not new, the global pandemic brought new attention and interest to ways in which researchers can gather primary data while adhering to health requirements such as social distancing. This study defined the basic steps and concept of Systematic Social Media Data Collection and how it can be adapted. Itused three U.S.-based projects as case studies for this overview, with topics including sandwich generation caregivers, birthing experiences during the pandemic, and infant and maternal well-being after Hurricane Ida. Systematic Social Media Data Collection highlighted unique strengths and limitations that must be acknowledged. For example, using social media as a respondent recruitment tool can save overall project costs, researcher travel time, and other resources. There are also more nuanced aspects of SSMDC that should be considered, such as empowering researchers with functional and access needs. Another critical issue is the ethical guidelines when engaging in SSMDC, such as transparency, adaptive scripts for contacting closed groups, and issues related to identity and reflexivity. Limitations of SSMDC include challenges in obtaining a diverse respondent sample and limitations regarding study generalizability.

This project investigates how the COVID-19 pandemic and unprecedented wildfire seasons in California have (1) elevated demands for greater emotional and mental health support among students and educational staff, and (2) reinforced the need for robust communication networks, deep compassion, and flexibility on the part of the education sector to adapt quickly in frontline communities. Through an in-depth case study of the educational community in Butte County, California, the author explores the role played by Communities of Practice before, during, and after disasters to encourage knowledge-sharing, innovation, and collaboration among educational staff. The findings from this study show that as climate-induced disasters increasingly disrupt day-to-day operations, sustainable and low-cost approaches for addressing the emotional wellbeing of staff and students are critical. Based on this study, the author argues that participation in these communities bolsters individual and organizational resilience.

The global COVID-19 pandemic upended the ways in which social science researchers engage with research partners, participants, and communities alike–particularly in multi-shock settings. Typical practices surrounding the training of enumerators and interviewers, engagement with translators, and other local partners were rapidly adapted (and in some cases abandoned), with many learning-by-doing through with the support of new technologies. In contexts where COVID-19 was another shock in a long list of compounding crises, researchers based outside of the field had to become quickly attuned to the challenges their research partners were encountering on a daily basis. Drawing on insights from a qualitative study conducted in Taiz, Yemen by researchers at Mercy Corps and the Feinstein International Center at Tufts University, the presentation will highlight lessons learned from the study team's experience designing, implementing, and analyzing the study in a context characterized by rapidly evolving economic and social conditions. This includes applying iterative approaches to interviewer training, data collection, and analysis, as well as working with local research partners to identify emerging and new avenues of inquiry. In particular, the presentation will discuss and present the ways in which we addressed the well-being of local researchers and partners, recognizing that their lived experience often reflects the very questions we seek to answer through our research.

The COVID-19 crisis has made data collection more challenging in both natural hazard preparedness planning and post-event forensic studies. Data collection events such as community meetings and interviews have been curtailed, necessitating alternative collection strategies. Process Network Diagramming is a new tool that allows field researchers and citizen scientists to collect data on their own and then combine their results electronically for analysis. This method enables socially distanced collection, synthesis, and analysis of natural hazard field data. As a method for analyzing complex and dynamic systems, Process Networks Diagramming can be applied to existing human-natural systems by non-experts, enabling cloud-based data collection and analysis before, during, and after natural hazard events. Process Network Diagramming also provides clear and simple steps for field researchers or citizen scientists to identify processes and their relationships in complex systems based on direct on-site observation. Additionally, it incorporates systems dynamics modeling and non-linear analytics to ensure accurate modeling. Process Network Diagramming provides a clear and comprehensible interface that makes it easy for experts and non-experts alike to adjust process variables and see outcomes change in real time. This presentation describes the methods associated with Process Networks Diagramming, its benefits and challenges, and case studies in its application.

The COVID-19 pandemic has proved that communities can remain functional during disruptive events and recovery phases by transforming many traditional in-person services into online ones. During the pandemic, community members have been able to meet some of their needs through realized access to products offered by organizations that could operate remotely. This revolution illuminates that in some cases, organizations can be functional without relying on the physical spaces previously thought imperative. It also showcases that while proximity and physical access are traditionally the most common proxies for measuring accessibility, they are not always the best indicators. Accessibility must also include availability, adequacy, acceptability, affordability, and awareness. This presentation dissects what accessibility is, scopes out its multiple dimensions and crucial role in a community’s functionality, and reconciles between the functionality of an organization and accessibility to its products.

Organizational functionality is the capability of an organization to offer its intended products. Thus, the availability and adequacy of products are directly related to organizational functionality. This presentation introduces staff, supply chain, and supporting infrastructure (specifically electricity and telecommunication) as critical components required for an organization to function, particularly in a remote environment; a lack of these components compromises organizational functionality and consequently community accessibility to critical social services. Lastly, functionality models of select organizations, including banks and schools, are presented. Findings offer new insight on ways to improve community resilience through securing reliable access to organizations delivering critical social services.

Volunteer rescue organizations have become significant contributors to disaster response and relief efforts across the United States. Such organizations, often identified under the umbrella of “The Cajun Navy,” have been frequently covered by news media organizations since the major disasters of the Baton Rouge floods in 2016 and Hurricane Harvey in 2017. They have since become significant players in major disasters across the southeastern United States and beyond for various types of disaster. In the midst of both disasters, a number of such organizations emerged. For this study, 384 official news media articles were analyzed discussing eight volunteer rescue organizations between the years of 2016 and 2021. This study compares how the organizations are covered in the media with the in the field observations from members of our research team as well as qualitative interviews with members of the eight organizations. Results reveal that coverage by media organizations often distorts the reality of how volunteer rescue organizations operate, the types of activities they participate in, and their role in various disasters. Volunteer rescue organizations are also more complex in terms of their operations, membership, and participation across disasters. However, the growing amount of coverage and details included in the articles demonstrate how the emergent organizations have changed over time and solidified their role in disaster management. This analysis is part of a larger study on volunteer rescue organizations in disaster.

Wildfires are increasing in intensity and frequency. At the same time, communities bordering wildlands are increasing in size and density and increasingly intermixed with burnable vegetation. These two events happening simultaneously create an increased risk within the Wildland Urban Interface (WUI). Home destruction due to wildfires can have immense impacts on communities economically and socially. During the 2017, 2018, 2020, and 2021 wildfire seasons within the states of California, Oregon, and Colorado, towns experienced significant damage to their water distribution systems, particularly service laterals, plumbing within the homes, and water meters. Eleven municipalities within California and Oregon discovered volatile organic compounds (VOC) within their water distribution system, which resulted in over a years worth of continuous water sampling and testing, replacement of service laterals at all destroyed or damaged properties, closing of businesses, and additional hurdles for rebuilding homes. This presentation summarizes this issue within communities across the Western United States and describes a research program at Oregon State University that is testing different materials of pipelines to understand how thermal degradation of pipelines influences the release of VOCs into the water distribution system. Further, a heat transfer analysis determined what surface conditions during a WUI fire create such contamination. This study also develops a methodology to determine exposure temperatures of homes during a wildfire based on data collected after the 2021 Marshall Fire that can be used to determine if pipes within a standing home need to be replaced due to a neighboring structure fire.

Risk communication is essential for mitigating flood risk. Discussions with stakeholders identified incentives and barriers to flood adaptation strategies such as freeboard, which refers to the elevation of a building’s lowest floor above the base flood elevation (BFE) requirements. The predominant barrier identified regardless of stakeholder type was the cost of mitigation. Optimizing the level of freeboard against expected cost and savings is a crucial decision for flood-vulnerable residents. While existing flood web portals provide helpful risk information, their information focuses on flood forecasting or general risk information such as flood insurance and risk, historical flooding, and flood zoning, at global to regional scales, for new residential developments. Web-based flood risk information tools rarely consider micro-scale (i.e., parcel or individual building-level) economic loss or exposure for existing homes to enhance decision-making. The interactive decision-making tool, Flood Safe Home (https://floodsafehome.lsu.edu/), helps individuals identify an optimal freeboard level by calculating monthly savings at the building level by assessing freeboard cost, insurance premium savings, and expected annual avoided flood loss based on user preferred/defined insurance coverage, deductible, and building attributes and external sources local flood characteristics, for three parishes (i.e., counties) in Louisiana, U.S. This web-based decision-making portal, with infographics to streamline communication, helps potential homeowners to make risk-informed decisions with various economic analyses and benefit information, ultimately enhancing protection of life and property, and therefore long-term resilience, to the ever-present flood hazard in one of the most flood-prone U.S. states.

Flooding in the United States is the most costly and common hazard in the United States with approximately $17 billion in losses annually from 2000 to 2018. Urban flooding, in particular, impacts human lives, livelihoods, infrastructure, financial assets and poses pressing questions of how quickly communities can recover before the next storm or flood occurs. Accurate estimates of both population and infrastructure exposure to flooding are critical to mitigate future damage and inform land use planning for future development or potential relocation of communities. Advances in the spatial and temporal resolution of remotely sensed satellite imagery and deep learning techniques capable of extracting detailed spatial features offer an opportunity to tackle the challenge of urban flood and damage monitoring. This research drew upon high-resolution satellite imagery to detect and map blue roofs, or the presence of blue tarps that are installed on building structures after a major storm and flood event occurred. Here, the authors compared the locations of detected blue roofs on the Louisiana Coast following Hurricane Ida with georeferenced data from Federal Emergency Management Agency’s Individual and Housing Assistance program to understand the spatial relationship between property damage and federal assistance. This relationship is useful for strengthening assessments of exposure and risk and also for tracking recovery as communities and individuals finance the repairs to damaged buildings following an event. The results of this research contribute insights into the advantages and unique challenges of satellite imagery to track the trajectory of urban damage and repair.

Disaster researchers have extensively studied communication strategies during tropical cyclone scenarios. However, few studies addressed communication strategies during consecutive tropical storms with concatenated effects that overlap and accumulate quickly. It is important to understand how communication practices work in such emergencies and whether communication shortfalls can trigger problems that amplify the impacts of natural hazards, particularly for vulnerable populations. In addition, among English-language literature related to risk and emergency communications, there is noted inequity in the countries of focus by the international research community. For example, despite the unique susceptibility of Guatemala to disaster impacts as well as known challenges related to rapid communication and information dissemination, little literature has addressed disaster communications in the country. Using data collected through digital archival methods from Twitter Application Programming Interface (API) (over eight-thousand tweets), organizational sources (two reports from the Guatemalan National Office for Risk Reduction), and digital news outlets (over 100 news articles from the major news outlet in Guatemala), this study examines disaster communication in Guatemala by government and non-governmental organizations and Guatemalan news organizations during Hurricanes Eta and Iota. These back-to-back hurricanes devastated several territories in Central America in combination with the ongoing COVID-19 pandemic. This research combines textual analysis and multiple data visualization methods to infer communication patterns and contextualize them with the overall technological uses and trends in Guatemala, the evolution of the hydrometeorological properties of the storms (e.g., storm track and parameters), as well as documented community and engineering impacts (e.g., infrastructure damage).

The September 2020 Oregon wildfires were widely described as an unprecedented event with respect to both geographic scope and the number of communities affected by both smoke and wildfire. This research used a process tracing approach to explore how potential focusing events–in this case, the Oregon 2020 wildfires–yield opportunities for public policy change. The study drew on interviews with policymakers and other stakeholders, providing empirical evidence about how extreme weather events shape public policy. This study contributes to ongoing development of focusing events theory, highlight strategies by which key policy actors engage in the policy process and influence policy decisions, and offers insight into post-event conditions that facilitate or impede policy change, including the effects of the COVID pandemic on public attention and engagement in the policymaking process.

Throughout much of its history, the sociological study of human communities in disaster has been based on events that occur rapidly, are limited in geographic scope and their management understood as phased stages of response, recovery, mitigation and preparedness. More recent literature has questioned these concepts, arguing that gradual onset phenomena like droughts, famines and epidemics merit consideration as disasters and that their exclusion has negative consequences for the communities impacted, public policy in terms of urgency and visibility, and for the discipline itself as the analytical tools of sociology are not brought to bear on these events. The researchers agree that gradual onset disasters merit greater attention from social scientists and have addressed the two most significant ongoing disasters that are gradual in onset, global in scope, and have caused profound impacts on lives, livelihoods, communities and the governments that must cope with their effects. These disasters are the COVID-19 pandemic and global climate change. We begin with an examination of the foundational work in the social scientific study of disaster that established a conceptual framework for future work and has guided the field throughout the latter part of the twentieth century and into the current period. The focus is on several components of the existing framework for defining and studying disasters which we term “borders”. These “borders” are temporal, spatial, phasing, and positioning which, in our view, must be expanded and redefined to accommodate the full range of disasters to which our globalized world is vulnerable.

Measures must be tested, validated, and adapted for use in new contexts and places, especially following disaster. The authors focused on what measures can be used to answer the questions: How did the authors’ housing recovery program improve access to safe and secure housing? How did this change impact the lives of participants, their families and communities? To examine these questions, the authors invited 40 stakeholders, including staff, program participants, and members of comparison communities to share their experiences of disaster and recovery in a day-long interactive community workshop. Next, the authors of this study convened a 12-person steering committee including representatives of each stakeholder group. Then, the authors examined secondary programmatic data and collected primary qualitative data in each area of impact identified in the evaluation research question. After analyzing findings from the Community Workshop, the authors learned that there are many varied definitions of vulnerability. The authors also learned the importance of understanding how social capital is developed, impacted by programming, or exchanged in a post-disaster context. The authors identified community empowerment as an essential concept to explore in understanding potential impacts of disaster recovery programming through interviews and focus groups. These findings informed survey design. Based on collection, analysis and interpretation of these data, the Committee developed and validated measures that are both generalizable enough to enable comparison and specific enough to generate detailed learning within specific contexts.

How can equitable disaster recovery be advanced through evaluation research during a pandemic? To answer this question, a participatory approach guided this study's methodological development. Program participants and members of research communities were invited to a community workshop to help ground the research question in local experiences of vulnerability and recovery. Diverse stakeholder groups formed a steering committee to collaboratively develop the research design and oversee implementation. Subgroups of the steering committee reviewed program data and collected participant interviews and focus groups to inform the design of a survey instrument.The final survey was presented, tested, and validated in a community workshop prior to broader use in the field. These methods relied on a combination of in-person and virtual meeting, cloud-based collaboration tools, chat messaging, and online surveys. Three primary findings associated with this work were identified: (1) including participant and external stakeholders is pivotal to representing vulnerability in the research design as a dynamic experience gets compound during cascading disasters, (2) there are gaps in efforts to provide equitable access to recovery, and (3) inclusive research design led to a new research question that adds local relevance to the treatment of vulnerability and addresses concerns such as how social capital is exchanged in a post-disaster context. Inclusion and equity practices in evaluation research design translate community experience into organizational learning. The resulting research framework advances recovery program design and the translation of scientific research for broader local use.

