Researchers Meeting Abstracts

In this study, we investigate whether housing conditions can serve as predictors of disaster-related protective behaviors. Specifically, we address the research question: Can housing conditions be used to predict individuals' preparedness for disasters? We employed a quantitative research design, utilizing a structured questionnaire administered to 909 participants across North, Central, and South Florida. The survey instrument was developed through a comprehensive literature review and was carefully aligned with the study's research objectives. To ensure content and face validity, the questionnaire was reviewed by five scholars from diverse disciplinary backgrounds. The survey focused on two key preparedness behaviors: (a) developing an emergency communication plan and (b) assembling an emergency supply kit. In addition, seven behavioral constructs were examined: risk possibility, risk severity, subjective norms, personal responsibility, perceived behavioral control, outcome expectancy, and behavioral intention. Data analysis involved multiple statistical techniques to identify behavioral patterns and quantify attitudes toward disaster preparedness. Among the examined constructs, perceived behavioral control, outcome expectancy, and subjective norms emerged as the only factors that significantly influenced individuals' intentions to engage in disaster preparedness behaviors. Furthermore, the findings revealed that only two housing-related conditions—the need for dwelling repairs and whether the residence was located on the ground floor—were significantly associated with differences in individuals' risk perceptions. By explicitly linking housing conditions to preparedness behaviors, this study addresses an important gap in the disaster preparedness literature and provides actionable insights that can assist local governments and emergency planners in designing more targeted and effective community preparedness strategies.

In this U.S. Department of Housing and Urban Development (HUD) funded project, I examine how increased adoption of off-site construction methods can support faster and more effective disaster recovery. Communities across the US frequently face natural hazards that damage housing stock and create urgent post-disaster rebuilding needs. Accelerating construction timelines while maintaining affordability is therefore essential to improving disaster recovery outcomes. I hypothesize that broader use of off-site construction and manufactured wood components can significantly reduce rebuilding time, improve labor efficiency, and enable rapid deployment of replacement housing following disasters. These benefits position off-site construction as a critical strategy for shortening recovery timelines while also supporting the production of affordable housing. The study includes two fully funded illustration-build events hosted by the Structural Building Components Association (SBCA) with support from industry partners. The first event compares traditional on-site construction methods with off-site systems—including floor trusses, wall panels, roof trusses, and other prefabricated components—using a time-motion study to generate quantitative evidence of speed and efficiency gains relevant to post-disaster rebuilding scenarios. The second illustration build supports a market study focused on disaster recovery applications. The market study engages observers from both events through interviews, surveys, and focus groups to identify barriers to adopting off-site construction for post-disaster recovery. Findings from both components will inform a stakeholder workshop aimed at developing a strategic market response framework to increase builder adoption of off-site construction methods and reduce the gap between intention and actual implementation, ultimately strengthening disaster recovery.

Housing restoration is critical to post-hurricane recovery at both household and community levels, yet limited research has examined how community-level actors perceive the challenges associated with long-term housing recovery. Local officials and representatives of non-governmental organizations (NGOs) play central roles in facilitating recovery, but their insights are often underrepresented in the disaster recovery literature. We employ a semi-structured interview approach to examine perceived challenges to post-hurricane housing recovery among community-level actors in eastern North Carolina, a region that has experienced ten major hurricane-related disaster declarations since 1999. We conducted in-person field visits across nine eastern North Carolina communities, and used purposive and snowball sampling techniques to identify 21 local officials and NGO representatives involved in housing recovery efforts. We audio-recorded, transcribed, and analyzed interviews using a two-cycle qualitative coding process to identify recurring patterns and themes. Findings indicated that housing recovery is frequently prolonged, with recovery timelines ranging from six months to six years. Four primary themes emerged as key contributors to delayed recovery, including (a) financing constraints, including disparities between homeowners and renters, (b) institutional and administrative barriers, (c) storm intensity and magnitude of damage, and (d) workforce and infrastructure limitations. By foregrounding the perspectives of actors directly engaged in post-hurricane housing assistance, this study provides actionable insights for local governments and service providers facing similar challenges. The findings contribute to community-engaged scholarship and advance understanding of the social and institutional dimensions of long-term disaster recovery.

In this paper, I critically reexamine disaster risk reduction (DRR) by situating it within broader questions of disaster risk governance, foregrounding Indigenous Knowledge systems and the power asymmetries that shape their recognition. Drawing on multi-sited ethnographic fieldwork among Wakhi communities in the Pamir region of High Mountain Asia, I develop the Wakhi Worldview as a theoretically grounded framework through which risk, environment, and social life are understood across interconnected layers of perception, preparedness, communal verification, moral restraint, and divine orientation. The findings demonstrate that local knowledge is not merely adaptive or experiential, but embedded in social institutions, ethical relations, and spiritual practices that collectively constitute a community-based system of governance. These are expressed through embodied environmental sensing, domestic material preparedness, and distributed authority anchored in trusted spaces such as the Jmotkhona. Such practices reveal anticipatory and relational forms of resilience that remain largely overlooked within dominant DRR paradigms. At the same time, I interrogate global disaster risk governance systems that privilege technocratic expertise while marginalizing or instrumentalizing Indigenous ontologies as supplementary inputs. Engaging with debates on epistemic justice and decolonial theory, I argue for a shift beyond integrationist approaches toward genuinely pluralistic and community-centered models of governance. By positioning Indigenous knowledge systems as dynamic, co-equal, and contextually grounded, I highlight their critical relevance in addressing layered vulnerabilities in the Pamir region under accelerating climate change. It calls for a fundamental rethinking of disaster risk governance to better align with local realities and knowledge traditions.

Communities with high social vulnerability often face compounded challenges in disaster recovery, including prolonged disruptions to vital services such as electricity. In this study, we ask: (a) Which dimensions of social vulnerability are associated with longer post-disaster power outage durations? and, (b) Do outage recovery times exhibit spatial clustering across Florida counties? Using Florida county-level outage duration data for major disasters since 2015, combined with hazard severity measures and social vulnerability indicators, we apply survival analysis with spatial diagnostics to examine drivers and patterns of restoration time. Preliminary results show persistent disparities in restoration speed, with longer outages concentrated in counties characterized by socioeconomic disadvantages and infrastructure constraints. Findings offer practical insights for equitable energy resilience, informing targeted grid hardening, restoration prioritization, and disaster planning for communities most likely to experience prolonged outages.

In this study, we examine the role of planning, preparedness, and other factors that impacted organizational recovery following Hurricane Maria in Puerto Rico. Drawing on preliminary findings from a large, interdisciplinary investigation and study led by the National Institute of Standards and Technology (NIST), we show that the presence of an emergency plan is associated with improved recovery outcomes for organizations across business, education, and healthcare sectors, underscoring the importance of preparedness for organizational functioning after disasters. By viewing planning indicators as signals of broader organizational capacity (rather than mere planning activities), we provide an interpretive perspective for understanding how organizations function within contexts of physical damage, prolonged infrastructure disruption, and interdependent service demands. We utilized a mixed-methods approach, integrating longitudinal surveys and semi-structured interviews to capture a holistic view of recovery. In addition to planning, other factors examined for their impact on recovery include repair progress, receipt of disaster assistance, social capital, and organizational size. This study is additionally informed by a systematic review of the disaster recovery and resilience literature, which shows that planning and preparedness are among the most frequently examined explanatory factors, though findings vary in how preparedness is conceptualized. Ultimately, this framing reinforces the empirical importance of planning while contributing to ongoing discussions about how preparedness supports recovery in complex disaster environments.

Financial resources are a critical tool for mitigating the losses from natural hazards and enabling recovery. Insurance, federal assistance, and personal savings are a few examples of the resources used during disaster recovery; however, there are a myriad of other possibilities, including social networks, charities, and grey markets that fill in the gaps. The question guiding this research is "Does utilizing multiple resource types promote a faster post-disaster recovery?" We utilize a survey that sampled 515 U.S. households from Florida and Louisiana affected by Hurricanes Ian and Ida to understand how socioeconomic factors and resource use affect recovery time. The logistic regression results indicate no significant association between socioeconomic factors and recovery time while controlling for the cost of damages. However, there was a significant association between the number of resources utilized and the time to recover. Surprisingly, there was no significant association between financial compensation from resources and recovery time. On average, resources received from a government agency were associated with slower recovery times. The findings also indicate that accessing multiple resources may increase disaster recovery times. This research examines potential interpretations of the findings and identifies avenues for future research to enhance their nuance and depth.

To advance disaster recovery practice from descriptive tracking to actionable decision-making, we map the translation layer between data and decision-making by analyzing how academic research benchmarks housing recovery. Managers frequently track housing indicators, such as rebuilding rates or permit issuance, but often lack the reference points needed to interpret them. Without explicit benchmarks, a metric like "60% rebuilt" remains descriptive and cannot reliably trigger assistance, prioritize neighborhoods, or evaluate equity. We conducted a scoping review of 58 peer-reviewed studies (2005-2025) across 16 country contexts, extracting 293 unique indicators. We coded 340 study indicator entries against a four-type taxonomy: prescriptive standards, empirical patterns, assessment thresholds, and planning targets. Results reveal a structural mismatch between research outputs and practitioner needs. Only 42.9% of entries reported an explicit benchmark. When present, reference points were overwhelmingly event-specific empirical patterns (80.1%), describing what did happen, rather than prescriptive standards (0%) or operational thresholds (16.4%) that define what should happen. Benchmark coverage was uneven, with the highest availability for housing/shelter (61.6%) but critical gaps in governance, services, and well-being domains. To close this gap, we propose a benchmark metadata schema and minimum reporting standards. We outline how researchers and practitioners can co-develop the decision-ready, equity-sensitive benchmarks necessary to transform recovery descriptions into recovery action.

In this study, we develop a novel analytic pipeline to address a critical gap in urban flood governance: understanding who owns flood-exposed properties, not just where they are located. Using all single-family residential parcels in Douglas County, Nebraska, we link property ownership records to regulatory floodplains as defined by Federal Emergency Management Agency (FEMA)s Special Flood Hazard Areas. Ownership is classified into five organizational forms—individuals, Limited Liability Company (LLC)s, corporations, trusts, and government or nonprofit entities—and by portfolio scale using a fine-tuned BERT transformer model. We used a modified Poisson regression with neighborhood fixed effects to estimate associations between ownership type and flood exposure, controlling for parcel-level attributes and micro-spatial heterogeneity. Results show that ownership structure is a significant predictor of parcel-level flood exposure. After adjustment, single-parcel LLCs exhibit the highest exposure, with a 72% higher risk than individually owned single-parcel properties. Multi-parcel corporations and multi-parcel LLCs also face significantly elevated exposure. Single-parcel LLCs are more exposed than multi-parcel LLCs, whereas multi-parcel corporations are more exposed than single-parcel corporations. This reproducible framework offers a transferable method for making ownership more transparent and supporting more effective floodplain governance.

Results from a survey in Vermont uncover concerns about the health impacts of wildfire smoke, widespread awareness of air quality alerts, and significant behavioral adaptation during poor air quality events, establishing a baseline understanding of public risk perception. Wildfire smoke is an increasing public health threat in Vermont, largely due to downwind exposure to large western U.S. and Canadian wildfires. At least one major wildfire smoke event occurred in the study area with max Air Quality Index (AQI) of 156 (Red) between late May and early September 2025. The survey instrument assessed public concern, vulnerability, and adaptive capacity related to wildfire smoke and compromised air quality during the summer season, when co-occurring climate hazards like extreme heat are also present and was distributed via online local message boards between August 14 and 25 (n=966). Concern about wildfire smoke ranked fourth among perceived climate-related threats, behind extreme heat and flooding. However, respondents reported high concern about smoke-related impacts on individual health (69.5%) and the health of other vulnerable people (67%), with lower concern for the safety of pets or livestock (25.2%) and children (15.8%). More than 80% of respondents reported seeing an air quality alert in the past year and indicated they were most likely to reduce outdoor activity when the AQI reached Orange (49%) or Red (27%) warning levels. Findings establish a baseline understanding of public concern, risk awareness, and adaptive behaviors related to PM2.5 exposure from wildfire smoke.

Where do people turn for information that aids decision making in the face of natural hazard events–and are they finding the information they need? Over the last two decades social media has become a primary source of real-time information—information that ranges from reliable to outlandish. Making the online landscape more complicated, AI-generated content is being shared with consequences for how people understand the severity of hazards and how emergency response resources are deployed. In an event that requires non-experts to make life-changing decisions, including protective actions like evacuation, navigating an information environment that is rapidly evolving can be difficult. In December 2026, an atmospheric river inundated the Pacific Northwest, resulting in wide-scale evacuations and road closures. In this presentation, we summarize research on the "information landscape" generated around this event: online posts about weather, flooding, and response as related to decision-making for risk reduction. Analysis indicates that local, state, and federal authorities, local and regional news media outlets, citizen groups, and individuals all contributed social media posts pertaining to life safety. Comments on posts from all sources can be used to understand information needs and to inform future information strategies. For example, questions and comments on posts of evacuation maps indicate a need for multiple or interactive maps with all streets within the evacuation zone represented, to enable people to identify zone limits. These findings can inform communication practices in a variety of hazards, including flooding, volcanic activity, earthquakes, and wildfire.

Researchers can support decision-making for risk reduction on all scales, from individual and household decisions up to policy enacted by national government and international organizations. But how do we do this effectively? In this presentation, we will discuss the opportunities and limitations of research to make a practical impact on the ground. Our focus will be on communications research, but all disciplines and perspectives are welcome. Our intent is to create a space to identify the role of partnerships in whether and how research informs practice, to share stories of both impact and missed opportunities, and to pose questions to other researchers about how to improve our applied research approaches. Through the discussion, topics for future research directions and collaboration will be identified.

Research on wildfire evacuation has largely emphasized warning compliance, departure timing, and route choice, while comparatively little attention has been paid to where evacuees ultimately go and how evacuation destinations vary across communities. We examine evacuation destinations, and trip chaining during two concurrent wildfire events in Los Angeles County in January 2025: the Eaton Fire and the Palisades Fire. Using anonymized mobility data, we identify devices and vehicles originating within official evacuation zones during warning periods and track trip trajectories over the day of the fires to determine destinations, intermediate stops, and potential "first points of safety." We compare evacuation behavior across the two fires, which differed in geography, timing, and socioeconomic context, by linking origin and destination locations to census block group characteristics and land-use classifications. Our findings reveal substantial heterogeneity in evacuation trajectories and destination choices across the two events, including differences in patterns of relocation across income contexts. These results highlight that evacuation outcomes extend beyond route planning and are shaped by unequal access to housing, resources, and social networks. By shifting attention from evacuation timing to evacuation destinations, this study contributes new insights for hazard planning, sheltering strategies, and equitable wildfire response. 

After a wildfire, impacted residents often seek air, water, and soil sampling and testing to understand the extent of contamination on their properties and to guide remediation efforts. Such environmental reports are completed by a variety of companies and stakeholders, often leading to differing recommendations despite similar testing outcomes. These inconsistencies often leave residents confused, limiting their ability to address hazards on their properties. Despite this trend being well-reported by community members, there has yet to be an analysis of the extent to which recommendations vary after a disaster event. To fill this gap, we aim to identify information misalignments between environmental testing results and remediation recommendations after the 2025 Eaton and Palisades Fires in California. To do so, we systematically analyze a representative sample of over 600 post-fire environmental testing reports received from impacted residents. Results from floor samples for 17 heavy metals typically tested in California (i.e., CAM-17) and their associated remediation recommendations are extracted from the reports and compared to identify discrepancies in recommendations for similar empirical results. Results indicate that recommendations vary drastically, including self-cleaning for lead levels that require abatement, and recommendations for similar results that range from throwing away soft goods to addressing hard surface contamination. The information misalignments identified in this research are not unique to these fires but rather represent a systemic problem across disasters. Results from this analysis will inform critical needs for evidence-based guidance for chemical hazards and exposures, thereby transforming the broader post-disaster recovery process.

Studies have been discussing the deservingness of target populations for government programs. However, these studies on the deservingness of target groups have been limited in examining the balancing of social equity with urgency and scarcity during crisis periods like the COVID-19 pandemic. Using the Control, Attitude, Reciprocity, Identity and Need (CARIN) scale for deservingness with the social construction theory, I applied quantitative content analysis techniques to analyze the press releases (n=477) from 50 U.S. governors and their vaccine distribution plans (n=50) from October 9, 2020, to March 31, 2021. The findings show that state governments neglected some critical, high-risk vulnerable groups during the COVID-19 vaccination program. This unexpected result contradicts the initial belief that vulnerable populations often receive state support and protection, even in times of crises. It enlightens policymakers that advancing social equity in emergency policies with complex policy contexts could warrant addressing apparent trade-offs and rekindling public service ethos. 

Governments routinely collaborate with community groups and organizations. Yet, effective collaboration still needs leadership. The literature suggests that such leadership can be particularly effective if related behaviors and functions are distributed among members, and rotated across initiatives or time. However, distributed leadership is often considered to grow naturally out of social interactions, giving little guidance to groups on how to activate and sustain such practices. In this study, I examine distributed leadership in a cross-sector, multi-organizational anti-displacement collaboration in Miami, fighting the impact of climate gentrification. The case study identifies social mechanisms through the reconstruction of narrative episodes. It illustrates how instrumental mechanisms can activate distributed leadership, but value-based mechanisms are needed to sustain it: Distributed leadership can result from a group's efforts to engage in pressure buffering but is unlikely to be sustained without the group completing the process of identity formation through internalization and construction.

In this presentation, we detail how academic researchers, community scientists, state employees, and mutual aid organizers work together as a coalition to respond to, recover from, and mitigate riverine flooding. Our team is connected by Culvert Crawlers, a community-science-based stormwater infrastructure monitoring program we co-designed. Each of us wears multiple hats, meaning we have multiple responsibilities to different communities, agencies, or sets of practices—which is quite common in rural areas. Our multiple hats allow us to learn from one another and teach one another, as we all have different expertise and perspectives, while also having a common purpose. We pose a key question in this paper: in the multisector coalitions necessary for flood response, recovery, and mitigation in rural areas, how do all the layers of the system play together? What do people think everyone's responsibility is, and how do people communicate about that? The paper includes brief stories about each participant's coalitional work for response, recovery, and mitigation, then describes the shifting roles played by each author in responding to rural flooding disasters. We examine roles and responsibilities for different groups, including community organizations, municipal management, the State, and the Federal Government - not just looking backwards, but going forward. We reflect on how wearing many hats out of necessity ends up building stronger coalitions. We argue that diversifying perspectives and expertise, and a willingness to listen in any conversation helps to build trust across systems and communities of practice.

Community engagement is widely recognized as an essential component of effective emergency preparedness and response planning. However, challenges continue to limit the widespread use of this process, potentially leaving communities underprepared for the next disaster. In this presentation, we will discuss the design and implementation of preparedness and response activities that incorporate a strong community engagement component. Using the example of a public health preparedness pilot project in Detroit, Michigan, USA, we will cover frameworks for community engagement, strategies for building multisectoral coalitions and partnerships, methodologies for collection of non-traditional, mixed methods data, lessons learned during implementation, and readily available tools and resources. The pilot project has potential applications to a broad range of other emergency preparedness and response activities. Implementing similar frameworks to engage communities early and often could improve planning for future emergencies before the next crisis occurs. 

As climate change intensifies, global population movements are expected to grow. Migration decisions result from multifaceted processes involving interacting social, political, economic, and environmental factors. While existing research has identified common decision-making factors driving migration, two gaps remain. First, there is a lack of understanding regarding how different factors interact to produce migration outcomes, especially related to their intersection with increasing climate hazards and vulnerabilities. Second, migration literature exhibits a mobility bias, leaving non-migration understudied. We address these gaps by examining Lower Plaquemines Parish, Louisiana, a community that has experienced multiple catastrophic hurricanes, environmental and economic stressors, and is increasingly vulnerable to saltwater intrusion and coastal inundation. This research is part of the Mississippi River Transition Initiative, funded by the National Academy of Sciences. We are conducting in-depth interviews with 150 residents who have migrated out of Lower Plaquemines Parish over the past 20 years and those who remain. We use Qualitative Comparative Analysis (QCA) to identify how multiple factors interact to shape migration and non-migration. Preliminary analysis of 15 interviews highlights distinct factors associated with migration vs. non-migration. Shrinking community, insurance rates, limited economic opportunities, service closures, and housing constraints were discussed as challenges. While some households cited these factors as reasons for leaving, those who decided to remain emphasized age, livelihood dependency, place attachment, and community ties as central to their decision to stay. This study will help to explicate the multi-dimensional nature of migration decision-making and evaluate both migration and non-migration as outcomes shaping coastal communities.

The United States is facing a wildfire crisis of ever-increasing risk to homes and communities in wildfire prone regions, particularly the western states.  As seen in several recent events, Wildland-Urban Interface (WUI) fires can impact communities quickly, leaving little or no time to evacuate. This increasing frequency of large-scale wildfire evacuations requires innovative pre-planning and training by emergency managers to carry out timely and effective evacuations. The Resilience Institute staff is addressing this need by assessing the expressed critical needs of our community stakeholders. In particular, we are assessing the adequacy of evacuation messaging and evacuation plans developed by key decision-makers and participants in wildfire response. We worked with three community emergency managers to co-produce FlamMap simulations that were subsequently used in tabletop training exercises conducted by local officials to identify key decision points and messaging. Next, we used these results to conduct community surveys that assessed local residents' expectations about how they would respond to wildfire evacuation warnings. We then focused on responses of the emergency managers using a semi-structured interview to assess their satisfaction with various aspects of the engagement process, such as sense of trust, sense of ownership, process/development participation, planning collaboration, and management/community collaboration. We expect that this style of community engagement will promote data-driven policies and decision-making for wildfire risk reduction within our study areas. We also expect this project to provide a better understanding of how to motivate and engage community partners and stakeholders in effective and resilient wildfire evacuations.

The longitudinal effects of flooding on individuals, households, and communities are challenging to study due to the limited availability of consistent, event-scale flood-extent data at high enough spatial resolution to connect with demographic and health surveys. We advance methods for reconstructing historic flood extents from fragmented observations using optical imagery (Landsat 5-9, Sentinel-2), human observation data through life history calendar methods, and water height levels from ground-sensing stream gauges. Our method identifies high-confidence flooded pixels across all sources, even in areas that are heavily obscured or have low reporting density, leveraging the advantages of each. For example, satellite imagery provides increased spatial coverage but is limited due to cloud cover and revisit time, whereas human observation and ground sensors provide highly accurate data points but a lower density of observations. Flood observations across sources are spatially expanded using physically based constraints to hydrologically plausible locations, resulting in a harmonized flood extent for a given historic flood event. The method is applied to Western Chitwan, Nepal—a region characterized by frequent cloud cover, complex and frequent riverine flooding, and limited historical flood documentation—for 35 large-scale flood events from 2000 to 2024, and a combined estimate highlighting areas of repeated flooding. Overall, this framework supports the retrospective analysis of the impacts of floods on health and livelihoods in regions where historical flood records are unavailable and offers a scalable pathway for reconstructing flood histories in data-limited settings. 

Power outages caused by tropical cyclones can disrupt critical infrastructure, posing serious risks to public safety and emergency response. A primary goal of the National Hurricane Center (NHC) is to reduce the number of indirect fatalities from extreme heat in the wake of a storm, and power outages exacerbate this risk. Furthermore, as hurricanes approach landfall, it is crucial to provide Impact-Based Decision Support Services (IDSS) to clearly describe the locally-relevant risks posed to communities. We develop an artificial intelligence (AI)-based model that forecasts county-level power outage severity based on official forecast products from the NHC, the Weather Prediction Center, and the Storm Prediction Center. We used ArcGIS Pro to collect and geoprocess archived probabilistic meteorological data from 35 landfalling Atlantic hurricanes between 2020 and 2024 by county. We incorporated additional variables such as social vulnerability, vegetation coverage, and power infrastructure characteristics to represent non-meteorological drivers of outage risk. The response variable was the maximum percentage of electrical utility customers without power in each county within 120 hours after a forecast time, sourced from the Department of Energy's EAGLE-I Outage dataset. An AI model using neural networks was trained to decipher the complex patterns between predictor variables and outage severity, and it was found to have great accuracy. Meteorological hazard variables proved to be the most indicative of the location and severity of outages, but the model performed significantly worse when exposure and vulnerability metrics were removed. Ultimately, this capability will increase the NHC's situational awareness of communicating hurricane-induced impacts.

Rural communities across the western United States face increasing exposure to natural hazards alongside rapid growth, infrastructure strain, and environmental change. Traditional hazard-focused assessments often fail to capture how development pressures, economic priorities, and environmental stewardship interact to shape long-term resilience in these contexts. In this paper, we present the West Central Mountains Resilience Assessment, a place-based, participatory process conducted in Valley County, Idaho, and its municipalities. The assessment integrated a modified City Resilience Index framework with surveys, focus groups, interviews, and community workshops involving local governments, emergency managers, economic development organizations, and residents. Rather than treating hazards in isolation, we examined how land use decisions, housing growth, infrastructure capacity, and ecosystem management influence both disaster risk and community well-being across the region. A key outcome of the process was the co-development of a strategic action planning theme, balancing growth and stewardship to advance a prosperous future, which emerged consistently across hazard, economic, and environmental domains. This theme reframed resilience from a narrow focus on emergency preparedness toward integrated decision-making that aligns development patterns with natural systems, infrastructure limits, and community values. The resulting action framework identifies near- and long-term strategies to reduce hazard exposure while supporting economic vitality and quality of life. This case demonstrates how participatory resilience assessment can support rural communities in navigating growth-related tradeoffs and translating complex hazard and development challenges into actionable, community-owned strategies.

As intensifying flood events reshape previously stable regions worldwide, understanding relocation decisions in newly at-risk territories offers critical insights for global adaptation planning. In this ongoing research, I examine why communities persist in flood-prone areas despite recurring disasters. I use Italy's Romagna region as a unique case study where unprecedented flooding along the Lamone and Montone river watersheds between May 2023 and December 2025 represents a new phenomenon that has exposed gaps in disaster preparedness infrastructure common to many emerging risk contexts. I employ surveys and walking focus groups to investigate how generational rootedness, economic constraints, topographic conditions, and institutional relationships influence residents' capacity and willingness to relocate. The design anthropology framework foregrounds lived experiences of landscape transformation and spatial adaptation under conditions of recurring risk. Preliminary findings reveal three patterns with broad implications for adaptation policy. First, relocation decisions correlate strongly with generational sense of belonging and economic ties to place, with outcomes varying significantly by topography and ecosystem type—valley settlements face different relocation constraints than hillside or historic urban centers. Second, considerable tension exists between residents and emergency management agencies, stemming from perceived institutional failures and communication breakdowns during crisis events. Third, the combination of insufficient infrastructure interventions and unprepared emergency systems for intensifying flood phenomena has generated deep institutional distrust, complicating recovery pathways. This research advances understanding of how communities navigate persistence and relocation when familiar landscapes transform into risk zones, offering transferable insights for regions worldwide confronting similar transitions from environmental stability to recurring flood exposure.

