Poster Session Abstracts

Preparing individual researchers and interdisciplinary research teams to carry out extreme events research that is ethically-grounded, methodologically sound, and scientifically rigorous remains a priority for the disaster preparedness workforce. The CONVERGE initiative provides a repository of resources to support the advancement of disaster research, including webinars, briefing sheets, and training modules.

In 2019, the Natural Hazards Center, in collaboration with the Centers for Disease Control and Prevention (CDC), developed two training modules as part of the initiative: (1) Socially Vulnerable Populations and (2) Disaster Mental Health. With a special emphasis on students and emerging and situational researchers, the training modules are designed to accelerate the education of hazards and disaster researchers. The curriculum begins with a set of learning objectives and is followed by written content that provides background on the module’s specific topic. Additionally, the training modules offer case study vignettes, identify gaps in knowledge, and include a list of publications for further reading. Tools and resources, such as topically related scales and measures, online resources, and data sets, complete the training modules after a short assessment quiz.

This poster highlights the progress with the training modules component of the CONVERGE initiative. Along with the (1) Socially Vulnerable Populations and (2) Disaster Mental Health training modules, the following are additional training modules currently in production and to be showcased in the poster: (3) Cultural Competence in Hazards and Disaster Research; (4) Institutional Review Board Procedures for Hazards and Disaster Researchers; and (5) Conducting Emotionally Challenging Research. Future modules will be developed on the following topics: (6) Ethical Considerations for Hazards and Disaster Researchers; (7) Social Science Methods and Approaches for Hazards and Disaster Research; (8) Interdisciplinary Methods and Approaches for Hazards and Disaster Research; (9) The Science of Team Science: Forming Interdisciplinary Teams for Hazards and Disaster Research; (10) Data Curation, Sharing, and Publication Protocols; and (11) Good Practices for Sharing Hazards and Disaster Research with the Public. The training modules, which will be developed over the next five years, will be freely available online via www.converge.colorado.edu.

The Minority Scholars from Underrepresented Groups in Engineering and the Social Sciences (SURGE) reconnaissance mission is built on a holistic service-learning model. Though not connected to one classroom experience, the mission is structured to provide minority graduate students with meaningful, real-world engagement, which is community-centric. In 2018, the SURGE scholars visited St. Thomas and St. John of the U.S. Virgin Islands to explore the effects of the 2017 hurricane season on the natural, built, and human environments. In 2019, the SURGE Scholars split into four groups and worked on: (1) a food security project in St. Thomas; (2) the All Hands and Hearts rebuilding project in St. John; (3) an infrastructure data mapping project in St. Thomas with the University of the Virgin Islands; and (4) a heritage research data collection project in collaboration with the University of South Florida. This poster highlights the activities and partnerships developed during the SURGE scholars’ reconnaissance field study experience.

The increasing frequency of natural disasters in recent decades has caused communities to suffer not only from direct economic losses, but also from longer-term hardships in recovery. The recovery of businesses plays a significant role in overall community recovery. In order to conduct risk-based assessments to make decisions for enhancing community resilience, improved prediction and quantification of business recovery is needed. This study introduces a stepwise, probabilistic modeling approach using Bayesian linear regression that is comprised of three steps: data collection, development of model forms, and model selection. First, a comprehensive dataset containing vital information used to describe a business and its recovery was obtained from a study conducted in 2016 after Hurricane Matthew by researchers from the National Institute of Standards and Technology-funded Center of Excellence for Risk-Based Community Resilience Planning in Lumberton, North Carolina. Then, based on a review of business disaster impact and continuity literature, a set of initial candidate model forms was developed. This included four complementary measures of business recovery: business cease operation days, revenue recovery, customer retention, and employee retention. These measures comprehensively describe a business’s recovery state post-disaster. Finally, the most appropriate model was selected through stepwise evaluation and elimination. One advantage of this approach is that it accounts for the interplay between businesses and households during the recovery process. The final models are applicable in risk-based resilience assessments of communities with similar socioeconomic characteristics and hazards.

Today, pet owners occupy 62 percent of households in the United States, and research suggests that this number is rising. This research explores the risk perception and preparedness of college students with pets in relation to a tornado event using Trigg (2015) and Thompson’s (2017) notions of “pets as a protective factor.” This notion holds that pet owners choose to act differently in emergencies because of the bonds to their pets. This research was conducted in the form of an onsite controlled experiment at Oklahoma State University (OSU). It involved 69 participants from OSU, who took part in a computer simulation and multiple-choice questionnaire. The simulation, powered by DynaSearch, examined participants’ decision-making in real time by monitoring their “clicking” patterns on an information screen. The questionnaire, which assessed each participant’s risk perception and choice of protective action after viewing the information, addressed three issues regarding pet safety: pet ownership, pet risk perception, and household preparedness planning. Our aim is to help emergency managers design messages that are targeted specifically to pet owners, thereby reducing the number of potential human and animal disaster victims in areas with high pet populations. 

This project utilizes an interdisciplinary approach to resilience by examining the linkages between water infrastructure and human health during storm surge flooding under future sea level rise in two East Coast cities: Charleston, South Carolina, and Morehead City, North Carolina. The project employed two phases of analysis—modeling exposure and susceptibility analysis of extant water infrastructure and populations to coastal flooding, and analysis of future vulnerability using a resilience matrix within a facilitated tabletop exercise for a landfalling hurricane scenario. Results point to differential exposure of the two study areas due to population and settlement patterns along coastal floodplains and future increasing vulnerability, particularly for human health and underground utilities. The utility of a resilience matrix assessment, scenario-based simulation, and tabletop exercises across sectors (e.g., water utilities, public health, emergency management, and urban planning) are highlighted. 