Currently, amateur radio operators (or ham) go to disaster site with their transceivers, antenna, and power supply to restore emergency communication. In the process, time is lost, when every second makes the difference between lives and deaths. The solution is to install amateur radio equipment within ~ 739 Indian district EOCs costing ~ $3 million in equipment. Ham will then go quickly during disasters to EOCs and maintain amateur radio emergency communication, reducing mortality, morbidity, and infrastructure losses. The Coalition for Disaster Resilient Infrastructure (CDRI) awarded Fellowship 2021-22 to the first author to research on the solution. The study involved two outputs. First, at a praxis level Malaviya National Institute of Technology Ham Club has obtained an amateur radio club license, HF transceiver, and will install antenna for going on the air. Second, at a theoretical level, writing a white paper for policy changes for installing amateur radio equipment within district EOCs. We conducted two in-person focus group meetings with 13 participants including ham, media personnel, and lay people. All other activities, involved participatory action research virtually, including collaboration with different stakeholders. We gave presentations in six international conferences and Hamfest India 2021. We published six papers and two will be published, written by first author individually or with one or more authors of this paper. The CDRI Fellow has submitted final report. He will convert final report into a white paper to persuade government for policy changes and to install amateur radio equipment in district EOCs. 

Wildfires have been increasing in frequency and intensity in recent years. In particular, California has experienced severe wildfire seasons lately, with the state accounting for 40% of the acreage burned in the United States in 2020. The goal of this study is to characterize causes of a subset of wildfires that occurred in California between 2010 and 2019. The California Department of Forestry and Fire Protection (CAL FIRE) is responsible for the prevention and suppression of fires in about 24 million acres of land, with an additional 3.4 million acres protected by local county fire departments in six contract counties. This represents over a quarter of the land in California, and much of the state’s wildland-urban interface, where human development meets wildland vegetation. The authors of this study obtained CAL FIRE’s annual wildfire activity statistics reports and extracted information on the acreage burned and causes of wildfires in these areas. Between 2010 and 2019, wildfires burned over 3.3 million acres in areas attended to by CAL FIRE and contract counties. Approximately 25% of the acreage burned was due to fires caused by electrical equipment, including power lines. Lightning was responsible for 12% of the acreage burned. As climate change continues to alter weather patterns, produce high temperatures, and impact drought conditions, catastrophic wildfires are likely to persist. Understanding patterns of causes of wildfires is essential for the development of targeted interventions for wildfire prevention. Interventions focused on stopping power lines from sparking fires will reduce wildfire impacts in California.

In this big data era, information dissemination during disasters increasingly takes place on social media platforms, making it a critical socio-technical component of community resilience. Many public agencies have adopted these platforms for crisis communications and coordination during disasters. Despite being widely adopted, there is a lack of understanding, empirically and theoretically, of how the local and regional organizations utilize these platforms for efficient communication in different phases of a disaster and how significant their roles are in enhancing community resilience. Although previous studies considered crisis communication among social media users in general, to the best of the research team’s knowledge, no study has focused solely on the local organizations and their cross-partnerships in social media. This study addressed five questions: (1) Which specific local agencies are leading in disseminating information during a disaster on social media platforms? (2) How does the dynamics of their information sharing change during different stages (preparation, response, recovery) of a disaster? (3) Which variables play a role in increasing the efficiency (i.e., attention gained) of crisis-communication posts shared by local agencies during a disaster? (4) Are the Emergency Support Functions (ESFs) of these agencies correlated with the efficiency of their crisis-communication posts? and (5) Are these local agencies, involved in disaster response, well-connected with each other in social media? This study considered Hurricane Irma as the case, Twitter as the social media platform, and the East Central Florida region including 8 counties as the study region.

Homeownership in the United States has declined over the last several decades as people increasingly rent their homes. Despite the trend away from homeownership, the literature on natural disaster impact and recovery has focused disproportionately on homeowners, leaving renter populations critically understudied even as they experience particular vulnerabilities to the effects of environmental and economic crises. This paper contributes to the connection between housing, disaster impact, and resource access by examining damage from flood disasters and access to federal aid during recovery among renters and homeowners. Preliminary findings from the Individuals and Households Program (IHP) dataset from the Federal Emergency Management Agency (FEMA) from 2002-2020, show that while homeowners fare worse on a number of critical damage and recovery measures, renters experience disproportionately high levels of flood insurance non-coverage and undercoverage, reported personal property damage, and utility outages and emergency needs while also facing lower eligibility, and amounts received, for temporary housing. Differences are also considered by place characteristics, including minority racial composition and share of housing occupied by renters. These findings suggest that quantitative sociological studies of disaster and resilience should consider the particular consequences among renter populations who make up an increasing share of the population, particularly in major cities prone to severe and repetitive flooding. Further, this paper suggests that more resources are needed to support renters who may face greater housing instability and loss of personal property in the aftermath of disasters with less equitable access to resources.

This study provides new evidence on the status of small- and medium-sized business planning, mitigation, and preparedness, as well as the effectiveness of these activities on loss reduction and business interruption. Data were analyzed from a longitudinal study of small- and mid-sized businesses (SMBs) in Lumberton, North Carolina conducted by members of the NIST-funded Center for Risk-Based Community Resilience Planning and researchers from NIST’s Community Resilience Program. The city of Lumberton was greatly impacted when in early October 2016 Hurricane Matthew crossed North Carolina as a category 1 hurricane and was followed less than two years later by Hurricane Florence in September 2018. Unlike previous studies, the researchers looked at SMB operators’ longitudinal decisions and outcomes across two hurricane events in addition to controlling for demographic factors (age, race) and business characteristics. The researchers’ estimated models explored the influence of mitigation on self-reported business profitability after Florence. The study found that each preparedness/emergency action taken by a business ahead of Florence increased the odds of being profitable by, on average, 1.7 times, holding everything else constant, in comparison to businesses that did not implement any actions. This study demonstrates that longitudinal investigations are critical for the study of community resilience, and SMB resilience, so that impact, decision, and recovery data are comparable across time and inferences may be made on the effects of decisions and impacts from one period on those in future periods.

Two cancer clusters near potential contaminated sites have recently been identified in Northeast Houston, Texas. A suspected cause of contamination includes a Creosote plume from the Englewood Rail Yards wood treatment processes that occurred from the early 1900s to the 1980s. This study demonstrated how spatial analysis and Geographic Information Systems (GIS) can be used to identify spatial relationships between five risk variables (total incremental lifetime cancer risk, Total concentration of polycyclic aromatic hydrocarbons (PAH), Benzo(a)pyrene, Naphthalene, Pyrene) and the Center for Disease Control and Prevention (CDC)’s 2018 Social Vulnerability Index. The results show that high-density clusters of high concentrations of all five risk variables were found surrounding the Englewood Rail Yard, and low-density clusters of low concentrations were found at the edges of the study site, which was expected. Clusters were also positively correlated with low socioeconomic status, which is in line with environmental justice literature. Complex spatial analyses that account for spatial autocorrelation produced more accurate results. Spatial representation of pollutants allows researchers to analyze the breadth of contamination and communicate findings to a broader audience. Further research is needed to incorporate a larger study area, investigate updated socioeconomic variables (2020 U.S. Census), and investigate other PAH pollutants. Spatial clustering analysis is a productive means to identify pollutant risks and potential sources of environmental contamination. Prospective designs implementing this technique are needed within environmental justice research.

Texas and Louisiana were heavily affected by Hurricane Laura while, at the same time, experiencing relatively high COVID-19 infection rates during August 2020. Adopting an abbreviated PADM, this study tests a serial mediation model exploring the drivers of evacuation decisions during Hurricane Laura as individuals cope with social distancing requirements and COVID-19 infection risk. The authors of this study collected data using a questionnaire distributed to a stratified random sample of households in two counties in Texas and two parishes in Louisiana (n=291). The authors tested the hypotheses using bootstrapping procedures to test for indirect effects. Findings suggested significant serial mediation for several predictors, including positive indirect effects of being in an evacuation zone (mandatory and voluntary), prior hurricane experience, negative affect toward the threat, and evacuation impediments on respondents’ decisions to evacuate, through perceived storm characteristics and subsequent expected personal impacts. Significant indirect effects were also found for official warning (positive) and family being at risk for COVID-19 (negative) on respondents’ decisions to evacuate, through expected personal impacts alone. Practically, findings provided insights for understanding the factors that promote and buffer evacuation decisions and the underlying mechanisms that explain these relationships. More specifically, findings suggested providing sheltering that includes protections and communicating to the public that they will be protected from COVID-19. The findings also contributed to the emerging research on pandemics and disasters, but with a focus on how the pandemic shapes protective actions within the context of built environment barriers and facilitators, which has generalizability to other compound events.

When disaster strikes, emergency response agencies can become overwhelmed, and communities must rely upon internal resources to survive. This research explores individuals’ willingness to share resources needed to carry out essential daily activities in two cultural contexts: the United States Pacific Northwest and Nagoya, Japan. Situated on either side of the Pacific Ocean, the two regions have similar geophysical conditions that put them at risk of extreme and isolating seismic “megaquake” events.

While previous research on sharing assumes a top-down, centralized approach to resource allocation, this study seeks to understand how willing people are to share resources locally according to the strength of the social ties connecting individuals as well as attitudinal factors like trust and place attachment. The authors ask how willingness to share varies between the two cultural contexts. Data is gathered from a sample survey administered in four communities across the two regions. Main factor analysis and an ordered response prohibit model are used to understand factors affecting willingness to share different kinds of resources.

Findings show that willingness to share varies by type of resource, strength of social ties between the provider and the recipient, and in some cases, by cultural context. Trust is found to be strongly associated with willingness to share in both countries. These findings suggest that building trust among community members is an important disaster preparedness strategy across cultural contexts, and that disaster preparedness strategies should be tailored to reflect local resource availability and attitudes toward sharing.

Despite previous research on the impact of emergency management public-private partnerships (PPPs) on community resilience, there have been limited studies on the organizational outcomes of such partnerships. The study aims to address this research gap by focusing on the following research question: do factors (e.g., institutional design, organizational characteristics) involved in the process of emergency management PPPs predict emergency managers’ and operators’ perceptions of organizational resilience? The paper proposes a sophisticated conceptual model to define the emergency management PPP process and collects data through a quantitative web survey from a purposive sample (N = 888) of emergency managers and operators in the public, private, and nonprofit sectors in the State of Florida. This study utilizes partial least squares structural equation modeling (PLS-SEM) as a powerful statistical method to identify factors that predict organizational resilience in collaborative emergency management processes. The findings of the study provide insights into the complex decision-making and planning processes of organizations involved in PPPs for emergency preparedness and response.

In August 2017, Hurricane Harvey hit Texas and triggered myriad mental health challenges among the survivors. Houses were destroyed, utility services were disrupted, necessary resources were inaccessible, and also resulted in adverse mental health outcomes to the thousands of Texas residents. As the frequency and magnitude of Hurricanes increase, mental health impacts are significant to consider in future Hurricane mitigation efforts. The objective of this study is to identify the key risk factors for escalating mental health challenges after a major Hurricane. For this analysis, household survey responses (n= 780) were collected from the Hurricane-stricken areas in Texas. Respondents were asked to document their socio-demographic, hurricane damages, utility disruptions, and mental health related experiences. This study applied multivariate regression models to identify both the determining and mitigating factors regarding mental health outcomes following Hurricane Harvey. Empirical findings reveal that people who were younger and who had higher disruption of phone and hospital services were more likely to experience mental health challenges. Additionally, people who had higher hurricane damages and did not receive adequate mental health counseling also experienced adverse mental health challenges. The outcome of this research is beneficial for the decision-makers to prioritize their resources to help recover the affected people. Besides, the findings have different policy implications of social and clinical interventions to mitigate disaster shocks from future Hurricanes.

Hazard reduction is an important goal when rebuilding after a sudden environmental change. To address this, a critical first step in the reconstruction process is to refine existing hazard maps based on new damage information and better-informed damage simulations. Land use plans and regulations can then follow this updated hazard information, and public and private investors can propose projects based on it. This model of rebuilding sequence, however, assumes that the disaster event immediately improves knowledge of the hazard. In this paper, the authors draw insights from the coastal recovery of Palu City and its vicinity after the 2018 Central Sulawesi Earthquake. The earthquake killed more than 4,400 and displaced about 170,000 people with the cascading phenomena of ground movement, tsunamis, landslides, and liquefaction. Most of the deaths resulted from the geophysically-unusual tsunamis and liquefaction, which are still not fully understood. Recovery could not wait for complete scientific explanations of these complex phenomena. Documenting the decision processes of recovery actions, identifying contested interpretations of coastal hazards between initial and later understandings, and investigating local community exposure to hazards in relation to new risk reduction projects provided the authors of this study with initial insights regarding the hazard-map based rebuilding process, in which early decisions on the hazard map guide rebuilding. The authors conclude that the current practice of hazard map-based rebuilding needs to be revisited to better understand the limitations and potential of the map itself, and how to implement map information more wisely in rebuilding.

Coastal Maine communities are vulnerable to impacts of coastal hazards. Geographically, these communities are located on narrow peninsulas, often with a single access roadway. Demographic characteristics of these communities, such as older residents living alone, high reliance on natural resource economies, and high levels of self-employment, combine to create a population that is at high risk from coastal hazards. Partners from federal, state, non-governmental, academic, and consulting organizations worked with eight communities to strengthen social resilience. Central to this vision was looking beyond the physical impacts of coastal hazards and focusing on how these physical impacts pose risks to socially vulnerable populations. The first stage of the project involved conducting focus groups with representatives from emergency management, social service, conservation, and municipal sector organizations. A second step was conducting an asset inventory identifying organizations key to planning for, responding to, and recovering from the impacts of coastal hazards. Informed by findings from the focus groups and asset inventory, the research team conducted a scenario planning exercise to test how organizations from the four sectors collaborate in the region to support socially vulnerable populations. Due to COVID-19, the planning exercise was conducted virtually. Participants completed pre- and post- exercise surveys to assess the impact of the scenario exercise and a story map guided breakout groups through the scenario. Survey data and transcripts from discussions sessions were analyzed to identify establishment of new partnerships and increased awareness of social vulnerability resulting from the exercise.

Large-scale extreme weather events like Winter Storm Uri, which affected the Southern, Central, and Eastern United States in February 2021 present an unprecedented risk to the safety and security of individuals and communities in areas lacking resilient infrastructure. While the paradigm for measuring social vulnerability to these events often involves aggregate population characteristics, such an approach overlooks the coping and response capacities of individual actors in adverse circumstances. This study developed an individual-centered approach to neighborhood characterization that allows for the simultaneous consideration of individual and social vulnerability in areas of Harris County, Texas affected by long-term power outages during Uri. To achieve this, the study produced census tract-level synthetic populations from the American Community Survey’s Public Use Microdata Sample (PUMS) and attributed them with a vulnerability index based on factors including income, race, and linguistic isolation to inform a social vulnerability index based on the cumulative properties of individual vulnerability within each neighborhood. Finally, using National Aeronautics and Space Administration (NASA) Black Marble imagery, the study assessed inequity in power loss resulting from Uri by comparing anomalies in nighttime light values during the storm and its aftermath to the social vulnerability and neighborhood environmental covariates. As such, the study provides avenues for discussion for adapting this approach to wider spatial and temporal scales, as well as a larger variety of recent extreme weather events.