We address the practical challenges of developing and sustaining interdisciplinary research teams focused on built environment resilience. As climate hazards intensify, effective resilience research increasingly requires collaboration across disciplines. Yet researchers often lack guidance in structuring teams that bridge disciplinary boundaries, integrate diverse methodologies, and maintain momentum over time. Through facilitated scenario planning activities, we will explore built environment resilience research that transcends individual disciplines. Each scenario will prompt participants to collaboratively develop strategies, drawing on their experiences and expertise. Our discussion will examine what organizational structures and practices enable productive interdisciplinary collaboration, how teams can adapt over time to address evolving needs, what strategies help teams navigate tensions between disciplinary priorities and funding priorities, and how interdisciplinary collaboration help facilitate community engagement and applied practice. Our focus is on faculty participating in resilience research teams, graduate students navigating interdisciplinary collaborations, and practitioners partnering with academic units. Expected outcomes include practical frameworks for structuring collaborative research, peer connections across disciplines, and collective identification of resources needed to sustain impactful teams. This dialogue advances understanding of how researchers can work together more effectively to address complex resilience challenges. 

Water system consolidations, or the managerial and/or physical merging of two or more systems, can be driven by a variety of technical, economic, and environmental factors; among these, natural hazards have received comparably little attention from scholars. We assess the role that natural hazards played in water system consolidations decision-making in California and the role consolidations can play in hazard mitigation. We find that hazards were an important factor in many cases and identify key gaps that could be addressed to improve the hazard mitigation benefits of water consolidation going forward. This study uses qualitative analysis of data collected for 11 case studies of completed water system consolidations in California. These data comprise 29 semi-structured interviews and 144 documents collected for analysis, including local, state, and county-level water system administration documents. We find that some hazards like drought put environmental pressures onto communities that necessitated consolidation. In other cases, consolidation was used as a mitigation strategy for anticipated hazards like wildfire. However, we also find that there were additional barriers to water consolidation as a mitigation strategy and that ongoing challenges post-consolidation limited mitigation benefits in other cases. These challenges will need to be addressed if water system consolidation is to become a more accessible mitigation strategy. These findings are significant for research and practice regarding hazard mitigation in particular.

Limited uptake of earthquake insurance and growing stress in traditional insurance markets motivate the exploration of alternative, community-based risk transfer mechanisms. We investigate the potential of reciprocal exchanges as a transparent and analytically grounded framework for managing earthquake risk, with particular emphasis on how physical risk-reduction measures can enhance exchange performance. We focus on the integration of seismic retrofitting and reciprocal exchange design as a coupled strategy for improving long-term affordability, stability, and resilience. We develop a modeling framework to examine the dynamics of a reciprocal exchange using a pilot study of single-family residential buildings with cripple-wall vulnerability in Los Angeles. We combine stochastic earthquake loss modeling with an explicit formulation of exchange operations, including premium contributions, surplus accumulation, claims payments, and participation in external insurance markets. We also conduct parametric sensitivity analyses to evaluate the influence of key design choices, including insured limits, deductibles, exchange premium rates, and the role of external market coverage. The results demonstrate that seismic retrofitting can materially alter exchange dynamics by reducing loss frequency and severity, accelerating surplus growth, and improving exchange capacity for partial self-insurance over time. More broadly, the findings highlight how integrating retrofit-based risk reduction with reciprocal exchange design can support more efficient, scalable, and transparent approaches to community-level earthquake risk management.

Wildfires impose significant economic, health, and social costs, including worsening air quality and severe disruptions to regional economies. In Utah, one in fourteen households is at moderate to high risk of wildfire. In this study, we examined how communities in Salt Lake City, Utah, perceive wildfire risk as well as the costs and benefits of pursuing land use and insurance-based wildfire mitigation options. We used a qualitative design including observations of public hearings and homeowners association (HOA) meetings, key informant interviews, and focus group discussions with key stakeholders such as state and local officials, HOA representatives, prominent nonprofits, and local community leaders. We analyzed the data using qualitative content analysis to identify critical themes related to risk perception (spatially and temporally) and costs and benefits of undertaking specific mitigation actions (e.g., short- and long-term financial costs or savings; regulatory requirements; and feasibility of changing market and policy landscapes). Findings from this study contribute to knowledge on the willingness of community members to engage with specific forms of wildfire mitigation and the reasons for this. This study is part of a two-year research project funded by the U.S. National Science Foundation that aims to develop models to test the effectiveness of mitigation strategies on wildfire risk reduction in Utah communities and to develop specific community-based toolkits to implement viable strategies. This approach responds to the need for tools that allow communities to evaluate mitigation strategies and to develop plans that balance wildfire risk reduction with the specific development needs of that community.

Frequent and severe wildfires across the Western United States pose severe risks to communities, ecosystems, and critical infrastructure, particularly power systems. Electric utility providers play a central role in both contributing to and responding to wildfire risk. This study develops a stakeholder-validated framework for utility providers to use for wildfire planning, particularly within the Western Interconnection region. We used a two-step methodology to develop this framework. First, we conducted a comprehensive review of emergency operations and mitigation plans developed by utility providers as well as academic literature and industry-established best practices to establish eight practices that promote wildfire resilience in power utility practices. The eight practices include: grid hardening and design, vegetation management, situational awareness, asset management and inspection, long-term data tracking and planning, wildfire emergency coordination and outreach, grid response operations, and public safety power shutoff. In the second step, we conducted 3 focus group discussions with representatives from utility firms, energy planning consultancy firms, and regulatory agencies to validate the framework components, and to identify challenges and opportunities for the adoption of the framework for wildfire resilience on the ground. The key challenges include resource constraints (e.g., financial, insufficient staffing or expertise, data availability, and technological), geographic or environmental constraints, regulatory hurdles, stakeholder opposition, and a lack of coordination. These outcomes of this study contribute to a more consistent understanding of wildfire mitigation practices and support improved decision-making in the power utility sector.

The COVID-19 pandemic caused significant shocks to businesses and economies around the world. However, the breadth and diversity of policy responses presents an opportunity to better understand the role of financial and non-financial assistance on business outcomes. In this presentation, we report on findings from a large, international collaborative effort to understand business recovery and the role of these supportive policies after the pandemic and related public health restrictions. In addition to gathering data on business recovery policies from 14 case study cities in nine countries (n=374 policies), we conducted a survey of businesses in 10 of those cities: Vancouver, Calgary, Miami, Los Angeles, Amsterdam, Auckland, Christchurch, Cape Town, Bangkok, and Osaka (n=3,435 businesses). Results from the survey provide insights into business impact and recovery; business responses; e-commerce and other business adaptations; access to financial and non-financial assistance; supporting program characteristics; and the role the pandemic played in changing business readiness for future disruption. Contrasts between the case study cities are also identified. Finally, we reflect on some of the methodological benefits, challenges, and lessons from this experience in international research collaboration. This study evolved out of a working group formed as part of the Natural Hazards Center's CONVERGE initiative in 2020.

Amid the growing frequency of rapidly intensifying hurricanes (RIHs)— characterized by a wind speed increase of at least 35 miles per hour over a 24-hour interval—realities often outpace forecasts, complicating evacuation planning that relies primarily on wind metrics and often underestimates associated flood risks. Using an expert survey of emergency management and transportation officials, we examine current practices related to RIH and investigates the determinants of evacuation lead time, defined as the number of hours before expected landfall when authorities recommend evacuation. Our findings reveal the rarity of specific plans for RIHs, with many respondents reporting operational delays in issuing evacuation orders, deploying transit resources, or opening shelters. Multilevel mixed-effects regression results indicate evacuation lead times increase significantly with hurricane category, and a positive association between evacuation lead times and flood risk. No evidence is found linking lead times to hurricane preparedness indicators or locational risk factors such as proximity to coast. These results indicate the overreliance on wind speed-based metrics in guiding evacuation decisions and highlight the need for impact-based, flexible frameworks that integrate storm surge, rainfall, and rapid intensification forecasts.

Recent studies show the promises of Micromobility Resilience Hubs (MRHs), community-based nodes that integrate micromobility options such as bicycles and e-scooters with local social networks, to enhance community disaster capacity. To understand the public's perception and willingness to use micromobility and MRH with social networks in disasters, we collected 911 survey responses from Seattle, Washington (n = 710), and Stillwater, Oklahoma (n = 201). Results indicate that mode choice varies by task, with vehicles still their primary evacuation mode in both cities (Stillwater 85.1%, Seattle 89.7%), underscoring the vulnerability of single-modal transportation systems, consistent with other recent disaster evacuations such as the 2024 Northern California tsunami and the January 2025 Palisades and Eaton wildfires. Walking and micromobility gain ground when the destination is a local hub (walking: 68% likely; micromobility: 32% likely), while micromobility plays a limited but observable role in the recovery phase.  Moderate and consistently seen levels of general social trust across both sites, together with similar average endorsement of MRH functions spanning both disaster response and everyday community use (Stillwater 65%, Seattle 68%), showed broad community support for neighborhood resilience hubs. Preferences for hub locations, particularly parks, recreation centers, and K-12 schools, further suggest public willingness to embed resilience infrastructure within familiar community spaces. These findings provide an empirical foundation for implementing the MRH framework and planning for disaster response and recovery.

Local Korean governments install smart technology-based monitoring systems in approximately 10% of the 22,300 rivers nationwide to improve disaster risk management in river basins, where most flood-related casualties occur. These systems measure real-time water surface elevation and discharge without direct contact with water, relying exclusively on surface velocity measurements. Accurate discharge estimation therefore requires conversion of surface velocity to mean velocity using an appropriate mean-to-surface velocity conversion coefficient. However, because direct measurements are often difficult or unsafe during flood conditions, a constant value of 0.85 or values estimated using empirical equations are commonly applied. Existing studies show that in steep and hydraulically complex rivers, the conversion coefficient is nonconstant, highlighting the need for diversified and river-specific estimation approaches. In this research, we determine the mean-to-surface velocity conversion coefficient using measured surface velocity distributions at different water depths while accounting for flow characteristics such as velocity distribution, bed condition, and slope. We evaluate the applicability of existing methods using derived coefficients. The equation based on Manning's formula seems most accurate among existing approaches. Furthermore, we developed a new prediction method that uses hydraulic variables easily measured in rivers. Depth–velocity profiles and hydraulic data were collected at 19 rivers using an RS-5 radar velocimeter. Dimensionless variables were derived through dimensional analysis, and a nonlinear regression model estimated the conversion coefficient. Validation results indicate excellent reproduction of the mean–surface velocity relationship in rivers. These results demonstrate the practical applicability of the proposed conversion coefficient method to smart technology-based river monitoring systems.

Hurricane preparedness is critical for disaster resilience, yet socioeconomic disparities create differential risks. In this study, I examine how sociodemographic and efficacy factors shape hurricane preparedness in diverse U.S. communities. The analysis was based on 2021-2023 FEMA National Household Survey data, which comprised 3,466 samples from 11 states. I applied a quantitative approach specifically to hurricane-related preparedness measures. Drawing on vulnerability and resilience scholarship, the analysis demonstrates that preparedness is not merely an outcome of individual choice but is structured by unequal access to resources, information, and institutional support. Intersectional modeling reveals that preparedness deficits are most pronounced at the intersection of gender, low income, social vulnerability, and efficacy. These effects are not additive but interactive, producing qualitatively distinct experiences of constraint. Importantly, the findings challenge single-axis explanations by showing that structural advantages do not uniformly translate into preparedness in the absence of efficacy, while high self-efficacy alone is often insufficient to overcome severe material constraints. By conceptualizing social vulnerability as structurally produced rather than individually possessed, this study advances disaster scholarship by highlighting how institutional arrangements, power relations, and intersecting social positions jointly shape hurricane preparedness.

Caregiving households often experience more complicated evacuations and worse disaster outcomes than non-caregiving households during extreme events, revealing critical gaps in emergency plans that assume all households can evacuate quickly and independently. We examine differences in evacuation processes and outcomes of the 2025 Los Angeles wildfires between caregiving households (defined as those with children under 18 years of age and/or members having conditions making travel difficult) and non-caregiving households. Using survey data collected from affected residents, we analyze evacuation preparedness, processes, and displacement outcomes. We hypothesize that caregiving households face compounding preparation, coordination, and mobility barriers that heighten risk of prolonged displacement. Initial data exploration suggests that caregiving households were less likely to have specific evacuation plans, required more preparation time, reported elevated worry about rescuing family members and finding shelter, and expended greater effort gathering household members and information than non-caregiving households. During evacuation, they were more likely to abandon vehicles, require rescue, and experience difficulties from depleted batteries or vehicle charge. Post-evacuation, caregiving households experienced higher rates of housing damage, utility loss, difficulties securing shelter, and prolonged displacement. We are currently conducting advanced statistical modeling of departure timing and return trajectories and performing qualitative analysis of open-ended responses to clarify evacuation needs among caregiving households. These findings will guide more inclusive evacuation planning, communication strategies, and support systems for caregivers in wildfire-prone regions.

The National Science Foundation RAPID project "Bridging the Data Gap: Capturing Ephemeral Data to Enhance Wildfire Resilience and Recovery after the 2025 Devastating Los Angeles Fires" is a collaborative effort bridging academia and practice to document perishable data essential for wildfire recovery and long-term resilience. By integrating interdisciplinary researchers with practitioners and nonprofit, community-based, and government stakeholders, the project addresses critical gaps in understanding post-disaster recovery processes, organizational responses, and evolving community needs, while fostering partnerships that strengthen equitable disaster preparedness and response. We present preliminary findings from 40 qualitative interviews with nonprofit, philanthropic, and community-based organizations engaged in response and recovery following the January 2025 Los Angeles wildfires. The analysis focuses on organizational roles, coordination, and the incorporation of community priorities during the early stages of recovery. Findings reveal a complex and rapidly changing organizational recovery landscape that raises fundamental questions about who is recognized as building disaster resilience and whether local communities and community-based organizations are afforded the capacity, authority, and resources to meaningfully shape recovery outcomes. These dynamics challenge conventional ways of categorizing "disaster organizations" and assessing contributions to disaster resilience, underscoring the need for more nuanced frameworks to understand organizational roles, collaboration, and community-centered recovery. Together, these findings have important implications for future disaster planning and policy, particularly for improving coordination, aligning resources with community priorities, and supporting more equitable and effective recovery strategies.

Field studies in post-event reconnaissance present an opportunity to gather perishable, real-world data. These data are nearly impossible to replicate in a controlled lab setting and support the development of impact simulation models, guiding strategies for community recovery, rebuilding, and resilience. However, a prevailing challenge is the lack of standardized, interdisciplinary methodologies for collecting and integrating crucial information. This necessitates a spatial framework to integrate all collected data, facilitating standardized, scalable, and interdisciplinary data collection across spatial, social, and temporal scales. There is a need for a system to populate secondary pre-event data, such as demographics, housing, community asset locations, and infrastructure networks, to enhance situational awareness during reconnaissance missions. In this talk, we present our work on developing a spatial data collection support system to integrate hazard characterization data, engineering-based damage assessments of buildings and infrastructure, and validate socio-economic survey and interview instruments.

As wildfires increase in frequency and severity, disaster management agencies are seeking to collaborate with community-based organizations (CBOs) to build wildfire resilience. This qualitative case study examines the critical role of CBOs as cultural brokers supporting Latin@/e/x wildfire survivors following the 2020 "Mobile Home Parks" Fire in rural Oregon. Drawing from 23 semi-structured interviews, three focus groups, document review, and participant observations, I investigate how Latin@/e/x CBOs navigated the complex cultural landscape between wildfire survivors and traditional disaster recovery frameworks. The study reveals significant challenges in wildfire recovery for Latin@/e/x communities. Institutionalized disaster recovery systems, designed for administrative efficiency, often overlook cultural realities of Latin@/e/x survivors. When disaster frameworks clash with local culture, those with the least institutional power bear the greatest risk. I propose an expanded conceptualization of cultural brokering, differentiating between one-way cultural brokering (improving access to existing systems through linguistic interventions and case management) and multi-way cultural brokering (circular cultural translation enabling communities to create new disaster recovery paths rooted in local culture). Multi-way cultural brokering offers a transformative approach centering community priorities, lived experiences, and cultural resources as essential rather than supplemental to recovery programming. The findings demonstrate that effective cultural brokering requires institutional commitment to long-term leadership development, institutional humility, sustained relationship-building, and robust accountability mechanisms. This approach positions Latin@/e/x community members as primary architects of their own recovery, offering a model for creating more equitable, responsive, and transformative disaster resilience strategies through shared governance between CBOs, government agencies, and funders. 

The way people understand risk depends not only on factual information, but also on how that information is communicated. Considering flood hazards, effective risk communication plays a critical role in motivating individuals to take protective actions for themselves and their property. Improving flood communication requires understanding the factors that shape flood risk perception and how these drivers differ across individuals and communities. In this presentation, we examine spatial and social factors that may influence perceived flood risk, including geodesic distance from the hazard, social connectedness to the affected location, and community-level vulnerability. We collected our dataset using a randomized controlled trial (n = 929) in which participants were randomly assigned to view different flood risk representations and self-report their perceptions and decision intentions. These representations included flood maps from government agencies and nonprofit scientific organizations, as well as crowdsourced flood imagery and plain geographic maps. Preliminary results indicate that participants who live farthest from flooding tend to report the lowest perceived risk. In contrast, social connectedness to the flooded location does not show an overall association with flood risk perception. Social vulnerability exhibits a non-linear pattern, with participants who perceive flood risk as either very low or very high being more likely to reside in counties with higher social vulnerability. Overall, our findings highlight the importance of accounting for spatial and social factors when designing flood risk communication and proactive risk management strategies for diverse audiences, including policymakers, emergency managers, and community leaders.

Immigrant disaster vulnerability stems from intersecting policy arrangements, structural barriers, and institutional exclusions rather than immigration status alone, yet disaster research predominantly focuses on undocumented populations while treating immigrants as passive subjects rather than knowledge producers. We employed modified photovoice methodology with 30 immigrant participants who experienced disasters between 2020 and 2025 to document how immigration status intersects with gender, race, class, English proficiency, and socioeconomic position to shape differential vulnerability and resilience across disaster phases. Systematic sampling captures heterogeneous experiences across naturalized citizens, lawful permanent residents, temporary visa holders, refugees, asylees, and other statuses experiencing disasters like pandemics, wildfires, hurricanes, and floods. Participants submitted at least five photographs per research question documenting their perceived vulnerability and resilience, factors influencing differential disaster risk, resource mobilization strategies for recovery, and policy recommendations for improving disaster management. Individual photo-elicitation interviews using the SHOWeD framework generated rich narrative data interpreting visual documentation. This methodology addresses critical research barriers by requiring no immigration status disclosure, accommodating language barriers through visual communication, and repositioning immigrants as active co-researchers, controlling which experiences to share. Our study advances disaster vulnerability theory by integrating migration scholarship through intersectional frameworks, revealing how multiple marginalized identities compound to create unique vulnerability patterns while immigrants simultaneously draw on cultural assets and transnational networks for resilience. Findings generate evidence-based recommendations for culturally responsive emergency communication, inclusive disaster assistance eligibility reform, community-engaged preparedness programs, and emergency management training while contributing methodological guidance for participatory visual research with marginalized populations during crises.

Policy arrangements, not disasters themselves, create systematic vulnerability for immigrant communities, yet these structural causes remain poorly understood in disaster management practice. In this study, I analyzed major United States news coverage from 2020 through 2025 to document how policies across multiple domains exclude immigrants from disaster protection. Through content analysis of the Wall Street Journal, Washington Post, New York Times, USA Today, Newsweek, and National Public Radio (NPR) across pandemic, wildfire, and hurricane disasters, I identified three policy failure patterns. First, assistance exclusions denied aid to nearly five million United States citizens under the Coronavirus Aid, Relief, and Economic Security Act (CARES Act) based solely on family members' immigration status. Second, enforcement policies created fear-based barriers preventing immigrants from accessing legally available assistance, with families avoiding shelters and aid sites due to perceived deportation risks, regardless of protective policies. Third, communication and workplace safety failures left farmworkers without language-appropriate warnings, with California employing one Spanish-speaking inspector per 192,308 Spanish-speaking farmworkers during wildfires, while Indigenous families received no K'iche' or Mam emergency alerts during Hurricane Helene flooding. Critical analysis revealed these exclusions stem from interconnected policy choices in immigration law, disaster assistance eligibility, enforcement practices, and emergency communication systems. Findings demonstrate that achieving disaster equity requires policy reform across multiple domains simultaneously, not simply improved outreach to immigrant communities. This research provides evidence-based documentation of exclusion mechanisms with direct implications for reforming disaster assistance eligibility, emergency communication protocols, and interagency coordination to ensure immigrant communities receive equal disaster protection.

In January 2025, a series of wildfires swept through neighborhoods in the Los Angeles region, destroying entire communities and resulting in numerous injuries and deaths. Existing wildfire studies provide limited insight into evacuation behavior during large urban wildfires such as those in Los Angeles. In this study, we examine the evacuation decision-making of residents affected by the two largest Los Angeles wildfires, the Palisades fire and the Eaton fire (which began approximately eight hours later). Adding to the wildfire evacuation literature, we model a little-studied outcome: those who evacuate multiple times. Drawing on survey data collected in July of 2025 in the two wildfire areas, we use sequential logistic regression to model the factors associated with multiple evacuations. Across both fires, 3.5 percent of the sample did not evacuate; however, approximately eighteen percent of the sample evacuated more than once. The model predicts the likelihood of evacuating as a function of personal and household characteristics (e.g., sex, race, age, income, presence of children, auto ownership), risk perception, prior evacuation experience, evacuation communication, and wildfire area. We then draw on responses to the open-ended survey questions to contextualize the model results. The findings extend the existing body of research on evacuation decision-making to capture the role of sequential factors and, in so doing, can help to enhance evacuation planning and communications during complex urban wildfire events.

What does it mean to be dis/abled in an area prone to repeated disasters? For marginalized groups in Eastern North Carolina, "ordinary" everyday realities are often experienced as emergency, crisis events. Everyday things people may take for granted, such as food and safe housing, can create everyday disasters for people with disabilities and other marginalized identities. Through 12 months of ethnographic fieldwork, I examined how experiences of disaster (broadly defined) shape perceptions of, and experiences with dis/ability in Eastern North Carolina. Participant observation with a Healthcare Preparedness Coalition (HPC) and Center for Independent Living (CIL), along with semi-structured interviews with HPC and CIL staff and consumers, revealed that marginalized groups face ongoing, chronic struggles while simultaneously preparing for, responding to, and recovering from disasters. Because "disasters" are often framed as immediate and acute, overlooking the extended time it takes marginalized communities to recover from such events, recovery is not always framed as dynamic and ongoing, which neglects interrogating "normal" pre-disaster conditions that recovery processes aim to return to. What is a "return to normal" when the normal everyday lives of people with disabilities often include harm and suffering? In addition to advancing anthropological engagement with disability and disaster, I contribute to disability and disaster studies, public health, and policy, and provides evidence of the importance of including marginalized groups in research and policy processes. Examining disasters through dis/ability narratives advances social justice and helps create environments that prioritize access for all stigmatized and devalued bodies and minds.

In the face of recurrent natural disasters, Florida stands at the forefront of proactive mitigation efforts aimed at fortifying resilience against the ever-growing impacts of climate-related events. In this presentation, we delve into Florida's strategic approach to mitigating disaster risks and its trajectory towards a more resilient future. As a disaster-prone state, Florida recognizes the importance of investing in mitigation projects to enhance resilience against the impacts of natural disasters. Since 2012, the state has been conducting Loss Avoidance Assessments (LAAs) after every presidentially declared disaster to evaluate the effectiveness of these projects by quantifying the dollar damages averted by mitigation efforts, crucial for garnering community support and securing federal and state funding. Notably, LAAs play an integral role in the state's Enhanced State Hazard Mitigation Plan, meeting FEMA's standard planning requirements. This enhanced status enables Florida to access additional funding and benefits through FEMA's Hazard Mitigation Assistance programs. The latest LAA examined 120 state and federally-funded hazard mitigation projects in response to Hurricanes Ian and Nicole in 2022. Utilizing a return on investment (ROI) formula (Losses Avoided / Mitigation Project Costs), the assessment gauged the effectiveness of these projects in safeguarding 586 structures. The results revealed significant ROIs for both hurricanes, demonstrating the substantial impact of these mitigation efforts in reducing disaster-related damages. This comprehensive analysis underscores the sound investment in mitigating natural hazard risks in Florida, further highlighting the state's strategic use of enhanced status to fortify resilience measures.

Resilience decisions made today can mean two very different futures for communities facing tomorrow's challenges. In this presentation, we discuss two fictional communities, Resilienton and Denialton, that demonstrate different approaches to resilience. Using real-world examples, we discuss how choices such as budgets, insurance, ordinances, and partnerships shape whether a community stays strong or could end up facing higher taxes, lost businesses, and costly consequences. Community leaders are responsible for governing decisions that include budgets, insurance, ordinances, and partnerships. These decision-making processes—or lack thereof—determines whether their jurisdiction's finances, operations, and residents are protected when disruption comes. We argue that stressful decisions and actions regarding current and future budgets, insurance, ordinances, and partnerships—not emergency response—are what ultimately determine their communities' resilience. This presentation and case discussion have implications for emergency management and governance. Specifically, practitioners can gain insight into how administrative decisions shape township resilience and long-term financial and community impacts; and practical steps to strengthen resilience and protect township resources without major new costs.

The Upper Colorado River Basin (UCRB) is experiencing its worst megadrought of the past 1,200 years. There is an urgent need to address growing water scarcity for millions of people in this region. However, understanding of available coping and adaptation options is clouded by lack of agreement on the underlying drivers and necessary actions. Examining news media helps to understand this evolving discourse on water availability, quality, and access, along with strategies for coping and adaptation to these challenges. We employed structural topic modeling (STM) to analyze newspaper articles (n= 6,503) discussing water availability and vulnerability topics in the UCRB from January 2000 to February 2024. The STM approach allowed statistical assessment of associations between automatically identified topics and their relationship to larger themes of vulnerability and water availability. Findings revealed the layered and complex nature of water availability – including its connection to conflicting presentations of underlying causes as long-term aridification; short-term hazards (flash drought and heat); and shifting population and economic patterns. Results showed distinct media representations of federal actions as primarily funding hazard response and mitigation, while state and local interventions were presented as longer-term policy actions and adaptation. Tensions between water preservation for the agricultural sector and water resource capture for growing municipalities were also revealed. The study highlights how underlying stakeholder world views, jurisdictional boundaries, and economic drivers have shaped the media representation of water availability and water security risks in the UCRB.