New Zealand is producing the country’s first National Climate Change Risk Assessment (NCCRA). As a first step toward drafting the NCCRA, an expert panel developed a risk assessment framework that enables a broad range of risks to be systematically compared. This poster outlines the framework produced, indicating where this “fits” within the wider context of planning for—and monitoring—adaptation to a changing climate. The framework will allow users to conduct national assessments of climate change risks across sectors with uniform degrees of detail and scale (although amenable to varying availability of data/information for different domains). The framework recommends the use of standardized assumptions of climate change and socioeconomic change across sectors, and a standard method to assign levels of risk. The framework will also provide a means for the NCCRA to compare and evaluate risks and opportunities in terms of urgency. This will inform New Zealand’s central government in prioritizing where further action is most urgent to reduce risks or take advantage of opportunities through preparation of a national adaptation plan. 

The Centro Nacional de Investigación para la Gestión Integrada de Desastres Naturales (CIGIDEN) is a government-funded interdisciplinary research center that was established after the 2010 Maule earthquake. The primary mission of this consortium of four universities is to produce research to reduce the impact of natural hazards in Chile. The ability to disseminate research among policymakers and the community is a key element of this mission. Our goal is to develop disaster and resilience scholarship and propose solutions to some of the factors that could lead us to repeat the 2010 events. To accomplish this, we have developed innovative methods to disseminate and engage with communities, both regionally and locally, to understand the complexity of disaster risk reduction. In this poster, we describe our methodologies and tools used to engage individuals and communities. We have developed virtual simulations of tsunamis and flash floods, evacuation modules, community mapping with drones, collaborations between local authorities and nongovernmental organizations, and an evolving communication strategy that includes traditional and social media. We are in the process of defining an educational strategy that integrates children and adolescents in the development of school curricula. We are also designing transdisciplinary case studies that allow us to reach and learn with the community. 

Studies suggest that healthcare workers often feel unprepared to respond to disasters because of their own poor personal preparedness, a lack of confidence in facility disaster plans, insufficient training, and an unwillingness to report to work. Most existing U.S studies have a narrow focus and are limited to clinical staff at one hospital setting. Accordingly, the Veterans Emergency Management Evaluation Center (VEMEC) conducted a nationwide web-based survey of 4,026 Veterans Affairs clinical and non-clinical employees’ perceptions of organizational readiness for disasters and employee willingness to prepare for disasters within the Veterans Health Administration (VA), the nation’s largest integrated health system. More than half of the respondents were confident about their facility’s ability to respond to disasters, 62 percent for natural disasters, 54.6 percent for epidemics, and 50 percent for manmade disasters. Compared to non-clinical staff, VA clinicians were less confident in their facility’s ability to respond. In contrast, clinicians were more likely to consider their role to be important during a disaster and more likely to seek additional training to prepare for disasters. Respondents that indicated greater household preparedness were also more likely to report greater workforce preparedness. Our study suggests both a desire and a need for additional training, particularly about epidemics and manmade disasters, and a need for better understanding of the role of non-clinicians during disasters. This is particularly important as prior research suggests that perceptions of the importance of their role are a significant factor in whether non-clinicians report to work during disasters.

Communicating about hazards on social media is challenging, especially during crises. Communicators, regardless of their roles, can help or hinder efforts to effectively convey the information most needed by the public, either supporting or creating more work for official agencies charged with public safety. The Communications Working Group (CWG) within the International Association of Volcanology and Chemistry of the Earth's Interior (IAVCEI) Commission on Volcanic Hazards and Risk is developing a set of professional considerations for people in the volcanology community who communicate via social media during volcanic crises. This effort builds off a similar set of guidelines developed for agency and non-agency scientists when responding to volcanic crises, which was developed by an IAVCEI working group in 1999. The new considerations are focused solely on social media communications and will address the impacts of well-intentioned but misinformed messengers, potential ways to coordinate messaging between agencies charged with public safety, and “amplifiers,” or individuals who can spread and support official messaging by speaking with the media and sharing preliminary results. The considerations are based on discussions from a social media workshop at the 2018 international Cities on Volcanoes meeting and a survey sent to the broad volcanological community in early 2019. A particular effort has been made to solicit input from professionals representing regional diversity. Respondents include agency and non-agency scientists, emergency response professionals, and journalists. CWG members represent the international volcanology community and recognize that these considerations will, necessarily, be very general to account for situational and cultural variances.

Located in the northeastern region of India in the state of Assam, Majuli is the largest inhabited riverine island in the world. Majuli is surrounded by the mighty river Brahmaputra on the south, its tributaries Kherkutia suti on the northeast, and Subansiri on the west. This unique river island is spread over an area of over 500 square kilometers and is home to 168,000 people. Majuli represents an almost unparalleled diversity in its cultural heritage, comprised of diverse tribal groups, satras (monasteries)—the seat of vaishnavite culture, and unique housing types. Majuli is severely affected by recurrent flooding and erosion. The flooding and erosion have affected the social, economic, and cultural fabric of the community, causing disruption of livelihoods, resettlement of people, loss of land, and loss of agriculture production. Because of the unique geography of the island, it is more susceptible to geomorphological changes along with demographic, social, economic, and environmental reshaping.

This poster explores the socioeconomic and demographic background of the population across different parts of the island. These areas are categorized as most vulnerable, vulnerable, and flood affected to identify the locations with high concentration of socioeconomic and demographic vulnerability for prioritizing disaster risk management. This research takes stock of existing short- and long-term coping and adaption strategies to understand what types of risk management is most prevalent in the communities. The advantages and disadvantages of these practices are analyzed and suggestions are made to augment practices with external help.

Infrastructure projects are the most significant and costly investments made by communities. Coastal communities must plan for growth, future development, and infrastructure degradation due to sea level rise, extreme weather events, and other climate hazards. While many coastal communities are facing this reality, the techniques and costs of incorporating adaptation and hazard planning into the capital planning process are unclear.

As infrastructure projects often directly impact the built and unbuilt environments, incorporating hazards planning into the infrastructure planning process could greatly enhance a community's resilience to extreme weather and sea level rise. Mainstreaming these techniques will help many coastal communities struggling with long-term adaptation.