The fundamental concept of modern seismic design of structures is achieving safety against collapse under severe earthquake ground shaking. This philosophy, by design, allows buildings to sustain significant damage that may deem them unsuitable to be occupied for extended periods of time. The COVID-19 pandemic has introduced new challenges to this well-established philosophy, because the displacement and possible injuries of building occupants after an earthquake, may lead to overcrowding of shelters and hospitals, and leave thousands of people substantially more vulnerable to the pandemic. These vulnerabilities have been noted in recent earthquakes, e.g., in Mexico, where crowded hospitals struggled to meet the cascading demands of earthquake injuries and COVID-19 patients. Such incidents raise the question: can structural engineers build better structures that enable occupants to shelter in place, in order to meet the new demands imposed by COVID-19? This study examined the impacts of a magnitude 7.0 earthquake in one of the most populous urban areas in the United States, the San Francisco Bay Area, with a particular focus on the expected damage to modern building structures. The author of this study used physics-based earthquake simulations in combination with structural simulations to predict the portion of buildings that may become uninhabitable after the earthquake, and how these conditions can increase the vulnerability of the residents to COVID-19. Finally, the author of this study explored possible improvements to the seismic design of the buildings that may reduce their susceptibility to damage, and increase their potential for immediate occupancy after the earthquake.

The COVID-19 pandemic has underscored the ways in which emergencies represent evolving epistemic situations. Knowledge about emerging hazards changes over time; what “facts” are socially accepted can adjust based on political, empirical, and ideological pressures; and what is “common sense” one day might be rejected the next. While this has been illustrated in several cases throughout COVID-19 (such as in changing viewpoints on masking or airborne spread), it also poses challenges to other forms of disaster research.In this paper, we discuss how the theory of “symmetry” can be adopted to underpin new methods of research design. Drawn from the field of Science and Technology Studies, the principle of “symmetry” refers to a commitment to explaining both “true” and “false” beliefs through analogous social processes, rather than an appeal to their ultimate truth (or lack thereof). This is particularly important when what is “true” changes - such as in research attempting to explain why people hold certain attitudes about mask use or coronavirus transmission. We begin by considering specific cases where the “facts” changed throughout the course of the COVID-19 pandemic. We then analyze the strengths and limitations of different questionnaire-based approaches to conducting surveys on these topics. In doing so, we illustrate the improvements to data quality, longitudinal designs, and consistency that can be obtained through adopting symmetric approaches. We conclude with a series of lessons for survey-based research in emergency contexts.

Public housing residents are often low-income, racial minorities, seniors, and people with chronic stress and disabilities who are dependent on governmental resources. These residents often struggle to access resources even in non-disaster contexts and disasters only serve to put extra pressure on them to find resources to meet their recovery needs. To better serve these populations requires a better understanding of the various institutions involved in their recovery and better ways to engage these institutions into the formal recovery processes Community-based organizations (CBOs) and nonprofits are key institutions in this process. CBOs provide critical housing, health, and social services to marginalized populations after disasters. However, CBOs may be limited in their ability to do so by impact from concurrent disaster events.

This study explores the challenges and opportunities of engaging CBOs that serve public housing residents into public health and housing recovery efforts after concurrent disasters and the extent to which concurrent disasters complicate these efforts using the case of the 2020 Southwest Earthquakes and COVID-19 in Ponce, Puerto Rico.The study employs a qualitative approach, using archival research and key informant interviews with community-based organizations that serve public housing residents, housing and health advocates, and public housing and public health administrators that are currently involved in post-earthquake and pandemic response and recovery in Ponce, Puerto Rico to investigate these questions. These findings help planners, and policymakers better understand and engage CBOs in public housing recovery, and through that improve public health outcomes of these socially vulnerable residents.

Real estate purchase is a major transaction for many households in the United States and the price of real estate properties may or may not reflect a wide range of associated environmental risks (e.g., geological and hydro-meteorological hazards). Earlier hedonic studies consider one or the other type of these risks and analyze how that affect the property values. In this study, the authors consider both geological hazards (sinkholes) and hydro-metrological hazards (hurricanes), and analyze their impacts on property values. This study investigates how the proximity of sinkholes affect the housing price using real estate sales data of almost 35,000 single family homes in Lake County, Florida from 2014 to 2018. Given that the area has experienced significant damages during a recent hurricane (Irma in 2017) and the state of Florida instituted a change in sinkhole insurance law in 2016, the analysis can exploit variations in a quasi-experimental setup and use difference-in-difference and regression discontinuity techniques to establish causal relationships. Findings show that houses near to known sinkhole locations experience a significant price discount, which is amplified following a hurricane event. The authors also analyze the effect of the new sinkhole insurance law on housing prices in Florida. Findings show that houses that are located close to known sinkhole locations face a subsequent price discount due to reduced sinkhole damage protection offered by the new insurance law.

Post-disaster supply chain disruptions are more likely to adversely impact small businesses because they are typically less-resourced, have higher rates of minority ownership than other size businesses, are limited in their ability to develop alternative supply streams, and have higher transaction costs due to lower bargaining power and reputation. Whereas franchises and large corporations may get assistance from their headquarters, local and small businesses have limited resources to overcome unexpected interruptions from disasters. Also, small businesses take fewer mitigative actions against disasters and are generally less prepared for such events. Lastly, small businesses not only have fewer programmatic options for post-disaster aid, but these programs often require significant and tedious paperwork that acts as a barrier to their participation. This study addressed this gap by examining the recovery experiences of small businesses after the 2020 Magna Utah (UT) Earthquake and the COVID-19 pandemic in Salt Lake City, UT. This study sought to more deeply understand whether the experience with current disasters has affected the resilience level of small businesses to future events. Specifically, the study addressed the following questions: (1) What was the impact of the COVID-19 and/or the Magna earthquake on small businesses operations within Salt Lake City? (2) What recovery actions are taken by small businesses to maintain operations in the face of disasters? and (3) To what extent does this recovery process inform future disaster preparedness and resilience?

Government acquisition of residential land has played a growing role in the reconstruction of housing in safer places and reduction of water-related risks. This paper explores how the rationales and processes of residential buyouts have resulted in different impacts for coastal recovery and mitigation after the 2011 Great East Japan Earthquake Tsunami in comparison with Superstorm Sandy in the United States. Using government documents, the characteristics of buyout programs were explored,consequences of mitigation and recovery through analysis of secondary data were identified, and the effects of community buyouts were deduced. Analysis revealed buyout results that included sustainable and resilient coastal rewilding, enhanced community livability through the recreation of blue-green areas, and building of community resilience, as well as projects that transfer the experience and narratives of disaster in buyout areas.However, buyouts have also contributed to community fragmentation, and limiting homeowners’ opportunities to make their own choices to stay or move away. Buyout effects are influenced by government interventions for post-disaster recovery which induced wide disparities between cities and communities, raising issues of social inequity and environmental injusticeIn light of the multifaceted role of residential buyouts emerging throughout long-term recovery processes, these results suggest that planners and disaster managers need careful consideration to redesign and manage property acquisition programs that not only increase regional resilience but also are equitable for affected residents.

Today, few communities can instantaneously see all of their subsurface utility data. This is often caused by a lack of digitization, incompatible formats, proprietary issues (spanning security, privacy, and competitive advantage), and/or abandoned assets. The inability to visualize (to say nothing about manage) subsurface data not only complicates capital planning, maintenance, and emergency response activities, but it poses major impediments to assessing and enhancing community resiliency. The advancement of an international data storage and exchange format through the Open Geospatial Consortium holds the promise to revolutionize this situation. This study provides (1) an overview of the model, (2) the implementation to date in the United Kingdom as part of the National Underground Asset Registry, and (3) insights gained during the pilot implementation at two testbeds in New York City as part of a National Science Foundation funded Civic Innovation Challenge project, which involved seven unique asset holders from telecom to transportation and directly involves all New York City based emergency services and first responder organizations.

Extreme weather events increasingly have greater impacts because of exposures, vulnerability, and, in some cases, increasing incidence or severity due to climate change. At the same time, meteorological science and communication technologies are advancing. To fully take advantage of these advances and improve hazard outcomes, the creation and communication of weather hazard information should center around meteorological justice. This study uses data from several focus groups conducted after four major hurricanes in the United States–Sandy, Matthew, Irma, Harvey–to make recommendations for building an ethical warning and forecast communication system. Rather than focusing on a vulnerability paradigm, this research flips the emphasis to focus on how justice can become more central to risk communications in ways that empower people to best access, and utilize hazard information.

In order to predict the loss and the damage from the hazards, such as debris flow resulted from dam failures, three important factors must be taken into account: the strength of hazard, the inventory and the vulnerability of the inventory to the hazard. In the case of the debris flow, the flow speed, the inundation boundary and depth, and the flow force can be the hazard. The inventory corresponds to the list of assets and demographic distribution while the vulnerability is the probability of the damage of each inventory by the specified hazard. In this study, the hazard is assessed from three-dimensional numerical simulation of the debris flow incurred by the dam failure. Since the detailed description and modeling of the inventory is nearly impossible, the present study utilized geographic information system-based regional assessment of the vulnerability combined with the inventory, in which the distribution of the inventory represents the exposure and the performance of the inventory such as age of building represents the sensitivity. The selected proxy variables are evaluated with predefined scoring criteria and nondimensionalized based on a standardization method. The resulting vulnerability is normalized for the relative assessment with the region of interests. The computed strength of the hazard is then convoluted with the normalized vulnerability and the results show the risk of the region.

Disaster risk reduction (DRR) policies, like building codes and construction regulations, protect the public. At the same time, public opinion may shape these policies and their enforcement. This paper presents findings from a National Science Foundation-funded collaborative research project. The study analyzed survey data (n=25,500) collected from 16 Latin American and Caribbean countries and the US using regression-style modeling to study the effects of three sets of factors on public support for DRR: 1. perceptions of disaster risk; 2. political trust / corruption perceptions; 3. experience with disasters.This study built upon public policy scholarship to answer a question central to the third set of factors: Do disasters, as potential focusing events, shift public opinion in ways that might open windows of opportunity for improving DRR policies/implementation? Disasters are often-mentioned focusing events, but the effects of experiencing disaster (whether, when, how) on perceptions of risk and support for DRR remain unclear. Here, the study tested the following hypotheses: (1) Experiencing a major hazard event in the past increases one’s support for DRR policies; (2)The more harm a person or their family experienced from a major hazard event, the greater their support for DRR policies; and (3)The more recent the experience of a major hazard event, the greater that person’s support for DRR policies. Preliminary results are consistent with H1 and H3 but not H2. Individuals’ support for DRR is bolstered by experience with disaster (especially if recent) yet is apparently not related to the severity of the harm experienced.

As the tsunami threat across the Pacific coast becomes better understood, it appears that structures designed to meet “life-safety” requirements during an earthquake may not survive subsequent fluid loading due to tsunami inundation and flooding. As such, it is crucial to examine the resilience of structures susceptible to sequential earthquake and tsunami loading. Design of structures subject to these demands requires careful consideration of the fluid-induced forces, which are largely dependent on the motion of the structure upon which the fluids impinge. Motion of a structure during the tsunami phase of loading relies upon the dynamics of the damaged state of the structure following an earthquake. Thus, it is imperative to account for the damage sustained during an earthquake in a structural model of a building subject to tsunami loads to evaluate the true performance of the structure in-situ. To address this, an open-source tool (FOAMySees) was developed for simulation of tsunami and wave impact analysis of post-earthquake nonlinear structural response of buildings. While FOAMySees was developed specifically for tsunami-resilience analysis, it can be utilized for other fluid structure interaction applications with ease. Through this coupled CFD-FEA program, tsunami and earthquake simulations can be run sequentially or simultaneously, allowing for the evaluation of nonlinear structural response to multi-hazard excitation. FOAMySees shows strong correlation with other fluid structure interaction simulation software when benchmarked against analytical and experimental test cases.

While the uncertainty of the COVID-19 pandemic is declining, the public’s critiques of response and noncompliance remain hot topics even among the most stringent institutions in East Asia. This study was able to detect the critical information content distributed by government stakeholders that made up people’s understanding of the risk. The authors also examined how the public’s collective norms and the risk information influence their attitudes toward stakeholders and protective actions. In particular, findings show how information content related to personal protections were the significant predictors of emergent norms during the early stage of the pandemic. Second, the conflicting information between governments and other sources can harm resident’s perception of stakeholders. This effect was strong especially when the norms (predecision processes and threat perceptions) were low. Next, the norms and stakeholder perceptions could positively explain the public’s attitudes toward the necessity of lockdown and the life satisfaction during the lockdown. Finally, the stakeholder perceptions were found to be positive predictors of self-reported protective actions.

The findings updated the protective action decision model (PADM) through the inclusion of the prolonged exposure to risk experienced during pandemics, which extends the formation period of social norms and perceptions. The research methods can assist researchers in better defining the boundary of communities by detecting informal and latent resident groups and measuring their perceptual and behavioral outcomes. The findings also provide valuable guidance for governments and decision-makers in targeting and tailoring warning messages to the varying social norms and levels of risk perception of the residents they serve.

Communities are at growing risk of coastal hazards due to continued urbanization of coastal areas compounded by climate change.This paper explores how the rationales and processes of residential buyouts have resulted in different impacts for coastal recovery and mitigation after Superstorm Sandy in comparison with the 2011 Great East Japan Earthquake and Tsunami. Drawing from government documents, literature, stakeholder interviews and field visits, the characteristics of various buyout programs implemented during recovery in Sandy-affected areas were explored, consequences of mitigation and recovery were identified through analysis of secondary data, and the effects of community buyouts were deduced. This analysis revealed buyout results that included sustainable and resilient coastal rewilding, but also contributed to exacerbating social inequities and environmental injustice and limiting homeowners’ opportunities to make their own choices to relocate. Conducted without post-disaster recovery land use planning including a vision of how to recover and mitigate following disasters, as merely a property transaction and mitigation measure, buyouts do not contribute to community recovery in Staten Island. In light of the multifaceted role of residential buyouts emerging throughout long-term recovery processes, these results suggest that planners and disaster managers need careful consideration to redesign and manage property acquisition programs that not only increase regional resilience but also are equitable for affected residents.