California's public safety agencies uniquely coordinate wildland fire responses through a connected system; however, interactions among first responders and emergency managers have yet to be systematically studied, limiting understanding of their actions under extreme conditions. To address this gap, we examined the alignment between guidelines and responders' experiences during three concurrent wildland fires in September 2024 and into January 2025 in Southern California, where two additional events resulted in catastrophic impacts. We developed an interview-based survey based on a pilot study of 27 government documents, which led to a coding book reviewed by academics and a practitioner. We then conducted semi-structured interviews with 20 respondents from the fire service, law enforcement, and emergency management across six primary domains, including a synopsis of response actions, the situation assessment process, protective actions, operational challenges, feelings experienced during the decision-making process, and a review of guidelines and experience. Using AI-assisted analytic tools, the research team applied a semi-open coding process to the transcripts to ground the key findings. Summarizing the California model, it highlights distinctive features of how public safety agencies organize their responses at the regional crisis management level, compared to a top-down coordination structure emphasized in federal guidance. Specifically, the model is characterized by pre-existing professional networks and cross-jurisdictional trust, which enable rapid resource deployment and complex decision-making during chaotic moments. At the same time, the case study also reveals system challenges in cross-disciplinary and cross-agency coordination, and guidelines that must be addressed further.

About 70% of dams in the United States have exceeded their service lives, and over 2,400 High Hazard Potential Dams do not have an updated Emergency Action Plan (EAP) in place. Several catastrophic dam and levee breaks in recent years have garnered public attention, but many Americans remain unaware of how their communities would fare in the event of a dam breach or failure. While local governments should assess and communicate the risks of dam breaks and failures in local hazard mitigation plans, many municipalities do not possess in-house capacities to model inundation scenarios and cannot afford to contract such studies or reports from engineering or planning consulting firms. Through a systematic plan review, we examined all fifty-state hazard mitigation plans to assess which policies and practices can best guide local governments to communicate dam failure risks to the public and incorporate dedicated risk reduction actions in local hazard mitigation plans. By examining intergovernmental dam risk assessment, we also asked, "what collective roles should local communities play in dam risk management?" Findings underscored the need for state governments and federal agencies to harmonize hazard classifications and develop more decision support tools to model and predict dam inundations, coordinate emergency procedures, and identify how damages and disruption would impact critical local and regional water and energy supplies. Recommendations include proposed criterion for determining when dam ownership transfer would best mitigate risks and could allow for public benefit, including recreational amenities and ecological reconnectivity for aquatic species.

Tribal, state, and local jurisdictions are responsible for developing plans that outline strategies intended to mitigate, prepare for, respond to, and recover from extreme events. However, there exists limited evidence about the effectiveness of included strategies. Over the past several years, methods have systematically assessed the content and quality of extreme event plans, and describe commonly adopted strategies and gaps across jurisdictions. There remains significant potential to use these results in epidemiologic investigations to assess the effectiveness of adopted strategies in reducing the health impacts of extreme events, with cascading benefits for evidence-based planning. Yet, several challenges in using these data for such analyses must be considered. For example, inclusion of specific strategies in plans does not necessarily confer their implementation, reach, or fidelity in response to a specific event. Additionally, jurisdictions with plans likely differ systematically, and in unmeasured ways, from jurisdictions without plans, making cross-jurisdictional comparisons difficult. Relatedly, it cannot be reliably assumed that jurisdictions without publicly available plans lack plans, reducing their utility as "controls" in analyses. In this study, I provide examples of data generated through recent plan analyses to identify data use opportunities and challenges in epidemiologic analyses and other evaluations.

Cardiovascular disease (CVD) is contributing significantly to rising death rates in the United States. Furthermore, areas susceptible to natural disasters face more challenges. When a disaster strikes, a part of the population seeks refuge in shelters where access to essential treatments is limited. The limitation of treatments results in a higher mortality rate among individuals with CVD. In response, our research presents an agent-based model to explore the repercussions of CVD patients lacking access to essential treatment. The model was built to represent the potential impact on health outcomes of individuals with CVD conditions that might be relocated to shelters during a hurricane event. The simulation results showed an average 14% rise in CVD mortality after hurricane occurrences, which approximately represent the rates observed from hurricane events in Texas. The model is an instrument to forecast long-term health outcomes and to plan for public health interventions associated with disaster relief. 

Stormwater infrastructure systems are critical for protecting people, property, and ecosystems from escalating climate risks, yet local officials managing these systems must make decisions under uncertainty, limited resources, and evolving regulatory and community expectations. A proliferation of climate services aims to support local decision-making about climate risks, yet a persistent gap remains between their development and use in practice, highlighting the need to better understand stormwater practitioners' decision-making and information needs. In this study, we present an institutional analysis of stormwater system management based on 15 semi-structured interviews with local practitioners across 14 cities in the San Francisco Bay Area and Hudson Valley watersheds. Practitioners reported limited use of formal climate services; instead, their understanding of system performance, prioritization of actions, and day-to-day decisions were shaped primarily by immediate operational demands—particularly maintenance needs, regulatory requirements, and accountability to elected officials and residents. We develop an integrative framework that characterizes three interrelated streams of activity shaping stormwater management decisions: tactical stormwater system maintenance, regulatory compliance, and strategic, long-term infrastructure planning. For each stream, we identify the key administrative roles involved, the biophysical, socio-political, and institutional pressures at play, and the information sources that inform decisions. Our results suggest that expanding the reach and effectiveness of climate services requires aligning tools with practitioners' decision contexts and professional responsibilities and disseminating them through trusted institutional and relational channels. The study offers practical insights for strengthening connections among researchers, service providers, and local decision-makers to support risk reduction.

In January of 2025, the Eaton and Palisades fires in Los Angeles, California, burned over 12,300 buildings. Previous studies have associated wildland-urban interface fires with physical and mental health impacts, such as respiratory issues due to chemical exposures and psychological distress due to uncertainty post-disaster. Since context-specific drivers can influence health risk perceptions, it is important to identify the specific needs of communities affected by the Eaton and Palisades fires. This information can guide public health interventions to meet household needs and strengthen community resilience. The present study addresses the question: What are the mental and physical health concerns among the households impacted by the Eaton and Palisades fires? To answer this question, we, with community input, designed and deployed a survey about four to six months after the fires. A total of 1,229 responses were obtained from the fire-impacted areas. The survey included seven open-ended questions, which were qualitatively coded following a hybrid inductive-deductive approach to identify health-related perceptions and experiences. Preliminary results showed that the most common physical health concerns were general health impacts (57%), followed by poisoning (34%), and respiratory issues (17%). For affective response (i.e., emotional reactions and feelings), fear, worry, or concern were the most common themes, cited by 94% of respondents that reported emotional reactions. Sadness or stress (10%) and anger or frustration (4%) were also mentioned by respondents. These results show how the community interpreted health risks, providing insight to inform public health response and communication strategies.

Historical housing policies have left enduring marks on patterns of environmental vulnerability in U.S. cities. We investigate the relationship between Home Owners' Loan Corporation (HOLC) redlining and contemporary flood exposure across 201 cities, linking the spatial legacies of 1930s racialized investment maps to present-day patterns of inundation. We integrate digitized HOLC maps from the Mapping Inequality project with building-level flood depth estimates from the First Street Foundation Flood Model to evaluate differences in flood exposure across HOLC grades (A, B, C, and D). Using nationwide, coastal versus non-coastal, and city-level comparisons, we examine two complementary dimensions of flood exposure: mean flood depth, and the proportion of buildings exposed to the 100-year flood event. Across all cities, we find statistically significant differences in flood exposure by HOLC grade. On average, historically redlined D-graded neighborhoods experience greater flood depths and higher proportions of exposed buildings than neighborhoods graded A, B, or C, reflecting persistent environmental inequities. Stratified analyses reveal that elevated flood exposure remains concentrated in redlined areas in inland cities, while in many coastal cities, A-graded neighborhoods exhibit greater flood exposure than D-graded areas. City-level analyses further demonstrate substantial geographic heterogeneity in the magnitude and direction of these disparities. Findings highlight how historical housing discrimination, hydrologic setting, and subsequent urban development interact to shape contemporary flood exposure. By situating redlining-related disparities within the policy-relevant 100-year flood zone, this study contributes new evidence to debates on environmental justice, urban planning, and climate adaptation in an era of intensifying flood hazards.

The United States experiences more tornadoes than anywhere else on Earth. Like many other natural hazards, one of the enduring questions that interests researchers is the potential impact a changing climate has on tornado behavior. Previous studies have looked to explain trends in tornado activity through environmental factors using reanalysis data or tornado reports through the national historical record. While these approaches have substantially improved a collective understanding of tornado frequency and the atmospheric conditions necessary for events, they offer limited insight into how tornado activity may evolve in the future or how projected changes intersect with future population distributions. In response, we present a newly developed modeling framework that leverages Coupled Model Intercomparison Project Phase 6 (CMIP6) data to generate gridded estimates of tornado activity in the United States through the year 2100 for a high emission shared socioeconomic pathway. These projections are further integrated with gridded estimates of future population for the same emission scenario to assess how potentially changing tornado behavior coincides with population change across the country. Together, these results offer actionable information for emergency response, risk management, city infrastructure planning, and community engagement to better prepare for, respond to, and mitigate future losses.

Despite more than a decade of the Sendai Framework for Disaster Risk Reduction, disaster losses, vulnerability, and exposure continue to increase unevenly across countries, which raises fundamental questions about why national disaster risk governance trajectories diverge under shared global commitments. In this study, I examine how political-institutional and economic change shapes variation in countries' alignment with global disaster, development, and climate frameworks. Focusing on disaster risk reduction, I situate the Sendai Framework within a regime complex linking risk governance, development planning, and climate policy. Rather than treating alignment as static compliance, I conceptualize alignment as directional change over time in policy-relevant indicators associated with the Sendai Framework, the Sustainable Development Goals, and the Paris Agreement. Drawing on insights from international cooperation, regime complexity, and political economy, the analysis argues that evolving domestic conditions, particularly governance capacity, income inequality, and economic structure, mediate how international commitments translate into observable policy movement. Methodologically, I employ a change-based cross-national design that emphasizes within-country year-to-year shifts rather than cross-sectional comparisons. Alignment is assessed separately for each framework using widely used indicators, including disaster loss measures, development performance scores and climate policy indicators, which are analyzed in relation to changes in governance effectiveness, inequality and fossil-fuel dependence using descriptive analysis and simple first-difference models. By examining disaster risk reduction alongside development and climate governance, the study highlights how persistent disaster risk reflects political and institutional dynamics along with technical and capacity limitations that shape uneven governance outcomes across countries.

Despite an ever-growing literature on post-disaster relocation, understandings of the interlinkages between household and other community stakeholders' decisions to relocate remains limited. Focusing on this gap in the literature, we examine the following questions: how do other households, businesses, and NGOs' relocation decisions affect households' decisions to stay and rebuild or relocate following a disaster event? The paper is based on case study research conducted in two communities in Puerto Rico, Comerio and Loiza. Both communities were affected by 2017 hurricanes Maria and Irma, and subsequently, Hurricane Fiona. The data comes from small-and large-group, disaster-based scenarios that informed exercises conducted with residents of these communities (n=49). Scenarios were developed through several rounds of input from policy makers and community leaders and were conducted as part of a larger research project on post-disaster relocation funded by the National Science Foundation. Our results indicate that household decisions to relocate after disasters depend on actions taken by other households, businesses, and NGOs. However, such dependency varies by disaster type, social capital, and the characteristics of businesses and NGOs in these communities. The study offers several avenues for future research and policy implications on post-disaster relocation.

Understanding how flooding shapes residential mobility requires observing people and their movements not only after single events, but across multiple events over time. For many populations, recurrent flooding associated with seasonal storms and hurricanes is becoming increasingly common with climate change. Existing research on flood-related mobility has largely focused on isolated events, typically major disasters that attract widespread media, government, and research attention. As a result, little is known about how prior flood experiences influence relocation decisions for repetitively flooded populations. We addresse this gap by linking novel longitudinal flood and mobility data in North Carolina. We draw on 73 flood extent maps from the North Carolina Flood Extent Archive (NC-FLDEX), covering floods between 1996 and 2019, and link these data to residential address histories for nearly 10 million NC residents. We find over 113,000 individuals experience at least one flood during the study period, with approximately 13% experiencing multiple events. While repetitively flooded individuals move less frequently on average, they exhibit elevated mobility responses relative to single-flood and non-flood populations following major disasters, such as Hurricane Matthew in 2016.To isolate the effects of individual flood events on mobility, we implement a deconvolution signal-processing approach to account for overlapping mobility responses when floods occur in close temporal and spatial proximity, such as Hurricanes Matthew (2016) and Florence (2018). Together, this dataset and approach provide new evidence on how repetitive flooding shapes residential mobility and offer a framework for studying mobility responses to increasingly frequent climate hazards. 

This project advances wildfire preparedness by demonstrating how tailored, in-home education can significantly improve household readiness in high-risk areas. The American Red Cross launched wildfire preparedness visits in 2023 to address the threat of wildfires, which burned nearly nine million acres in 2024. These visits aim to increase preparedness by helping households create evacuation plans, sign up for emergency alerts, and protect their property. The National Opinion Research Center (NORC) at the University of Chicago surveyed all 799 households that received a wildfire preparedness visit between September 2023-2024 and a nationally representative comparison group in high wildfire risk areas to assess implementation and outcomes. NORC used chi-square tests to assess differences between intervention and comparison groups. Participants were more likely to participate in preparedness behaviors than the comparison group. For example, 64% of respondents signed up for local emergency alerts after the visit, 84% had a planned evacuation route (vs. 60% in the comparison group*), and 60% reported having a Go Bag (vs. 30% in the comparison group). Additionally, 92% engaged in at least one activity to prepare their property for wildfire following the visit. Attitudinal changes were notable; 87% felt prepared to respond to a wildfire after the visit, compared to 63% in the comparison group. Nearly all participants found the visit useful, and 88% would recommend it. These findings highlight the value of personalized outreach for advancing wildfire resilience and underpin opportunities to expand this approach to other high-risk communities.

An engineering-centric focus on structural vulnerability (without considering how buildings function and how communities rely on them for essential services) limits decision-makers' ability to anticipate earthquake impacts. In Los Angeles alone, roughly 1,500 non-ductile reinforced concrete buildings were constructed before modern seismic codes. Often called "killer buildings," they threaten life safety and can trigger social and economic disruption when the services they house are interrupted. Especially with public buildings, it is not clear which buildings will be retrofitted first and based on what criteria. Therefore, we created a retrofit prioritization framework that reflects stakeholder preferences through community surveying. First, to identify target partner communities, we analyzed demographic characteristics of populations using public buildings that do and do not require retrofits. Asian students are disproportionately overrepresented in schools that require seismic retrofits and in census block groups with at least one residential building requiring retrofit, low-income residents comprise 33% of the population, compared to 23% overall. These disparities motivate a community survey targeting residents who are disproportionately represented in high-risk areas. In partnership with neighborhood councils, we are distributing a survey to elicit stakeholder preferences for how public-use buildings should be prioritized for seismic retrofits. We estimate prioritization rankings using best–worst scaling with a balanced incomplete block design, which helps us quantify how communities weigh the relative importance of different public buildings and services during seismic events. We also collect information on how frequently respondents use these services in non-disaster conditions and assess their earthquake preparedness.

Disaster recovery presents a window of opportunity to mitigate or eliminate structural, economic, and social vulnerabilities. However, recovery is rarely planned in advance for an event, resulting in restoring pre-existing risks and vulnerabilities, especially among ethnic minorities. Based on the study of the 2015 Nepal Earthquake recovery experience of ethnic minorities in Dolakha District, Nepal, we find that the insufficiency of recovery finances, top-down recovery governance, and livelihood-incompatible building designs were significant challenges, resulting in displacement, migration, and the emergence of new forms of vulnerability among ethnic minorities. A 7.8 magnitude earthquake shook many densely populated districts of Nepal on April 25, 2015, followed by a major aftershock of 7.3 magnitude on May 26, 2025. A decade after the event, the Bhimeswor Municipality of Dolakha recovered, but many questions on vulnerability and resilience remained unanswered for marginal groups. We explore the transcending nature of social and economic vulnerabilities among ethnic minorities in the aftermath of the 2015 Nepal Earthquake. Using binary logistic regression models on 400 household survey data and purposive coding of 13 in-depth key informant interviews, this study concludes that informality played a critical role among ethnic minorities in recovery from the earthquake, compared to other demographic groups. The study recommends empowering local governments, strengthening community capital, and enhancing informal institutions among ethnic minorities as the most effective strategies for building resilience against future geological and climatic disasters in Nepal and the Global South.

This interdisciplinary, mixed-methods investigation examines the intersection of mass incarceration and disaster vulnerability by integrating a scoping review of multidisciplinary literature describing disaster vulnerability near sites of incarceration in the United States with quantitative analysis of associations between social vulnerability to disasters and prison presence in Appalachia, and a critical disaster studies historical analysis of the 2022 Central Appalachia floods. The review revealed a dearth of literature describing disaster vulnerability and incarceration, but limited existing research demonstrated that intersecting social factors affect disaster vulnerability near incarceration sites, incarcerated persons experience disproportionate vulnerability, and incarceration site location, conditions, and practices exacerbate vulnerability. The quantitative analysis demonstrated clear associations between social vulnerability to disasters and the presence of prisons in Appalachian census tracts, with higher social vulnerability scores associated with increased odds for prison presence, particularly among racial/ethnic minority and housing/transportation factors. The historical analysis revealed harrowing descriptions of disaster vulnerability and loss tied to historic themes of coal mining, as well as more recent prison proliferation and production of disasters like the 2022 floods. These findings have significant implications for disaster science, which should be informed by a deeper investigation of specific contexts in which disasters occur. Improved contextual understanding of upstream social, political, and economic drivers of disaster vulnerability can assist researchers, emergency managers, and policy makers alike in their efforts to identify and disrupt the circumstances that produce disasters, their associated negative outcomes like morbidity, mortality, and economic loss, and their disproportionate impacts on marginalized populations. 

Adaptations to extreme weather events are increasing in prevalence across the globe. Adaptation actions can produce co-benefits, which are ancillary positive outcomes beyond the primary objectives of reducing risk and enhancing resilience. Despite growing attention to co-benefits in the adaptation literature, systematic analyses across adaptation types, sectors, actors, and regions remain limited. We use the Global Adaptation Mapping Initiative (GAMI) database to investigate the associations between adaptation characteristics and the presence of co-benefits. We processed and coded 1,684 articles, to categorize co-benefits and analyzed them alongside other adaptation characteristics through descriptive statistics, chi-squared tests, and logistic regressions. We found generally weak associations between adaptation characteristics and the presence of co-benefits. Further, the regression results showed that there were no significant differences between the associations with co-benefits of adaptations that occur across economic sectors. Ecosystem-based responses were found to be more likely to be associated with the presence of co-benefits than human behavioral based responses. Technical or infrastructural responses were found to be less likely to be associated with co-benefits than human behavioral responses. Additionally, while some of the chi-squared associations and regression associations trended together, they also differed, leading to ambiguity in the types of adaptations that often have co-benefits present. These findings highlight the need for improved co-benefit assessment frameworks and enhanced co-benefit documentation. Enhanced documentation could better inform adaptation planning and maximize the ancillary benefits of adaptation actions.

Over the past two decades, predictive indices of community resilience have received notable attention. Multiple indices have been developed to serve as a guide for measuring the status of communities in relation to hazard events and other disturbances. The suggested use of these indices has also been expanded in research and practice, including criteria for policymaking and funding allocation. Despite their widespread use, the credibility of predictive indices remains in question due to validation challenges. Without proper assessment, it is difficult to determine whether a community with a high index score will indeed withstand a disruption and recover more quickly. The National Institute of Standards and Technology has been conducting a series of studies to provide evidence-based support for the use of indices through construct validation methods, particularly emphasizing the use of content and predictive validation. The former utilizes structured expert judgment on indicator choices, while the latter involves establishing longitudinal post-disaster community outcomes as dependent variables and examining how well predictive indices correlate with these outcomes. We selected thousands of U.S. counties that had presidential disaster declarations in the past two decades. Population, employment, and life expectancy outcomes were transformed to measure two key aspects of resilience: the ability to mitigate initial impacts, and the trajectory of community recovery. Utilizing Structural Equation Modeling, commonly used predictive indices were tested against different levels of disturbances and community outcomes. The resulting method serves as a quality assurance tool, including recommendations for the enhancement of existing indices. 

This project examines how Disaster Recovery Centers can be allocated more equitably across federal, state, county, and local decision levels following a major disaster. Using Hurricane Helene as a case study, I ask three questions: how county-level capacity targets and local siting constraints shape recovery center placement, how access to recovery services varies across communities under demand uncertainty, and how equity considerations can be incorporated into allocation decisions without sacrificing operational feasibility. I develop a data-driven framework that integrates county-level capacity constraints, locally feasible facility locations, population and vulnerability indicators, and transportation network travel times. I represent post-disaster demand uncertainty through multiple scenario-based demand realizations. To address the computational challenges of large-scale facility location models, the framework employs a reinforcement learning search that iteratively improves recovery center placement through quota-preserving adjustments. The objective minimizes worst-case access cost across scenarios while penalizing disparities in access between high-vulnerability counties and other counties. Results demonstrate that different policy priorities, including risk-focused, equity-focused, and cost-focused objectives, lead to distinct spatial allocation patterns under the same capacity constraints. Comparisons with observed Federal Emergency Management Agency recovery center locations show where alternative placements may improve access or reduce inequities while remaining consistent with jurisdictional targets. The implications of this work lie in its ability to support coordination across decision levels. The framework provides emergency managers with interpretable maps and access metrics that can inform pre-disaster planning through the identification of viable candidate sites and post-disaster decision-making through scenario-based evaluation of access and equity outcomes. 

Urban development in flood-prone areas continues to increase exposure to flooding from hurricanes and extreme rainfall. While the Federal Emergency Management Agency’s (FEMA) Flood Insurance Rate Maps (FIRMs) delineate 100-year flood zones used for regulation and insurance, they often fail to capture areas made flood-prone by recent land-use or hydrological changes. In this study, we investigated development trends within both FEMA-designated flood zones and federally-overlooked (FO) flood zones, areas identified as flood-prone by Fathom's high-resolution U.S. Flood Map but located outside FEMA's 100-year floodplain. We analyzed trends across 1,597 census tracts in the Houston metropolitan area from 1985 to 2020. We applied the Floodplain Development Index (FDI) and the Floodplain Housing Index (FHI) to assess both the extent and concentration of floodplain development. Our analysis revealed that federally-overlooked zones, which disproportionately encompass low-income and socially marginalized populations, experienced diffuse yet persistent development (Mean FHI = 1.14; Median = 0.30), while FEMA floodplains exhibited more concentrated and intensive growth (Mean FHI = 2.60; Median = 0.96). Mean FDI values reflected similar contrasts (FO = 0.87; FEMA = 1.16). This research fills a critical gap by extending floodplain development analyses beyond FEMA-designated zones to include federally-overlooked flood areas–an aspect largely absent from prior studies. Furthermore, it explored how social and economic factors, including race and income, shape patterns of urban development across different flood-risk contexts. These findings expose major gaps in current federal flood-risk delineation and highlight the need for more inclusive resilience planning in rapidly urbanizing regions.

Hydrological models involve numerous methodological decisions, including spatial resolution, model scope, data sources, and coverage, that affect how flood risk is perceived, mapped, and mitigated. What may appear to be purely unbiased, objective technical decisions can render flood risk visible for some communities while obscuring it for others. As a result, hydrological models play a powerful political role in shaping whose risks are recognized, how flood mitigation strategies are defined, and how benefits are measured. Despite their influence, there remains a need to understand how modeling choices and their associated limitations impact the development of effective and equitable flood mitigation solutions. We seek to contribute to this gap through a co-production approach with modelers, planners, and decision-makers across South Central Louisiana. Working collaboratively with local knowledge holders through a series of workshops, we co-design hydrological modeling approaches that integrate community perspectives, needs, and priorities. The analysis centers on illustrative multi-jurisdictional flood mitigation project concepts developed through this process. Each concept is modeled using alternative modeling designs and counterfactual scenarios to examine how different assumptions, resolutions, and evaluation metrics influence risk identification and projected outcomes. By comparing these scenarios, we demonstrate how modeling decisions shape not only technical results but also perceptions of risk and distribution of benefit. The findings will serve as a scalable framework for future planning processes across the region and will also help modelers design analyses that identify and address potential adverse impacts, minimize disproportionate or unintended consequences, and support transparent, informed decision-making.

As climate risk intensifies, a growing range of strategies are proposed to reduce long-term exposure in hazard-prone communities. Among them, retreat from hazard-prone areas has received increasing attention. Yet, it represents only one approach among a broader set of strategies that reduce exposure. Some of these risk avoidance strategies target moving population and assets and are managed through government programs, while others work through regulatory signals, infrastructure decisions, or market dynamics. Existing research often evaluates these strategies in isolation rather than comparison, primarily through outcomes alone, obscuring the decision drivers, implementation steps, actors, consequences, considerations, and challenges that shape how risk avoidance unfolds in practice. We develop an implementation-focused matrix that provides a new typology of risk avoidance strategies and then expands these strategies to their sub-strategies, implementation steps, and associated actors. In the matrix structure, we then test each sub-strategy against dimensions that define and shape the experience and outcomes, such as decision drivers and consequences, at each stage of implementation. By structuring evidence around implementation rather than end states, this matrix enables systematic comparison across strategies, scales, and contexts, highlighting the tradeoffs, tensions, and inequities inherent across retreat approaches. The matrix serves as both an analytical tool for researchers synthesizing evidence, while also functioning as a diagnostic tool for practitioners, policymakers, and affected communities to identify gaps, misalignments, and overlooked considerations in risk avoidance design and implementations. The matrix aims to support more effective, equitable, and context-sensitive implementation of risk avoidance strategies.

Determining disaster-related deaths is challenging because of insufficient investigation information, lack of awareness on proper death certification, and incomplete death certificates. We assessed awareness, knowledge, and use of two CDC guidance documents: Reference Guide for Certification of Deaths in the Event of a Natural, Human-induced, or Chemical/Radiological Disaster (Disaster VSRG) and Death Scene Investigation After Natural Disaster or Other Weather-Related Events Toolkit (Disaster MDI Toolkit). We conducted a mixed-methods study, combining a quantitative web-based survey with qualitative focus groups. We used a census approach to contact 2,910 ME/C offices in the National Directory of Law Enforcement Agencies. Survey responses were weighted for a nationally representative sample. We analyzed key themes. The web survey received 610 responses, and 10 focus groups with 40 participants were conducted. Awareness of the Disaster VSRG was 48.8%, while 41.5% were aware of the MDI Toolkit. Among those who reviewed the Disaster VSRG (n=549), 77% used it in practice, and 69.2% of those who reviewed the MDI Toolkit (n=419) used it. Barriers to use included prior knowledge and the perceived complexity of the documents. Focus group participants expressed frustration with documenting disaster-relatedness on death certificates. Awareness of both guidance documents was low, but adoption was high among those who had reviewed them. Based on these findings, we developed and disseminated training materials to improve practices for identifying and certifying disaster-related deaths, as well as reference materials like disaster-specific fact sheets for certifiers. We aim to discuss and broadly share the training and reference materials.