This National Oceanic and Atmospheric Administration (NOAA)-funded project has developed a variety of techniques to help practitioners incorporate climate, flood, and hazard data into local and regional infrastructure and capital improvement plans. The primary takeaways for participants will be: (1) how to improve and enhance community capacity to incorporate data, research, and information related to coastal hazards and extreme weather into infrastructure planning, (2) how to implement resilience and adaptation measures in coastal infrastructure and public buildings, and (3) how to understand and quantify the costs associated with the replacement, protection, or improvement of public buildings and infrastructure.

Although many of the world’s deadliest earthquakes have occurred at night, the present earthquake literature lacks understanding of household immediate protective actions during nighttime earthquakes. This study surveyed households about their experiences during the 2018 Eastern Taiwan Earthquake, a magnitude 6.4 earthquake that struck Taiwan at 11:50 p.m. local time, and examined the correlations of structural and environmental contexts, social contexts, household context, perceived shaking intensity, risk perceptions, earthquake experience, preparedness, and demographic characteristics with respondents’ immediate protective actions. The results indicate that the majority of respondents did not take any protective actions. The remaining respondents were more likely to have taken an alternative action (e.g., they stayed beside solid furniture or took cover near soft items), followed by evacuation and property protection actions, and were less likely to have taken the recommended action (i.e., drop, cover, and hold on). Conversely, few people who took the recommended action were injured, whereas 30 percent of respondents who took an alternative action were injured. Additional analyses revealed that none of the conditional contexts had a significant effect on household protective action decisions. Whether or not respondents took a protective action, as well as the type of protective action they took, were found to be influenced by their perceptions of shaking, alerts, stress, and risk. 

In response to the growing number of mass fatality incidents (MFIs) across the world, and specifically in the United States, there is a need to both understand nuances and create a better prepared and more resilient community. Recent terrorist events, natural disasters, and transportation accidents highlight opportunities to determine lessons learned, and to utilize data from after-action reports to create collaborative increases in future preparedness.

This proposed research focuses on the need presented by the Scientific Area Committees for Forensic Science at the National Institute of Standards and Technology. It recognizes the urgency to move from anecdotal, single-event data to a comprehensive view of multiple MFIs to determine if there are ways to predict the surge in communication and data requests following an event. The predicted surge can then be utilized to promote policy and budgeting adjustment/changes in both the disaster management and first responder fields. The predictive model aims to provide objective data to support requests for additional personnel, technology, and infrastructure in any community, both nationally and potentially globally.

When there is a disaster, emergency management and risk communication become critical. In today’s interconnected world, social media applications, such as Twitter and YouTube, are being used to communicate risk. Social media apps are used to create synchronous, two-way communication channels between emergency managers and their populations. The process of understanding what people are communicating through social media apps can be daunting, especially because there is so much information to process. This issue is especially problematic for emergency management agencies and offices that already have limited resources and personnel to do their jobs. Fortunately, technological advances from the data science field have provided the emergency management community with an innovative way to manage information shared on social media. TextualETL, a textual analytics tool created by William Inmon—considered the father of data warehousing—provides an innovative solution to better understand what people are communicating through social media and transform social media information into knowledge that can be used to better inform risk communication in disaster management. As emergency managers, we must understand what people are saying, especially in public social media forums, to better understand how to more effectively communicate risk with the populations that we serve. TextualETL provides emergency managers with an information management tool to achieve this goal.

This poster presents a framework for funding and managing Hawaii’s coastal zone through the training and development of the next generation of experts. The program will allow experts to proactively and more efficiently complete required environmental impact statements (EIS) for beach cells at risk and will include an improved one-stop permit shop and online e-permit system. This framework provides a proposed solution for overcoming the three identified major obstacles to managing Hawaii’s coastal zone: completing coastal zone EIS proactively, reducing the time and cost of the multi-jurisdictional permit process, and creating a place-based cadre of experts to undertake critical EIS studies before major erosion events. The framework is based on the creation of a state funding mechanism for grant managers, project managers, and professors in the University of Hawaii system to create curriculum for graduate programs to produce experts that can manage Hawaii’s coastal ecosystems. This program of place-based scientists, engineers, planners, and cultural practitioners will create a cadre of locally-educated experts for all islands, which will be vital to managing Hawaii’s shorelines in response to climate change. A goal for the Next Generation Coastal Zone Managers Program is to have EIS studies completed before damaging erosion events occur, eliminating two years from the current permitting process. The creation of a streamlined and transparent e-permit system that combines all required best management practices and mitigation conditions for federal, state, and county jurisdictional agencies is expected to save an additional year of permit planning time, reducing costs and time by approximately 75 percent. 

Emergency management practitioners and other public agency communicators have increasingly leveraged the potential of online social media to directly engage with the public and disseminate information across the disaster lifecycle. While these messaging strategies have become increasingly sophisticated, much remains to be learned about how they play out in different contexts. Since 2010, the Hazards, Emergency Response, and Online Informal Communication (HEROIC) project has collected data to further our understanding of formal and informal communication in response to hazardous events. In this poster, we highlight recent work that focuses on:

• Engagement strategies employed by National Weather Service (NWS) offices in their use of Twitter to communicate with the public in threat versus non-threat periods;

• The context-dependent relationship between NWS messaging strategies and message retransmission for meteorological hazards;

• The relationship between messaging strategies used by public health agencies and retransmission in a high-ambiguity health hazard; and

• Online interaction dynamics among NWS offices in threat and non-threat conditions.

Our findings suggest that while many elements of message style, content, and structure have stable effects on retransmission across contexts, some features do matter more—or have different effects—in high-threat or high-ambiguity situations. Taken together, these results can inform organizations tasked with public safety in both quotidian and high-threat contexts on how to better use social media to communicate with the general public.

Flooding is the most common and costly natural hazard in the United States. Established through the National Flood Insurance Act of 1968, the National Flood Insurance Program (NFIP) is intended to encourage the wise use of floodplains and empower property owners and communities to proactively address the risk of flooding. Now serving almost 20,000 communities across the nation, the NFIP operates thorough four primary programmatic pillars, including (1) flood insurance, (2) flood mapping, (3) floodplain management, and (4) hazard mitigation.