Inundation events, such as tsunamis and storm surges, pose a significant threat to coastal communities and infrastructure globally. Damage is not only caused by flowing water, but also by debris from collapsed structures, vegetation, and anything mobilized by the inundation flow. Previous studies have investigated wave- and flow-induced loading, but there are few that investigate the influence of debris fields carried by flows. While single-debris impacts are fairly predictable and reproducible, debris field impacts are chaotic in nature and require a statistically-driven approach. A comprehensive experimental program at the Natural Hazards Engineering Research Infrastructure Wave Research Lab’s Large Wave Flume was designed to generate a statistically-representative data set to understand structural loading during a debris-laden inundation event and inform numerical modeling efforts. The large volume of data is intended to shed light on the non-deterministic phenomena underlying flow-driven debris fields.The study will systematically analyze data collected and present the following advances in understanding: (i) interactions between a debris-field, fluid, and stationary structure; (ii) comparison between a single debris tests and large-scale debris field; (iii) trends in three-dimensional structural loading from a debris field as correlated with initial condition; and (iv) rigid body dynamics of a flow-driven debris field. Ultimately, this study will inform how a flow-driven debris field may interact with the coastal built environment during large-scale inundation events, and seeks to increase coastal community resilience.

Studies from a variety of disciplines reveal that humor can be a useful method to reduce stress and increase compassion, connection, and empathy between agencies and people they serve during times of crisis. Despite this growing evidence base, humor’s use during geohazard (earthquake, volcanoes, landslides, and tsunami) humor’s utility to aid scientific agencies’s crisis communication response has been rarely studied. A broad literature review of humor in crisis, and an exploratory examination of several case studies, reveals that scientific organizations, specifically those that respond to geohazards, can harness the power of humor to help create connection and empathy with the publics they seek to serve. Evidence reveals that the use of humor acknowledges a shared human experience, reducing the barriers between public officials, scientists, and the people most impacted by crisis. Public statements made by scientists and public officials during the U.S. Geological Survey (USGS) response to the Kīlauea eruption in 2018 in Hawai’i and GNS Science/GeoNet response to the M7.8 Kaikōura/North Hurunui earthquake in 2016 in Aotearoa New Zealand, are used to inform the development of this conceptual model. The researchers then posit a conceptual model which unifies concepts from the literature with case studies to help provide potential guidelines for those crisis communicators working for science agencies on how best to use humor during times of crisis to help people cope with crisis. This model can be further tested for future research to determine its effectiveness and utility for scientific agencies responding to geological crises. 

How will now routine seasons of wildfire destruction affect human settlement patterns? This study examines building destruction and post-fire reconstruction among the approximately 23,000 buildings located within the burn footprint of the 2018 Camp Fire. Drawing on a range of geospatial data, results show that mobile homes, lower-value, and renter-occupied dwellings were significantly more likely to be destroyed compared to single-family, higher-value, and owner-occupied dwellings. Analysis of buildings’ physical characteristics indicates that differences in destruction outcomes were most attributable to level of building density. These findings suggest that filtering of housing took place, wherein less valuable residential buildings were physically more susceptible to hazard damage.

To map post-fire building reconstruction patterns, a support vector machine algorithm was developed in Google Earth Engine using data from the National Aerial Imagery Program and Microsoft’s Building Footprints database. Applying this algorithm to post-fire imagery, buildings that had been reconstructed 20 months after the fire were identified. New buildings were located on properties that were more likely to have been owner-occupied prior to the fire, and, on average, had higher pre-fire value. These findings provide evidence for the theory of cost-burden climate gentrification, wherein wealthier households have a greater capacity to remain in a hazardous area than less affluent households. Collectively, the study’s findings suggest a need for wildfire hazard mitigation strategies that operate at a neighborhood and regional scale, are better tailored to the needs of low-income residents and renters, and are designed for dense developments, especially mobile home parks.

Housing is one of the most basic social needs that determine the quality of life and wellbeing of its residents. In an ever-changing and fast-pacing urbanizing world, affordable housing has not kept up with citizens’ demands. Those from low- or middle-income classes face more challenges finding houses that fit their needs and budget. Most housing development programs are designed to meet affordability requirements rather than to consider sustainability principles. Moreover, construction practitioners are primarily concerned with profit as opposed to sustainability. Therefore, outcomes of most projects do not lead to “safe and just space for humanity.” Ironically, some of those projects end up overshooting the ecological ceiling (causing problems such as global warming and air pollution) and/or result in trapping social foundations (e.g., increasing inequity in access to basic needs). This study reviewed the barriers and challenges of integrating affordable housing and sustainable practices in Jacksonville, Florida. The study was based on an extensive literature review and by engaging local communities through the five cycles of Participatory Action Research (PAR). According to the collected data, challenges were identified in different categories, such as environmental, marketing, current policies, societal, economic, and techniques. Then, Critical Success Criteria (CSC) was utilized to analyze sustainable affordable housing performance. This approach can help develop sustainable affordable housing policies to improve the status quo, and anticipated affordability crisis, for the targeted area to make effective decisions regarding built-environmental development.

Households living in the Gulf Coast areas have recently suffered from multiple crises, which caused a severe and long-lasting impact on their well-being. While many of them are still recovering from the impact of two consecutive hurricanes (Harvey and Irma) by resuming their jobs or seeking new employment opportunities, COVID-19 hit the entire world and created another round of health and economic shocks. The authors of this study present comparative analyses to understand the extent and distribution of workforce disruptions from a pandemic and a natural disaster based on two hurricane survey datasets. According to the Hurricane Harvey survey of 780 residents in Texas, 24.23% experienced a reduction in income, and among them, only half (50.79%) were able to recover when COVID-19 hit again. The average working hours per week reduced from 21.2 hours before both disasters to 18.6 hours due to the hurricane and 19.9 hours due to the COVID-19. According to the Hurricane Irma survey of 768 residents from Florida, the average working hours reduced from 20.1 hours before both disasters to 18.4 and 18.2 hours due to the hurricane and COVID-19, respectively. Spatial and demographic information of the respondents were also collected in both datasets to identify the most impacted areas and vulnerable groups. This study provided policymakers, disaster management, and public health agencies with reliable information about the distribution of the impact to help inform the relief and recovery efforts.

In this study, the authors conducted the first comprehensive, nationwide analysis of multiple federal post-disaster and pre-disaster grant programs. The authors of this study assessed the social equity of disaster aid distribution by examining the relationship between a county’s receipt of federal aid and its recent disaster damages, socioeconomic, demographic, local government, and geographic characteristics. We estimate a two-stage random-effects tobit model to examine the correlates of disaster aid by program and address the endogeneity of disaster damages (from floods and storms) using instrumental variables and the control function approach. Based on a panel data set of over 3,000 U.S. counties from 1990-2019, the authors found that disaster damages are influenced by local socioeconomic conditions and they are also a key driver of federal disaster aid, particularly those requiring Presidential Disaster Declarations. For all disaster programs, the authors showed that more federal aid is disproportionately allocated to counties with larger populations. For public disaster programs (Federal Emergency Management Agency’s Public Assistance, Hazard Mitigation Grant Program, and Flood Mitigation Assistance), more aid is received by counties with better economic conditions (higher median housing values, lower poverty rates, or more fiscal resources), suggesting a potential mismatch between the need and federal assistance. Private disaster relief programs (Individuals and Household Program and lower-interest disaster loans) show relatively better social equity performance, as their aid is more distributed to counties with lower incomes and greater social vulnerability. The results also indicate that counties located at higher-risk regions receive more disaster aid conditional on their recent damages.

Online social networks play a critical role in major disasters such as hurricanes where different agencies and the public interact and communicate the underlying risks and protective actions. This study revealed such crisis communication patterns in the aftermath of hurricane Laura compounded by the ongoing COVID-19 pandemic. The study utilizes large-scale Twitter data continuously monitored during Hurricane Laura which was one of the strongest (Category 4) hurricanes on record to make landfall in Louisiana. Twitter followee-follower relationships were used to generate the online social graphs. Using network science theories and advanced community detection algorithms, the study split such networks into eleven communities and several isolates were also identified. The team applied different natural language processing techniques (i.e. word clouds, bigrams, topic modeling) to those communities to observe risk taking or averting behavior of users. Findings showed that social media presence on Twitter from local stakeholders such as news channels, radio, universities, sports pages, among others, interacted closely with local residents during this major crisis. In contrast, emergency management and planning units in the area engaged less with the public. Findings of this study provide novel insights into the design of efficient social media communication guidelines to respond better in future disasters.

In recent years the word 'emergency' has found a place in everyday life. In a world increasingly interdependent and marked by planetary disruptions, emergencies are not only on the rise but are becoming a kind of emergency condition punctuated by a series of catastrophic events. The author of this study posed several questions from an architect and urban planning perspective including: what is understood by ‘emergency’ and how does it operate? What does it mean for spatial practices to act under emergency conditions? How can researchers identify these processes and draw them? Can we find cases in which ‘emergency conditions’ have already been integrated into built-environment practices?

From these questions, the author started a thesis regarding emergency architectures, and during the first year, they created the ‘Emergency Atlas’. The recomposition from the emergency appears to critique the spatial practices solutionism. In addition, preparedness and post-emergency research have been carried out from process or event casuistry as discrete entities or isolated phenomena. In this way, the Atlas is constituted of different information layers, at different scales, in a methodology of diffraction and recomposition. The Atlas retraces past, present, and future events in different entities that overlap to create an emergency orography and with it a reading of events that modify our experience with the territory and create vectors of action and relationship. Consequently, this Atlas proposed methodologies to navigate the numerous environmental ruptures and their changing dynamic realities from all the scales supporting life and its frictions.

It has been widely demonstrated that rental units are slower to recover than owner units, since post-disaster funding has traditionally favored owners. Existing housing recovery models focus on owner units and do not capture the unique challenges related to the recovery of rental units. This study introduces a methodology that considers both owner and rental units for the simulation of housing recovery and applies it to a hypothetical M7.0 earthquake affecting Alameda, California. In some communities, a significant portion of the rental properties are owned by local landlords. Past disasters have shown that landlords prioritize repair to their own homes, further exacerbating the differences in the recovery processes of renter- and owner-occupied homes. Rental units and the address of their landlord are identified by comparing parcel addresses and mailing addresses recorded for each parcel in the local tax assessor database. Landlords with addresses within the city can experience damage to their own homes as well as their rental units. When landlord occupied homes are damaged, repairs to their rental units are delayed. The most likely financing avenues available to repair rental units are identified considering damage to the home of the owner. Preliminary results show that the proposed model captures the disparity in the recovery of rental and owner housing and helps us better understand the contributions of various factors to that disparity. Thus, the proposed approach can be used to design effective strategies to reduce inequity between renters and landlords after disasters.

This study combines multiple impacts of sea-level rise on natural hazard risk due to climate change. The authors provide a new method of comparing impacts of different hazards on the built environment and possible mitigation strategies. Coastal areas are impacted by a variety of geohazards with a range of physical impacts and frequencies, from nuisance flooding to tsunamis. Based on current climate change scenarios, sea level rise will cause flooding in many coastal areas. However, the compound impacts of multiple hazards must be considered to understand the risk to a community. In addition to overland flooding, groundwater table shoaling can cause emergent groundwater and an increased probability of soil liquefaction during earthquakes. Quantification of combined impacts to the built environment allows for the comparison of the contribution of each hazard to the risk of the community and the design of multi-hazard mitigation strategies. A case study is presented for Alameda, California, which is a low-lying island in the seismically active San Francisco Bay area, susceptible to liquefaction and coastal inundation. The projected impacts of sea-level rise are contrasted to demonstrate the difference in considering hazards individually and with the proposed multi-hazard method.

Resilience bonds are a type of insurance-linked security meant for disaster mitigation. The authors of this study aim to develop a framework for coastal community resilience bonds (CCRB) for Charleston, South Carolina, a low-lying coastal city facing major flood risks. The authors designed a community engagement process before the COVID-19 pandemic with two groups of participants: Charleston’s local resilience stakeholders and finance and bond experts nationwide. The pandemic posed specific obstacles that required the authors to reframe research methods in response to travel restrictions and social distancing policies related to COVID-19. The authors switched to be fully remote and conducted the outreach and interviews virtually on Zoom, though the original design of the community engagement process was intended to be in person. There are several implications that the adapted engagement methods have for methodological advancements in engaged qualitative research. First, building a common understanding around a novel concept—such as resilience bond—is challenging in online communications, can take a longer time, and more patience on both sides.Second, as outsiders, the research team did not have adequate prior connections with Charleston stakeholders or finance experts, making it difficult to get responses from potential informants to participate in the research. Third, finding alternative ways to learn about footprints of previous disasters in Charleston, building trust with local and finance industry participants, and understanding various and sometimes conflicting local perspectives on political, social, and environmental issues related to flood resilience became necessary due to postponing field visits due to travel restrictions.

Improving the resilience of a community is a complex and costly process with many stakeholders competing for various potential sources of funding. One of the novel financial tools to fund such efforts is a resilience bond. While disaster resilience bonds are discussed in the literature as a mechanism for communities to protect themselves against catastrophe, the concept is often focused solely on mitigating damages to physical infrastructure, a necessary but insufficient condition for community resilience. Working with the City of Charleston, South Carolina, a historical city on the Atlantic coast facing a myriad of disaster risks, the authors of this study hypothesized that resilience of coastal communities can be substantially improved if community-driven equity and social infrastructure priorities are considered in tandem with physical infrastructure in the resilience bond development process. This study draws on one part of a larger project that aims to develop a resilience bond for Charleston. An in-depth qualitative analysis of semi-structured interviews with local resilience stakeholders and financial experts, media articles, and local organizations in Charleston documents various dimensions of resilience investment priorities, opportunities, and constraints within a historic context of development and disasters in the study area. Findings from this community-engaged qualitative research informs development of a combined physical and social bond framework that enables both the repair of physical systems and valuation and investment to enable timely and equitable recovery of social infrastructure in Charleston, prioritizing equitable hazard mitigation for socially vulnerable communities.

East North Carolina (ENC), a predominantly rural region, often experiences the intersection of fluvial, pluvial, and tidal flooding, which lead to complex and impactful outcomes, such as morbidity, mortality, economic disruption, and loss of livelihood. Managing these connected hazards is challenging, especially as climate change drivers will likely lead to a greater incidence of compound coastal water events. Funded by a 2019 National Oceanic and Atmospheric Administration Coastal and Ocean Climate Applications Sectoral Applications Research Program grant, a key objective of this research is to assess the perceived risks and needs of the hazard management and planning community in ENC. This paper focuses on the mitigation tools used by ENC communities and the barriers they face in coordinating mitigation efforts.

The project focuses on rural counties in ENC located along the coast and those adjacent to it that share estuarine environments or linked riverine systems. Data was obtained through focus group interviews conducted with 41 planners and emergency managers during a flood workshop held on campus in February 2020. Study findings show that ENC communities use a combination of public education, regulations, and floodplain planning and management to minimize exposure to floods. Communities also actively seek funding to clear streams and channels, to implement buyouts and elevation projects, and for infrastructure maintenance. At the same time, ENC communities face several barriers to mitigation including lack of adequate funding to meet their needs and conflicting priorities among governmental entities.