In this presentation, we will briefly summarize a series of our recent studies on enhancing subseasonal-to-seasonal (S2S) predictions of hydroclimate extremes, including wildfire, drought, and extreme rainfall/flooding, through a combination of physics-based Earth system modeling and AI/ML-based techniques. Accurate prediction at the S2S timescale is a grand challenge for the scientific and operational communities right now. We use state-of-the-art high-resolution Earth system models to investigate key factors and mechanisms driving the evolution and predictability of hydroclimate extremes, and leverage AI/ML and satellite observations to establish efficient prediction models for these extreme events, which shows adequate accuracy for application. This work has important implications for end-users and stakeholders from hazard prevention and resource management communities. This presentation intends to stimulate relevant discussions and collaborations on improving S2S predictions of hydroclimate extremes and enhancing effective communications with stakeholders to make the prediction product/tool more beneficial for society.

In this project, we develop a new methodology to measure and analyze causal reasoning in complex systems. Traditionally, researchers use manual, labor-intensive methods like interviews and focus groups to gather conceptual models from stakeholders, which is inefficient. We engineer a solution using AI agents to automate the extraction and analysis of causal relationships from this qualitative data. This significantly improves efficiency and reduces the manual burden. Automated reliability metrics are under development to quantify similarities and differences between these models. This capability enables advanced downstream tasks like modeling, visualization, classification, and regression. By bridging the gap between stakeholder insights and technical modeling, this workflow facilitates better understanding of forward-looking building codes and enhances community resilience. It allows for the proactive identification of systemic vulnerabilities, transforming a time-consuming qualitative process into a scalable, data-driven workflow that ensures infrastructure can adapt to evolving environmental and social stressors.

Traditional cost-benefit analysis (CBA) frequently undervalues resilience investments by focusing primarily on avoided losses (AL). We develop a unified theoretical and analytical framework to define and value resilience dividends (RD)—the ongoing welfare gains in economic, social, environmental, and institutional dimensions that accrue even in the absence of a disruptive event. By shifting the evaluative scope from loss minimization to welfare maximization, we explicitly link resilience design to the standard welfare economics tradition. The proposed framework formalizes the resilience dividend as the expected welfare difference between a resilient and a non-resilient system. Key components of the framework include: Multidimensional Dividend Categories: Identification of economic (reliability), social (well-being), environmental (ecosystem co-benefits), and institutional (adaptive learning) gains; Welfare Maximization Model: A formal model where optimal investment equates marginal cost to the combined marginal benefits of both avoided losses and dividends; Distributional Weights: Integration of welfare weights to account for income and vulnerability; and Uncertainty and Learning: A dynamic formulation that connects resilience dividends to the Value of Information (VOI) and adaptive management theory. Through a stylized model of a water system, we demonstrate that including RD increases the attractiveness of hybrid or green-gray designs, even when hazard probabilities are low. We conclude that resilience dividends should be treated as core components of welfare rather than secondary co-benefits, providing a foundation for future empirical applications and the refinement of global resilience standards.

Flooding is a recurring and intensifying hazard in South-Central Louisiana, posing a substantial risk to communities. This low-lying region has a complex environmental and socioeconomic landscape that often amplifies the impacts and burdens experienced by residents during flooding. Therefore, quantifying spatial disparities in people's capacity to prepare for, respond to, and recover from floods is critical to informing more comprehensive risk assessments and equitable adaptation planning. While approaches to integrating social vulnerability into flood hazard mitigation plans and risk assessments vary, one common method is to utilize federally produced social vulnerability indices (SVIs), such as the United States Centers for Disease Control (CDC) SVI. However, these indices are typically generalized across multiple hazards and broad geographic extents (e.g., state or national). Recognizing these limitations, we co-developed a Flood Hazard Social and Environmental Vulnerability Index (FH-SEVI) with local non-profits, community organizations, and regional-to-state stakeholders across South Central Louisiana. Through a collaborative workshop and focus group discussions, participants identified and weighted flood-specific social and environmental characteristics to develop a regionally relevant FH-SEVI. The index was further refined with participants and critically compared with other widely used SVIs and flood exposure estimates derived from a region-specific hydrodynamic flood model. This research highlights key themes of flood vulnerability and demonstrates how to calculate and translate such information into a usable format for stakeholders.  Overall, the FH-SEVI provides a flood-specific vulnerability tool for South-Central Louisiana localities to help address vulnerabilities and support more targeted and effective preparation and adaptation strategies.

There is a growing literature on first responders' burnout, including how burnout affects their health and well-being as well as their organizations and the services their organizations provide. Our understanding of the link between first responders' burnout and their organization's community relations is not yet well understood. First responders are street-level bureaucrats who interact with the public daily. Their organization's strained relationships with the public may negatively affect their burnout, while relationships that are based on trust and reciprocity can reduce their burnout. To better understand this link, we ask: (a) How are first response agencies' community relations associated with burnout among their employees? (b) To what extent does organizational support explain the relationship between first response agencies' community relations and burnout among their employees? (c) How does the relationship between first response agencies' community relations and burnout differ across types of first responders (police, fire, and emergency medical services)? The study is based on a U.S.-wide survey (n=3517) and interview data (n= 91) collected from first responders during the COVID 19 pandemic. Analyses of survey data include ordinal least squares and mediation methods, while the interview data were analyzed using NVivo software. The paper offers lessons for first response agencies on how they can reduce burnout among their employees through organizational support and strengthened community relations. 

Predicting the spread of wildfires is essential for effective fire management and risk assessment. With the fast advancements of artificial intelligence (AI), various deep learning models have been developed and utilized for wildfire spread prediction. However, there is limited understanding of the advantages and limitations of these models, and it is also unclear how deep learning-based fire spread models can be compared with existing non-AI fire models. In this work, we assess the ability of five typical deep learning models integrated with weather and environmental variables for wildfire spread prediction based on over ten years of wildfire data in the state of Hawaii. We further use the 2023 Maui fires as a case study to compare the best deep learning models with a widely-used fire spread model, FARSITE. The results show that two deep learning models, i.e., the ConvLSTM and ConvLSTM with attention models, perform the best among the five tested AI models. FARSITE shows higher precision, lower recall, and higher F1-score than the best AI models, while the AI models offer higher flexibility for the input data. Using an explainable AI method, we further identify important weather and environmental factors associated with the 2023 Maui wildfires.

Severe weather events are increasing in the southwestern United States, posing growing risks to populations that may be unfamiliar with these hazards. College students represent a vulnerable and diverse group, often relocating to areas with environmental risks they have not previously experienced, which can impact their perceptions of risk and protective behaviors. On March 31, 2023, Arkansas was impacted by two EF3 tornadoes that struck Little Rock and Wynne. Following these events, we surveyed 498 college students from Arkansas State University and the University of Arkansas at Little Rock, both located in affected regions. We examined how students' places of origin influenced their tornado experience, risk perceptions, knowledge, and protective action behaviors. Results indicated that students' home regions shaped their understanding of severe weather risks. Based on these findings, the study recommended incorporating an educational session in First Year Experience (FYE) classes to freshman students on local natural hazards and their risks to better prepare incoming students. In fall 2025, Arkansas State University implemented this recommendation through an immersive educational experience funded by the Weather Ready Research to Operations Award. All FYE students participated in a professional educational video featuring local, state, and federal emergency management officials discussing regional hazards, available resources, and preparedness strategies. Students also engaged in a Virtual Reality tornado simulation and received a comprehensive educational PowerPoint with safety guidance and resource links. This presentation highlights the original study and demonstrates how implementing its recommendations enhanced student preparedness and campus resilience at Arkansas State University.

In this research, we leverage cross-sectoral professional expertise to explore ways in which micromobility (lightweight vehicles including, but not limited to, bicycles and scooters) can support more equitable place-based disaster response and recovery while enhancing everyday community resilience. We also seek to learn how community facilities, including resilience hubs, can best connect people with transportation options and other important resources in disaster scenarios. Focusing on the characteristically different study communities of Seattle, Washington and Stillwater, Oklahoma, we conducted 17 in-depth, semi-structured interviews with experts in emergency management, shared mobility, access/functional needs, public health, transportation planning, and bicycle and pedestrian advocacy. Our research questions include: What are the opportunities and challenges associated with micromobility use in disaster scenarios, and how might barriers be overcome? How can incorporating micromobility into disaster preparedness, response, and recovery support more equitable outcomes for impacted communities? Findings include a summary of commonly cited barriers and potential solutions as well as suggestions for integrating micromobility into disaster preparedness by actively engaging local organizations and community networks as well as guiding the strategic location and support of resilience hubs. Additionally, interviewees shared specific examples of ways in which micromobility might be leveraged to support more equitable disaster response and recovery processes and outcomes. We summarize the results as a series of recommendations for practical application that include strategies for more effectively integrating physical and social networks to support micromobility in both everyday and disaster scenarios with attention to specific tasks, disaster phases, community infrastructures, and micromobility modes.

As the incidence and impact of disasters continue to grow, volunteer relief provided by place-based responders is becoming an increasingly valuable resource. We compare two unique community-based disaster response initiatives: the Seattle Emergency Hubs network in Seattle, Washington, and the Sendai Bosai Leaders program in Sendai, Japan. Historically linked, though structurally distinctive, both share a common purpose of complementing official disaster response structures, drawing upon local skills and knowledge to prepare for catastrophic events in which residents cannot expect to be rescued by officials. In the case of Sendai, this is based on the experience of the 2011 Great East Japan Earthquake. In the U.S. Pacific Northwest, a similarly scaled magnitude 9.0 Cascadia Subduction Zone earthquake is expected to impact the Seattle region in the foreseeable future. Emergency managers anticipate that communities will need to be self-sufficient for up to three weeks following such an event. Drawing upon the results of member surveys of both organizations, we highlight similarities and differences in participant roles and lessons learned as well as feedback on specific successes and challenges. In particular, we focus on the ways in which the two initiatives leverage networks of partnerships with government agencies, local institutions, and other place-based organizations and the processes by which they conduct trainings, build collective capacity, and seek support for their work. In sharing the practices and strategies of these two exemplary place-based preparedness initiatives, our findings suggest strategies for better supporting place-based disaster preparedness efforts more broadly.

Public health emergencies, ranging from those caused by natural hazards to technological disasters, are increasing in their scale and impact. As these emergencies unfold, public health leaders must make critical decisions with limited or conflicting information, coordinate efforts amongst many groups of people, and maintain public trust. Now more than ever, opportunities to foster flexible and adaptive leadership skills in the public health workforce are imperative. While crisis leadership trainings exist, many of these programs emphasize individual professional development and may neglect opportunities to strengthen organizational capacity and collaborative response. Moreover, many programs are offered nationally and do not account for local or regional contexts. In response to this growing need, the Northwest Center for Evidence-Based Public Health Emergency Preparedness and Response at the University of Washington designed and delivered an inaugural Crisis Leadership Institute (CLI) that invited a cohort of 16 public health practitioners from local, state, Tribal, and Indigenous organizations and agencies across Region 10 to participate in a series of six modules, each focusing on a component of crisis leadership. The CLI sought to foster opportunities for mentorship, professional network growth, and peer learning. The evaluation of the inaugural CLI employed mixed-methods to assess individual knowledge change of module learning objectives, self-reported capacity to affect organizational change, as well as lessons learned for future iterations of the CLI. Results from the evaluation found increases in knowledge gained for all learning objectives and intent to incorporate content from the CLI into organizational practice by the majority of participants.

Disasters alter the natural environment and the overall functioning of a community, leading survivors to a decision point: should they stay in the community or relocate elsewhere? Past studies identify how place attachment, risk perception, age, insurance coverage, and familial ties impact survivor decisions to move after a disaster. However, most of these studies have been conducted on homeowners, leaving behind renters, who make up approximately 34 percent of United States households and contribute significantly to their communities.  Understanding renters' post-disaster decision-making process is critical to ensuring an equitable post-disaster housing recovery that will help survivors return to their communities. Building on the modified push-pull theory of migration, we investigate how place attachments, risk perceptions, and other external factors contribute to the post-disaster residential choices of renters after wildfire. We apply photovoice, a participatory research method, with renters from Butte and Sonoma counties in Northern California. Initial findings suggest that social capital, place identity, place dependence, risk of recurrence, and insurance have an outsized influence on the overall decision-making process of renters. Policy implications from the primary analysis include an embargo on evictions and rent control after a disaster, transparency in environmental remediation after fires, funds for mitigation of rental units, and protections from utility price increases after disasters.

Coastal neighborhoods are showing various spatial forms 14 years after Hurricane Sandy. While some show high density with elevated houses, others remain vacant lots without much rebuilding. Transformed spaces are the result of housing adaptation policies used since the reconstruction phase of Sandy, including the Housing Elevation Program aligning with the Federal National Flood Insurance Program (Build It Back Program: NYC) and the Buyouts and Acquisition Program (NY Rising Program: NY State). We examine the effects of housing adaptation policies on flood-affected neighborhoods, with a focus on four neighborhoods along the east coast of Staten Island. We use mixed methods: Statistical analysis using publicly available data (e.g., spatial data providing NYC's land use and building elevation information) and qualitative inquiry using data acquired through semi-structured residents' interviews in the targeted neighborhoods. Our quantitative analysis targets high-hazard risk areas (Special Flood Hazard Areas) and suggests that, first, there are no relationships between the areas bought-out and the degree of flooded areas in a neighborhood, and second, larger numbers of elevated houses are found in neighborhoods with a high degree of flooded areas. This reveals that the scale of damage is not a factor in deciding whether to stay or leave. Initial analysis of interview responses confirms that most residents made decisions based on their financial situation. The interviews also highlight that approximately one-third of the residents have moved into the targeted neighborhoods post-Sandy at different times in the past years.

Environmental disasters are increasingly exposing the uneven mental health burdens borne by families in socially and geographically marginalized regions, yet the experiences of rural Appalachian communities remain largely absent from the research literature. The flooding caused by Hurricane Helene in Western North Carolina offers a critical opportunity to examine how place-based vulnerability, pre-existing trauma, and climate-related perceptions shape the post-disaster well-being of parents and caregivers of young children. we employ a mixed-methods design, utilizing an online questionnaire administered to young Appalachian families impacted by Helene. The survey captures storm experiences, risk perceptions, protective actions, trauma histories, and climate attitudes, and links these data to indicators of physical and social context across the region, including community-level measures of vulnerability and exposure. Using inferential and geospatial analyses, we explore how these factors intersect and vary across geographic settings. Early results indicate meaningful differences in experiences and stress responses that appear to be tied to both local environmental conditions and individual histories of adversity. By foregrounding the perspectives of young families in an understudied mountainous region, this work advances theoretical and empirical understanding of disaster vulnerability, contributes to emerging conversations about climate impacts in areas previously viewed as climate refuges, and offers actionable insights for preparedness, mitigation, and long-term recovery planning. The project ultimately aims to support community partners and practitioners working to reduce mental health disparities and strengthen resilience among children and caregivers following extreme weather events.

Although enhanced building codes are widely recognized for their potential to reduce hurricane-related losses, there remains a limited understanding of how equitably these benefits are distributed across communities. In this study, we apply spatial regression methods to investigate social disparities in wind-damage reduction under a hypothetical scenario involving the combined enforcement of four mitigation strategies: improved roof deck attachment, secondary water resistance, installation of shutters on doors and windows, and strengthened roof-to-wall connections, across 708 census tracts within the Louisiana Coastal Zone. We analyze four tract-level outcomes: absolute and percentage reductions in expected annual structural damage and expected annual damage in dollars. Using a dataset that links damage reduction estimates to sociodemographic data from the 2018–2022 American Community Survey and the 2020 Decennial Census Demographic and Housing Characteristics file, we assess associations between multiple social vulnerability indicators and the four damage reduction outcomes. Preliminary results indicate larger absolute and percentage reductions in tracts with higher proportions of non-Hispanic Black residents, college-educated residents, and owner-occupied housing, and lower proportions of rural residents. Our findings suggest that while building code enhancements may yield substantial benefits for some communities, including those with higher concentrations of Black residents and renters, such benefits may not equitably extend to all vulnerable populations, such as tracts with larger shares of rural residents. Extending prior exploratory environmental justice analyses, we contribute a more rigorous assessment by applying spatial regression to measure how social vulnerability indicators are associated with mitigation benefits.

When, how, and why do people living somewhere decide that it is time to pick up and leave? In this paper, I focus on two climate-driven disaster events: Hurricane Helene in western North Carolina, and the Hermit's Peak-Calf Canyon Wildfire in Northern New Mexico. I recruited three major groups: those who moved, those who stayed, and community leaders with a stake in disaster management, response, and recovery. I conducted 18 interviews, with a total of 22 people. Interviews covered themes surrounding the experience of disaster events and recovery, community change, disaster preparedness and response, climate change, and adaptation. Based on these data, I suggest that several interconnected factors lead to the decision to migrate or not. As one disaster event occurs, the forces that shape emergence shape decisions to leave or stay, mitigated through personal attachments, institutions, and physical forces.

The National Institute of Standards and Technology (NIST) has recently concluded a project that enhances the capacity of lifeline infrastructure system owners/operators to implement actions that accelerate the recovery of key services after earthquakes. The framework is published in two NIST Special Publications, which provide an overview of functional recovery performance, explain complex system operations, outline procedures for identifying enhancements and upgrades to current systems, and present case studies illustrating improvements to post-earthquake recovery outcomes. The framework is the result of a contract with the Applied Technology Council, which convened a project technical committee and a project review panel, with leadership from Dr. Craig Davis, to integrate existing information and develop best-practices guidance. This framework is applicable to multiple types of infrastructure systems and is expected to be extended through subsequent development to multiple hazards. The framework makes a novel contribution to the performance design of infrastructure systems by proposing a process for socio-technical coordination that integrates technical asset design with actions to promote effective behavior and planning across organizational groups. Therefore, this effort is at the forefront of advancing functional recovery performance for both building and infrastructure systems by introducing system-level procedures that operationalize the assessment of organizational actions key to the success of enhanced performance for physical systems.

Standard cost-benefit analysis systematically underprices coastal adaptation policies by ignoring catastrophic tail risk, leading communities to dangerously underinvest in flood protection. In this study, we develop a machine learning framework that analyzes policy optimization with hidden costs. We assume that when catastrophic risk is properly valued, the timing and structure of federal program support matter more than dollar amounts. This has direct implications for programs such as the Federal Emergency Management Agency's Hazard Mitigation Assistance. In testing this assumption, we address two questions: (a) Does standard cost-benefit analysis systematically misrepresent adaptation policies by undervaluing catastrophic tail risk? (b) How do federal program features (subsidy timing, grant structure, and information provision) alter the value of delaying versus accelerating project investments? We develop a computationally tractable machine learning framework trained on simulations combining sea-level rise projections with statistical models of extreme storm surge events, then apply tail-risk pricing to separately value catastrophic exposure of alternative strategies. This sheds light decision problems at realistic scales while ensuring rare but extreme events are included in adaptation strategies. The framework unfolds in two steps. First, we identify strategies that minimize expected costs. Second, we price each strategy by valuing full risk exposure, including catastrophic tail events. This study offers guidance for policies accounting for low-probability, high-consequence events that are not often emphasized in conventional adaptation planning.

Federal, state, and local institutions are increasingly expected to "advance equity" in preparedness, mitigation, response, and recovery, but they often do so without shared conceptual boundaries, consistent operational definitions, or common standards for evaluating progress. This fragmentation poses a challenge for disaster governance: when equity is treated as an aspirational goal rather than a clearly specified construct, coordination across agencies weakens, accountability becomes difficult to establish, and opportunities for collective learning are limited. We address this gap by developing an ontology-based framework that treats equity as a multi-dimensional policy construct with clearly defined components, relationships, and decision-relevant implications. Drawing on a systematic synthesis of public administration, policy, and disaster governance scholarship, we construct an equity ontology that nests core dimensions (distributive, procedural, structural, and recognitional/restorative) alongside their normative rationales and the policy mechanisms through which they are commonly operationalized. The ontology is translated into a mutually exclusive and exhaustive codebook that enables consistent identification of equity claims and their associated implementation commitments within policy texts. The framework is applied empirically to Federal Equity Action Plans developed under Executive Order 13985, many of which directly shape disaster preparedness, mitigation, and recovery programs. The analysis reveals recurring patterns in how equity concepts are blended, simplified, or unevenly operationalized across agencies. By providing shared analytic infrastructure, this framework supports coalition-building in disaster risk reduction by enabling clearer purpose, stronger coordination, and more consistent evaluation of equity commitments across the disaster governance system. 

The Ministry of the Interior and Safety installed and operates automatic discharge measurement equipment, known as the Small Stream Smart Measurement System, at 880 small streams nationwide for flood management purposes. Although this framework for small-stream flood management has been established, further research is required to verify the accuracy of the measured data and to improve the maintenance and management of field-installed measurement equipment. It remains necessary to examine how actual field conditions affect equipment performance when installed in real small streams. Accordingly, we conducted field measurement experiments in 2025 at small streams equipped with the smart measurement management system. Discharge measurements obtained using an Acoustic Doppler Current Profiler (ADCP) were compared with data collected at the same cross-sections by the installed equipment. A comparison of flood-time measurements at Daeyeoncheon Stream in Busan showed similar velocity distribution patterns across the cross-sections, indicating that the automatic discharge measurement equipment can achieve sufficient accuracy under field conditions. In addition, field inspections conducted at 21 small streams during this study identified various operational and maintenance issues. Most notably, to ensure measurement accuracy, maintenance activities such as the removal of in-channel vegetation and periodic cross-sectional surveys to account for changes in channel bed geometry are essential. Looking ahead, continuous verification of the accuracy of field-installed measurement equipment through ongoing field experiments, combined with systematic maintenance inspections, will contribute not only to real-time flood management but also to improved accuracy of small-stream flood prediction technologies through the accumulation and utilization of long-term datasets.

In this study, we analyze how a local society and economy recover after years of vacancy caused by a post-disaster evacuation order and explore the tipping point measured by the speed of recovery. In the theory of spatial economics, temporary shocks to the population are expected to have a fatal impact on regions already experiencing population decline. However, rapid reconstruction may prevent such a tragic outcome. Using the great variation in the reconstruction speed in the extensive former evacuated zone of the Fukushima nuclear power plant accident following the Great East Japan earthquake and a 500m mesh census dataset, we conducted a series of quasi-experimental analyses on the effect of the reconstruction speed on the long-run population and business densities. We found that the adverse impact of the six-year delay in lifting the residence restriction offset the positive inertia the hubs have. Our data also indicated that, unlike reconstruction speed, the quality of recovery had little impact on long-term population size, provided that the essential infrastructures were restored. Furthermore, we observed that hubs created by the faster lift of the restriction became long-lasting hubs only for business densities, whereas pre-disaster hubs remained a major predictor of the long-run population and business densities. The central policy implication from this study is that swift reconstruction is essential, and the residential restriction should be lifted in less than six years.

Human-wildlife interactions are commonly attributed to the spread of zoonotic diseases; however, climatic-related events such as extreme heat associated with high temperature, floods, deforestation, wildfire, and drought are noticeably contributing to the emergence and spread of zoonoses due to pushing wildlife closer to humans and livestock. In Nigeria, zoonoses are a significant public health challenge besides being ranked among the top ten globally with the highest burden of infectious diseases, critical zoonoses including rabies, avian influenza, Ebola virus disease, swine influenza, and anthrax are being reported. The approach to surveillance, monitoring, and control somewhat restricts identification of zoonotic causative pathogens largely to animals, which leads to the question: could there be a knowledge gap at the intersection of zoonoses with climatic hazards? This study highlighted the intersection of climatic and environmental factors with the emergence and spread of zoonoses. Adopting a scoping literature and document review of publications on the emergence and control of zoonotic diseases in Nigeria, the study found that a knowledge gap exists on the intersection of climate hazards with the emergence and spread of zoonoses. The three implications of this study include (a) awareness creation on the intersection of certain climatic hazards with the increasing zoonotic diseases, (b) spotlighting of extreme weather events as driving factors in the emergence and spread of zoonotic diseases, and (c) offer of actionable response in targeting the control of zoonotic diseases and transmission preparedness.

Baltimore County, Maryland, faces growing risk from climate-related hazards such as flooding, severe storms, and hurricanes. Proximity to the Chesapeake Bay and extensive coastal and riverine areas heighten exposure, with disproportionate impacts on socially vulnerable residents, including low-income households and people living in subsidized housing. These groups often experience delayed or incomplete recovery due to systemic barriers. Improving interagency coordination is therefore essential to achieving equitable housing recovery outcomes. I examine the current state of housing recovery coordination in Baltimore County, identify barriers to effective collaboration, and explore strategies to strengthen support for at-risk populations. Using a qualitative design, the research combines a structured review of county disaster management plans with semi-structured interviews with key informants across emergency management, housing and human services, public health, and community-based organizations. Interviews focus on how agencies coordinate before and after disasters, where breakdowns occur, and which practices promote effective cross-sector alignment. Preliminary findings indicate broad agreement on the importance of collaboration, yet persistent challenges continue to constrain its implementation. These include fragmented communication, unclear or overlapping roles, limited staff capacity and resources, and inequities in access to recovery assistance. Results highlight the need for policy and procedural reforms that clarify responsibilities, improve information sharing, and center equity in recovery operations. Findings will inform local policymakers, emergency planners, and housing advocates seeking to strengthen interagency collaboration and advance more equitable post-disaster housing recovery.

Bangladesh is increasingly experiencing severe heatwaves, which remain an overlooked climate threat, notably harming people with disabilities because of structural, environmental, and social barriers. We explore how people with disabilities in urban and suburban areas manage extreme heat events through a decolonial framework that centers lived experiences, community knowledge, and daily forms of coping and adaptation. We collected qualitative data through in-depth interviews with participants, including individuals with mobility, intellectual, and speech disabilities, as well as caregivers, when direct communication was not possible. In-depth analysis of the findings highlights interlinked patterns, including impacts on physical and mental health; limited access to mobility, healthcare, and essential services; dependence on family and community support; and the role of cultural and spiritual practices in strengthening resilience. The study illustrates that accessible, community-based knowledge and rights of people with disabilities are vital in adapting to heatwaves.  Considering heatwaves as formal disaster risks and embedding inclusive measures for people with disabilities into planning are significant for climate resilience. 