In collaboration with the Association of State Floodplain Managers (ASFPM) and with support from the U.S. Federal Emergency Management Agency (FEMA), the Natural Hazards Center is conducting a retrospective and prospective analysis to assess the NFIP’s 50-year evolution and look toward its future.

Through conducting oral histories with six of the nine living NFIP administrators, case studies of flood prone communities, and other geographic and sociological analyses, this project highlights the centrality of the four pillars of the NFIP. This project will culminate in a book-length report highlighting the progress and the challenges with the program. This poster describes this project in greater detail and highlights a complementary database developed to showcase the NFIP in action. The database catalogs FEMA-funded mitigation initiatives designed to help individuals and communities reduce flood risks and increase resiliency. Specifically, it outlines the types of flooding that FEMA’s various programs have helped to address, specific interventions utilized, measures of success, stakeholders and partners involved in these efforts, and the ways in which the four pillars of the NFIP are manifested across the spectrum of these flood risk reduction activities. These data help to illustrate the thinking and design that have undergirded this long-standing program while demonstrating the tangible benefits it has delivered to the American public.

Road networks are exposed to and affected by natural hazards. The potential social and physical impacts of these events warrant the need to incorporate an integrated risk evaluation process for the management of road networks. Available risk management systems, applied to critical infrastructure, commonly consider physical risk in terms of direct economic losses caused by the damage of a certain component or asset without considering a systemic approach or social aspects. This article proposes a methodology to integrate the social vulnerability of the exposed population and the physical risk of road infrastructure from a network perspective. The methodology considers social vulnerability in terms of sociodemographic variables, the dependency of the society to the road network, and the density of available critical infrastructure. Physical vulnerability is analyzed considering infrastructure failure and its operational consequences. The main outcome is the assessment of risk integrating physical and social vulnerability, including considerations of hazards modeling and study area. The proposed methodology is applied to a road network exposed to volcanic lahar flows in southern Chile. Outcomes demonstrate that incorporating social aspects modifies risk assessment when contrasted to only physical vulnerability. The proposed methodology may help road agencies and municipalities address the effects of social vulnerability in their mitigation decisions.

Near-field tsunamis related to Cascadia Subduction Zone events are a significant concern for coastal communities in the Pacific Northwest region (Washington, Oregon, and Northern California). Low-lying coastal communities with limited access to high ground need effective evacuation strategies (e.g., Vertical Evacuation Shelters, or VES) to reduce loss of life during tsunamis. Considering the limited number of effective risk assessment tools for evaluating potential evacuation outcomes, this paper advances previous work by exploring the FN-curve from an agent-based simulation to assess the risk of different year recurrence interval tsunamis to prioritize infrastructure improvement and VES location in a coastal community. Two series of simulations and in-depth analyses for Seaside, Oregon, demonstrate evacuation results in terms of risk in different year-recurrence-interval tsunamis, the location of evacuees caught by tsunami waves, how placing VES influences mortality rate, and the best location options for multiple VES. Findings demonstrate that placing VES close to the shoreline saves at-risk populations nearby; however, VES can be dangerous for those located in the dilemma zone. While placing a VES at the population center seems plausible when locals can invest in only one VES, clustering two VES close to a population center cannot provide the best outcome. The lowest mortality rate is reached when multiple VES exist with consideration of the geographic shape, population distribution, and tsunami inundation of a city.

Biomedical libraries can play a vital role during public health emergencies. Helping healthcare and emergency response organizations be prepared is critical for decision making during these events. A literature review and interviews with organizational partners were conducted to uncover data needs and address how librarians can provide support.

Reoccurring themes surrounding data barriers were exposed:

• Awareness: Work is being duplicated across organizations.

• Policy: Data sharing policies and data use agreements are time-consuming.

• Quality and timeliness: Standards are not widely adopted, and shared data is often not peer-reviewed.

• Recognition: There is a hesitancy to share data before findings are published. Reused data may not be properly credited. Furthermore, there exists a lack of promotional or tenure incentives to share data.

Outcomes identified ways librarians can assist with data needs, including data sharing during outbreaks, offering research data management services, and providing digital toolkits. In addition, many specific tasks were identified for librarians: links to grey literature; summaries of response efforts across organizations; tips for relationship building; data-use agreement guidelines; curated datasets and databases; and instructions for proper data citation. By anticipating the needs of patrons, librarians can assist in rapidly and efficiently responding to outbreaks.

After disasters, we expect that the most vulnerable will be prioritized. For children, however, designation as a vulnerable group creates what the National Commission on Children and Disasters has called a “perverse benign neglect,” where children receive less, rather than more attention in disaster planning and management. My research aims to understand how policy takes shape for children after disasters, as well as the factors that contribute to child-sensitive post-disaster policy. At the federal level, I review the significant policy advances made for children since Hurricane Katrina and the unfinished agenda. The review sets the stage for a case study on policy agendas and decisions in the aftermath Superstorm Sandy, focusing on three policy areas—childcare, mental health, and housing. All three were salient issues for children and families in the aftermath of the disaster but had varying policy responses. At the same time, the storm’s widespread impacts provide the possibility to explore variations in the policy response; the case study compares the experiences of New Jersey, New York (state), and New York City, in supporting children and families. The research is motivated by an interest in longer-term recovery and resilience, during which underlying vulnerabilities and inequities can be addressed or exacerbated. In addition, the research focuses on policies that support children from low-income and minority families, who face the most severe and persistent risks.