Implementation of non-pharmacological protocols in the protracted COVID-19 pandemic has significantly curtailed travel and face-to-face interactions necessary for data collection. The implications of which include the loss of ephemeral disaster research data and missed opportunities of engagement among communities and researchers. Ethical considerations, inconsistent public health responses, and institutional statutes have also contributed to impeded disaster research in the last two years and are ongoing. While the call for the co-creation of knowledge through collaborative research with communities and local researchers is not new, the imperative has come in sharp focus during the pandemic. Additionally, decolonizing research agenda-setting and methodologies relevant to addressing global disaster hazards and threats deserve prioritizing. Thus, an opportunity exists to fundamentally adjust how disaster research is conceptualized and operationalized.

The authors of this study present the process of enhancing existing and coherent community non-academic systems for disaster-informed research. During the ongoing pandemic, the authors partnered with a Kenyan non-profit organization initially to develop pandemic response strategies which evolved to community initiated participatory action research training. The latter being a function of community intent to attain disaster justice and sovereignty in the face of perceived disaster injustice. In this presentation, the authors seek to share the central role of community initiating and implementing the project, the process of virtually training the Kenyan team while in the United States, the significance and value that this and replicated models would have on travel and contact restricted disaster research, and finally implications for institutional and funding support.

The challenges of conducting research in and with diverse communities are exacerbated by simultaneous and cascading disasters. Tools such as community-based participatory research (CBPR) principles provide guidance for working within marginalized communities from conception of research projects to implementation of findings; however, there is limited guidance on how to conduct research during emerging crises. This study examined how researchers can disrupt and alter the participatory research principles to conduct research as crises emerge in Black, Indigenous, People of Color (BIPOC), immigrant, and refugee communities–those most impacted by the global pandemic, racism, economic volatility, housing instability, and more–in Washington State. The authors conducted monthly meetings over six months with members of the Community Health Board Coalition and Vietnamese Health Board to understand how researchers can engage grassroots health boards in mixed-methods research as disasters unfold. The authors found that members of the health board were aiding their community as well as experiencing some or all the crises themselves, hence they seek partnership with researchers who will “do the work with them.” This means being actively engaged with response activities such as advocating for equitable distribution of resources and systemic change. These types of engagement activities build trust and could lead to meaningful and sustained research projects. Findings from this study resulted in a researchers’ toolkit that provides guidance for researchers who are interested in work with BIPOC communities who are often disadvantaged, marginalized, stigmatized, and under-represented.

The frequency and intensity of natural hazards are exacerbated by climate change which increases the exposure and vulnerability of communities to severe storms. While there have been multiple studies on the risk, post-hazard functionality, and recovery assessment of communities exposed to natural hazards, current methodologies stop short of comprehensively investigating the collective relationship between the interdependent community, namely households, buildings, and networks. This includes the physical, social, and economic interdependencies between the different building sectors. In this study, a novel comprehensive post-hazard functionality assessment approach was developed to account for the total post-hazard functionality of the different building sectors within a community. This was done by modeling the interdependencies between households, buildings, and networks along with propagating the inherent uncertainties through the different components/systems. A portfolio of building archetypes was used to model different building typologies within the community. Detailed models for power, water, and transportation networks were also developed and linked with the buildings models to quantify the physical functionality. A synthetic allocation algorithm was also developed using the United States Census data by allocating households to housing units and labor/staff to businesses. The developed approach accounts for the impact of the social and economic disruption on the total post-hazard functionality of the different building sectors which can better allow better risk- and resilience-informed decision making.

The notion that disasters, and subsequently disaster management, is becoming more complex has become axiomatic in policy dialogues. However, there is a lack of precision in understanding how the changing nature of disasters translates into changes in the complexity of the task environment being managed. Without a more nuanced understanding of these changes over time, it is difficult to assess the extent to which and ways in which the institutions we have created for responding to disasters are keeping pace with the changing nature of the challenges they are being asked to manage. Institutional change is generally neither fast nor linear. This begs the question, are we managing today’s disasters with yesterday’s disaster response systems? If so, are these systems adequate for the new challenges they are encountering? No more is this true than in the rapidly evolving context of wildfire. In this presentation, preliminary findings are presented from the creation of a 20-year longitudinal study aimed at understanding the pattern of changes that have occurred related to the institutional complexity of the United States’ most complex wildfire events. Changes in incident size, duration, and scope as well as values at risk, cross-boundary jurisdictional complexity, and resource scarcity associated with the National Preparedness Level are considered. By framing the problem using geographic boundaries in addition to administrative complexities, the extent to which and ways in which the task environment is more complex is examined. Implications for institutional readiness to respond to complex wildfire events across US regions will be discussed.

Disaster warning messages represent a fundamental form of risk communication that aims to provide at-risk individuals with the information they need to make protective action decisions. Therefore, warning messages should help warning recipients perceive danger by indicating the severity and probability of experiencing a hazard while simultaneously motivating them towards action. This information may also cause negative emotions, such as fear and anxiety, in message recipients. With a particular focus on emotion, this study further explored the theoretical mechanisms behind warning response by experimentally manipulating hazard and guidance language in tornado and flash flood warning messages. This study then determined how variables such as perceived threat, perceived efficacy, fear, and anxiety, related to one another and explained information seeking and protective action intentions. The results of this study indicated the degree to which each concept predicts protective action and information-seeking intentions and how these intentions are influenced by emotion. These findings can help warning message designers and practitioners better prioritize the type(s) of information to include and/or highlight in a warning message.

Over two years into the global COVID-19 pandemic, childcare providers in the U.S. continually respond to the challenges of new variants, changing mitigation strategies, and the cumulative impacts of worker instability, low wages, and heightened risk among children who remain ineligible for vaccination. In response, this study surveyed childcare providers in the five-county metropolitan Rochester, NY region to understand the impacts of COVID-19 on their facilities between late September 2021 until mid-winter 2022. This timeline reflects the intermediate stage of an ongoing disaster landscape, reflecting the impacts of both the Delta and Omicron variant waves on providers. Using quantitative and qualitative indicators, the study investigated changes to staffing, operations, responses to positive cases, federal and other assistance in the first year, morale, and financial stability. Preliminary data revealed a heavy burden upon providers, as demands for personal protective equipment (PPE), safety protocols, closures, and other requirements are rarely backfilled with resources to meet those evolving standards. Staffing was also a challenge as this critical infrastructure pays very little, yet demands so much, especially given the exposure risks among unvaccinated children. To promote COVID-19 safety, this study engaged in fully remote data collection through electronic surveys, zoom based research team meetings, and electronic communications. Lastly, from a reflexive perspective, this study was designed from a “what you know standpoint.” The Principal Investigator is a mother with two children in daycare and primary school settings, designing research to support those in the community who directly enable her own ability to work. Childcare reverses the Shedemic.

United States infant health outcomes remain poor relative to other high-income countries, especially among minoritized infants. A large and still growing literature demonstrates the negative impacts of toxic exposures on minoritized infants at birth, suggesting that direct exposure to environmental toxins, in utero, negatively impacts infant health. However, other environmental stressors may play a role in birth outcomes, such as toxic maternal stress resulting from disasters. Given known racial/ethnic inequalities in disaster experiences, disasters may drive and/or exacerbate racial/ethnic inequalities in birth outcomes. Drawing on National Vital Statistics System Natality (NVSS-N) data, this study assessed the impacts of different types of hazards (hurricane, earthquake, wildfire, flood) on Black, Hispanic, and White infant birth outcomes using a unique panel spanning U.S. counties, collected annually from 1980-2002. Based on five and ten-percent samples of births at the county level, this study estimated two-way fixed effects models for multiple indicators each of birth weight and gestational age. This presentation provides insights into the impacts of different types of hazards on infant birth outcomes in disaster-affected regions, and it will provide important insights into pernicious racial/ethnic disparities in birth outcomes by comparing birth outcomes before and after a disaster. As disasters increase in frequency and intensity, it is critical to understand the mechanisms between disasters, racial inequality, and birth outcomes for improving public health preparedness and disaster response.

Known for its diverse cultural heritage and fishing opportunities, southeast Louisiana faces a range of environmental threats. Fishers, because of their affinity and familiarity with marine ecosystems, notice these environmental changes and feel firsthand the impacts of climate change mitigation strategies, such as river diversions. As such, their perceptions of the environment, fishing practices, and responses to climatic change can provide key information for future environmental policy and resource management. Within this group, but potentially overlooked, are recreational fishers, a subgroup with diverse motivations and experiences. This presentation will show preliminary findings from the National Academies Gulf Research Program-funded Survey of Recreational Fishers (SuRF), a web-based survey of recreational fishers in and around Plaquemines Parish, Louisiana. The SuRF was developed by an interdisciplinary team, including social scientists, ecologists, biologists, and legal scholars. It includes a range of questions ascertaining environmental risk perceptions, changes to fishing practices, responses to ongoing and proposed river diversions, and socio-demographic questions. The presentation will provide descriptive analyses of findings and will discuss research methods and analytic strategies implemented due to the COVID-19 pandemic and Hurricane Ida. Among the impacts of these disasters, the team opted to conduct recruitment and data collection remotely with the help of key stakeholders and Hurricane Ida occurred during that data collection. Both disasters shaped the participants of the survey and their responses, a dynamic which will form part of the presentation’s discussion. The presentation will conclude with lessons learned for researchers and policymakers.

COVID-19 presents a direct threat to the population, with concerns for health being cited as justification for a range of protective action recommendations and governmental policies. At the same time the economic impacts associated with both protective measures and falling ill have also featured prominently in debates of how the pandemic should be handled. Research has identified several factors that can shape risk/threat perception and concerns regarding hazards from exposure to and familiarity with the hazard to demographic characteristics. While the widespread presence of COVID-19 has produced large-scale exposure to the pandemic, the wide variation in direct exposures to the virus, variations in individual and household vulnerability, and governmental approaches to managing the pandemic has resulted in substantial variation in pandemic experiences. It is possible, then, that this variation in experiences has led to not just different levels of concern about the pandemic, but also to meaningful differences in what people are concerned about. This study builds on existing research on threat perception by looking at two types of threat perceptions for COVID-19: health concerns and economic concerns. Using survey data of adults in New York, Louisiana, and Washington, this study explored both people’s level and type of concerns in the COVID-19 pandemic. Specifically, it examined the level of concern for health and economic impacts to themselves and their associations with experience of several pandemic-related stressors. Findings and implications for practice in the COVID-19 pandemic and disasters in general are discussed.

This paper aims to call for change in disaster research through a metis-based approach that values practical skills and knowledge (versus technical knowledge) derived from responding to ongoing changes in the natural and human environment. This paper is based on metis from Miami-Dade County that is prone to an array of climate-related disasters. Metis is supplemented by a review of secondary sources (e.g. newspaper articles, government reports). There is a need to reconceptualize disaster phases in disaster research: preparedness, response, recovery and mitigation. For many members of marginalized communities of color, this paper depicts preparedness and mitigation as luxuries and response as a time of worry about financial obligations and survival after the disaster. It suggests that even communities that are not on a hurricane’s path could have post-disaster experiences. It also highlights ongoing risks to marginalized communities’ physical and mental well-being that are in addition to the mental health impacts of the disaster during the recovery phase. This paper’s originality is twofold: (1) underlining the importance of metis, a less-studied and understood concept in disaster risk reduction, prevention and management literature and (2) questioning disaster researchers’ technical knowledge with respect to each of the four disaster phases in light of metis.

The COVID-19 pandemic has continued to be devastating for incarcerated persons and their loved ones. The state of Texas has continued to lead the United States in deaths of incarcerated persons and prison staff due to COVID-19. In 2020, the Texas Prisons Community Advocates partnered with The Campaign to Fight Toxic Prisons to reach out to incarcerated persons to document how they were being impacted. Together they collected 428 surveys from incarcerated persons describing their direct experiences with pandemic prevention policies and exposure to the virus. Results from the surveys reveal significant issues relating to the response to the pandemic in prisons and highlight environmental justice concerns stemming from the treatment of incarcerated persons as expendable in the pandemic. Incarcerated persons described inadequate and disparate access to resources (quality healthcare, Personal Protective Equipment, sanitizing/hygiene supplies, nutrition, free movement, etc.) and detrimental mental and physical impacts from the enforcement of pandemic prevention policies in prisons (medical lockdowns, quarantine procedures, and movement restrictions). Results also demonstrate the complex vulnerability of incarcerated persons by highlighting differences in experiences by gender, race, age, health, ability, religious identity, and security status. These findings further illuminate the ongoing social violence of incarceration within the context of the enduring COVID-19 pandemic and suggest further engagement with proposals to decarcerate prisons and invest in strategies to address harm which do not compound harm.

Disasters bring considerable economic and social ramifications by disrupting public utility services, for example, power outages, phone service disconnections, and transportation interruptions. This study seeks to understand the performance and resilience of critical infrastructure systems using Hurricane Harvey (2017) as a case study. This work is centered on data collected from a household survey of 500 respondents residing in the Houston Metropolitan Statistical Area (MSA) during Hurricane Harvey’s landfall. This study investigated the number of households that experienced various types of utility services disruptions (electricity, water, waste disposal, phone/cellphone, internet, public transport, educational institutions, workforce/business office, financial institutions, hospital/doctor's office, pharmacy/medical stores, medical test centers, and grocery stores) as well as the duration of each type of disruption. Around 69% of the respondents reported having electricity disruption while half (49%) of the respondents had no water supply for up to six days. Two thirds of the surveyed households did not have internet access and 47% had their phone services disconnected. Finally, around 60% of the respondents could not commute and 53% of the respondents could not visit hospitals for medical emergencies. The household survey responses were then incorporated into the Dynamic Inoperability Input-Output Model (DIIM) to estimate the extent of inoperability and economic losses in multiple linked sectors. Understanding the resilience of each sector and the inherent interdependencies across the sectors can provide useful input to policymakers for disaster risk management, notably preparedness and recovery planning for future events.

There is a major gap in understanding how hazards and disasters impact researchers and research processes. Conceived at the onset of the COVID-19 pandemic, the “Research on Researchers‚” or RoR project focused on learning about the activities and experiences of disaster researchers across a variety of disciplines who are examining social dimensions of COVID-19. The project systematically studied and documented the impacts of the pandemic on a purposive sample of this group of researchers. The research questions included: “how is the current and ongoing situation influencing researchers’ thinking about disaster research?” and “how is the pandemic affecting researchers both professionally and personally?” Using a qualitative approach, the research team conducted telephone or Zoom interviews with 30 individuals during the summer of 2020, and follow-up interviews in early 2022. This session will cover findings of the study, ranging from the methods and theoretical frameworks used by researchers in their ongoing studies to how they adapted their strategies to conduct their work to the researchers’ goals and objectives. Moreover, it will shed light on various effects of the pandemic on work life balance including efforts to attend to professional obligations and personal responsibilities. The presenters will also address how the Black Lives Matter movement, following the killing of George Floyd, impacted ongoing studies with respect to affecting or explaining patterns in data; affecting research teams, students, organizations, or administrations; informing research agendas or processes; and causing reflection and thinking about future research.