In this presentation, we summarize recent research on the use of scenario planning as a tactic for understanding and managing uncertainty arising from natural and technological hazards and threats. In addition to summarizing its use in military, corporate, and government settings, some examples and projects specific to climate, transport, urban development, and hazard mitigation will be described and evaluated using familiar plan quality metrics and indicators (fact base, authority, engagement, outcomes, etc.). The purpose of this review is to stimulate interdisciplinary discussion and consideration of quantitative and qualitative data as well as attention to values and consideration of equity, fairness, and social justice. A key component of the research involves understanding pathways for learning and adapting systems, processes, policies, and structures relevant to safety, security, resilience, and sustainability. We also report on recent projects from diverse domains and settings focused on scenario planning and community building. While the review draws most heavily on urban planning and emergency management perspectives, connections to futures studies, economic modeling, and geospatial analytics and decision-support tools and platforms help to understand the trajectories, prospects, and perils of scenario planning for building and sustaining resilient communities.  In addition to climate scenarios, examples from flooding and responses following hazard events and recovery operations provide further insights for researchers interested in understanding and applying scenario planning as a potent research approach.

Timely information from rapid needs assessments during a disaster can improve situational awareness and guide disaster response and recovery efforts. The Centers for Disease Control and Prevention's Community Assessment for Public Health Emergency Response is a household-based survey that provides information about a community, allowing public health leaders and emergency managers to make informed decisions. While Community Assessment for Public Health Emergency Response have proven successful, the traditional 30x7 design is not always feasible or ideal. Centers for Disease Control developed and field-tested new rapid needs assessment methodologies as alternatives to Community Assessment for Public Health Emergency Response, field-testing four approaches: combined stratified systematic and purposeful sampling, simple random mail-based sampling, targeted sampling based on damage assessment data, and a combined 1- and 2-stage cluster sample method. The stratified sample yielded representative data but required extensive planning and analysis. The one-stage cluster sample reduced selection bias in a cost- and time-efficient manner but had lower internal validity and necessitated oversampling of clusters. Purposeful sampling allowed for intentional selection, but integrating these results into the broader rapid needs assessment proved challenging. The multitiered approach offered flexibility for tailoring to specific populations, but complicated weighting and analysis. Based on lessons learned from these field tests, Centers for Disease Control continues to refine rapid needs assessment guidance for situations where this approach is not the best fit. This session will share the new methodologies, their advantages and limitations, potential uses, and lessons learned.

The COVID-19 pandemic and escalating climate-related disasters have underscored the interdependence of human, animal, and environmental health, elevating One Health from a conceptual framework to a practical governance challenge. This paper presents the first systematic, cross-state comparison of One Health initiatives across all 50 states, examining how states institutionalize One Health principles and integrate them into public health preparedness, disaster response, and resilience planning. Using a mixed qualitative–comparative design, I analyzed state statutes, executive orders, preparedness plans, and agency coordination structures. The analysis reveals substantial variation in the scope, durability, and political embedding of One Health initiatives. Some states have formalized One Health through interagency task forces, statutory mandates, and sustained funding streams, while others rely on ad hoc coordination activated primarily during crises. These differences matter for preparedness: states with institutionalized One Health arrangements demonstrate greater capacity for early threat detection (e.g., zoonotic surveillance), cross-sector data sharing, and coordinated decision-making during compound disasters. Conversely, fragmented governance structures often reproduce federalism-driven conflicts, jurisdictional ambiguity, and competition for authority during emergencies, undermining resilience. Situating One Health within the literature on disaster politics, the paper shows that adoption is shaped not only by epidemiological risk but also by political institutions, crisis histories, and intergovernmental relationships. The findings contribute to debates on public health preparedness by demonstrating that effective One Health implementation is less about technical capacity alone and more about institutional design, political commitment, and the management of interagency power during disasters.

In this study, I examined how United States state governments responded to the COVID-19 pandemic, with a focus on the institutional challenges, collaborative successes, and future opportunities for improving public health preparedness. Using a qualitative research design, I analyzed elite interviews conducted between 2020 and 2023 with senior state-level bureaucrats responsible for pandemic decision-making. Interview data are supplemented with systematic analysis of state public health preparedness plans, statutes, and policies, as well as key federal and intergovernmental documents from agencies and organizations such as the Association of State and Territorial Health Officials and the National Conference of State Legislatures. I used interview data from 46 bureaucratic elites across 25 states and all 10 Federal Emergency Management Agency regions, including officials from state public health agencies, state emergency management agencies, and governors' offices. Findings revealed substantial variation in relationships between emergency management and public health agencies across states. Prior to COVID-19, coordination across agencies–and with governors' offices–was minimal, with limited awareness of public health emergency capacities. Public health responses were frequently siloed both across agencies and within public health departments themselves, contributing to early communication failures during the pandemic. Where improvements occurred, they were often driven by individual bureaucratic "policy entrepreneurs" who cultivated informal relationships to overcome institutional fragmentation. Persistent challenges include workforce burnout, unclear roles and responsibilities, legislative backlash limiting emergency powers, inadequate and unsustainable funding, and the absence of durable public health emergency response capacity and infrastructure.

This study explores how rural communities in Bangladesh understand and respond to flood early warning systems. We conducted the study in four flood-prone villages in Sirajganj district, Bangladesh. These villages were selected due to their repeated exposure to seasonal flooding and their diverse social and economic characteristics. To understand how community members perceive and respond to flood early warning systems, we used qualitative research methods. Data was collected through 20 in-depth interviews and five focus group discussions. Findings suggest that many respondents had heard of flood early warning systems, but very few received clear, timely messages. Most did not understand the warnings or take action. Instead of relying on formal systems such as interactive voice response or water-level gauges, people used indigenous knowledge. They observed natural signs such as river color, wind direction, and rainfall patterns. These methods were passed down through generations and often proved more useful than official alerts. Community members also shared information through neighbors, relatives, and local markets. The study found that formal warning systems do not meet the community's needs. Many people lack access to phones, TVs, or the internet. Messages are unclear or not local enough. To improve flood preparedness, early warning systems must include local knowledge and be designed with the community. This will help people take action and reduce damage during floods.

Structural constraints, funding availability, mismatched timelines, job insecurity, and public engagement fatigue undermine practitioners' flood mitigation work and make decision-making tools such as scenario visualizations irrelevant. In this case study, I revealed and discussed these factors from the perspectives of practitioners engaged in flood mitigation projects in an East Coast neighborhood. Through an institutional trust lens, preliminary analysis indicated that these factors control practitioners' abilities to build timely progress on flood mitigation projects. In turn, slow progress reinforced public skepticism and distrust, undermining the credibility of government agencies and external actors. The study site was a low-lying, urbanized area built on former wetlands, facing escalating flood risk from a history of failed flood support attempts and inconsistent investment. I interviewed eight participants using a semi-structured format and open-ended questions to explore how visualizations fit into practitioners' work during the mitigation project. Participants were selected based on their active involvement in community flood mitigation efforts. Interview data is analyzed using inductive thematic analysis. Practitioners expressed that scenario visualizations are not prioritized or valued during the current phase of work, and that incorrectly timed visualization-use may confuse or frustrate residents, reinforcing distrust. Instead, practitioners expressed that a trusted messenger has more influence on the credibility of a visualization than the agency they represent. This study has significant implications for flood risk planning professionals, reminding experts not to lose sight of the value of time, process, and relationship-building at the individual-to-community scale for building trust and momentum.

Research has shown that climate-induced hazards are occurring more frequently; however, studies show that emergency management responses to disasters serve people disproportionately, especially along lines of race. The emergency management community has infrequently tracked data on race and is thus unable to fully account for how communities of color are affected throughout the disaster cycle. Given these inconsistencies, we ask what are the harms of being race-neutral in practice and policy and assess the interconnectedness of being race neutral and ahistorical.  We conducted 48 interviews with community members and emergency management personnel, transcribed and analyzed these interviews, and identified 124 emerging codes. Emergent themes from these interviews highlighted the deflection of race and absence of historical knowledge. Findings indicate that emergency management personnel often fail to recognize the historical context of harm and disinvestment in communities of color. The historical racism and disinvestment continue to impact how communities respond to environmental hazards negatively as emergency management fails to include race in their data. These findings emphasize the need to build the bridge between emergency management and the community in various ways, including depending on community-based knowledge, networks to inform efforts, and to examine disaster planning with special attention to race and equity.

Project delivery methodology remains critical to construction management success, yet Post-Disaster Reconstruction presents unique challenges, including time compression, community consultation requirements, and urgent communication needs. While collaborative approaches like design-build logically address these demands, a persistent problem exists in aligning client values with procurement-related activities during construction. Current research lacks understanding of how owner values shape contractor selection and Request for Quotation/Request for Proposal (RFQ/RFP) development, with existing client value indices geographically limited and failing to provide comprehensive models for U.S. disaster recovery contexts. In this research, we address two main questions: how global post-disaster recovery values transfer to U.S. reconstruction contexts and how these values are communicated during the procurement phase of design-build construction projects. By surveying approximately 30-50 U.S.-based owners and developers, we recontextualize global value indices for American PDR settings. We then use research charrettes to map values to procurement practices, making it possible to build a model of how client expectations are met during construction procurement. Our findings include a U.S.-specific client value hierarchy emphasizing recovery challenges in the United States, as well as systematic patterns revealing value concentration in evaluation criteria rather than technical project specifications. Specifically, it is expected that these evaluation criteria are reflected in the RFQ/RFP, though other methods may be used to communicate owner values during the procurement process. This research contributes a comprehensive framework for mapping client values to procurement elements in design-build PDR projects, thereby providing actionable tools for embedding collaborative delivery principles and improving disaster recovery outcomes. 

The increasing frequency and intensity of climate disaster necessitates planning and governance that enable effective response and support equitable recovery. Permitting manufactured housing in the floodplain offers insight into the complex dynamics of disaster recovery. Recent studies highlight challenges within permitting practices, planning rationales, and the need for more effective disaster governance. Our research asks: How do permitting processes in flood hazard areas influence community resilience and what role does disaster planning play in the construction of risk? Focusing on Buncombe County, North Carolina, this study examines permitting dynamics before and after Hurricane Helene, with particular attention to new residential and manufactured housing permits within the designated flood hazard zone. GIS mapping of pre-/post permitting of manufactured housing in the flood plain, semi-structured interviews with regional planning officials, and examination of policy shifts through review of recovery plans reveal three key findings. First, how immediate disaster impacts can compound longstanding and preexisting socio-economic inequities; second, how complex governance topologies complicate disaster recovery efforts; and third, how new risks are manufactured in relation to regulatory frameworks that are not designed to address compounding crises. These findings suggest that disaster preparedness planning and governance play a critical role in the formation of disaster risk, response, and recovery. We contend that developing integrated disaster governance frameworks, utilizing data-driven tools to anticipate compounding risks, and aligning disaster preparedness with broader resilience goals are essential steps toward more just recovery.

Since the rise of the social vulnerability paradigm in disaster/hazard studies, a robust body of theories and evidence has emerged recognizing social capital as a critical factor in promoting community hazard resilience. Meanwhile, land-use planning possesses much agency to enrich social capital. In particular, a wealth of studies has established causal relationships between specific physical-environment features (e.g., walkability) and the capital. Nonetheless, there has not been any synthesis of this burgeoning literature. This gap generates multiple problems, one being policymakers' uncertainty about how to shape the built environment for social capital when scholarly evidence does not align. We address this gap by carrying out a systematic literature review of the planning literature on social capital, answering three questions: (a) what planning literature has studied about social capital, (b) what physical-environment features whose influences on social capital have been empirically tested, and (c) in what direction these influences are. When addressing questions (b) and (c), wherever the evidence does not align, this review refers back to the samples and the designs of the primary studies. Based on that and aided by relevant theories, it derives the reasons why the inconsistent evidence emerges, thereby generating a more detailed specification of when a given physical-environment feature may promote/weaken social capital. We show that there has been consistent evidence verifying features such as the perceived quality of communal spaces as beneficial for social capital. Features such as development density and park accessibility, meanwhile, have a non-linear relationship with social capital.

Wildfire disasters are increasing worldwide. In Chile, this trend is driven by the effects of climate change, changes in landscape management, and formal and informal urban expansion into the wildland–urban interface (WUI). Over the past decade, wildfires destroyed hundreds of thousands of hectares in this country, resulting in severe human, material, and environmental losses. While existing research has examined the relationships between WUI expansion and wildfire risk at regional-metropolitan scales in Chile, comparatively little attention has been paid to the morphological characteristics of affected areas, their evolution over time, and their role in shaping damage patterns. We examine two of the most destructive WUI wildfire disasters in Chile: Viña del Mar in 2024, which caused 138 fatalities, destroyed more than 8,000 buildings, and affected approximately 11,000 hectares; and Penco–Lirquén in 2026, where fires resulted in 21 deaths, impacted over 1,800 constructions, and burned nearly 25,000 hectares. Using a mixed-methods approach, we reconstructed the historical expansion of the affected areas and characteristics of their built environments. These attributes were then compared to observed damage levels, as well as early resident-led reconstruction processes. The findings highlight how terrain, urban patterns, and building characteristics can be associated with differential damage outcomes, providing actionable insights for improving urban planning instruments and building standards for wildfire risk reduction in Chile and other wildfire-prone regions. 

Coastal hazards, like flooding and erosion, are growing more frequent and intense due to changing storminess patterns and rising sea levels. In Kiholo Bay State Park Reserve on Hawai’i Island, these hazards threaten cultural and historical legacies, vulnerable ecosystems, and active recreation and cultural practice opportunities. Effective adaptation requires both localized projections of future environmental changes and a clear understanding of community priorities and planning contexts. Here, we present a convergent approach that integrates probabilistic coastal hazards modeling and focus group and interview data to support ongoing planning processes for the Kiholo community. Future hazard impacts are simulated using a climate emulator to produce probabilistic hydro-met-ocean conditions, which are then downscaled to local impacts using a nearshore wave model and combined with climate scenarios. Outputs from this model framework are used in hazard exposure analyses that track community-relevant metrics related to the resilience of culturally and ecologically important sites. These metrics are designed based on focus group and interview data, in which community members described their personal connections to Kiholo, observations of change, and hopes for the future. Community-identified values, locations, and concerns directly shape hazard impact analysis and communication. By integrating quantitative hazard projections with qualitative place-based knowledge, this approach advances understanding of coastal risk and community relationship to place, while producing decision-relevant information that can be more readily incorporated in local adaptation planning.

The Rhode Island Coastal Hazards Analysis, Modeling, and Prediction (CHAMP) system is a hyper-local impact forecasting tool developed to support and protect the critical infrastructure of Rhode Island. Studying CHAMP provides insight into effective communication of operational forecast systems that can be applied to a wide range of forecast tools. CHAMP is unique because it uses end-user generated and vetted qualitative data to develop Hazard Consequence Thresholds (HCTs). HCTs are the measurable point at which a hazard will cause a consequence. This approach creates a "closed loop" system that increases transparency and legitimacy perception of its forecasts amongst emergency managers and planners. Since CHAMP was born out of research capability, not expressed need, I conducted semi-structured interviews of key CHAMP project members to illuminate the opportunities and challenges the team faces to improve prediction and communication to end users. Through inductive thematic analysis, the results reveal that the same qualities end users praise are also the source of CHAMP's main challenges, particularly the feasibility of managing end-user-driven data and securing funding. Other challenges involve complex modeling parameters, legal liabilities, and reliance on specialized expertise. While all team members agree on the importance of improving communication, the priority tasks that each member identifies to do so link closely with their role on the CHAMP project, demonstrating the need for a conversation to create a cohesive communication plan. 

The purpose of this research is to explore how nonprofit managers mitigate the potential physical and psychological risks faced by volunteers and promote their safety and well-being during disaster response. While there is broad consensus that nonprofit organizations and volunteers play a critical role in disaster response and recovery, post-disaster environments are inherently hazardous and can place substantial strain on the physical safety and mental well-being of volunteers. Despite the significant role nonprofit managers play in organizing, protecting, and supporting volunteers, existing research paid limited attention to how managers perceive volunteer risk and take proactive strategies to mitigate it. Using 22 interviews with nonprofit managers engaged in disaster response and recovery in the Texas Gulf Coast region, I employ grounded theory and inductive coding to identify the strategies of nonprofit managers in avoiding liability issues and promoting the well-being of volunteers. By incorporating risk management into volunteer management, this study first makes theoretical contributions by providing the insights of nonprofit managers and exploring the psychological dimensions of volunteer management in disaster response. It also provides practical implications for nonprofit managers and emergency managers by outlining strategies to protect volunteer well-being, which in turn strengthens the resilience of nonprofit organizations and the community.

As federal disaster recovery funding has expanded in scale, duration, and administrative complexity, understanding how governmental capacity shapes recovery performance has become a critical policy concern. We examine how state and local governments in the United States differ in their capacity to manage federal CDBG-DR funds and how these differences influence disaster recovery outcomes. Using quarterly data from Quarterly Performance Reports covering federally declared disasters in 2003-2023, we construct comparable indicators of fund-management performance, including ratios of disbursed-to-obligated and expended-to-obligated funds, and measures of governmental attributes, expenditure stability, and program completion. We apply covariance-based Structural Equation Modeling (SEM) to estimate latent constructs of governmental capacity and recovery outcomes, using separate and multi-group models for state and local governments. Results consistently show that governmental capacity has a positive and statistically significant effect on recovery outcomes across all model specifications. However, the magnitude and stability of this relationship vary systematically by government level. State governments exhibit a strong and highly stable capacity-outcome relationship, indicating a near one-to-one translation of administrative capacity into recovery performance. In contrast, local governments display weaker and more heterogeneous capacity returns, with recovery outcomes more sensitive to expenditure volatility and implementation constraints. Robustness checks, including outlier removal and alternative model specifications, confirm that these differences are not driven by extreme cases but reflect underlying institutional dynamics. Overall, the findings highlight persistent, scale-dependent capacity disparities in disaster recovery governance and underscore the need for level-specific federal program designs that prioritize execution stability and targeted administrative support for local governments.

Los Angeles County residents face various health and safety concerns in the aftermath of the 2025 Eaton wildland-urban interface fire, which damaged or destroyed over 10,000 homes, businesses, and community spaces. Thousands of surviving standing structures were inundated with toxic ash carried by the wind, impacting communities miles downwind from the fires. We explore the extent to which indoor contamination is sufficiently reduced through remediation practices using Eaton Fire Residents United (EFRU) data. EFRU is a coalition of concerned residents and scientists formed in the wake of the fire, whose mission is to ensure a safe, just, and transparent public health recovery process for the Eaton Fire-affected communities by using real-world data to drive recovery. The EFRU indoor contamination mapping project tracks reported contamination levels of Eaton Fire-generated heavy metals and pollutants in the interiors of standing structures, as tested by professional industrial hygienists before and after remediation. Utilizing data from over 350 pre- and/or post-remediation environmental testing reports, we examine the extent of indoor contamination, including wildfire debris, asbestos, and CAM-17 heavy metals, within and outside the burn scar. Analysis shows widespread contamination after the fire, with 100% of homes testing positive for wildfire debris and lead, often at levels a thousand times over the Environmental Protection Agency Dust Lead Action Level, requiring abatement. Even after remediation, six in ten homes are not safe to inhabit due to persisting lead and/or asbestos levels. Numerous health crises may unfold if this remains unaddressed.

Along with a history of disasters, Japan has an active culture of passing down disaster experiences. As part of efforts to make sense of the massive and complex Great East Japan Earthquake, tsunami, and nuclear disaster of March 11, 2011 (hereafter 3.11), and unprecedented damage and loss, there have been many initiatives to tell the stories of disaster experiences and pass down lessons to future generations. Along with disaster storytellers and memorial facilities, more than 130 picture books that deal with stories of 3.11 have been published by large and small publishers, through crowdfunding, or self-published by authors.  Many authors or illustrators have personal connections to disaster-affected areas, and many books are based on true stories and people's experiences during and following the disaster. Most books focus on the experiences of the tsunami, although about 1/3 deal with the nuclear accident and its impacts. Common themes include stories of evacuation and disaster safety; loss and grieving; community bonds and hope for the future; and animals and nature. Focusing on the role of disaster images within 3.11 picture books, I consider how the 3 disasters (earthquake, tsunami, and nuclear accidents) are shown (or not) in the pictures, the style and size of the disaster images, and the relationship to the text. Through this visual image analysis, typologies of disaster images depicted in 3.11 picture books and correlations across narrative themes are clarified.

Since Klinenberg's seminal work on the Chicago heatwaves, it has been widely accepted that social isolation substantially increases vulnerability to heatwave-related mortality, particularly among older adults, due to the absence of social monitoring and timely emergency assistance. Strong social ties enhance resilience by facilitating communication, sharing cooling resources, and providing emotional and physical support during periods of extreme heat. Interventions that reduce social isolation have been shown to lower excess mortality during heatwaves, underscoring the protective role of community connections. Consequently, social connection programs and community-based networks are increasingly recognized as cost-effective adaptation strategies for mitigating health risks associated with extreme heat. Urban planning and housing policies that reduce social isolation, especially among older adults living alone, therefore play a critical role in long-term adaptation to heat-related risks.However, existing research highlights the complexity of how social networks influence adaptation. Moreover, many studies rely on aggregated mortality data, while everyday heatwave adaptation practices remain relatively understudied. In this study, we investigate how social network density influences adaptation to heatwaves, while accounting for contextual factors such as health status, sociodemographic characteristics, building type, ambient temperature, and thermal comfort in Warsaw, Poland. We conducted 300 interviews with residents aged 50, across three local climate zones. Data were gathered using questionnaires, a mobile application, and a temperature-measuring device, through which participants reported their thermal comfort over a three-week period. The study aims to inform and support the development of Warsaw's climate adaptation policy.

Strong local leadership and flexibility to experiment with erosion-control measures emerged as key enablers of adaptation in North Cove, Washington, while grant competition, regulatory barriers, and a sense of individualism constrained efforts. Understanding what enables and constrains community-based adaptation can help practitioners support rural communities facing coastal hazards with limited institutional support. This case study of coastal, rural adaptation to erosion in North Cove, Washington examines how residents have adapted to chronic erosion over more than a decade. We conducted 32semi-structured interviews with residents and practitioners that centered around lived experiences with erosion, adaptation decision-making, and perceived efficacy of adaptation projects.  Our thematic analysis also identifies how collective efficacy—people's perception about their collective abilities to overcome challenges facing their group or community—shapes the community's overall adaptive capacity. Effective community-led erosion projects have strengthened some residents' beliefs in their collective impact and motivated continued participation, while negative experiences with institutionally imposed regulations have reduced collective efficacy among others. Early findings indicate that adaptation outcomes are not shaped simply by access to resources, but by governance systems and relationships within and beyond the community. Residents possess critical place-based knowledge of hazards and hazard mitigation and are often responsible for the long-term maintenance and success of their communities' adaptation efforts. The dynamics in North Cove reveal tensions between top-down adaptation projects and locally grounded approaches, underscoring the need to more meaningfully incorporate local perspectives into adaptation projects. 

In this study, we highlight the role of response-phase, child-centered interventions as a critical yet often overlooked component of post-disaster recovery architecture. Our findings demonstrate that beyond food, water, and shelter, families require social and emotional support that helps restore routines, a sense of safety, and psychological stability for children. Traditional post-disaster response is organized around restoring physical infrastructure and meeting immediate material needs but far less attention is paid to the broader set of social and emotional infrastructures required to support children and families as they navigate disruption, displacement, and uncertainty. Using surveys and semi-structured interviews with families, volunteers, camp staff, and community partners, we examine the benefits of Project:Camp, a pop-up program that operated during the response phase to both Hurricane Helene and the L.A. wildfires. While damage to buildings like homes and schools manifests in a visible, tangible way, disasters erode equally critical forms of support systems including routines, childcare systems, and children's sense of safety and stability. As a result, the emotional and psychological safety needs of survivors are often treated as secondary concerns, addressed later in the recovery phase when there may be less available resources or attention. By providing structured, child-focused spaces during the response phase, Project:Camp offers a model for addressing the infrastructure gaps in disaster response systems that are typically oriented toward physical recovery of adults rather than the psychological and emotional wellbeing of children and their families.

Effective disaster risk reduction depends not only on technical knowledge, but on strong, trusted coalitions among researchers, decision-makers, and local knowledge holders. This need is especially pronounced in rapidly changing rural-urban regions, like the Treasure Valley, Idaho, where population growth, land-use change, and evolving energy-water demands are reshaping exposure to natural hazards while stretching governance and planning capacity. In this research, we use the co-development of alternative future scenarios as a coalition-building tool to support locally grounded decision-making and governance capacity. Focusing on interconnected energy and water systems, we brings together planners, utilities, natural resource managers, and other key stakeholders to examine current stressors, emerging vulnerabilities, decision tradeoffs, and governance dynamics that influence both resource management and hazard risk. Using a "listen-first" approach, we integrate engaged workshops, surveys, and semi-structured interviews to elicit place-based knowledge of energy-water stressors, coordination challenges, and decision constraints. These insights are integrated with regional datasets and machine-learning models to develop baseline scenarios of water use, energy demand, and land-use change. Through iterative engagement with an expanding stakeholder network, these baselines inform the co-development of alternative future scenarios that reflect different management and decision-making pathways. The resulting data tools and web-based visualizations illustrate how energy-water governance choices shape regional vulnerability and resilience. By centering local knowledge, co-producing decision-support tools, and fostering sustained collaboration, this work enhances regional capacity for hazard mitigation, response, and recovery. More broadly, it demonstrates how scenario-based research can function as coalition-building infrastructure, building long-term resilience by helping communities plan stronger futures together.

Flood emergency communication often relies on technology (e.g., digital flood risk maps, app alerts), posing accessibility barriers for isolated groups without stable internet or telephone service. Understanding information flows between local governments and residents in digitally-dependent systems can reveal communication bottlenecks that foster institutional distrust and contribute to disproportionate financial and environmental flood impacts in underinvested regions. This study examines how residents in flood-prone neighborhoods of the Lower 9th Ward and New Orleans East access, prioritize, and share flood emergency information across preparation, response, and recovery phases, and where existing systems fail to reach the most vulnerable populations. Using a community-based participatory research approach, trusted relationships and local expertise guided the co-development of research questions, recruitment strategies, and data collection methods. We analyzed quantitative community-based survey (n=181) and qualitative focus group (n=19) data deployed through trusted organizational networks to contextualize perspectives often underrepresented in broader surveying efforts. We compared these localized findings with citywide survey results (n=400) to identify shared and unique information needs, gaps, and priorities. Results indicate that long-time residents, particularly those shaped by events such as Hurricane Katrina, often rely on informal networks due to a degradation of institutional trust and a lack of localized flood emergency information. These dynamics suggest actionable strategies for cities, including leveraging trusted community partnerships and non-digital communication channels, and aligning official messaging with informal networks to ensure timely dissemination. Coordinating digital and non-digital communication approaches may strengthen the effectiveness and reach of institutional emergency communication efforts nationwide.