Natural disasters place tremendous pressure on critical infrastructure systems by testing their service reliability under extreme conditions. System failures are inevitable during harsh events, and prolonged disruptions could pose serious risks to the mental, physical, and emotional well-being of community residents. However, research has shown that infrastructure service disruptions will not be experienced the same way by the subpopulations in the community, and socially vulnerable groups tend to suffer more from such disruptions. The research suggests that different sociodemographic factors, such as income, race/ethnicity, education, age, medical conditions, house type, homeownership, and years of residence, could magnify disaster impact. The objective of this research is to identify the sociodemographic factors most influential to the variances in the disaster impact experienced by households due to the service disruptions. The temporal and physical contexts of this study were the transportation, communication service, water, and power outages during and in the immediate aftermath of Hurricane Harvey in 2017. The research concluded that specific social groups were disproportionately affected by the service disruptions. The findings demonstrate the need to integrate the social dimensions in disaster mitigation and planning practices in order to make improvements to the current condition of the infrastructure systems and to address inequalities in risks experienced by the residents due to natural disasters. 

Risk management of road networks commonly evaluates the physical vulnerability of road assets exposed to natural hazards. However, the social role of roads in connecting a population to its networks and providing access to other critical infrastructures is often neglected. This article proposes a methodology to assess social vulnerability and critical infrastructure accessibility in the context of road networks management. The methodology is applied to the central region of Chile, which concentrates half of the population of the country and the most relevant economic activities. Data to assess socioeconomic variables were obtained from the 2017 Chile national census, the 2017 National Socio-Economic Characterization Survey, and georeferenced data of critical infrastructure locations. Data were analyzed using principal component analysis, from which the social vulnerability index was estimated including critical infrastructure density. In addition, accessibility to critical infrastructures, such as drinking water, schools, and healthcare facilities, among others, was studied in the context of road network topology. For this, roads providing access to critical infrastructures in a radius of 2 km were evaluated in terms of their lengths and annual average daily traffic. Then, an importance index was estimated and linked to each road. The two indexes were finally applied to the case study, allowing for a sensitivity analysis in decision making when considering the effects of social vulnerability and access to critical facilities.

Post-disaster studies are critical for measuring impact to communities, from physical infrastructure performance and vulnerabilities, to interdependencies with social and economic systems of the affected communities. The data collected in these studies are also used to drive community resilience modeling. Disaster studies aimed at capturing perishable data are often time-sensitive and intensive, logistically rigorous, and expensive due to the typically large impacted areas. Research entities regularly undertaking these studies, such as the Disaster and Failure Studies program at the National Institute of Standards and Technology, would benefit from enhanced methods for the execution of these activities.

Development of enhanced data collection and processing tools allows disaster researchers to prioritize time in the field, capture larger datasets expending less effort and fewer resources, and streamline the extraction of meaningful information from collected data. A methodology is presented to capture a large, passive, geospatially located dataset in affected areas using vehicle-mounted 360° video cameras, elicit serviceable datasets by extracting images from the videos at locations of interest (LOI), and apply these images to a machine learning approach for broader image classification. Algorithms have been developed to extract images from the videos using an input suite of locations of interest, e.g. damaged buildings, and associate the extracted images with their LOI. Extracted images can be labelled with identifying features (e.g. level of damage) to facilitate the creation of machine learning algorithms for image classification, which typically requires large amounts of labelled data before accuracy convergence can be achieved.

A case study is presented for Dauphin Island, Alabama, a barrier island in the Gulf of Mexico prone to hurricane winds and flooding. All roads of the island were driven and images were extracted at all building locations. A subset of extracted images were labelled either “elevated building”, “non-elevated building”, or “other.” A Convolutional Neural Network was trained using the labelled images, then used to classify all buildings on the island to determine areas of flood vulnerability. The methodology of collecting broad, passive datasets, extracting serviceable datasets, and training machine learning algorithms to automate processing can be extended to many disaster study applications to limit time and resource allocation and allow more focused field efforts.

Individuals who reside in the United States are subjected to a multitude of natural hazards. Thus, an overall integrated warning system (IWS) is needed that assesses natural hazards, disseminates meteorological information, and spurs protective actions in an effort to reduce risk, save lives, and build sustainable communities from an impending disaster. Storm spotting is one method that has contributed to the advancement of a natural hazard warning system. This proposed research aims to conceptualize the role of storm spotters, as part of an IWS, by being the first of its kind to analyze these community groups from the perspectives of the storm spotters themselves, as well as the three primary IWS actors: National Weather Service employees, emergency management personnel, and broadcast meteorologists. There are three primary goals of this research: to formally define storm spotters, to better understand the characteristics of storm spotters, and to understand the perceived roles of storm spotters as independent, yet complementary actors within a holistic, natural hazards IWS. To accomplish this goal, an exploratory sequential design mixed-methods approach is utilized. The first phase involves focus groups and semi-structured interviews with all four proposed actors. The second and third phases involve quantitative survey development and subsequent research methods to evaluate the generalizability of the collected qualitative data. This research aims to better understand the perception of the diverse network of storm spotters within the context of a natural hazards IWS to potentially improve overall effective practices.

An enduring challenge for practitioners and researchers is the difference between their communication styles. Researchers tend to communicate their findings to one another, which inadvertently creates access and comprehension barriers for others. To address this challenge in the disaster science and management field, the Disaster Research Center (DRC) developed “DRC It!,” a convergent communication initiative that consolidates known research findings for a specific topic and repackages the content for quick and scalable understanding. For each topic, an extensive literature review was conducted, and common themes were identified across the sources. Four public-facing products were created: a full bibliography, a thematic synthesis, an executive summary, and a short, animated video. These products allow emergency managers, media, and other stakeholders to access their desired level of detail on the subject. Topics are chosen based on expressed need by both practitioners and researchers. The two completed topics are hurricane evacuation decision making and business recovery following a disaster. Additional topics are being developed to continue building a bridge between practice and research.