The conceptualization of social equity in public administration and emergency management has been subjective to the user of the term. While the vagueness of the concept provides fertile grounds for intellectual debate, the failure to arrive at a single, formal definition leads to confusion and an inability to measure it as a programmatic or policy goal. For emergency management, how scholars define social equity within research is profoundly important for assessing and making recommendations related to governmental practices that have social equity as its guiding principle. To address this detrimental definitional situation, this research analyzes 15 years of social equity peer-reviewed articles in highly ranked emergency management journals. This manuscript concludes with a proposed working definition of social equity, recommendations to measure the concept, and a discussion of its implications for future research and practice.

The Marshall Fire destroyed 1,084 homes and damaged 149 more in the communities of Louisville, Superior, and unincorporated Boulder County, becoming the most destructive fire in Colorado’s history. For these and the growing number of communities facing a new set of risks due to climate change, key questions now emerge: will these events strengthen communities’ resolve to build climate resilience, or erode public support for such policies in favor of reducing short-term recovery costs? And do community members who live through rebuilding and recovery shift their opinions of and support for such resilience policies? This project focuses on policies that contribute to disaster and climate resilience of communities but that can make development more costly, particularly in the short-term. Prior work has shown that public support and the availability and requirements of external resources and programs shape these local government decisions, but it is unclear whether and to what extent this support changes after a shock such as a disaster, and how these policy decisions are affected by and affect individual preferences and decisions, including the decision to rebuild or relocate post-disaster. This project studies resilience policies after disaster events by collecting perishable data on resident recovery and local government decision-making after the Marshall Fire. This presentation will provide a first look at survey, interview, and document analysis of the processes, policies, and recovery from the Marshall Fire.

The Pacific Northwest is subject to devastating earthquakes but lacks a strong “earthquake culture.” Among the least prepared residents are 18-29-year-olds, who are generally overlooked in traditional messaging. Video games resonate with the media consumption habits of this age group and can put earthquake preparedness into an engaging context. The authors conducted an experiment with 125 young adults in Portland, Oregon to compare learning and self-reported self-efficacy, intent to act, and steps taken to prepare for earthquakes following a participant-determined period of up to 45 minutes of video game play versus web searching. This custom-made game included at least three solutions to each of four earthquake-related challenges. Those in the web search condition could browse at will or use three starter links to emergency management websites. Surveys were administered before and after the experiment task, and three months later. Game players engaged in their task significantly longer than web searchers (31 versus 19 minutes, p = 0.001) and found it significantly more enjoyable, challenging, and frustrating (p = 0.01, 0.001, and 0.03). They also reported higher immediate increases in self-efficacy around obtaining clean water and managing bodily waste (p = 0.05 and 0.001). Number of valid solutions recalled were equal for the two conditions, as were levels of trust in, and perceived reliability and applicability of, information learned. After three months, self-reported steps taken to prepare increased significantly for six out of eight specific actions among both game players and web searchers.

The lack of population representativeness has been a long-standing issue in disaster informatics research when incorporating social media data, establishing a known bias in using social media activity to assess areas affected by disasters. There are also confirmed links between health disparities and social systems which show that these disparities can be aggravated and intensified during a disaster. Recent studies reveal there is little known about the effects of existing data analytics approaches, and the fairness these techniques have in relation to vulnerable and underserved populations during disasters. Typical machine learning-based solutions can lead to negative impacts by exacerbating societal inequalities through biased training datasets, and raising privacy concerns. Related research has mainly focused on systems that address fairness assessment criteria, and concentrate on enhancing performance for a certain task or measure, but none account for the performance impact while including fairness and considering the public health issues that can worsen a community. In this study, a framework to account for sampling bias and health disparities in communities is introduced. A case study was performed for disaster event detection using social media (i.e., Twitter) data to evaluate the framework. The results of this study provide an assessment for the trade-off between fairness and performance, and establish a foundation for a machine learning event detection approach that accounts for health equity in disaster informatics. This contribution connects public health concerns and engineered systems to advance emergency management decision-making processes.

Communities in hazard-prone regions are often called upon to evacuate, shelter-in-place, or perform actions to prevent harm to themselves or their families. It is vital for these communities to have an effective and functioning warning system to receive critical information in a timely manner to prevent negative public health outcomes. There is growing evidence that not all communication strategies produce equal results and that a locally involved strategy that takes into account communal needs and social spheres can fare better than traditional methods. This study pulled from two research projects that collected rapid and critical data assessing individual willingness to abide by warnings. Two cross-sectional studies collected data from (1) the City of Deer Park, Texas, following a petrochemical fire at the Intercontinental Terminals Company that seeped carcinogenic compounds into the neighborhood and initiated several shelter-in-place commands along the Houston Ship Channel, and (2) data from seven Oregon counties that initiated evacuation warnings due to wildfires. This study discusses the findings, and argues that including residents as participants in educational and outreach efforts prior to hazard events is a critical component to ensure warnings are followed and trusted by communities.

Exposure to wildfire smoke is a global public health issue with serious short- and long-term consequences. Evidence also suggests that smoke exposure is associated with an increased COVID-19 case fatality rate, which is important as mask wearing is an adaptation behavior for both threats. Complicating these overlapping threats are the ways that adaptation decisions are socially situated, through the normative signals and support from close others that may influence intentions and behavior. Through an analysis of a nationally representative survey of U.S. adults in 2020, researchers investigated how social processes are related to wildfire smoke and COVID-19 adaptation behavior. Questions asked include, 1) to what extent are greater adaptation intentions associated with greater threat and efficacy perceptions, and perceived social norms?, and 2) to what extent is the relationship between self-efficacy and adaptation intentions moderated by social support? Findings suggest that greater adaptation intentions are associated with greater threat and efficacy perceptions in wildfire smoke and COVID-19 contexts, providing support for literature drawing on the Extended Parallel Process Model. Also, greater reported social norms are associated with greater adaptation intentions for both threats. Contrary to initial hypotheses, findings suggest that increased levels of social support have a weak, but significantly negative moderating effect on the relationship between self-efficacy and adaptation intentions. The findings suggest that by understanding the social processes involved in adaptation behavior, researchers may better predict adaptation intentions and inform the design of effective public health communications and policy interventions promoting adaptation to hazards.

There are more than six decades of research on how people respond to risk information. However, less is known about information seeking in the context of competing official and non-official sources of risk information available through social media. Moreover, there is a dearth of disaster research–moreso research on risk communication–in Puerto Rico. For the last two years, the authors of this study have been conducting research to understand how–in the context of navigating multiple disaster cycles simultaneously–emergency management-related personnel and residents make decisions about what actions they should take to protect themselves and others against the possibility of a significant earthquake in the future. In this study, the authors reviewed theories related to information-seeking behavior. The authors also used data collected through in-depth interviews with key informants (n=51) and a semi-structured survey (n=446) to explore how emergency responders and residents in Puerto Rico access, interpret and use available information to inform protective actions decisions. This research is vital to effectively promote situational awareness and preparedness for future earthquakes in Puerto Rico.

Schools play a significant role in disaster education to children. This study investigates the research works about school-based education programs in order to discover challenges and best practices. To do so, the author of this study conducted a systematic review of English language papers published in peer-review journals. The search identified 2577 publications and 61 articles meeting review selection criteria. Reviewed studies indicated that disaster education in schools is effective yet insufficient in many countries. Lack of equipment, financial resources, policy gaps, and teachers’ knowledge are common problems in disaster education programs. Main outcomes of this systematic review show methods used for health emergency preparedness of children of different ages, gender differences in school-based disaster preparedness, and differences in their life-saving skills in disasters. This study shows that some disaster education programs reported in the papers reviewed were low-quality, which may lead to insufficient preparedness of children in disasters and consequently may put their health at risk.

Disasters often exacerbate the pre-existing socio-economic inequalities in society. Previous research shows that minority and low-income populations are disproportionately impacted by disasters with respect to housing damages and economic losses as well as slower recovery trajectories. These disparities in impacts and recovery are further heightened in low-attention disasters due to the lack of assistance funds and temporary shelters, whereby vulnerable households rely on their own limited resources and networks to work on repairs at a slower pace and lower quality. This paper presents an in-depth qualitative analysis of semi-structured interviews with impacted households, media articles, local organizations documents, and social media posts about housing recovery in Marshalltown, IA after the July 2018 EF-3 Tornado. Findings showed that lack of attention from media and public agencies and denial of government assistance to individual households in need prolongs severe disaster impacts in a community when the majority of affected households are vulnerable due to compounding factors. For example, households at the intersection of minority immigrant status and low-income avoid applying for local governmental assistance due to the perceived fear associated with immigration status and opt to stay in a damaged cold house during winter or work on repairs without safety skills endangering their health. For many renters and those ineligible for governmental assistance, one-time recovery aid from non-profits and church communities provide short-term relief but does not account for the recurring costs of a damaged unsafe home during the cold winter when a landlord delays or ignores repairing critical damages.

Climate change has posed serious risks to coastal cities around the world. Effective urban disaster management calls for coordination between the local government and residents. The authors of this study describe a comprehensive framework to study urban disaster resilience under climate change with New Orleans of Louisiana in the United States as the study area. Municipal hazard mitigation must be sufficient to mitigate these hazards. Resident’s risk perceptions are a vital component of social vulnerability and can shape public decisions to increase disaster resiliency. Because climate change is expected to intensify, it becomes important to ensure that resident’s risk perceptions are considered when developing municipal plans to maximize regional resiliency. This research identifies a gap in the hazard mitigation process that can be addressed to better prepare communities to manage coastal hazards. To achieve this, an online survey was distributed in the New Orleans metropolitan area to determine residents’ risk perceptions and expectations of local government action. Policy analysis was conducted to identify the priorities held by municipal planners in these issues. Although there is no gap in the perception of risk and municipal mitigation of current coastal hazards, a substantial gap was identified between the municipal approach to climate change mitigation and the concern and expectation of actions residents held regarding the future effects of climate change. The approach to climate change should be reconsidered on a municipal level and new small-scale personal resiliency incentives should be promoted to maximize resilience toward coastal hazards in the future.

In this study, the authors developed the Geo City River Platform (GeoCRP), which regularly collects data on the web and provides the analyzed information by performing flood analysis based on the collected information in order to effectively perform smart city river management. GeoCRP was analyzed and tested at the Eco Delta City (EDC) area in Busan, Korea. This platform applied flood analysis in EDC using a watershed runoff analysis model, a river flow analysis model and an urban runoff analysis model. GeoCRP can obtain more reliable results by taking a step-by-step approach to urban overflow that may occur in smart cities through the applied model. In addition, since all analysis processes such as data collection, input data generation, and result data storage are automatically performed on the web, analysis can be performed and results can be viewed if an environment that can access the web is established without special equipment or tools. The displayed analysis result is provided visually so that the user can intuitively confirm the information, so it is easy to understand the analysis results. Through this, smart city managers can effectively manage rivers, and it is expected that educational institutions will be able to use it as educational material on urban runoff.

Exposure to hazards, as well as household and community level resources to cope with and recover from a given disaster are heavily mitigated by structures of oppression and exclusion. Relief programs, ironically, may perpetuate these same structures of oppression and domination, furthering disaster disparities. While this mismatch has been named by environmental and climate justice advocates, it has not been formally quantified, nor easily replicable methods of evaluating it have been adopted. Using fields from publicly available Federal Emergency Management Agency (FEMA) aid application data in combination with census data, outlier and cluster analysis as well as hotspot analysis reveal and confirm an extensive mismatch between homeowner relief funds and vulnerability at the census tract level. This geospatial exploration both illuminated significant room for improvement in FEMA aid distribution and offered a replicable evaluation technique for potential aid-vulnerability mismatches following other U.S. disaster events.

This session will overview three applied hazard mitigation research projects undertaken at North Carolina State University that can inform evolving community resettlement strategies in an era of climate change. First, the research team will discuss the assessment of innovative housing acquisition strategies across the United States. Semi-structured interviews were conducted in communities spanning differing capacities, sizes, and locations across the United States. Questions emphasized internal determinants (local characteristics affecting innovation) and policy diffusion (how well innovative ideas are shared with others) measures. The session will then focus on the creation of an open space management guide being developed for buyout properties funded by the Federal Emergency Management Agency. The guide, developed in partnership with a national panel of experts, blends policy and design approaches to meet the needs of those engaged in the management of open space. The third line of work involves a project tied to the development of hazard mitigation and climate change adaptation design typologies to protect historic properties and culturally significant landscapes. This research will produce a digital guidebook and a set of training modules containing design strategies, tips, resources, best practices, and case studies for use by local governments. In each project, findings will be framed in a way to help unpack policy questions (and potential answers) surrounding the still emerging community resettlement debate.

Efforts to understand and respond to threats posed by crises such as earthquakes, floods, and tropical storms are often informed by complex disaster risk models produced by experts from physical science and engineering. These models provide powerful understanding of the potential impacts of disasters and allow identification of appropriate interventions. However, quantitative disaster risk analyses (QDRAs) that use these models provide limited, if any, understanding of differentiated disaster impacts on different demographics, the uncertainty associated with current risk, and future risk predictions. The authors of this paper explore how, and how often, QDRAs that have been funded by the World Bank–a central producer of risk information globally–have taken into account questions of equity, dynamic risk, and uncertainty reporting. The authors performed a close read of 54 QDRAs and analyzed their contents through a framework composed of various relevant metrics. Findings indicate low rates of uncertainty reporting and indirect economic loss representation in the assessed QDRAs, as well as very few instances of disaster impact disaggregation by the parameters of race, class, gender, disability, and age. The study interprets these findings in context of the current state of practice in this area, discusses their implications, and outlines how the results of this work can guide the design of future disaster risk assessments.

In this study, the authors of this study considered a typical logistics planning problem for prepositioning relief items in preparation for an impending hurricane event. In this problem, the goal of the decision-maker is to preposition relief items over multiple periods at a set of supply points to minimize the total cost of unmet demand and logistics operational cost. The total cost consists of the logistics costs and a penalty for failing to satisfy the demand (if any shortage is present). This study assumes the demand for relief items can be derived from two of the hurricane characteristics: its intensity and landfall location. However, since these characteristics and their evolution over time cannot be known in advance, the authors assume that this stochastic process can be modeled as a Markov chain. This study considered this problem in two different settings, depending on whether the time of landfall is assumed to be deterministic or random. A fully adaptive multi-stage stochastic programming (MSP) model for each case is proposed. The MSP model allows the decision-maker to adapt their logistics operational decisions sequentially, over multiple stages, as the characteristics of the hurricane become more and more clear. A case study is conducted, which compares the performance of each MSP model to the clairvoyance solution and alternative decision policies. The numerical results provide several key insights into the value of MSP in adaptive disaster relief logistics planning.