A host of effective behavioral variables such as biases and misperceptions, motivations and incentives, and group culture create challenges across coastal governance contexts. Traditionally, research has focused on what roles behavioral variables play in community members' decisions to participate or sustain engagement in governance practices. A need remains to decipher how government staff behave when designing, managing, and supporting practices that involve members of the public, and those from under-represented groups. Specifically, the research question driving this study was: What are planning and policy pathways to increase and sustain involvement for underrepresented groups in coastal governance? We used semi-structured interviews to engage with 16 local stakeholders on the Oregon coast. The interviews focused on capturing stakeholders' experience with coastal hazards planning, policies, and government effectiveness in hazard mitigation and preparedness. The interviews were recorded and transcribed for systematic coding and content analysis. Attitude biases were coded and will be considered when conducting final analysis. The research aims to provide clarity on how the community and government can work together to co-create planning and policy guidelines to ensure safety during a natural disaster, especially for disadvantaged communities. The data and analysis provide opportunities for improved hazard mitigation, community engagement, resources for disadvantaged communities, and government efficiency during a hazard event.

In this study, we present a systematic typology of wildfire misinformation narratives, with significant implications for post-hazard sensemaking and risk communication aimed at effective recovery and adaptation. Wildfires pose unique challenges for hazard risk communication due to the often ambiguous, contested, and politically charged nature of their ignition sources and causal pathways, creating conditions that fuel recurring misinformation. We conceptualize wildfire misinformation as patterned narrative forms that merge through rumoring during and after fire events. While prior studies have documented misinformation dissemination during hurricanes, floods, pandemics, and wildfires, less attention has been paid to the specific content and themes within distinct hazard contexts. To fill this gap, we employ topic modeling techniques to analyze wildfire-related social media discourse from recent United States wildfire events, including the 2021 Dixie Fire, the 2021 Marshall Fire, the 2023 Maui Fires, the 2024 Smokehouse Creek Fire, and the 2025 Eaton and Palisades Fires. The analysis reveals recurring thematic clusters forming the basis of a systematic typology of wildfire misinformation narratives. Preliminary findings indicate that these narratives are consistent across events and predominantly revolve around contested attribution, institutional mistrust, and moralized blame. By focusing on narrative structures rather than individual false claims, the typology aims to provide a framework for anticipating and responding to misinformation that persists across wildfire events. These insights contribute to risk communication strategies by identifying persistent misinformation themes to address in recovery efforts.

Gaps remain in understanding electric power restoration challenges for communities impacted by sequential disasters. In this study, we focused on Hurricanes Helene and Milton and the Florida counties that qualified for the Federal Emergency Management Agency’s public assistance categories A-G (an approximation of hurricane impact on utilities and infrastructure). The primary focus was to explore the factors that reduce or prolong electric power disruptions in counties that experience sequential hurricanes compared to other similar counties that experience only one hurricane. Using a difference-in-differences design, and available power restoration rates by county, we found that counties impacted only by Milton experienced an average of 5.6 days without power, while those that were hit first by Helene and then by Milton (12 counties total) experienced an average of 6.6 days without power. We are completing data collection at the census tract and county levels to conduct more sensitivity analyses, including additional measures of hurricane strength (e.g., wind swath and storm surge). We examine more granular levels of electric power restoration rates in select areas using the National Aeronautics and Space Administration's nighttime light data at the census tract level. Other variables of interest include electric power outage accounts and utility providers, degree of urbanization based on strength of commuting ties, and socioeconomic characteristics. The findings have practical applications for disaster management and adaptation planning and contribute to multiple strands of inquiry on compound hazards, infrastructure restoration and recovery, and methodological approaches to account for sequential spatial-temporal impacts of hurricane disasters.

More than a decade after the Fukushima Daiichi Nuclear Power Plant accident, evacuation orders have been lifted in many affected areas. Nevertheless, a substantial number of evacuees continue to live outside their hometowns without permanently returning or resettling elsewhere. I examine such post-disaster lives through the concept of commuting--repeated movement between evacuation destinations and places of origin that do not culminate in full return. Based on qualitative fieldwork with former residents of Obori district in Namie Town, Fukushima Prefecture, I explore how commuting has emerged as a sustained response to long-term evacuation. While official recovery policies largely frame recovery in terms of either return or permanent relocation, evacuees' experiences reveal a more complex reality shaped by radiation risk perceptions, memories of forced evacuation, and enduring attachments to land, work, and community. The findings showed that commuting enables evacuees to maintain social and economic ties to their hometown while avoiding full-time residence in areas perceived as unsafe or unstable. At the same time, this practice entails significant physical, financial, and emotional burdens, particularly as institutional support diminishes over time. By conceptualizing commuting as a durable form of post-disaster recovery rather than a temporary transitional stage, I challenge policy assumptions that equate recovery with spatial closure and settlement. It argues that long-term evacuation should be understood as an ongoing social condition requiring flexible and sustained support. Recognizing commuting as a legitimate mode of post-disaster life has important implications for designing recovery frameworks that better address uncertainty, risk, and prolonged displacement.

Public health emergency preparedness (PHEP) requires local health departments to interface with communities and community organizations to prepare for all hazards (e.g., wildfires, pandemics, and train derailments). However, a major knowledge gap exists about the facilitators and barriers to PHEP community engagement and how local health departments can implement evidence-informed community engagement practices. In this qualitative study, we examined successes and challenges to partnerships between local health departments and community-based organizations (CBOs) for PHEP, with a focus on Health and Human Services Region 10 (Alaska, Idaho, Oregon, and Washington). We conducted three virtual focus groups among 10 local health department PHEP coordinators and 12 virtual interviews with representatives from CBOs and asked questions about their partnership practices, how they define a successful partnership, and any suggested improvements. These discussions were recorded, transcribed, and thematically analyzed. Preliminary findings indicated that economic and political conditions affect how PHEP partnerships are prioritized, with the ongoing challenge of resource scarcity inhibiting long-term partnership growth. We identified key recommendations from both local health departments and CBOs on how PHEP partnerships can be developed and maintained. The findings from this study can be used by local health departments and CBOs alike to foster sustainable PHEP partnerships.

Natural disasters cost Louisiana an estimated $50 billion by 2050, with manufactured housing particularly vulnerable to wind and flood hazards. FORTIFIED® certification, developed by the Insurance Institute for Business & Home Safety, has demonstrated 55–74% reductions in insurance claims. However, while prior research has examined FORTIFIED® costs and insurance benefits, the supply chain processes required to achieve certification—particularly in factory-built housing—remain unstudied. In this study, we conducted site visits to two manufacturing facilities, semi-structured interviews with industry stakeholders, and benefit-cost analysis across three housing types. Results revealed that documentation—not technical construction—is the central bottleneck, adding approximately 45 minutes per unit. Using thematic analysis, we identified four barriers: photo documentation not integrated with quality assurance systems, installer knowledge gaps, transport damage causing certification delays, and coordination failures across supply chain actors. Manufactured homes under the Department of Housing and Urban Development Code are limited to roof-level certification only. The study proposes traveler add-on sheets and operational protocols, with benefit-cost analysis demonstrating positive returns across housing types.

Disaster food assistance is critical to household recovery following natural hazards, yet limited research examines how affected residents experience different assistance modalities. We investigated how disaster-affected adults in Orleans Parish, Louisiana experience Disaster Supplemental Nutrition Assistance Program (D-SNAP) benefits and in-kind emergency food assistance, focusing on the timing of D-SNAP receipt relative to household food depletion, perceived cultural and nutritional adequacy, and variation across neighborhood food environments. We used a cross-sectional survey of approximately 400 disaster-affected adults. Descriptive analyses compared experiences with D-SNAP and in-kind food assistance, including perceived helpfulness and adequacy. Multivariable regression models estimated associations between D-SNAP timing, food depletion, and perceptions of assistance adequacy, while geographic information system–based measures linked respondents to neighborhood-level indicators of social vulnerability and food access. We used interaction tests, which assessed whether associations vary across neighborhood contexts. Findings inform disaster recovery policy and planning by clarifying how the timing and design of food assistance programs shape resident experiences and equity in post-disaster food access.

Cruise ports are highly susceptible to coastal hazards. Coastal hazards can be both widespread and region-specific, impacting the cruise industry in pre-disaster planning, during-event operations, and post-event recovery. The severity and frequency of these hazards will increase with climate change, presenting an essential need to understand the cruise industry's capacity, perspectives, and planning for cruise port resilience. In this research, we conducted a baseline, global survey of cruise industry practitioners to determine industry perceptions around coastal hazard impact to operations, future perspectives, and concerns, and best practices regarding hazard impacts to ports, destination communities, and operations. In partnership with the Cruise Line International Association and a steering committee composed of relevant industry experts, the survey yielded 89 usable responses from cruise ports, cruise lines, trade associations, and other relevant practitioners. Key findings included that 98% of respondents are concerned about coastal hazards impacting cruise operations in the next 50 years, while only 39% felt sufficiently informed about these hazards. This research identified gaps in cruise capacity and cross-industry collaboration, which can direct focus to areas of shared investment to mitigate risk and advance adaptation efforts. Future research should build upon this study to highlight cross-industry collaboration with local communities, as well as within-industry education and mitigation of coastal hazard impacts.

As the levelized cost of storage continues to decline, battery energy storage systems are no longer just market tools for managing intermittency; they are becoming the backbone of grid survivability. In this study, I analyzed grid resilience and recovery through the lens of Hurricane Milton and utilized a multi-scenario framework to evaluate how distributed energy resources function as a "shock absorber" for the community. By examining coupled solar-plus-storage systems, I also provided a dual-use value proposition of these systems through the resilience dividend of steady-state operational benefits, such as peak shaving and load shifting. The analysis further explored the recovery timeline, investigating how islanding capabilities reduce the restoration burden on utilities and allow for a faster return to steady-state operations. Finally, a macro-level analysis of national trends compared the deployment trajectories of behind-the-meter residential assets with front-of-the-meter utility-scale installations, arguing that decentralization is a fundamental requirement for community recovery. 

Natural and nature-based features (NNBFs) are increasingly promoted in the literature as viable strategies for flood risk mitigation, yet their implementation in engineering practice lags behind traditional gray infrastructure. A key barrier to broader adoption is uncertainty in estimating the benefits provided by different NNBF types, particularly ecosystem services that extend beyond flood risk reduction. To address this gap, we conducted a comprehensive literature review of ecosystem services associated with major coastal and floodplain NNBF types, including wetlands, dunes, and beaches, seagrass meadows, barrier and deltaic islands, and forested ecosystems. The review identifies key ecological and environmental factors, carbon capture, coastal land loss reduction, hazard risk reduction, socio-economic and cultural factors, and economic and financial services, along with the range of metrics currently used to quantify these benefits. Building on these findings, we propose a framework that systematically links NNBF types to ecosystem services and corresponding performance metrics for NNBF project evaluation. The framework is intended to support more consistent benefit estimation for cost–benefit analysis, improve comparability across NNBF projects, and reduce uncertainty in planning and decision-making. Ongoing work focuses on refining the framework, identifying data gaps, and incorporating cost and limitation information.

The Multiple Streams Framework (MSF) is a leading theoretical approach for understanding agenda-setting in climate change policymaking. Despite decades of application, no systematic review has evaluated how the MSF has been operationalized in climate policymaking research, leaving gaps for theoretical advancement. We addressed this gap through a systematic literature review of 73 peer-reviewed studies that apply MSF to understand climate change policymaking processes. Findings revealed persistent limitations: a heavy reliance on qualitative methods, a focus on national-level cases in advanced democracies, and limited attention to subnational and Global South contexts. Core components of the framework also remain under-theorized. By synthesizing this literature, we identified key patterns, lessons, and areas requiring further exploration. We concluded with a research agenda emphasizing quantitative approaches, multi-level analysis, and broader geographic coverage to strengthen theory, methodological diversity, and insights to advance climate policymaking.

Climate change poses unprecedented challenges for democratic governance, yet little is known about citizen perception of the government's climate responsibility and performance and its implications for democratic legitimacy. We examined how responsibility attribution and climate performance affect democratic satisfaction using Afrobarometer Round 9 data from 31 African countries (n=30,825). Employing multilevel fixed effects, we tested the conditional legitimacy theory. Our results revealed that government climate performance significantly enhances democratic satisfaction, while citizens attributing climate responsibility to government show lower baseline satisfaction. Critically, these effects depended on climate experiences. Among climate-vulnerable citizens, responsibility attribution became less negative and performance rewards were amplified. Conversely, perceived climate impacts reduced performance effects, indicating problem visibility raises standards. The findings revealed "experience-based conditional legitimacy," where lived climate impacts, rather than perceived awareness, drive democratic evaluations. This suggests targeted climate action for vulnerable populations may maximize both environmental and democratic returns.

This project advances hazards and disaster research by examining how communities navigate compound environmental risks in a coastal region where storm exposure, coastal flooding, and space-industry infrastructure converge. Focusing on communities in Merritt Island and Titusville, Florida, near the National Aeronautics and Space Administration Kennedy Space Center, we analyzed how residents interpret environmental change and everyday risk in a landscape shaped by space launches, conservation policy, tourism, and recurring storm threats. Using qualitative documentary analysis of more than 50 sources, including newspapers, government and non-governmental reports, peer-reviewed scholarship, and historical texts, alongside interviews and field visits, we examined how overlapping infrastructural and ecological pressures shape access to land and water, perceptions of responsibility, and experiences of uncertainty. Rather than treating storms, industrial activity, and environmental regulation as separate hazards, residents understood them as intertwined forces influencing livelihood, identity, and belonging. By centering these perspectives, we show how risk is lived and negotiated beyond technical hazard models, offering insight into how space infrastructure becomes embedded in everyday practices of coping with coastal vulnerability.

Disaster risks are increasingly complex and interconnected due to rapid environmental and socioeconomic changes, challenging traditional assessment methods. Existing frameworks often fail to capture cascading effects across socio-cultural, economic, and environmental systems. Here, we propose a multi-step systemic disaster risk assessment framework that integrates classical risk components, hazard, vulnerability, and exposure with modern systemic risk concepts emphasizing interconnectedness and multi-hazard dynamics. Our framework involves system identification and interconnection mapping, comprehensive risk analysis, strategy development with impact assessments, qualitative and quantitative scenario modelling, and adaptive prioritization. Supported by interdisciplinary collaboration and stakeholder engagement, our approach enables a nuanced understanding of cascading risks and informs more effective disaster risk reduction strategies. By addressing the complexity and dynamism of contemporary disaster risks, our framework contributes to enhancing resilience and sustainability in disaster management practices. 

This project advances community-centered climate and hazard resilience by documenting lived risks and strengthening inclusive, locally led adaptation in vulnerable Nigerian communities. I examined how community-driven strategies reduce exposure to climate and health risks, ensure marginalized groups are not left behind, and translate practical actions into measurable resilience gains. Communities face seasonal flooding, extreme heat, environmental decline, emerging health challenges, and livelihood pressures, disproportionately impacting low-income households, women, children, older persons, and outdoor workers. The project was guided by three core research questions: (a) how do lived experiences shape risk patterns, (b) how can co-designed responses prioritize equity, and (c) how can local actions scale for broader impact? The project employed mixed methods, participatory vulnerability mapping, and community engagement to co-develop solutions. Activities included risk documentation dialogues, locally led conservation (tree protection), small enterprise livelihood support, climate-health awareness campaigns, and leadership resilience training. This aligns with Natural Hazards Engineering Research Infrastructure CONVERGE (NHERI-CONVERGE) principles of equity, interdisciplinary collaboration, and ethical engagement, integrating natural hazards engineering with social sciences. Expected outcomes include reduced forest pressure, stronger shock preparedness, improved household stability, heightened climate-health awareness, and validated participatory mapping tools. The project generates evidence for Nigeria's National Adaptation Plan and contributes interdisciplinary case studies to NHERI-CONVERGE's global knowledge sharing for hazard-prone regions.

Communities rarely experience disaster risks in isolation; hazards and vulnerabilities often co-occur, yet their interaction remains understudied. When disaster risks converge, they amplify each other, compounding the burden on already vulnerable populations. In contexts where critical systems such as health infrastructure are weak or absent, disaster outcomes are exacerbated. Florida, one of the most hurricane-prone and socio-demographically diverse states in the United States, provides an interesting case for examining these compounding risks and the spatial heterogeneity of social vulnerability drivers. Using a hotspot analysis and a Geographically Weighted Regression, we investigated areas of risk convergence and the spatial patterns of the relationship between social vulnerability drivers and hotspots across Florida. Many census tracts were simultaneously social vulnerability hotspots and hurricane hotspots, with some standing out as "triple risk convergence" zones. Contrary to prevailing assumptions, hotspots were not dominated by elderly populations but by children and youth. Results also revealed the spatial heterogeneity of social vulnerability drivers across the state. The findings emphasize the need for child- and youth-centered disaster planning in Florida, while also addressing healthcare gaps and the root causes of social vulnerability. We recommend a multi-hazard assessment framework, along with continuous monitoring of risks for effective disaster risk management. We further call for context-specific vulnerability modeling practices to ensure policies and interventions are well-targeted and do not exacerbate conditions for already vulnerable groups or inadvertently lead to misallocation of resources.

In this study, we explore the compound hazard of building corrosion in seismic zones as it is experienced by residents of Hilo, Hawai'i, a city exposed to atmospheric corrosion, coastal flooding, and high seismicity. Recent research suggests corrosion from saltwater inundation and atmospheric exposure can weaken the capacity of buildings to withstand seismic shaking. Based upon a mixed method study composed of interviews with 24 homeowners, property managers, and businesses; 38 digital questionnaires; and visual inspection of 86 buildings, we found that corrosion was a major concern in Hilo, especially for buildings directly adjacent to the shoreline. Homeowners noted rusted connection joints and roofing in wood frame buildings. Property managers of multi-storied concrete buildings identified staining, cracking, and spalling as a constant maintenance issue. Damage to building components were especially notable on the underside of lanais (elevated porches and patios), roof edges, and the bases of columns. Actions to mitigate corrosion hazard ranged from fatalistic acceptance and a focus on maintaining aesthetics to extensive and frequent structural repairs. Despite the ubiquitous nature of corrosion, few residents considered how corrosion could undermine the structural integrity of their buildings in earthquakes. Similarly, corrosion risk was not a focus of wider disaster preparedness and planning education. More research is needed to quantify the impact of coastal corrosion on building deterioration and the implications this deterioration has on building safety, planning, and emergency management in high seismic zones.

Incidents of wildfire evacuees abandoning their vehicles to flee on foot have become increasingly common, with major examples in the 2018 Camp Fire, the 2022 Maui Fire, and most recently the 2025 Los Angeles wildfires. During the Palisades Fire, live media coverage showed first responders bulldozing abandoned vehicles to access active fire fronts, while public officials urged evacuees to leave keys behind to facilitate rapid clearance. Despite the operational risks posed by vehicle abandonment, little is known about why evacuees make these decisions or how emergency managers can mitigate their impacts on responder access. We examined these questions through 25 semi-structured interviews with evacuees from the 2025 Palisades Fire, roughly half of whom abandoned their vehicles. Data were analyzed using a deductive, interview-guide–based thematic approach. Preliminary results suggest that many more evacuees consider abandoning their vehicles than ultimately do so. Factors associated with abandonment include knowledge of prior fire fatalities in vehicles, observing others abandon their cars, being immobilized by surrounding abandoned vehicles, and the presence of fire immediately adjacent to the roadway. Factors associated with remaining in vehicles include disability, older age, concern for dependents or pets, and beliefs that returning home may be safer. Several participants reported voluntarily leaving keys behind prior to public messaging, while others justified retaining keys after moving their vehicles off the roadway. These findings highlight the need for public education on safe vehicle abandonment procedures to reduce loss of life while preserving first responder access during wildfire evacuations.

Urban wildfires have become increasingly frequent and destructive over the past half century in the United States and globally. Consensus assessments attribute this rise to three interacting forces: climate change, the expansion of urban development into the wildland–urban interface, and decades of fire suppression policies that have increased fuel loads. Public understanding of these causes plays a critical role in shaping support for wildfire mitigation policies. Yet, little experimental research has examined how causal framing affects public opinion, for example, attributing urban wildfires to climate change versus government mismanagement. We focused on the California wildfires of 2025, using a national survey experiment (n = 2,143) with an oversample of California residents (n = 1,193) to test how different emphasis frames influence perceptions of cause and responsibility. The experiment also investigated how partisanship and exposure to local media alters message strength. Results showed that causal frames significantly shape perceptions of wildfire causes, though these effects vary by political identity and regional media exposure. Climate change frames increased support for emissions reductions, while management frames heightened support for land-use and firefighting measures. However, in California, extensive prior exposure to wildfire information dampened framing effects overall. These findings clarified how communication strategies influence public understanding of hazard causation and reveal the limits of persuasive messaging in politically polarized and media-saturated contexts.

For mobile workforces such as long-haul truck drivers, official warning systems and organizational communication channels often fail to account for mobility, isolation, and real-time decision-making demands. We examined how migrant long-haul truck drivers in Atlantic Canada experience and navigate risk communication during hazard-related disruptions, including road closures, infrastructure failures, and emergency incidents. Drawing on semi-structured interviews with Ukrainian migrant drivers, we identified persistent gaps between formal risk messages and the operational information drivers require to make safe decisions on the road. Participants reported delayed or inconsistent updates, unclear guidance from employers or authorities, and difficulties interpreting jurisdiction-specific risk information while in transit. In response, drivers relied heavily on peer-to-peer communication networks to translate, verify, and contextualize risk information in real time. From an emergency management perspective, these peer networks functioned as de facto, informal communication infrastructures that shaped how official messages are interpreted, prioritized, and acted upon during crises. Drivers used these networks to rapidly disseminate information about route closures, enforcement changes, weather conditions, and on-the-ground hazards, often well before such information reaches formal systems. One practical implication for emergency managers and crisis responders is the need to recognize, monitor, and strategically engage with these peer networks as complementary channels for situational awareness and message amplification. Rather than viewing truck drivers solely as message recipients, integrating peer-informed communication pathways into emergency planning can enhance real-time intelligence, improve message uptake, and reduce cascading risks across transportation and supply chain systems.

Small, hazard-exposed communities often operate with limited formal disaster risk reduction (DRR) capacity, relying instead on informal coordination, peer-to-peer networks, and cross-sector collaboration. In this paper, we asked: How does participatory Community Asset Mapping (CAM) function as a governance mechanism shaping coordination, accountability, and collective decision-making in small, hazard-prone communities? We drew on qualitative and spatial data generated through facilitated group mapping, collaborative annotation, and structured discussion with community organizations, service providers, and residents. Findings showed that CAM produces governance-relevant insights that extend beyond inventories of resources or service gaps. While critical assets, particularly social, relational, and knowledge-based resources, were widely distributed across the community, the mapping process revealed expansive governance patterns, including diffuse responsibility, fragmented coordination, and uneven continuity planning during disruptions. Recurrent challenges related to communication, accessibility, and service reach, especially for marginalized populations, highlighted that governance weaknesses stemmed less from asset scarcity than from limited mechanisms for alignment, role clarity, and shared accountability. Analytically, the paper positions CAM as a form of low-resource participatory governance infrastructure that makes interdependencies visible and supports role negotiation in contexts with limited institutional capacity. Practically, the findings demonstrate CAM's value for strengthening local DRR governance by enabling communities to translate relational knowledge into coordinated action. The paper contributes to disaster governance scholarship by illustrating how participatory methods can function simultaneously as empirical research tools and as governance interventions in small, hazard-exposed communities.

In this study, we examined the role of group-based social networks in smoothing household food consumption in the aftermath of climate-related shocks. Using data from the Bangladesh Climate Change Adaptation Survey, Round II (2012), we analyzed a sample of 762 farming households distributed across seven major agro-ecological zones as defined by the Bangladesh Centre for Advanced Studies. The survey contains detailed information on household demographics, social and network participation, exposure to climatic shocks, and food consumption outcomes, allowing us to explore both coping behavior and welfare implications. We focused on major climate shocks—floods, droughts, and cyclones—and investigated how disruptions to food consumption influence farmers' decisions to participate in social networks. Our findings showed that households experiencing climate-induced consumption shortfalls are significantly more likely to join or intensify participation in group-based networks. Moreover, network participation was associated with improved post-shock food consumption, suggesting that social networks function as an informal insurance mechanism in the absence of complete formal financial markets. These results highlight the dual role of social networks: they provide short-term support through informal financial assistance and mutual aid, and they facilitate longer-term resilience by enhancing information sharing and access to agricultural extension services. The findings carry important policy implications for disaster-prone, agrarian economies. Strengthening community-based organizations and leveraging existing social networks can complement formal safety nets, enhance adaptive capacity, and improve food security in the face of increasing climate variability. 

In this paper, I examine the communications role and organization of labor in post-hurricane disaster recovery, focusing on the period from one month to three years following a disaster. Drawing on disaster communication and labor studies, the paper analyzes how different forms of labor (including paid emergency workers, nonprofit staff, volunteers, communicators, and affected residents) communicate and interact across recovery phases and how these dynamics shape recovery outcomes and equity. Using qualitative evidence from post-hurricane recovery contexts, including case-based and interview-informed research, this study explores how labor roles shift from immediate response to long-term rebuilding, highlighting tensions between volunteer and professional labor, disparities in access to paid recovery work, and the often-invisible communicative labor required to coordinate recovery efforts. Attention is given to marginalized and underrepresented groups, including immigrant workers and residents, who may face structural barriers to safe, compensated recovery labor despite being essential to rebuilding processes. The findings demonstrate that recovery labor systems are unevenly structured over time, with short-term surges in support giving way to fragmented, privatized, or informal labor arrangements during extended recovery periods. These shifts have implications for worker safety, compensation, and community resilience. The paper contributes to hazards scholarship by advancing a labor-centered framework for understanding long-term disaster recovery and its equity implications and concludes by outlining policy and practice recommendations for more equitable labor coordination in post-disaster recovery contexts.

This paper advances disaster research by arguing that when citizenship status is excluded from disaster management and planning, undocumented communities are rendered invisible through assumptions that reproduce marginalization across preparedness, response, and recovery. We introduce undocumented critical theory (UndocuCrit) as a necessary theoretical intervention in critical disaster studies, demonstrating how undocumented status fundamentally shapes disaster experiences and challenges dominant citizenship-based disaster paradigms. We present a conceptual and theoretical analysis grounded in interdisciplinary disaster, immigration, and critical theory literatures and the documented narratives of undocumented immigrants, using four UndocuCrit tenets, fear, liminality, parental sacrifice, and acompañamiento, as an analytical framework. Rather than collecting new empirical data, we synthesize and critique existing disaster scholarship to examine how legal exclusion produces distinct forms of disaster vulnerability. Applying the UndocuCrit lens reveals that fear of deportation, legal liminality, and reliance on informal support networks structure how undocumented communities prepare for, navigate, and recover from disasters, while remaining systematically excluded from formal disaster systems. The analysis further demonstrates that although disaster and immigration literatures occasionally intersect, undocumented epistemologies and experiences remain largely uncentered within the hazards field. By advancing an undocumented scholar-led framework, we contribute a reorientation of disaster ethics, methodology, and theory that challenges universalized notions of vulnerability and clarifies what equitable disaster preparedness and recovery could mean for undocumented immigrants.