While existing models of infrastructure resilience focus on systems’ performance, they fail to integrate necessary human-centric considerations. As a result, the extent to which infrastructure service disruptions create disproportionate risks and experiences for sub-populations of a community is often neglected. This study, therefore, investigates the effects of disruptions in electricity, water, transportation, food, solid waste, and communication services on the subjective well-being of households. Understanding these details is critical for minimizing inequities in service quality both during and after disasters. Utilizing empirical data collected from household surveys in the aftermath of recent natural disaster events (e.g., Hurricane Harvey and Hurricane Michael), correlation analysis is used to quantify the disparity in well-being risks across households exposed to service interruptions. The resulting analysis aims to characterize well-being experiences regarding different infrastructure services and sociodemographic factors. Overall, the findings confirm an empirical relationship between well-being exposure and experience with respect to vulnerable sociodemographic groups. This highlights the need for integrating societal objectives into standard system performance measures and models of infrastructure resilience.

This study explores the issue of disaster preparedness among Native American communities in the United States. Its goals are to empirically measure disaster preparedness levels within these communities, identify disaster preparedness challenges, and explore various sources of resilience. While past studies have sought to measure preparedness at the individual, household, and organizational levels, few have focused explicitly on the community level, and even fewer have specifically examined Native American populations. Thus, this research is transformative in that it refines and extends existing disaster preparedness frameworks while simultaneously broadening the field's horizon and reach through the inclusion of a historically underrepresented group. To better understand and describe disaster preparedness in Native American communities, this study employs a triangulated, mixed-method approach: surveys of Native American emergency management officials allow us to assess the effects of hazard exposure, past disaster experience, risk perception, and emergency management capacity on community preparedness; semi-structured interviews with these emergency managers provide thick description of a variety of unique preparedness challenges and solutions in these communities; and geospatial analysis identifies and maps the major natural and technological hazards faced by these communities. In addition to identifying the multiple natural and technological hazards facing these communities and the many preparedness challenges they confront, preliminary analyses also reveal a variety of creative and unique solutions to these problems that point toward important sources of resilience in Native American communities.

Hurricane damage to coastal communities is anticipated to increase because of climate change and resulting sea level rise. Buildings in the inundated region often suffer severe damage from waves and storm surge. Currently, most building flood damage models, including HAZUS, do not account for the effect of waves. To quantify damage as a function of storm surge level and wave height for a community, hurricane wave-surge fragility surfaces for five residential wood building archetypes has been developed. While most of the hurricane fragilities were developed based on empirical data, which represents specific sites, the physics-based fragilities developed in this study provide direct load-damage relationships using computational fluid dynamics models for each building component. Therefore, intermediate damage state fragilities can be obtained, which is the novelty of this approach. From aerial image analysis using Google Maps and building footprints provided in GitHub, buildings were mapped and assigned the best archetype for both shape and size to create a spatial distribution of buildings within the subject community. By integrating spatial building fragilities with hazard maps for hurricane waves and surge, risk assessment maps for a community with an array of building archetypes can be obtained. This study aims to provide information that can be used to determine flood insurance rates and help homeowners understand the probabilities that their homes will experience damage from each hurricane category. Finally, a case study of risk assessment was performed for Galveston, Texas, to demonstrate the method presented in this study.

Physical activity may support psychosocial recovery following disasters. We conducted a mixed methods study to examine the feasibility of using physical activity data as an indicator of disaster recovery and the experiences of physical activity-promoting organizations during the disaster recovery period. Using bike and pedestrian path data from four Houston trails, an interrupted time series analysis was conducted to examine the immediate impact of Hurricane Harvey on physical activity and t-tests were performed to examine pedestrian and bicycle counts as indicators of post-disaster recovery. In-depth interviews with 16 representatives of fitness, parks, and active transportation organizations in Houston, Texas and Santa Rosa, California following recent disasters were thematically analyzed to understand the role of such organizations in disaster recovery. Harvey landfall had a significant negative impact on physical activity following landfall, but activity returned to pre-Harvey levels 6 weeks post-landfall. Further research is needed to examine the feasibility and validity of these data as a proxy for recovery and wellbeing. Findings suggest that physical activity is de-prioritized in disaster-impacted communities during the recovery period, though fitness, parks, and active transportation organizations may promote social and physical wellbeing during the disaster recovery process.

The newly formed National Center for Integrated Coastal Research at the University of Central Florida integrates science with societal needs to find solutions for current pressing problems. The group combines geospatial data and scientific processes to create evidence-based, whole-community approaches for increasing coastal resiliency, decreasing vulnerability, and mitigating ever-changing hazards. The overarching goals of the Center are to support empirically based planning and decision-making by leveraging science and technology into time savings for thinly stretched public servants, and provide unbiased and apolitical analyses of hazards and disasters from the perspectives of victims. To meet these goals, the Center has been actively involved in five main research projects centered on social vulnerability, disaster impacts, and mitigation planning. These research projects have:

• Supported community disaster recovery in Florida and Puerto Rico through the development of the Community Development Block Grant Disaster Recovery Program (CDBG-DR) Impact and Unmet Needs Assessment.

• Built the Vulnerability Mapping and Analysis Platform (VMAP), automating social vulnerability mapping and analysis for any state, county, city, congressional district, or watershed in the nation.

• Developed a national hazard location system that shows risk zones for 18 different hazard types, in order to automate the Disaster Mitigation Act of 2000 risk assessment updates.

• Analyzed disparities in federal disaster loss estimates and funding in relation to social vulnerability.

• Explored links between social vulnerability and a variety of outcome measures with multiple stakeholders.

This work, borne out of convergence between public, private, and non-profit entities, supports all levels of decision making before, during, and after disasters.

Design thinking builds on the existing creative, resourceful, and optimistic capacities of youth to drive place-based and human-centered social change. It relies on a simple method to address overwhelming problems: partnering with local community efforts and small businesses to develop twenty-first-century skills, civic engagement, creative confidence, and deeper content learning for all ages. The self and community efficacy that arises as the community solves problems together builds momentum and motivation to tackle climate change adaptation to reduce risk at home and to contribute to a growing global culture of climate innovation. Further, authentic community-based problem solving strengthens integrated decision-making capacity, improves overall readiness, heals and builds mental and physical health, and strengthens social and cultural bonds. Training teachers to build design thinking mindsets and twenty-first-century skills into everyday curriculum empowers them and their students by adding meaning to their work and generating a sense of opportunity optimism. Future generations see themselves as leaders and changemakers in a world where everything depends on our ability to manage change. This poster describes the authors involvement with school, university, and community center educators in Puerto Rico to use design thinking to create integrated solutions for complex problems.