Federal Emergency Management Agency (FEMA) grant programs extend financial aid to qualifying applicants in both the pre-disaster mitigation and post-disaster response and recovery stages as communities attempt to advance disaster resilience. This research serves two purposes: (1) to understand who has access to FEMA’s Hazard Mitigation Assistance (HMA), Public Assistance (PA), and Individual Assistance (IA) programs, and any underlying inequities associated with that access, and (2) to develop an optimization model to aid local decision makers in determining what type(s) of mitigation projects will be most beneficial in advancing community resilience given the specific challenges of their community. To evaluate access and effectiveness, we examined how poverty rate, race, housing, and Social Vulnerability Index (SoVI) affect a community’s receipt of grant funding and we analyzed how different types of mitigation projects, including both structural and non-structural approaches, affected post-disaster funding distributions in the study area after Hurricanes Matthew and Irma, respectively. Considering the storm paths of Matthew and Irma, all counties in Florida, Georgia, and North Carolina were included in the study area for county-level analysis. A novel optimization algorithm was designed to allocate available budget to different types of mitigation that can best help a community mitigate long-term risk considering its hazard exposure and social vulnerability. This work points to non-structural mitigation approaches, such as land acquisition, as being most effective at the local level to move development away from high-risk areas, especially in socially vulnerable neighborhoods that suffer to a greater degree from the adverse effects of disasters.

Longitudinal investigations are critical for studying resilience so that impact, decision, and recovery data and progress can be tracked over time. Longitudinal studies thus require significant resources. When disruptions occur, the research team must determine whether to continue the longitudinal study, and if so, how to account for the disruption in assessing the study’s original goal. This presentation will describe how a cross-disciplinary team continues to navigate field study decisions for adapting their longitudinal study of Lumberton, NC during the COVID-19 pandemic, including how to control for the impact of compounding events, the ethical and safety considerations for if/when/how to continue, and employing novel methods effectively to document change over time when conditions are radically altered.

The longitudinal study began in November 2016, approximately one month after Hurricane Matthew flooded the community. Lumberton was still recovering from Matthew and 2018 Hurricane Florence, and waiting on significant federal recovery resources when the pandemic started. Findings from May 2019 revealed increased mitigation and preparedness employed by individuals, businesses, and the City for Florence compared to Matthew. Thus, there was still much to learn from Lumberton’s recovery, despite the pandemic.

Previously, the team relied on annual deployments to conduct damage inspections, door-to-door household and business surveys, and stakeholder interviews. Given the pandemic, the team revised its approach to documenting resilience with continued questions on how the pandemic is influencing long-term recovery from the hurricanes. Strategies learned can help with natural hazards and pandemics regarding which actions and policies have been useful across events.

The disaster literature references a vast host of variables to explain which communities and individuals are disproportionately likely to experience the adverse effects of a disaster. These variables range from family composition, economic opportunities, housing status, and ethnicity to strength of community connections and prevalence of intergenerational trauma. While great success has been made using these variables to understand disaster experiences and how they shape disaster recovery, understanding how these factors differ based on community type (i.e., urban, rural) is limited. This presentation uses Statistics Canada data for British Columbia (weighted N= 3,846,331) to answer the question: Are factors widely associated with community resilience expressed differently in different types of communities (i.e., urban, rural)? Seven factors will be explored within the presentation: 1) prior hazard experiences, 2) perceived exposure to hazards, 3) preparedness, 4) beliefs of personal efficacy, 5) community connections, 6) financial flexibility, and 7) confidence in support agencies. This research aims to test if, and how, these factors are expressed in different community types; information which will be consequential in how disaster preparedness and management are undertaken and resources allocated across the province.

Risk communication in times of disasters is complex, involving rapid and diverse communication in social networks (i.e. public and/or private agencies and local residents) as well as 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, let alone compounding risk events. This study analyzed large-scale datasets of crisis mobility and activity related social interactions and concerns available through social media (Twitter) for Oklahoma communities that experienced major tornadoes (Oct. 2021) and an ice storm (Oct. 2020). First, the study used natural language processing and advanced classification techniques to quantify crisis narratives (i.e. tweets). Next, geo-tagged tweets were mapped into co-located road networks using traditional GIS techniques. Finally, it generated insights using network science theories and quantified social narratives to interpret different elements of road networks (e.g. local roads, freeways, traffic lights etc.) for the Oklahoma communities that were impacted by those two events during the pandemic.

This study asks, how do Cascadia Subduction Zone (CSZ) researchers study and make knowledge claims about the CSZ in a context of data scarcity? Data has been difficult to obtain about the CSZ and the potential earthquakes, tsunamis and landslides that it could generate. Through semi-structured interviews with 31 CSZ researchers and inductive qualitative analysis of interview transcripts, we suggest that over the last 35 years researchers from different fields have shared many different types of evidence to understand this multi-hazard phenomena. As CSZ researchers from different disciplines and diverse methodological approaches share a variety of evidence, different interpretations of evidence have led to controversies about CSZ dynamics. Incommensurate interpretations of evidence have been addressed and sometimes settled through a number of synthetic research practices. Collaborations amongst researchers and experts in city planning or geotechnical and structural engineering further has helped consolidate consensus. These shifting relations of sharing, friction and synthesis are called “evidence-oriented relations” in this study and present a heuristic for analyzing sharing, frictions, and consensus-building work that occurred around CSZ evidence.

Recent wildfires in California and Oregon caused significant damage to water distribution systems which resulted in water that was contaminated with benzene and other volatile organic compounds (VOCs). To understand the risk of exposure to contaminants, the authors of this study developed a machine-learning model based on neural networks that accurately (85%) predicted water contamination after wildfire using publicly available spatial predictors such as topography, land cover, fire weather, type and density of infrastructure, and physical soil properties.

The authors then developed the Wildfire Vulnerability Explorer, a user-friendly interface that combines the water contamination predictive model with a place based social vulnerability assessment utilizing census, parcel, and land use data. Using fuzzy-logic Environmental Evaluation Modeling System (EEMS), the Wildfire Vulnerability Explorer provides an interactive visualization of areas most vulnerable to post-firewater contamination. The explorer allows the user to visualize how social and environmental factors interact to more accurately reflect on-the-ground vulnerability conditions. Stakeholders from both Santa Rosa and Paradise, California have provided key suggestions and validation of the explorer for their respective communities. Through this explorer, wildfire-prone communities will not only be able to better understand water contamination risk, but also identify feasible adaptation strategies that best meet their community needs.

Climate change and the urban heat island (UHI) effect is increasing the number of dangerously hot days and the need for communities to equitably plan for heat resilience. Planners face many barriers including a lack of research-based heat guidance for planning processes, underdeveloped regulatory structures, and siloed research, decision-making, and community plans. Planning for heat resilience requires an integrated planning approach that coordinates strategies across plans and uses the best available heat risk information to prioritize mitigation strategies for the most vulnerable communities. Supported by National Oceanic and Atmospheric Administration and in partnership with the American Planning Association, this project developed a methodology for assessing heat resilience planning. We adapted the Plan Integration for Resilience Scorecard approach, originally developed for flood risk, to the unique challenges of extreme heat. The resulting Plan Integration for Resilience Scorecard for Heat (PIRSH) methodology was applied in five geographically diverse United States communities that participated in the National Integrated Heat Health Information System and Climate Adaptation Planning and Analytics UHI campaign (Boston, Massachusetts; Baltimore, Maryland; Fort Lauderdale, Florida; Seattle, Washington; and Houston, Texas). The authors of this study developed a typology of heat mitigation strategies through an expert elicitation and literature review process. The study evaluated how policies would affect the UHI and mapped them to city districts to evaluate their spatial distribution and net effect on heat mitigation. The resulting PIRSH scorecard was then combined with heat hazard vulnerability data to assess policy alignment with heat risks and identified opportunities for improvement.

The objective of this study is to illustrate how to create and maintain a healthy flow of carefully curated communication between a research team and a community of interest using the principles of cultural brokering. The authors began with a brief critical review of the existing literature on cultural brokering between immigrant and academic communities. The authors followed this review with a case-study of cultural brokering between a research team at Tulane University and the Vietnamese American immigrant community in New Orleans. The authors drew upon specific examples from an ongoing collaboration to illustrate types of cultural conflict that are common to academic and community collaborations, and discuss how different types and features of cultural brokering have helped to resolve these conflicts and keep the project(s) moving forward. The authors summarize a set of traits, potential roles/contributions, and training and operating strategies for a cultural broker.

In the aftermath of a disaster, a community’s recovery can be gauged by its housing recovery. The literature has shown that the rate and the completeness of housing recovery have consequences for economic, business, household, and emotional recovery after disasters. Housing restoration and recovery pose a large financial burden to households, as well as access to financial resources–in the form of insurance payouts, federal aid, loans, and assistance from non-governmental organizations or family friends–determine whose home and which neighborhoods recover quicker. This study attempts to answer two key questions: what role do financial resources play in housing recovery, and how does the availability of financial resources determine the rate of housing recovery? The authors of this study used the data from field studies conducted in the aftermath of Hurricane Matthew in Lumberton, North Carolina. It is estimated that Hurricane Matthew, and the subsequent flood, damaged 88,000 homes across North Carolina, and displaced 335 households in Lumberton alone. This paper presents the results of an in-depth quantitative analysis of the impact of financial resources and assistance on housing recovery. Findings show interesting variations in the effect of different financial resources–insurance, federal, and non-federal assistance–on housing repair and reoccupancy times. The results also show that resident socioeconomic status, and the level of house damage determine housing recovery progress. Based on the results, the authors offer policy suggestions to improve post-disaster housing recovery in a more timely and equitable manner.

The mission of the NSF-supported Natural Hazard and Disaster Reconnaissance (RAPID) Facility, based at the University of Washington, is to “enable transformative research by providing investigators with the instrumentation, software, and support needed to collect, process, and analyze perishable data from natural hazard events and from disasters.” Aligned with this mission, we aim to increase RAPID’s social science user base by understanding and addressing their research and support needs. We are planning multiple forums and mechanisms for feedback from the social science community. This presentation will begin with an overview of the RAPID Facility and its capabilities followed by examples of how social scientists on interdisciplinary teams have utilized instrumentation and data from the RAPID Facility, including use of its Streetview-like imagery equipment to explore COVID-19 pandemic impacts and recovery and use of the RApp application for survey and interview data collection to understand tornado sheltering behavior and shelter access. We will then introduce upcoming opportunities to provide input to the Facility about ways it can support individuals and teams planning or conducting social sciences and interdisciplinary research and reconnaissance, including through training, resources, and technical assistance.

Prior research has demonstrated the need to tailor risk communication strategies, and community engagement approaches to reach at-risk populations, but little is known as to if or how at-risk communities are incorporated into emergency management and risk communication planning. This study aims to enhance understanding of how individuals experiencing homelessness are considered in emergency and risk communication planning, including how extreme weather risk communication is both delivered to, and received by, those experiencing homelessness. To address these aims, the authors conducted in-depth semi-structured interviews with 29 individuals representing homeless-serving organizations (n=10), emergency managers and city/county-level employees involved in emergency planning for unhoused individuals (n=6), and those with lived experience with homelessness (n=13). Study findings identify: (1) communication strategies used by homeless-serving organizations and local/county-level agencies involved in emergency planning for unhoused communities to disseminate extreme weather risk, (2) challenges associated with disseminating and receiving extreme weather risk information, such as difficulty communicating with individuals who have limited or no access to cell phones, and providing actionable information and viable options for protective actions to people experiencing homelessness, and (3) recommendations for improving risk communication and emergency planning for individuals experiencing homelessness. Results from this study identify an immediate need for plans to consider how unhoused individuals can reasonably act upon risk information, in terms of the modes of communication and content of communication. Emergency plans should be enhanced and made more equitable by inviting and engaging people with lived experience with homelessness in planning.

Nonprofit organizations are important sources of aid and assistance in the aftermath of disasters contributing to response and recovery efforts. Despite immense support these organizations provide across the US, issues of inequity in resource distribution have been documented. Of interest in this work are inequities faced by the Latinx communities, who face disproportionate impacts to disasters and unique barriers to recovery. Using an environmental justice framework (comprised of recognition, procedural, and distributional dimensions), organizational processes that influence equity for Latinx communities impacted by disasters are explored. Specifically, the following three research questions: To what degree do local leaders involved in disaster recovery recognize the Latinx population? How does the process of recognition manifest among these leaders? And how is recognition by these leaders related to procedural and distributional equity? In this study, 20 semi-structured interviews were conducted with local nonprofit and government organizations involved in disaster recovery in Wilmington, NC after Hurricane Florence (2018) to answer these questions. Results reveal four key findings (1) leaders recognize Latinx communities to varying degrees (2) leaders facilitate the process of recognition for themselves and others in diverse ways (3) limited organizational engagement of socially vulnerable groups hinders recognition (4) there is a positive correlation between recognition, and procedural and distributional equity. Because the COVID-19 pandemic started midway through project implementation, the study also reflects on how the pandemic influenced our research process and the data that was generated.

The use of tools utilizing machine learning, and in particular text mining and natural language processing (NLP), is rapidly growing in areas including manufacturing and engineering, weather forecasting, communication, community resilience, and climate adaptation. However, there are yet unexplored applications of these methods for the semi-automated analysis of large text datasets such as reports, planning documents, and other media related to hazard and resilience topics. Here, the study described the interdisciplinary development of a tool for automated text analysis of long-form technical documents (in the form of community resilience and adaptation guidance documents and plans) and compared it to content analysis, a more traditional text analysis process used in social scientific research. This study also provides tips to others interested in analyzing and exploring their own text datasets using NLP-type methods from our experiences working at the intersection of computer science and applied social scientific research. For example, emphasizing the necessity of qualitative aspects of the process to computer science and IT specialist team members, and the technical aspects and limitations of the process to social scientific and research team members. The hope of this study is that this overview will facilitate future collaborations in this extremely fruitful and exciting new area of research.

Development in coastal areas and floodplains, in combination with increasing precipitation and rising sea levels, have intensified flood vulnerability across many communities in the United States. Flood hazard mitigation policy instruments play an important role in reducing the frequency and severity of flood impact. Recently, studies have researched the property buyout program from different perspectives. This program specifically focuses on removing properties away from the floodplains with high flood risk. However, limited literature focused on the property buyout program from a government perspective. Consequently, there is a need to examine factors motivating local governments, as the primary actors for flood mitigation, to actively pursue and adopt property buyout as a way to mitigate flood risk and improve community resilience. Data was collected through a survey of local floodplain managers in Virginia counties. Based on the 59 responses received and secondary data sources, logistic regression models were employed to explore the relationships between the uptake of the property buyout policy and flooding problems, social vulnerability, institutional capacity (e.g., individual, organizational, and system capacities), policy diffusion, and policy environment. Semi-structured interviews were conducted with flood mitigation experts to identify qualitative insight pertaining to the buyout program. The findings demonstrated that individual capacity, including floodplain managers’ experience, awareness of flood risk and buyout benefits, and policy innovative ability, positively influenced the adoption of buyouts at the local level. Policy recommendations for the Federal Emergency Management Agency’s property buyout program were suggested in light of the findings.