General tornado warnings frequently fail to prompt immediate protective action, with recipients often engaging in "milling" behavior, or seeking to confirm the warning rather than immediately taking shelter. In this study, I examined how enhancing the content of tornado warnings can improve understanding, thereby reducing milling and encouraging timely protective actions. In collaboration with Vanderbilt's Data Science Master's program and the Vanderbilt Office of Emergency Management, a prototype warning portal will be developed to deliver enhanced tornado warnings. The warnings will include supplemental content such as radar imagery, photographs, links to external weather resources, images depicting sheltering behaviors, and an AI chatbot using Retrieval-Augmented Generation. The chatbot will allow recipients to ask questions about the warning and tornado safety. The warnings may also integrate the National Weather Service Application Programming Interface to align messaging with federal standards. Participants completed an anonymous survey assessing message comprehension, emotional response, the trustworthiness of the message, and their anticipated self-reported protective behaviors. Survey data was analyzed using mixed-methods (i.e., qualitative and quantitative) approaches. There are two research aims: (a) to contribute to hazard communication by evaluating how visual, interactive, and AI-enhanced warning elements may influence recipient warning comprehension and protective decision-making; and (b) to inform emergency management and weather agencies in developing effective, user-centered warning systems that can reduce injury and loss of life.

Hispanic communities often experience prolonged displacement in post-disaster events driven by unequal access to federal disaster aid, living in hazard-prone areas, and pre-existing socioeconomic disparities. Despite these concerns, there is a lack of research on how multi-federal disaster aid assistance shapes migration patterns, particularly for Hispanic communities. In this study, we address how multiple federal disaster aid programs shape post-Hurricane Harvey, and how these patterns differ across migration communities identified through an origin–destination community detection framework. To accomplish this goal, we integrated annual county-to-county migration flows from the Internal Revenue Service for the year before and after Hurricane Harvey (2016–2018), which capture address changes reported on individual tax returns and household moves between all U.S. counties. Moreover, to measure the severity of hazard, we integrated hazard losses from the Spatial Hazard Events and Losses Database for the United States. We used a national origin–destination (OD) network framework in which all U.S. counties act as nodes and directed, weighted county-to-county migration flows as edges. We used community detection to identify clusters of counties with the strongest migration ties one year before and the year after Hurricane Harvey. Comparing these migration networks across 2016-2018 and linking them to variations in disaster aid portfolios across federally declared counties highlights where recovery resources may be correlated with shifts in migration networks. This is the first network-based analysis to use the OD community detection framework to discover migration community structure in a disaster context by focusing on Hispanic individuals.

The Idaho Resilience Tool is an applied research project that aims to: (a) advance scholarship on effective community engagement for webtool development and (b) build a statewide resilience-planning support tool that provides rural Idaho communities with translated, accurate, and relevant geospatial data. This is critical and timely work as local decision makers in rural communities must balance intersecting and competing pressures such as economic development, hazard mitigation, environmental protection, natural resource extraction, and tourism, among others. Unfortunately, the data they need to make informed decisions is often fragmented, inaccessible, or inaccurate. Many rural communities lack the human, economic, and technological capacity to find and analyze necessary geospatial data, creating barriers to long-term resilience planning. Funded by the Federal Emergency Management Agency's Cooperating Technical Partners Program, this project works to reduce data barriers by collaborating with decision makers across scales, including federal agencies, state agencies, non-profit organizations, counties, and municipalities. Through focus groups, interviews, user-centered design workshops, and extensive beta testing, the Idaho Resilience Tool was built from the ground up to address the specific needs of Idaho communities. This presentation will: (a) introduce the Idaho Resilience Tool, (b) share methods and insights for webtool co-development, and (c) highlight how the Idaho Resilience Tool can build rural capacity through a case study in Valley County.

In the aftermath of Hurricane Helene in 2024, mountain communities in western North Carolina relied on broadcast radio and emergent organizations as a part of ad hoc networks for disaster response. Through exploring that story, this project helps local emergency managers and state Voluntary Organizations Active in Disaster (VOADs) better understand the capacities that non-VOAD organizations may be able to offer in response to complex catastrophes. After the storm, traditional response organizations were hindered by compounding infrastructure breakdowns and initially struggled to reach cut-off communities. In that vacuum, non-VOAD organizations created temporary networks for mutual aid and many requests for aid and offers of support were broadcast over radio. In this project, I ask, "who was helping whom, with what, and how did they hear about it?" Data analysis consists of qualitative coding of radio transcripts to support a network analysis identifying key organizations, their role in the network, and resource specializations by each group during the first weeks of response and recovery. Examples include local farms, public libraries, recreational facilities, temporary collectives, and so on. Analysis showed that many local organizations fulfill relief roles expected by past research while others, especially emergent local organizations, step-up to meet critical needs in sometimes unexpected ways. The analysis also reinforced the importance of local leadership in disaster relief even when outside aid is valued and essential. Firsthand observations and interviews enrich the project narrative. The results of the analysis help to contextualize the capacity and connectivity of non-VOAD organizations active in disaster response.

Broadly, the barriers to implementing evidence-based strategies and interventions have been well documented in public health literature. However, few studies exist that have examined barriers and facilitators to the implementation of strategies before, during, and after public heath emergencies supporting disproportionately affected populations. Noted barriers to implementation include lack of tailored evidence-based strategies, lack of awareness of existing evidence-based strategies, insufficient staffing, and lack of additional resources. These barriers are often exacerbated when concurrent disasters happen, such as a natural disaster occurring during a novel disease outbreak, stretching already limited resources. The Interactive Systems Framework (ISF) is used in dissemination and implementation science to understand and focus on the systems needed to address the gap between science and practice. It offers a guide for understanding the needs, assets, and barriers of three specific systems: synthesis and translation, delivery, and support systems. These systems all play a critical role in supporting populations disproportionately affected by public health emergencies. In this presentation, we will discuss innovative ways to implement evidence-based strategies and interventions within communities and populations disproportionately affected by emergencies. Outcomes of this presentation include learning how the ISF can support research and practice, sharing lessons learned, and building professional connections.

The increasing frequency and severity of natural disasters underscores the need for robust, context-specific resilience tools to support decision-makers, practitioners, and rural communities. Resilience indices and assessment frameworks play a critical role in translating abstract resilience concepts into actionable, evidence-based insights, enabling comparisons across time and place and informing the allocation of scarce resources. However, while the literature identifies 36 existing resilience frameworks, few adequately capture the specific contexts, capacities, resources, and priorities of rural communities. Meanwhile, parallel literatures on rural community development, community well-being, and sustainable development have evolved largely independently from rural resilience research, despite substantial conceptual overlap. These literatures are rarely integrated in ways that inform the design of resilience assessment tools that are both flexible and appropriate for rural contexts. Rural development is often treated as a broad and relatively neutral concept, whereas rural resilience and sustainable frameworks are more explicitly normative, emphasizing adaptation, positive change, and long-term viability. In this presentation, we demonstrate how a shift from hazard- and disaster-focused approaches toward multidimensional frameworks encompassing economic, social, environmental, and institutional dimensions offers a more holistic and customizable approach for rural resilience assessment. We argue that rural resilience functions as a bridging concept that links rural development, sustainability, and community well-being. Integrating these perspectives is essential for developing resilience assessment tools that are context-sensitive, participatory, and empirically grounded, and that can better support rural communities in navigating both acute shocks and long-term challenges.

Micromobility offers a potential pathway to strengthen urban disaster resilience by enabling the distribution of emergency supplies when infrastructure or transportation networks are compromised. We examined the viability of micromobility vehicles for post-disaster relief via agent-based modeling and simulations in Stillwater, Oklahoma. Focusing on the early recovery phase with exclusive micromobility deployment, we evaluated three operational variables: deployment rate, load capacity, and per capita supply allocation. The analysis revealed that optimal efficiency occurs when 2% of the population participates in distribution efforts, each vehicle transports 18-kilogram loads, and deliverers provide 4 kilograms of supplies per community member per delivery. The results indicate that a relatively small but dedicated and efficient group of residents could achieve feasible recovery processes in a small city solely using micromobility. To better support communities, the research suggests that micromobility be integrated into disaster relief plans and processes, that local cycling organizations be valued as key partners for disaster resource distribution, and that human-powered vehicles offer a flexible and nimble alternative for effective response in select situations. This research contributes to the emergency management literature by quantifying operational thresholds and demonstrating micromobility's feasibility as a supplemental disaster response mechanism when conventional systems fail.

Hazard mitigation and climate adaptation increasingly depend on complex chains of models, datasets, and policy frameworks, yet these components remain fragmented across institutions, software platforms, and professional disciplines. We present a new agentic artificial intelligence (AI) architecture that integrates climate data, hazard mitigation planning, and infrastructure representation into a single, end-to-end decision support system. The framework links multiple AI agents to external scientific and policy tools, including a state climatology database, Hydrologic Engineering Center's River Analysis System hydraulic models, a state hazard mitigation plan, and a 3D digital twin of buildings and infrastructure. Through structured tool-calling and retrieval-augmented reasoning, the system can autonomously select data sources, configure and execute flood simulations, extract scenario-based flood depths, retrieve relevant mitigation actions and funding pathways, and connect those outputs to specific physical assets. We demonstrate the approach using Louisiana case studies drawn from state hazard mitigation and climate data systems. Results showed that the agentic workflow can run analyses of flood depths and mitigation options, while preserving traceability to the underlying models, datasets, and policy documents. This enables reproducible, evidence-based decision support across traditionally siloed domains such as climate science, engineering, and emergency management. Beyond computational efficiency, the system represents a new model for human–AI collaboration in hazard science, where domain experts remain in control while AI agents ensure that relevant models, data, and policy constraints are consistently integrated. The presentation will describe the system architecture, tool-grounded reasoning methods, and implications for scaling this approach to support multi-agency coalitions for disaster risk reduction.

A primary objective of the Stakeholder Engagement for Water-Hazard Science and Response Project (SERP) is understanding how the U.S. Geological Survey (USGS) Water Mission Area data, tools, and services can best support stakeholders in disaster risk reduction before, during, and after water hazard events. SERP research explores: (a) how community partners perceive USGS’s role in emergency management, (b) USGS's ideal part in the water hazard management cycle, and (c) what barriers limit engagement with USGS expertise and data. To address these three questions, SERP developed a multiscale, stratified purposive sampling approach to capture geographic and cultural variations nationally while enhancing representation of populations historically most at risk. While efforts for external coalition-building were anticipated, the team encountered the need to form unexpected internal coalitions and faced unexpected setbacks that ultimately affected engagement with priority communities. Inclusive strategies were developed to approach interest-holders, in addition to protocols for leveraging internal partnerships to access communities through local USGS Water Science Centers. Preliminary findings indicated a need for outreach and education targeting community-scale and county emergency managers, as well as affected communities. USGS data, tools, and services were considered valuable across all stages of emergency management, with particular emphasis on mitigation and preparedness. Strengthening internal and external coalitions and tailoring outreach to local contexts enhances the integration of USGS resources into emergency management and improves resilience and reduces disaster risk.

As Los Angeles declared their community "debris-free" in September 2025, three manufactured housing communities (MHCs) remained burdened with burnt-out cars and structures. Impacted by the Palisades Fire, these MHCs struggled to recover due to lack of clarity concerning who is responsible for their recovery. This stems from their divided tenure, whereby MHC residents often own their homes but not the lands underneath, leasing them from private or corporate owners. As such, MHCs are typically one "property" upon which many homes share infrastructure and utilities, an arrangement contributing to confusion regarding post-disaster responsibilities. Existing scholarship shows substantive gaps in post-disaster outcomes for MHCs, largely stemming from lack of clear responsibility and social stigmatization. Given the post-Palisades Fire experiences of MHCs, the first part of our ongoing research project asked: What are the response and recovery experiences of low- and moderate-income households in MHCs following the Palisades wildfires? Working alongside community partners at Neighborhood Partnership Housing Services, we conducted a resident survey in the three affected MHCs, semi-structured interviews with key government officials and community leaders, and content analysis of relevant government documents covering MHC disaster response and recovery. From this data, we identified the core policy and governance barriers preventing equitable recovery for MHCs. We found an absence of consideration of MHCs in much of the post-Palisades Fire response and recovery, leading to confusion regarding who is responsible for their recovery. Finally, we offered policy recommendations to communities, academics, and policymakers to redress these gaps.

In recent decades, the frequency of tropical cyclones undergoing rapid intensification before landfall has increased, amplifying the complexity of forecasting and risk communication efforts. Spanish-speaking populations in the United States face unique challenges when responding to tropical cyclones. However, a research gap exists in understanding their experiences with rapidly intensifying tropical cyclones. Language barriers, limited access to resources, and unfamiliarity with protective actions can hinder timely decision-making, particularly during shortened warning windows. We examined how rapidly intensifying storms influence risk perception and protective action decision-making in Spanish-speaking communities and explores the role of past experiences in shaping future responses. Using a nationwide survey of 631 Spanish-speaking adults, familiarity with and understanding of the term rapid intensification were compared between coastal and non-coastal respondents using a paired t-test. To assess perceptions of storm impacts, we used a non-parametric rank test to compare the perceived likelihood of personal, financial, and infrastructure impacts from tropical cyclones that rapidly intensify versus those that do not. Preliminary results suggest that while Spanish-speaking respondents are familiar with and demonstrate understanding of rapid intensification, forecasts of rapid intensification do not consistently influence perceived likelihood of hurricane-related impacts. This study aimed to inform risk communication strategies tailored to Spanish-speaking populations. By understanding these challenges, forecast meteorologists, emergency managers, and local leaders can develop more effective communication strategies to enhance resilience and ensure that critical information reaches populations most vulnerable to rapidly intensifying storms.

Hurricane Helene had devastating implications for both mental health and community resilience in Western North Carolina. We examined the psychological effects of Hurricane Helene on community volunteers who responded to the disaster. We surveyed 408 community volunteers six to eight months post-hurricane to investigate the impact of hurricane exposure, volunteering, and coping strategies on the mental health of community volunteers. Content analysis aimed to contextualize findings and capture volunteers' perceptions of disaster response and recovery. Survey results highlighted significant mental health challenges among volunteers, with more than half reporting at least one poor mental health indicator, roughly one in five reporting all four generalized anxiety disorder indicators, and one in 20 reporting all four post-traumatic stress indicators. Hurricane exposures, such as fear, and coping strategies, such as engaging with substances, were the most consistent predictors of reporting poor mental health. Engaging with greenspace to cope was associated with protective effects. Contextual analysis suggests prolonged distress related to exposure to Hurricane Helene. Positive reflections shed light on the unique ways in which community members united to respond to this unprecedented event. Findings from this work contribute novel insights into Hurricane Helene's impact on the mental health of volunteers and first responders.

Approaches to community engagement for disaster risk reduction vary widely, and there is growing interest among governments and emergency management organizations in integrating these efforts to strengthen disaster resilience and reduce risk. Community engagement can enhance the validity of risk assessments, reduce social vulnerability, and support the co-creation of effective disaster mitigation and climate change adaptation solutions through convergence science. However, a deeper understanding of community capacities, the challenges communities face, and the mechanisms needed to build and sustain respectful, reciprocal relationships remains essential. We outline the key challenges that motivated our collaboration within the Rising Voices, Changing Coasts: Earth and Indigenous Science Puerto Rico Hub. We describe the engagement process we have developed and share insights on how meaningful, long-term relationships contribute to resilience. We will also highlight our practice of convergence science, an approach that weaves ideas, methods, technologies, and knowledge systems to drive innovation and discovery that are locally grounded, inclusive of the whole community, and meaningful. This process intentionally incorporates the perspectives of residents, government officials, and private-sector representatives from the outset and throughout the scientific process.

In this research, we draw on qualitative data from the Emergency Management Organizational Structures, Staffing, and Capacity Study to examine how stakeholder understanding of emergency management roles and responsibilities shapes the authority and legitimacy emergency management agencies are granted and how the authority affects agencies' ability to access resources and sustain mission-aligned work. Analysis indicated a reinforcing pathway: as stakeholders better understand emergency management functions and constraints, emergency management leaders reported increased credibility, clearer boundaries, and improved standing in decision-making processes. In turn, these conditions enabled more effective resource advocacy and reduced pressures to engage in non-mission related work. Analysis also suggests that authority and legitimacy in emergency management are multi-dimensional and develop over stages. This work proposes a stage-based model linking stakeholder understanding, authority/legitimacy formation, and resource mobilization, offering a theoretical framework that can inform capacity-building strategies, interagency coordination practices, and Federal Emergency Management Agency efforts to strengthen local and state emergency management.

In this research, we describe how survey respondents to the Emergency Management Organizational Structures, Staffing and Capacity Study report allocating time over the preceding year and explores the factors that inform how time is allocated. It also describes how respondents report wanting to allocate their time, especially where the ideal allocation differs from the actual allocation and how it differs from an idealized model of emergency management practice (i.e., comprehensive emergency management and involvement in preparedness, response, recovery and mitigation). We also explore how agencies—identified through qualitative data collection—have pushed back against requests to go beyond their identified mission and what features of these agencies distinguish them from those respondents who have expanded their activities beyond a traditional emergency management mission set.

Following large earthquakes, science agencies produce probabilistic aftershock forecasts using statistical models, which are used by different organizations to support decisions on earthquake response and recovery. These forecasts can be communicated through different visual products, including tables, graphics, and maps. The design choices for these products can affect how the forecasts are used and understood by different users. In prior work, we held workshops with users of aftershock forecasts, including emergency managers, engineers, public communicators and others. We identified numerous–and sometimes conflicting–user needs for several forecast product types. We proposed a three-category typology for product design that can cover much of the variability observed in user needs. We next designed numerous user-centered prototypes for several key forecast products: barplots to show forecast trends by magnitude, lineplots to show forecast trends over time, and heat maps to show the spatial distribution of forecasted shaking. We used a novel method for user-centered design that involves first decomposing a product into its key design elements and then translating user needs into choices for specific design elements. Using this method, we developed numerous product prototypes that captured key user needs expressed in the user workshops or were aligned with our product design typology. Prototypes were presented to the original workshop participants in a focus group and then refined based on their feedback. We will discuss these user-centered and -tested prototypes for aftershock forecasts products, and their implications for the visual communication of hazards information.

PREPHUB 2.0 is an international and interdisciplinary research project that addresses the need for more inclusive and socially grounded approaches to urban disaster risk reduction. Building on the Emergency Preparedness Hub (PREPHUB) concept developed by the Urban Risk Lab at MIT, we adapt and expand this model to the Latin American urban context through a situated, intersectional gender perspective. In this project, we challenge predominantly technocratic approaches to urban risk management by emphasizing the role of everyday practices, care networks, and social inequalities in shaping community resilience. We propose the design of an urban preparedness node that functions both as emergency-support infrastructure and as a meaningful component of daily urban life, particularly for women and other marginalized groups who experience risk disproportionately. Methodologically, we employed a research-through-design framework, combining qualitative and quantitative methods across three phases: critical analysis of the original PREPHUB and comparable international precedents; participatory, multidisciplinary co-design and fabrication of a localized PREPHUB 2.0 prototype incorporating gender-sensitive and culturally situated criteria; and in situ testing and evaluation of the prototype in a public space within Lomas Latorre, an informally developed and wildfire-affected neighborhood in Viña del Mar, Chile, assessing community interaction, perceived usefulness, and technical feasibility. We expect our findings to contribute practical design knowledge and policy-relevant insights for integrating gender-sensitive urban infrastructure into disaster preparedness strategies, supporting more equitable, resilient, and context-responsive DRR practices applicable to vulnerable urban settings globally.

The burden of illnesses, injuries, and deaths due to cold and extreme cold events is increasing in the United States despite a warming climate. While extreme cold events (ECEs) are unusual in western Washington, they have the potential for significant impacts on human health and infrastructure. In addition, populations that face a higher risk of negative health impacts from the cold, including people experiencing homelessness, people unable to afford adequate heating, and older adults, are growing. Evidence-informed planning and decision-making during extreme cold events, coupled with information provided by local NWS forecast offices, have the potential to aid in mitigating the health impacts of extreme cold. Yet research available to inform extreme cold response and risk communication strategies is lacking. Through a series of open-ended interviews with forecast meteorologists and response agency staff in and around King County, Washington, we explored practitioner "mental models" of extreme cold to understand how practitioner knowledge, experience, and beliefs influence their decision-making during response to extreme cold events. The ultimate goal of this convergent research is to improve early warning and response processes by describing the ways in which different agencies and organizations approach ECEs and information needs. Findings from this research will inform local extreme cold response and inform the development of decision support and forecast tools for use in Washington state to improve awareness of extreme cold impacts and response decision-making.

Housing resilience broadly refers to the ability of a community’s housing systems to anticipate, absorb, recover from, and adapt to slow (e.g., sea level rise) and rapid onset (e.g., hurricanes) disasters. There is a growing interdisciplinary literature on housing resilience that highlights the importance of collaboration across diverse sectors, including the public, private, non-profit sectors, and academia. However, there is limited understanding of the factors that enable or hinder such collaboration. Focusing on this gap in the literature, we examine the role of institutional logics—belief systems, norms and values that shape how organizations decide and act within the broader system (e.g., market, communities)—in multi-sector collaboration in housing resilience. We conducted semi-structured interviews (n=50) with representatives across diverse sectors as part of a National Science Foundation-funded study on housing resilience in Miami. The findings show that while the institutional logics in the public, private and non-profit sectors hinder collaboration in housing resilience, the professional logic in the academic sector (prioritization of professional competence) is more conducive to multi-sectoral collaboration. The institutional logics that hinder collaboration include: a dominant market logic (e.g., Miami as a global capitalist hub) which contributes to leadership deficits and resistance to influence within the public sector, and to disengagement in the private sector due to insufficient incentives; and, an ideal state logic (i.e., public service informed by meaningful participation) in the nonprofit sector that fosters distrust toward policy makers (and private sector actors).

Snow avalanches pose serious risks to people, infrastructure, and economic activity in mountainous regions. Risk reduction relies on public safety services such as highway avalanche mitigation, ski area safety programs, professional training and operational protocols for worksites, and public avalanche forecasting and education initiatives. While these approaches are widely implemented, there is no systematic framework to support decisions about how to design, revise, or discontinue services. Typical performance measures, particularly fatality counts, provide limited insight into what works well, in what circumstances and why, and how interventions can be improved. Disciplines facing similar challenges, including disaster risk reduction and public health, offer abundant frameworks for designing and evaluating risk reduction interventions. However, unique aspects of avalanche risk, including the potential for human triggering, characteristic uncertainty in hazard assessment, and wicked feedback environments, limit the direct transferability of existing approaches. Addressing this gap requires synthesizing interdisciplinary research into a domain-specific, practice-oriented approach. In this presentation, we introduce a framework that integrates insights from relevant literature into a comprehensive, practical tool tailored to avalanche risk and service-based design decisions. Organized into guiding principles, a systems-based representation, and an implementation process, the framework is intended to ground how decisions are made, justified, and coordinated within established bodies of knowledge and support interventions that are effective in practice. Although developed for avalanche contexts, the framework offers insights for researchers and practitioners in other hazard domains by illustrating how interdisciplinary theory can be synthesized for practical risk reduction service delivery. 

While many people come to understand natural hazards like earthquakes through popular media, these media often prioritize entertainment, action, and emotion over accurate or realistic depictions of earthquake monitoring and warning. We conduct a content analysis of how early warning, aftershock forecasting, and prediction are depicted in popular media formats including film, television, and podcasting. Audiences of this presentation will hear findings such as how earthquake early warning, aftershock forecasting, and predication are scientifically portrayed, how much time is spent on each concept, and the significance of each to the plot of the story. Initial findings show that prediction in particular plays an outsized role in storylines. They are often used early in stories to stage danger and establish heroes and then throughout as a plot device that advances the story. Aftershock forecasts play a smaller role and are typically not accurately represented. Earthquake early warning hardly makes appearances in any of the studied content. Communication, education, and outreach professionals working to help people understand the realities of earthquake science and how to utilize tools like early warning will benefit from understanding the quality of information made available through popular media. This study can also inform the hazard community's collaborations with media makers, who frequently solicit information from the science and hazard communities. Because stories told in popular media formats often include multiple hazards, implications from this research can extend to other hazards such as volcanoes and tsunamis.

In highly seismic central and northern California, The Pacific Gas & Electric Company (PG&E) uses an earthquake notification and response prioritization tool called Dynamic Automated Seismic Hazard (DASH).  This tool was developed for rapid and reliable dissemination of earthquake characteristics (magnitude, location, shaking levels) and initial estimates of damage for each asset system (gas, electric, generation, buildings) to guide initial inspections and restoration crew efforts.  DASH also is used to develop scenario earthquake models for emergency exercise support. We provide an overview of the DASH platform, and earthquake response examples.

Local officials seeking to maintain the trust of increasingly skeptical and polarized communities may downplay climate related hazard information while also working to mitigate those hazards and protect long-term safety. Narrative media such as animations, short videos, and zines provides one way to elicit community concerns, observed environmental signals, shared experiences, and values that form a basis for engaging people with that work. We present examples of animations, storyboards, and zines from three case studies in the Blackstone Valley of Rhode Island, New Bedford, Massachusetts, and Erie, Pennsylvania. These works demonstrate the potential of narrative media to widen audiences, engage diverse and conflicting points of view, and foster constructive engagement. The narrative media was created by landscape architecture students at Penn State University and funded by the National Oceanic and Atmospheric Administration (NOAA) and Pennsylvania Sea Grant as part of larger research and engagement projects. We ground the application of this media in the larger context of literature on hazard visualization and climate communication, noting the bias towards increasingly complex and technical visualization tools. We conclude that narrative media provides a valuable compliment to other forms of visualization that may inform more conventional processes.