This study explores how top officials at humanitarian organizations are influenced by their own understanding of reality when making decisions about personnel deployment during a disaster. This study expands on disaster literature by exploring how organized convergence originates. Semi-structured interviews were conducted with 30 decision makers. The conversations were transcribed and analyzed using an inductive approach. The article used institutional logics as the theoretical framework. Data from Mexico, the Dominican Republic, and Puerto Rico in the aftermath of the 2017 earthquakes and hurricanes was combined with collected field research data. Primary findings indicated that top officials' decisions were based on the information at hand, as well as how they relate to others. The interpretations and conclusions about the need and purpose of interventions can be seen as biased. Additionally, the research shows that certain symbolic constructions serve as a magnet to make humanitarians search for one another in order to pursue information, validation, and support, which then prevents them from fully understanding the real needs of the affected population.

Warning and alerts studies have repeatedly been found to be ineffective in soliciting protective actions in real emergency situations. These findings have prompted researchers to identify factors that might promote/impede evacuation actions. The Protective Action Decision Model (PADM) is a widely utilized formalization of such decision-making processes. However, the whole causal chains of protective action decision constructs have not been empirically examined. Furthermore, a construct of disaster schema, which guides and orients individuals to construct certain phenomena as disasters, has not received enough attention in protective action decision studies. This study aims to demonstrate the effects of both PADM-postulated and disaster schema-related constructs upon evacuation actions. The subjects consisted of those who were residing in the landslide-prone and recently impacted areas of Hita, Nakatsu, and Tsukumi cities in Oita Prefecture (N=1934, response rate: 32.2 percent). Structural equation modeling was conducted to predict risk averse behavior (evacuations and protective action recommendations) by disaster information; environmental, social, and psychological factors; and disaster schema. As a result, disaster schema demonstrated the strongest total effect on protective responses among all the variables. Furthermore, disaster schema activated System 1 thinking and thus directly provoked protective actions. This indicates that those who were equipped with well-formed disaster schema initiated behavioral responses without involving much slower System 2 PADM causal chains. The results imply that effective planning for public protective action initiation should be based on proper disaster schema building through disaster risk reduction education.

This poster will describe Hagerty Consulting’s “reaching resilience” initiative, our approach to supporting communities in their efforts to become more resilient, and how this initiative has been implemented to support communities in their efforts to enhance resilience. We will first describe the initiative, including our definition of resilience—a community’s ability to withstand, recover from, adapt to and/or advance in spite of acute shocks and long-term stressors—and how we arrived at this definition. Next, the poster will describe in detail our five tenets of resilience: empower a champion, share a unified vision, define metrics for success, create and implement a culture, and maintain accountability. The poster will also describe what is meant by each tenet, outline each tenet’s contribution to reaching resilience, and describe how this initiative is supporting communities in enhancing their resilience nationwide. For each tenet, we will provide in-depth information about a specific community we have worked with, how this tenet shaped the work the community did to enhance resilience, and any challenges this tenet elucidated regarding community resilience. Lastly, the poster will discuss lessons learned to evaluate and monitor the initiative in order to better serve communities and advance resilience research.

Having a preparedness kit and consistently practicing household emergency preparedness, including planning in advance for evacuation and family reunification, is one of the most effective ways to mitigate the effects of natural hazards. Thus, it is imperative to understand preparedness behaviors, barriers to preparedness, and the most effective communication strategies for motivating disaster preparedness and alerting communities to imminent threats. Relying on five years of data from the Chapman University Survey of American Fears, a nationally representative survey, this poster explores the best predictors of preparedness, explains why people choose not to evacuate, and provides details on the best way to communicate emergency information. The analyses pay special attention to vulnerable populations and are broken down by disaster-linked socio-demographic characteristics, such as race/ethnicity, disability status, income, age, education, number of dependents, and region. The poster concludes with concrete suggestions for improving outreach to the public through the use of targeted messaging and specific communication channels.

Earthquakes have become more common in Oklahoma. In the face of new hazards, previous studies suggest that communities should incorporate resilience strategies into their planning in order to cope with emerging hazards. Because of their novelty, particularly in relation to other hazards in the state, we know little about how emergency managers are planning for earthquakes. This research uses in-depth interviews to explore how key stakeholders in local governments are adjusting to this newfound risk across the state. In general, participants describe using strategies that reinforce the status quo rather than building resiliency in response to the shaking. We conclude with recommendations for how local governments can build more resilient communities when faced with new hazards.

Post-disaster recovery and funding can provide opportunities for the promotion of historic preservation efforts. While considerable research exists on both disaster recovery and historic preservation, insufficient research has attempted to clarify the relationship between preservation and its impacts on post-disaster recovery. Moreover, recovery from catastrophic flooding in the American Midwest region has received comparatively limited attention. The present poster addresses these two gaps, through an examination of the relationship between pre-evaluation of historic properties and post-disaster recovery. Using Cedar Rapids’ flood recovery after the 2008 floods as a case study, we evaluate the role of officially designated historic properties in shaping the process of post-disaster recovery. Building upon existing scholarship on disaster recovery and historic preservation, this analysis uses geographic information system property and flood data, Federal Emergency Management Agency and state government records, and interviews with local stakeholders to explore the role of historic preservation in recovery outcomes. Preliminary findings indicate that the presence of designated historic buildings and districts affects recovery efforts and goals in demonstrable ways. The goal of this mixed-methods analysis is to expand our understanding of the challenges and opportunities that preservation plays in the aftermath of a disaster. Additionally, we aim to make policy recommendations for communities facing disaster threats to their historic buildings. 