Professionals from different disciplinary backgrounds tend to talk over each other within the field of disaster studies by using the same labels, such as disaster, vulnerability, and exposure, to refer to different conceptual domains. This is not only because of different disciplinary practices of defining concepts but also to a large extent due to the way of academic inquiry in which a label is identified first before meanings and conceptual domains associated with the label are being grown and transformed. An example of such an inquiry is the classical debate around the question of “what is a disaster.” However, empirical research, based on solid evidence and data, requires ontological development that follows the opposite direction of scientific inquiry in which a conceptual domain with specific meanings is first considered necessary before an appropriate label is attached to the conceptual domain. This presentation highlights this paradigmatic difference between the label-first and meaning-first approaches to scholarly inquiries in disaster studies. It then proposes the meaning-first approach with an applicational example of identification of three qualitatively different types of phenomena that are fundamental in disaster studies. The identified three fundamental types of phenomena can be labeled as the disaster, risk, and crisis types, respectively. The presentation further calls for scholarly efforts towards the development of an ontology for empirical disaster studies with the meaning-first, instead of label-first, approach.

Hurricane Ida reached the United States on August 29th, 2021, causing 95 deaths and at least 50 billion in damages. The remnant of the hurricane caused an unusually intense and deadly impact on New Jersey because the soil was saturated after Hurricane Fred and Henri. Manville Township was one of the areas that was severely affected by flooding induced by Hurricane Ida in New Jersey. This study aims to figure out the damage caused by Hurricane Ida at the Manville Township and estimate the vulnerability of communities. A high-resolution two-dimensional hydrologic model was employed to simulate the flood conditions. It is constructed based on Hydrologic Engineering Center's River Analysis System (HEC-RAS) coupled withHydrologic Modeling System (HEC-HMS), forced with the best available forcing data including measured streamflow, precipitation, and storm tide. The computational mesh is created based on high-quality point cloud datasets. The observed High Water Marks were used for model validation. This study also employed a physics-based human instability model to quantify the residents’ vulnerability in flash flooding. It uses the simulated flood characteristics (flood depth and flood velocity) to calculate the vulnerability of buildings and residents to better understand the damages caused by flash flooding induced by Hurricane Ida. The real building first-floor elevation extracted from point cloud datasets is used for building vulnerability assessment. This study highlights the contributions of high-quality point cloud datasets and the building-scale flood model on flood vulnerability assessment in urban areas.

Businesses are a crucial component of community resilience and provide the goods, services, and wages needed for recovery and economic stability. Surveys remain an important tool in assessing the resilience and recovery of businesses since they are flexible in their design, deployment, and sampling strategy. This can provide nuanced information on recovery dynamics through time, as well as potential differences across ownership, geographical, and operational characteristics of the business. Despite their importance, conducting surveys can present a variety of practical challenges. This has only been exacerbated during the COVID-19 pandemic, where the demand for information on businesses was high across academia and practice, yet the methods and modes for collecting data were limited due to safety, health, and ethical concerns. This research addressed these challenges by combining the experiences of previous business survey efforts to collect best practices. It drew from cross-sectional and longitudinal research conducted across several countries after hurricanes, fires, earthquakes, floods, and COVID-19. These surveys were conducted by a diverse group of organizations with a range of research questions and objectives. The lessons learned from these surveys can provide guidance for researchers or practitioners who wish to conduct their own business resilience surveys. To that end, the study includes four broad takeaways: recognition that response rates will be low, consideration of disaster dynamics in the research design, the need to address research bias, and acknowledgement of the unique ethical considerations of disaster resilience surveys in the business and economic context.

How have Black and White communities differed or resembled each other in their response to the COVID-19 pandemic, through the lens of disaster studies? With support from the NSF, the research team collected 438 in-depth ethnographic interviews from March 2020 through December 2021 with respondents from New Orleans and Baton Rouge, of whom 104 were Black and 279 White. Students at LSU to conducted remote interviews with people they knew, a method that enabled us both to obtain in-depth interviews while observing pandemic safety measures, and to leverage quota sampling methods traditionally used in European polling. The diversity of LSU students produced a diverse non-probability sample that reflects the ethnic, age, and gender diversity of the region. The semi-structured interview protocol covered themes of safety, stress, family life, vulnerability, interpersonal conflict, mutual assistance, partisanship, and social justice. The researchers transcribed all interviews and after making a first-pass examination of themes, the team assembled a sub-team of a half-dozen Black students who met weekly with us to identify themes significant to the Black community. Several important themes have begun to emerge. Among Black interviewees, themes include a strong sense within the community of mutual support, shared responsibility, caring for others, and support for social justice, but also feelings of vulnerability and distrust based on a long history of discrimination and injustice. The researchers set the findings against the backdrop of large-scale public opinion polling and official statistics, to probe for deeper meaning in our qualitative data.

Flash floods (FFs) have consistently resulted in more fatalities, injuries, and damages than any other flood type in the United States (US). FFs usually have small spatial extents and high-water flow velocities, making them a localized phenomenon. Furthermore, many factors such as land use pattern, terrain characteristics, meteorological conditions, and infrastructure quality may contribute to FFs. However, due to the localized nature of FFs, factors contributing to these flood types may differ from one region to another. Therefore, this localized behavior necessitates assessing FF susceptibility dynamically, allowing us to estimate the FF susceptibility on a region by region basis. This work centers on a Geographic Information System (GIS) tool developed to estimate the FF susceptibility of census tracts in the US. The GIS tool accepts a group of census tracts (area of interest) as input, determines the area’s physical features (e.g., lithology, ground slope, among others) that significantly influence FF likelihood, and then returns susceptibility scores for each census tract in the area. The features that influence FF are determined using a logistic regression model (dynamically generated for each area of interest) and statistical analyses. This tool offers an opportunity for targeted intervention enabling emergency officials and floodplain managers to assess FF risk at the census tract scale and develop preparedness and mitigation plans.

Power companies are enacting public safety power shutoffs (PSPS), a type of indirect climate adaptation policy, to reduce the risk of wildfires during periods of favorable wildfire conditions in places like California. While PSPS may reduce the risk of wildfires, it may also impact health and normal daily activities. Little is known about how the public views PSPS, and how personal wildfire experience and existing vulnerabilities may relate to those views. The aims of this study are to understand how personal perceived wildfire-related exposure, personal negative wildfire outcomes, prior health and current health, risk appraisal, hazard worry, and PSPS policy worry are interrelated and associated with PSPS policy support. Thus, a representative sample of 1,108 Californians was surveyed in December 2020 to address these aims. The survey revealed moderate levels of support for PSPS among Californians. More personal wildfire experiences in terms of health, hazard exposures, and hazard outcomes were positively related to wildfire risk appraisal and hazard worries, which in turn was associated with more support for PSPS. However, worry about possible negative impacts from PSPS was negatively related to support for PSPS. Improved communications may provide targeted information to households that are the most exposed or vulnerable to wildfires and their impacts, as well as on ways exposure can be reduced through household or policy measures. Climate adaptation policies informed by understanding the climate hazard and climate solutions risks the public faces may reduce unintended consequences, while also allowing for more rapid adaptation to climate change.

Direct experience with climate change-related extremes may be associated with attributions to climate change, as well as with pro-environmental attitudes and behavioral intentions. Given the United States’ (U.S.) recent experience with devastating wildfires and widespread coronavirus disease 2019 (COVID-19), the authors of this study conducted two national surveys of U.S. adults examining how subjective attribution is associated with personal experience with extremes, and concern with and subjective knowledge about climate change. The authors also examined whether these factors are associated with pro-environmental attitudes and behavioral intentions and whether these relationships are amplified or attenuated by subjective attribution. The authors found that personal experience, concern, and subjective knowledge are associated with the belief that wildfires and COVID-19 are attributable to climate change. These are all associated with pro-environmental attitudes (carbon tax policy support) and behavioral intentions (electric vehicle car purchase). Subjective attribution of COVID-19 amplifies the positive relationship between personal experience with COVID-19 and electric vehicle car purchase intentions (study one and two), and subjective attribution of wildfire or COVID-19 amplifies the positive relationship between subjective knowledge and carbon tax policy support (study two). These results provided insight into cognitive processes connecting climate-related extremes personal experience, subjective knowledge, concern, and subjective attribution with pro-environmental attitudes and intentions. Understanding whether subjective attribution, especially in the context of COVID-19, amplifies or attenuates important precursors to pro-environmental behavior deepens knowledge of the psychological processes aiding sustainability and potential pathways for more effective science communications and policymaking.

During the first wave of COVID-19 across Canada and the United States, on-campus evictions disproportionately affected out-of-province/state and international university students in both countries. From a gender lens, this cross-national research partnership examined disaster-specific leadership of female university students, who were generally treated as passive victims during emergency eviction and evacuation. This research recruited 20 out-of-province/state and international students (10 from each country respectively) for in-depth virtual interviews. This qualitative study discovered female university students’ leadership behaviors during three eviction stages: (1) before eviction - positive mental health leadership: their self-maintenance and self-preparedness strategies enabled them to support their peers’ social, emotional, and psychological well-being; (2) during eviction - facilitation leadership: their attitude and leadership behavior encouraged them to facilitate their peers’ eviction process and address their different challenges; and (3) after eviction - community outreach leadership: they united their peers through virtual connections and conducted virtual community outreach activities to raise the general public’s awareness regarding the diverse challenges of vulnerable and marginalized community members. The female university students' leadership that emerged during the eviction process became complementary to and even augmented the universities’ official efforts and beyond. Their leadership features evidence-based strategies that contribute to the existing disaster-specific literature on gender and leadership by demonstrating that female youth are empowered actors rather than merely passive victims. Empowering vulnerable groups’ leadership illuminates an inclusive research and practice approach to leverage diverse stakeholders’ expertise to contribute to the community-based disaster efforts in-person and virtually.

Wildfires are increasing in frequency and severity, especially across the western United States. This persistent disaster is especially pronounced in California, which has experienced eighteen of its twenty largest fires in the twenty-first century, causing massive destruction and economic damage. Employing a mixed-effect multilevel regression analysis, this study utilized data covering 2015-2020 from the Federal Emergency Management Agency (FEMA) and the American Community Survey to examine how social vulnerabilities influence the level of recovery funds that wildfire-affected households received from FEMA’s Individual Assistance Housing Program. Indicators, including household size, presence of young children in the home, and being of low-income status show positive relationships with receiving funds. Elderly applicants are, however, less successful. This study made important contributions to social disaster research, considering recovery programs related to wildfires, integrating measures of social vulnerability, and illuminating distinct relationships between vulnerable populations and federal disaster relief in the United States.

Wildfires continue to create severe challenges for emergency services and communities in the wildland urban interface (WUI). To reduce wildfire risk and enhance the safety of WUI communities, improving the understanding of wildfire evacuation decision-making is a pressing need. This study proposed a new methodology to analyze wildfire evacuations by leveraging a large-scale global positioning system (GPS) dataset. This methodology included a proxy-home-location inference algorithm, an evacuation-decision inference algorithm, and a destination-choice inference algorithm to systematically assign wildfire evacuees to different groups: self-evacuee, shadow evacuee, evacuee under warning, and ordered evacuee. Finally, the authors analyzed evacuees’ corresponding destination choices to a higher level of detail than past research. The authors applied the methodology to the 2019 Kincade Fire in Sonoma County, California and found that, among all groups of evacuees, self-evacuees and shadow evacuees accounted for more than half of the evacuees during the Kincade Fire. Furthermore, more than 50% of the evacuees chose destinations within 30 km (18.6 miles) of their proxy home locations, and residential areas were the most common destination type. The findings of this study can be used by emergency managers and transportation planners to better prepare WUI communities for future wildfire events.

Understanding how people make decisions about protective actions during earthquakes is important for improving emergency preparedness and response. In this study, the authors leverage more than 100 CCTV data and video collected from social media (i.e., Twitter, Reddit, TikTok, Instagram) during the 2018 Anchorage Earthquake to model and interpret individuals’ protective action decision-making using machine learning. The machine-learning model is theoretically grounded in Lindell and Perry’s Protective Action Decision Model (PADM) and examines the relationships between various input factors (e.g., environmental and social cues, warnings, individuals‚Äô characteristics, building types) and people’s protective actions (e.g., Drop, Cover, and Hold On; evacuation; going into a doorway). More specifically, the authors first applied a popular machine-learning algorithm, i.e., random forest, to predict people’s protective actions with the selected input factors. Then, the authors adopted machine-learning explanation methods (such as feature importance and partial dependence plots) to identify the importance of the input factors and extract the nonlinear relationships between these factors and a decision-maker's probability of taking protective actions. Particularly, the authors examined whether decision-makers with different roles (e.g., leaders vs. non-leaders) exhibit heterogeneous behavioral responses during earthquakes. The authors also explored different optimal protective actions, given timing and building characteristics, combined with decision-makers’ roles and relationships. This study offers new findings and insights to determine optimal protective actions in a variety of situations.

In the past two years, the COVID-19 pandemic has significantly impacted people’s travel behavior and activities. To better forecast and understand person-activity during the COVID-19 pandemic, this study used an emerging data source from the MobIntel Platform, which tracks mobile device locations along Clematis Street in downtown West Palm Beach. The dataset spans March 1, 2020 to December 31, 2021. The data provides a proxy for person-activity and is examined longitudinally during multiple important phases of COVID-19, including during government-mandated lockdowns, the early vaccination period, the Delta outbreak, the Omicron outbreak, and others. The study applied state-of-the-art deep learning methods to forecast the person-activity based on numerous factors, including different phases of the pandemic, regulatory lockdown information, weather conditions, time-of-day, day-of-week, important events, and other factors. The resulting model generated accurate predictions for person-activity levels (inclusive of pedestrians and vehicles) at a major corridor in West Palm Beach, Florida during the pandemic. This model can also help transportation planners and engineers better understand how different social factors may be impacting person-activity levels over time in a downtown district, and thus facilitate well-informed policy-making.

Homeownership is tightly linked to various dimensions of social inequality. For example, racialized barriers to homeownership have played a key role in perpetuating racial wealth disparities. Homeownership is also linked to disaster vulnerability and preparedness in multiple ways. Homeowners tend to have resources for hazard preparedness that renters often lack. Moreover, in the United States and elsewhere, flood protection and disaster aid policies often target homeowners. Past experience of flooding also has also been consistently found to increase flood preparedness. These two factors may compound each other, or flood experiences might motivate protective measures in households otherwise unlikely to take them. The authors of this study evaluated interacting effects of homeownership and flood experience on two different measures of household protective measures for flooding using survey data from cities in the northeastern United States. The analysis differentiated renters, owners with mortgages, and owners whose homes are fully paid off. The authors also assessed how these factors intersected with other axes of inequality. Initial results showed differentiated effects of experience across groups, with especially high adoption of protective measures among homeowners with flood experience. As homeownership correlates strongly with race and income, these findings have important implications for providing more equitable access to flood preparedness measures, whether within households or through community-oriented approaches.