Disaster relocation and recovery processes involve multiple actors, yet disaster literature predominantly examines coordination mechanisms from institutional perspectives. It overlooks community expectations, specifically how affected communities envision the roles of diverse actors in post-disaster contexts. We examined the expectations of residents of two Hurricane Maria-affected communities in Puerto Rico (Comerio and Loiza) (n=24), through a three-year, longitudinal PhotoVoice project. It is based on a content analysis of photographs documenting the community's expectations regarding disaster interventions (n=167). The analysis organized these interventions into five major categories of actors: (1) citizens, individuals, and informal networks; (2) NGOs and community organizations; (3) government agencies; (4) businesses; and (5) cross-sector collaborative initiatives. It showed that government actors dominated the interventions, accounting for 65% of recommended interventions, followed by 34% for community organizations and NGOs, 34% for citizens and informal networks, 14% for cross-sector collaborative initiatives, and 4% for businesses. Communities' expectations from government agencies were mostly related to infrastructure development, relocation processes, planning activities, policy and regulatory formulation, and environmental management. NGO and community organization expectations emphasized their roles in education and awareness raising, advocacy, and regular maintenance, while the roles of citizens, individuals, and informal networks centered on being informed and prepared. Communities articulated the need for cross-sector collaboration in carrying out the relocation processes. There were relatively less expectations from businesses, and these expectations mainly focused on continuity of operations and providing new scopes (opportunities and space). Our findings emphasize a bottom-up and participatory approach aligned with community expectations for collaborative responsibilities across public infrastructure, private sector engagement, nonprofit capacity-building, and individual preparedness.

Housing resilience planning has become an important focus for governments seeking to promote public safety, risk reduction and environmental sustainability. While the expertise of stakeholders is crucial to the design of robust housing resilience strategies, the extent to which these considerations meaningfully translate into or are reflective of governmental plans remains significantly underexplored. Using the case of Miami Greater Miami and the Beaches region, we contributed to addressing this gap by evaluating the extent to which the values and priorities of key stakeholders measure against official housing resilience plans in the region. Data was obtained by interviewing 50 respondents across public agencies, private industries, nongovernmental organizations, academia, and community residents in the region. We developed a housing plan evaluation matrix to compare support for values based on 60 selected housing plans and analyzed the data using NVivo. Preliminary findings suggest that the alignment between multi-stakeholder priorities and government housing resilience policies is mixed and nuanced, underscoring the need for this gap to be more explicitly acknowledged and addressed in resilience planning. We argue that such integration is crucial not only for the effectiveness of housing planning but also for ensuring government responsiveness and accountability. Findings of the study hold key insight for resilience planning and long-term sustainability.

The notion of strong local community partnerships remains a critical opportunity and ideal standard of practice for response organization engagement after a disaster. Strong relationships built upon contextual understanding, acknowledgment of local capacity, funding priorities, and cultural sensitivity can greatly improve response and recovery outcomes by promoting a shared understanding and vision. Localization is one approach for response organizations working with communities to acknowledge and work toward accomplishing these goals. How localization as a concept is applied, or is relevant, in domestic disaster response work has yet to be explored in depth. Using a hybrid of several localization frameworks, we conducted 17 semi-structured in-depth interviews across six states with rural community actors who have had, currently have, or will have a funding partnership with Save the Children USA. These interviews aimed to better understand the needs and expectations of local community members who serve children and/or families in rural areas. This report concludes with key quotes from respondents, ideal interactions between response organizations and local community organizations, and guiding questions to be asked introspectively by the response organization. Overall, the application of an adjusted localization framework for domestic disaster response has great utility. Findings suggest the addition of a new localization dimension of "rural context" to include social-cultural factors, physical environment, historical and economic factors, and public health factors. This additional dimension will help promote contextually appropriate response decisions meant to support and bolster local capacity and capabilities.

In the United States, government-funded property buyouts have become a widely used strategy for adapting to climate-related risks, such as flooding. However, scholars have noted that while buyout programs focus primarily on homeowners, renters, who are also displaced, often remain overlooked. Similarly, renters have received comparatively limited scholarly attention in existing literature. We employed a case study approach to explore drivers and outcomes of floodplain buyouts of rental apartment complexes which is an understudied perspective and may provide new insights into the challenges of buyouts, specifically, and managed retreat generally. The Multifamily Voluntary Buyout (MVB) program in Houston, Texas, is one of the only programs in the country specifically designed to support renters and the acquisition of multi-family housing. Through policy document analysis and preliminary Key Informant Interviews with stakeholders involved in the production and implementation of Houston's MVB policies, we provide valuable insights about new and adapted policies. Moreover, the program was executed just prior to a 2024 legislative reform in the Uniform Relocation Act of 1970 that provides stronger protection for renters involved in buyouts. The case therefore also provides data on how renters were protected prior to those changes and insight into the type of programmatic changes that are likely to be needed in the future to comply with the new regulations. Findings from this study contribute to informing other jurisdictions and offer lessons on drivers, administrative feasibility, funding models, benefits and challenges associated with multifamily buyout programs which can improve policy learning.

In the aftermath of Hurricane Ian, communities across Southwest Florida experienced widespread flooding and wind damage, which resulted in extensive housing damage and displacement, generating urgent and long-term recovery needs. While formal recovery systems, including federal assistance, SBA loans, and insurance payouts, play an important role in post-disaster recovery, access to these resources is often constrained by eligibility requirements, documentation burdens, benefit caps, and administrative delays, among other factors. These challenges result in delayed and insufficient assistance for many affected households. Due to these barriers, crowdfunding platforms like GoFundMe were significant informal mechanisms for accessing recovery support following Hurricane Ian. Some households and individuals sought financial support by creating campaigns for housing repairs, debris removal, temporary relocation, and other recovery needs. Unlike formal assistance based on eligibility criteria and documented need, crowdfunding is driven by raising awareness about needs and deservingness through voluntary donations. This means access may depend on factors beyond disaster exposure, including social networks, digital literacy, and narrative persuasiveness. In this study, we examine the distribution and mobilization of crowdfunding as a recovery resource following Hurricane Ian. Our study draws from and further extends theories of post-disaster therapeutic community, social capital, and social vulnerability.  Using over 700 publicly available GoFundMe campaigns, primarily from Lee County, we conducted spatial and qualitative analysis of campaigns' locations, narratives, and outcomes. We further assess how initiation and success of campaigns are related to hurricane extent and type, pre-disaster social vulnerability, and household social and financial capacities.

With natural disasters becoming more frequent and warning times potentially shortening, it is essential for households to be ready for the hazards common to their area. We conducted a study to determine if there were any associations between household preparedness levels and demographics, disaster experience and knowledge, and beliefs on preparedness. We used a survey to ask respondents who were impacted by Hurricane Ian, a Category 4 storm, about prior disaster experience, preparedness plans their households had taken, and their usage and contents of an emergency supply kit (ESK). We used a concurrent dual-mode approach to administer the cross-sectional survey, combining both mail and web-based responses, with an added incentive for those completing online to improve response rates. The survey sample consisted of 5,000 households in evacuation Zone B, selected using address-based sampling. While the majority of respondents indicated having some sort of preparedness plan, very few have adopted all five of FEMA's recommended household plans. Older adults were less likely to be prepared, even though they have a higher risk of negative impacts from a disaster. There was a positive association between almost all of those that had experienced a disaster and household preparedness for a future disaster. These findings demonstrate that perceptions and lived experiences can impact a household's preparedness levels, suggesting that customized messaging for individual populations is necessary to enhance preparedness.

Near real-time data are essential to determine the number of disaster-related deaths and identify current causes of death during disaster response. However, traditional disaster-related mortality surveillance faces challenges, such as data accuracy and timeliness. Therefore, the Centers for Disease Control and Prevention (CDC), Disaster Epidemiology and Response Unit (DERU) developed a protocol to track disaster-related deaths using media sources. DERU members update a disaster-specific Excel database daily during each response, using a written protocol that outlines how to scan the media for disaster-related deaths using Google or a similar search engine to search for key terms. Since 2021, this protocol has been used in over 45 disasters, including the 2021 Pacific Northwest heatwave, Winter Storm Uri, and several hurricanes, tornadoes, and wildfires. Daily reports were sent to CDC response leadership and communication teams to guide response efforts. The number of deaths in the incidents tracked ranged from one (Tropical Storm Elsa, 2021) to 262 (Hurricane Helene, 2024). Cause and circumstance of death were available for most deaths per incident. Slightly more than half reported age, and almost 70% reported sex. Name and race/ethnicity had the lowest percentage of data available from media reports per incident. Data were most complete for lightning-related deaths. Media mortality surveillance offers timely disaster-related death data that can be used for an all-hazards response approach and help inform overall response planning, support public health decisions, and understand the health impact of a disaster.

There is relatively little research addressing emergency supply kit (ESK) ownership immediately after a disaster strikes to help determine their overall effectiveness. The goal of this study was to determine the efficacy and public health impact of ESKs among the population affected by Hurricane Ian. Following Hurricane Ian's landfall, we conducted a cross-sectional survey between October and December 2022. The sampling frame included 5,000 households among evacuation Zone B in five Florida counties. We used address-Based Sampling with options to complete the survey online or in paper form. We conducted descriptive analyses to examine distributions of demographic characteristics, having an ESK, items in a household ESK, and items most needed. A total of 1,342 respondents completed the survey (~29%). The majority had heard of an ESK prior to the survey (87.3%) and 60.3% had an ESK at the time of Hurricane Ian. The top item that was in household ESKs and deemed most helpful was stored water at 93.2% and 64.4%, respectively. When asked about items that were needed from their ESK during or immediately after the storm, stored water was needed the most (62.9%). Roughly half (54.2%) of respondents had a member of their household leave home within 72 hours of impact for at least one item. These data serve as a foundation for understanding the landscape of ESK ownership, most used, and most needed items immediately after a hurricane's impact. ESK item lists can be modified in hurricane-prone areas to prioritize the most essential items.

Current sociological research rarely discusses disabled people's livelihoods amidst climate change and natural disasters, obscuring disabled experiences with those of able-bodied people. In this project, I highlight disabled people's different modes of survival during the Los Angeles wildfires, as many fatal victims were disabled. I conducted in-depth interviews, digital ethnography, and participant observation to investigate how these Angelenos utilized their formal and informal networks to survive, as evacuation orders and emergency responses often do not consider disability. The climate change scholarship that addresses disability frames it as a consequence of climate change, rather than centering disabled people and their experiences with climate change. I seek to innovate sociological understanding of climate crises and survivorship through the application of disability theory. Furthermore, I adopt an intersectional approach as Los Angeles' segregated nature resulted in varied emergency responses within ethnic communities. How can public policy and first responders address this gap and work to ensure the safety of vulnerable populations if social scientists continue to neglect disabled people as well? This project addresses this gap through community engaged ethnography with disabled Angelenos, activists, and community centers to further understand disabled survival under immediate crisis. Pushing the concept of survival forward, it questions what disabled survival looks like during natural disasters. When does the smoke clear? At what point is "survival" reached when medical equipment and homes with accessible structures, things these Angelenos need to survive their daily lives, have been turned to ash?

Risk information refers to the data and knowledge that helps us understand the potential dangers and consequences associated with floods and other environmental hazards. Understanding how climate risk is priced in housing markets, particularly how public availability of risk information influences buyer behavior, is essential for hazard risk management and has gained increasing attention. We conducted a mixed-methods study examining homebuyers’ risk information behaviors and how access to risk information influences their decision-making. We conducted qualitative, in-depth interviews with real estate agents in Washington and Elizabeth City, North Carolina, to understand risk disclosure practices, the impact of such disclosure on pricing, and the role of third-party websites in risk information disclosure. Quantitative aspects include geospatial data acquisition and analysis using descriptive statistics and Pearson's correlation coefficient. The Pearson's correlation coefficient of 0.115 showed a weak positive relationship, which indicates that high flood risk is associated with slightly longer selling times, though the effect is limited and other factors like price, condition, and demand matter as well. Qualitative results show that homebuyer behavior is shaped by past hazard experience, income, and physical condition of the property. Climate risk, particularly flood risk score, does not directly affect property prices in isolation; rather, insurance costs serve as the primary channel through which risk influences property values. We report that buyers’ concerns focus less on risk itself and more on the financial implications of required insurance premiums. Our findings show the importance of considering the influence of climate risk on housing decisions. 

Many emergency management (EM) agencies operate with one or fewer full-time equivalent (FTE) staff, creating significant capacity challenges. In this presentation, we use survey data from over 1,700 local, state, territorial, and tribal EM agencies and 19 listening sessions to describe how limited staffing constrains local EM capacity. Understaffing limits the ability of EM agencies to carry out their missions, producing coverage gaps, heightened burnout, and succession vulnerability. We then examine how collaborative arrangements, such as regionalization, mutual aid agreements, shared-service models, and more informal collaborative approaches, can help address these gaps. We also draw on case study interviews with local EM agencies to highlight key lessons from those that have built creative, supportive arrangements to expand coverage, share expertise, and strengthen continuity despite limited staffing.

Accurate estimation of wind pressure distributions on building envelopes is essential for assessing structural risk and resilience under extreme wind hazards. Wind tunnel testing remains the primary approach for obtaining such data, but it is costly and time-consuming, which limits its applicability for large-scale or portfolio-level hazard analyses. As a result, many buildings of interest are not explicitly represented in available experimental databases. In this study, I present a data-driven framework for estimating wind pressure fields on low-rise buildings using a Conditional Neural Network. The proposed approach is designed to interpolate existing wind tunnel data across both spatial coordinates on building surfaces and global building and wind characteristics. By conditioning model parameters on building geometry and wind conditions, the method captures complex, nonlinear aerodynamic behavior while maintaining a unified and computationally efficient representation. The framework was validated using wind tunnel measurements for low-rise buildings under a range of wind directions and configurations. Results demonstrate that the Conditional Neural Network accurately reproduces key statistical characteristics of wind pressure distributions and provides reliable estimates for building configurations that were not explicitly tested. Compared to conventional interpolation approaches, the proposed method improves accuracy while substantially reducing the need for additional wind tunnel experiments. The findings highlight the potential of machine learning–based interpolation to enable scalable wind pressure estimation for hazard and resilience applications. By reducing dependence on extensive experimental testing, the proposed approach supports rapid fragility assessment, portfolio-level risk analysis, and data-driven evaluation of wind-exposed infrastructure.

In this study, we investigate the potential for large language models (LLMs), in concert with human expertise, to address persistent methodological challenges in survey research with a particular focus on disaster settings. We examine how LLMs can augment core stages of the survey process, including adaptive questionnaire design, targeted sample selection, synthetic data generation for pilot testing, and rigorous data quality interventions, especially in environments affected by rapidly shifting populations and urgent timelines. For qualitative analysis, we evaluate LLMs' capacity to summarize, cluster, and extract latent patterns from large volumes of open-ended responses, increasing the interpretive depth of disaster research. Highlighting capabilities such as zero-shot, few-shot learning and knowledge graph retrieval-augmented generation, we outline a framework for effective human‚ AI collaboration and empirically benchmark the performance of LLMs against traditional and manual methods using a rapid post-disaster survey on disaster preparedness and time stress conducted during the 2024 hurricane season. The results point to practical strategies and foundational limitations for the integration of LLMs, balancing efficiency gains with the need for representativeness, ethical oversight, and analytic rigor in crisis contexts.

Agricultural regions across the United States face growing exposure to severe wind events. These storms can cause damage to agricultural infrastructure, including grain bins, pivot irrigators, and outbuildings, and disrupt farm operations. We examined the experiences of agricultural producers in Iowa, Nebraska, and Florida following the 2020 Midwest derecho, severe storm events in 2022, and Hurricanes Helene, Debby, and Idalia. Using a mixed-methods approach that integrates survey data, semi-structured interviews, and observational data from farm site visits, we compared patterns of damage, recovery, and decision-making across these regions and disaster contexts. Findings highlight variation in access to resources for rebuilding or replacing damaged infrastructure, producer decision-making under uncertainty, alignment between producer-reported damages and insurance assessments, and the cumulative effects of repeated storms compounded by broader economic pressures. Together, these results contribute to a more nuanced understanding of the evolving risk landscape facing agricultural communities and inform the development of more effective extension, research, and policy interventions.

The intensity and frequency of heavy precipitation in the United States has increased over the last several decades and is projected to increase in the future. As decision-makers seek to increase flood resilience, they do so within specific institutional cultures and structures that both constrain and facilitate their efforts, acting as sources of uncertainty and shaping whether and how flood resilience is built. Department of Defense decision-makers are being tasked with increasing flood resilience on military installations and in surrounding communities to protect life, safety, and infrastructure as well as for mission assurance and national security. Through interviews (n=46) and participant observation with military and civilian infrastructure decision-makers at two US Army installations (the Detroit Arsenal, Michigan, and Fort Hood, Texas), we identified key constraining and facilitating factors that are sources of uncertainty and influence decision making. Prominent among these are elements of military institutional culture (including mission assurance, the chain of command, and the hierarchical nature of decision making), competing and shifting priorities, economic factors, political and policy changes, and perceptions of risk. We show how understanding institutional constraints on decision making can affect whether and how flood resilience is built both on military installations and in surrounding communities, and we identify practical barriers and pathways to increased flood resilience. As civilian and military infrastructure is interconnected and the management of stormwater crosses jurisdictions, understanding military institutional culture and structures is important also for non-military organizations seeking to collaborate with the military to increase flood resilience.

Federally assisted housing developments provide essential housing for millions of low-income households, yet evidence from studies of affordable housing suggests that many such properties are disproportionately located in flood-prone areas, exposing residents to displacement, property damage, and long-term health impacts. We examine how flood hazard exposure varies across housing programs, property characteristics, and geographic regions to inform equitable climate adaptation and disaster mitigation strategies. Using a national dataset of federally assisted housing developments, we integrate property-level attributes, subsidy type, and geospatial flood hazard data to identify patterns of vulnerability. Preliminary analyses compare public housing, Section 8 Project-Based Rental Assistance, and Low-Income Housing Tax Credit properties, highlighting the role of property age, maintenance status, and location in shaping flood risk. By disaggregating exposure by program and property characteristics, this study advances understanding of structural and policy drivers of flood vulnerability in affordable housing and provides actionable insights for planners, policymakers, and housing authorities to prioritize resilience investments.

In this study, we examine how participatory community asset mapping can strengthen disaster resilience for older adults, with particular attention to how community members negotiate differing perspectives during the mapping process. While asset mapping is important for developing disaster resilience, there is limited research on participant dynamics during the asset mapping process. We focus on one sub-research question: how does negotiation surface non-traditional or overlooked community assets? Workshops were held in northeastern (Fall 2025) and eastern Nova Scotia (Spring 2026) with emergency managers, service providers, community organizations, and older adults. Using qualitative methods, each workshop used guided small-group mapping activities that blended expertise with lived experiences. Using conversation and thematic analyses, we analyzed field notes, worksheets, audio recordings, and co-designed maps to identify themes related to resources, service gaps, and group dynamics. Preliminary findings from Fall 2025 showed that disagreements (such as certain supports "counting" as assets) instigated new expansive definitions of "assets". These negotiations helped participants recognize assets beyond formal services, such as networks of older adults with deep life experience, strong community relationships that enable informal support, and the role of humor and shared laughter in maintaining morale during crises. Participants also noted gaps in communication, navigation, and regional coordination linked to funding limitations. Results demonstrate that participatory mapping not only identifies community resources but also uses negotiation and shared tools to broaden understandings of resilience. Practical implications include strategies for facilitating collective decision-making, integrating non-traditional assets into planning, and strengthening partnerships to support older adults during disasters.

Renters are vulnerable to natural hazards and disadvantaged in disaster assistance programs. However, their disaster outcomes are less studied due to limited records and recovery indicators on the renter population. Using an individual-level credit panel dataset on residents in Eastern North Carolina, we show that hurricane flooding worsens young renters' financial health and delays their transition to homeownership. First, we developed and validated a loan-based method for renter identification, tracking half a million young renters from 2015 to 2020, including during Hurricanes Matthew (2016) and Florence (2018). Next, we applied difference-in-differences regressions to quantify the effect of hurricane flooding on estimated incomes and credit scores. We used survival analysis to compare renter-to-homeowner conversions between affected and unaffected renters post-storm. Preliminary results indicate that flooding has a persistent, negative effect on renter incomes and a small negative effect on credit scores. Flooding also causes a sustained delay in renter-to-homeowner transitions, potentially because flooded renters drain their savings in the recovery process. In this study, we demonstrate how credit data can be used to understand renter recovery at scale. Eastern North Carolina is a relevant case for many hazard-prone communities in the United States, as it features repeated flood exposure across rural and urban settings. Furthermore, this study extends hazards research on financial impacts, which has largely focused on homeowners and property recovery. By showing how flooding disrupts wealth-building and pathways to housing security, the population-scale evidence highlights the need for renter-inclusive assistance and recovery policies. 

Getting an accurate assessment of damage at a regional, state, and national scale can present challenges for social science modelling, given that data can vary across organizations, events, geographies, and spatial scales. In this study, I developed a multidimensional damage assessment in order to provide a more comprehensive picture of damage in service of a more equitable disaster recovery process. Hence, I ask two research questions: How can hurricane housing damage be more accurately captured by incorporating a broader range of damage variables? How can the outcomes be utilized to have a more equitable disaster plan? I focused on Hurricane Irma (2017) in Florida, integrating hurricane characteristics, including storm surge, flood inundation, wind speed, rainfall, and the built environment information, including building age and housing values at the census track level, to construct a composite Damage Matrix. To capture the damage, I applied regression models to estimate associations and test sensitivity to indicator selection, and train machine learning and deep learning models to capture relationships and improve predictive accuracy. I compared the results with observed post-disaster information for sensitivity analysis and evaluating the reliability. The result is a detailed damage map, based on which different levels of damage and damage hotspots can be categorized accurately. To have a more equitable disaster recovery planning, the overlap between high damage areas and low federal assistance can be captured for further policy attention. This analysis provides a basis for prioritizing equitable assistance allocation and outreach in equitable recovery planning.

Event attribution science is increasingly positioned as a tool for linking climate change to disaster risk, yet its relevance for decision-making depends on how it is interpreted within policy and practice settings. In this study I examine how policy-relevant stakeholders in Louisiana interpret attribution science in the aftermath of Hurricane Ida, focusing on how scientific claims are translated, evaluated, and situated within existing governance and professional contexts. The study draws on semi-structured interviews with local and state planners, emergency managers, regulators, and nonprofit practitioners involved in disaster recovery, mitigation, and resilience planning. Analysis centers on how participants make sense of attribution science in relation to their institutional roles, responsibilities, and decision environments, rather than on scientific accuracy alone. Findings indicate that attribution science is interpreted less as abstract climate knowledge and more as applied risk information. Stakeholders are more likely to engage attribution claims when they align with observed impacts, are presented through visual or comparative formats, and are translated into economic or planning-relevant terms. Attribution framed in this way supports discussions about future risk trajectories and long-term preparedness. At the same time, interpretation is conditional and shaped by scale mismatches, institutional capacity, and governance constraints. Overall, the analysis highlights how attribution science is filtered through professional experience and institutional context, shaping its perceived relevance for planning, advocacy, and policy action. These findings contribute to ongoing discussions in hazards research about usable climate science and the conditions under which attribution knowledge informs governance and decision-making.

In this study, I examine the spatial interrelation of housing tenure and expected annual wildfire loss across California census tracts. Drawing on social vulnerability, risk, and environmental justice frameworks, I assess whether renter-dominated communities experience differential wildfire risk in terms of relative and absolute expected annual losses. Using generalized spatial two-stage least squares spatial autoregressive models, the study estimates direct, indirect (spillover), and total effects of two rentership indicators. Wildfire loss outcomes include total, relative total, and disaggregated components—agricultural, buildings, and population-equivalent expected annual loss— of expected annual loss, sourced from Federal Emergency Management Agency's National Risk Index, and American Community Survey data. Results reveal substantial spatial dependence in wildfire losses and consistently demonstrate strong associations between rentership and lower expected annual losses. These inverse relationships reflect differential asset ownership rather than reduced social vulnerability, indicating less agricultural land in renter-dominated communities, fewer owned structures, and lower exposure to asset-based loss metrics. Importantly, spatial spillover effects frequently exceed direct effects, underscoring the need for wildfire risk conceptualization as a regionally interconnected process across boundaries and borders. Findings challenge assumptions that higher expected losses necessarily correspond to greater social vulnerability and highlight limitations in asset-based risk metrics for capturing renter precarity, displacement risk, and recovery barriers. The study contributes to environmental justice and disaster research by distinguishing between exposure to loss and vulnerability to harm, requiring more renter-centered mitigation, recovery policy, and alternative measures of wildfire impact.

This study investigates how hazard experience shapes individuals' risk judgments and the timing of protective action decisions through the Protective Action Decision Model (PADM). Prior PADM research has largely emphasized whether experience predicts taking protective actions, but when people act is equally consequential: timely, correct decisions can reduce injury and damage as threats escalate. Empirical findings on experience–timing relationships remain mixed, in part because hazard experience level is multidimensional, and the pathways from these components to risk perception and action timing are not well specified. To address this gap, we conducted a two-factor mixed social experiment using DynaSearch, a web-based platform that simulates the storm warning process across stages. Participants (n = 270) were recruited from Bellevue, Washington (a low tornado-exposure setting) and Denton, Texas (a higher tornado-exposure setting). The experiment simulated an escalating tornado scenario across five alerts, from a thunderstorm warning to a tornado emergency. Results indicate that hazard experience is associated with higher risk judgments at select stages and earlier adoption of recommended protective actions. Participants with higher overall experience tended to move to the safest in-home shelter earlier than those with lower experience. Direct experience—such as having felt hazard impacts and previously received warnings and acted—showed the clearest relationship with earlier protective action timing.  These findings refine PADM by clarifying how experience dimensions translate into decision timing and offer practical guidance for emergency managers communicating uncommon hazards in low-risk regions.

The July 2025 Central Texas floods, which claimed at least 27 lives and destroyed Camp Mystic, highlight the urgent need to assess and mitigate flood risks at youth camps. Our research addresses a critical gap in hazard planning by conducting a statewide flood risk assessment of licensed youth camps in Texas. Using GIS-based spatial analysis, we evaluated approximately 1,240 youth camps licensed between 2000 and 2025 in relation to both Federal Emergency Management Agency (FEMA)-designated and Texas Water Development Board (TWDB) floodplain data. FEMA maps are widely used but incomplete in rural areas. TWDB provides more comprehensive but also expensive coverage. We introduce a six-tier flood risk ranking system based on environmental exposure, land use zoning, and infrastructure vulnerability. In Texas, nearly half (49.3%) of youth camps have buildings located within the 100-year floodplain according to TWDB data, compared to only 24.0% based on FEMA floodplain maps. Recent legislation in Texas has introduced stricter safety requirements for youth camps, including mandates to reduce flood risk in order to maintain permit eligibility. We estimate the financial cost is approximately $855.88 million of removing or retrofitting buildings located in floodplains to comply with these new regulations. By integrating spatial analysis, climate projections, and policy review, our research generates actionable insights to inform zoning reform, emergency preparedness, and equitable disaster response. We emphasize the unique vulnerabilities of non-residential land uses in rural settings and supports the development of a scalable flood risk assessment framework for youth-serving institutions.