This research explores household residential dislocation after three hurricanes: Andrew (August 1992), Ike (September 2008), and Matthew (October 2016). The decision to dislocate may be driven not only by damage to structures, but also by social factors related to vulnerability, such as ethnicity, income, and tenure. Surveys, including structural damage inspections, face-to-face interviews with residents where possible, and interviews with neighbors or residential managers where necessary, were conducted with randomly sampled housing units after each hurricane. The responses from three surveys are pooled into a single dataset to allow for analysis and comparison of factors influencing the probability of dislocation. While each hazard event was a hurricane, the actual hazard driving damage was quite different. Hurricane Andrew was a Category 5 hurricane and primarily a wind event; Hurricane Ike was primarily a surge event; and Hurricane Mathew was a riverine flooding event. The analyses attempt to capture the major factors that shape household dislocation: damage from wind, flooding, or both, and socio-economic factors. While structural damage may be a primary factor in dislocation, other factors appear to have consequences as well. Damage type, flood depth, building type, race-ethnicity, income, and tenure were the significant factors affecting the probability of dislocation. The development of this dislocation model will help identify vulnerable residential areas in communities that have a higher probability of population loss due to household dislocation. 

The Pacific Northwest is a very dangerous place when the ground shakes. It is best not to be in contact with the ground when it does. What options are there for avoiding contact with the shaky ground? Flying would shield us from seismicity, but so would being aboard a vessel at sea. Ships, like aircraft, are not in contact with the ground, and tsunamis are a shallow water phenomenon. Mariners have long known that at-sea vessels survive earthquakes and tsunamis. Pacific Northwest mariners, in particular, are noted for surviving fierce storms. The Pacific Northwest mariner infrastructure, with its global affiliates, is therefore in a position to help save lives in the event of a Cascadia megaquake and hasten response and recovery to seismic events. It is therefore critical for local, state, federal, and international Pacific Northwest mariners to become allies in governmental disaster response and take full advantage of nautical and marine resources for emergency executions. It is theorized that the development of a nautical/marine/air/land Pacific Northwest natural disaster government execution stratagem would reduce Cascadia Megaquake casualties and cost estimates. 

This research describes how the U.S. Department of Veterans Affairs (VA) and non-VA home-based long-term care (LTC) environments prepared for and secured the safety and well-being of elderly and disabled persons in the wake of Hurricane Maria, which struck Puerto Rico on September 20, 2017. In-person interviews, home visits, and field observations were conducted in Puerto Rico from January through March 2019. Staff from three of the VA Caribbean healthcare home-based LTC programs were interviewed, as well as caregivers in non-VA LTC environments. Veterans, family members of veterans, family members of VA caregivers, national VA leadership, and community members with expertise in disaster recovery were also interviewed, for a total of 62 interviews and 12 home visits. Preliminary results of qualitative content analysis show that VA and non-VA LTC environments prepared residents and caregivers for Hurricane Maria through the provision of education and recommendations in advance, including whether enough medications and food were on hand. Participants described Hurricane Maria not simply as a disaster, but a crisis, unlike any storm they had experienced. The interconnected nature of the VA seemed to provide a stronger support network compared to non-VA environments that were often independently run. The health of Veterans and non-Veterans was reported to be mostly stable during recovery. Perspectives from VA and non-VA entities allowed for a fuller picture to emerge around how Hurricane Maria impacted homelike LTC environments in Puerto Rico. This research can inform policies and procedures for facilities that care for elderly and disabled populations in areas prone to disasters.

Natural hazards impact communities and people in disproportionate ways, and Iowa is no exception. The goal of this research was to examine how Marshalltown, Iowa was affected by an Enhanced Fujita-3 (EF-3) tornado on July 19, 2018. We examined the impacts of low-attention disasters with respect to damages, specifically renter and socially vulnerable households. Marshalltown was extensively affected by the EF-3 tornado but was not a presidentially declared disaster until roughly two months after. Marshalltown has a higher than average amount of renter-occupied housing and many immigrant community members for a small town. Research has shown that in disaster events, socially vulnerable and low-income neighborhoods tend to recover more slowly than more affluent neighborhoods. Our team conducted surveys of a random sample of 660 households and recorded structural housing damage in the tornado path from September to November. The preliminary findings of our analysis indicate that houses in low-income neighborhoods were more heavily damaged compared to other neighborhoods, especially renter-occupied houses. Despite a lack of alternative housing options, these neighborhoods had not made significant progress toward repairs because of their limited access to financial recovery resources.

Hazards challenge us to reconsider our safety measures, leaving key stakeholders to reduce risks and thereby curtailing vulnerability. When addressing risks, college students and households might have different approaches and views. Person-relative-to-event theory indicates that differences in exposure, experience, and demographic factors shape risk perceptions and influence hazard adjustments. To test this theory, this study uses survey data collected from households and college students in Oklahoma, exploring their responses to earthquakes across the state. Our results show some interesting findings across both populations, and considerable differences between the student and household samples. The results indicate that households have higher levels of hazard intrusiveness and risk perceptions. To capture how residents are adjusting to the earthquake risk, the survey also asks respondents to report their intention of adopting 12 earthquake hazard adjustment actions. Among these items, five of them show significant differences between the two samples. We find that households are more interested in undertaking these activities to mitigate their earthquake risk. Curiously enough, students are more likely to believe that state government will lessen the risk of earthquakes. These findings suggest that, while students are considered a vulnerable population, they offload responsibility for adjusting to the risk to the state government. In addition, students are more optimistic about earthquake hazards and believe state government will lessen the risk in the next few years. 

This study examines the role of social capital on tornado resilience among international students at a university located in tornado alley. International students were recruited to participate in semi-standardized interviews to gain an in-depth understanding of their knowledge and experience with tornadoes, as well as the forms of social capital they depend on for tornado information and warnings. The results of this study illustrate the importance of social capital for international students and its role in tornado resilience. The result suggest that while international students may lack tornado knowledge or experience, the bonding, bridging, and linking of social capital afforded to them contributes directly to their risk perception and resilience. This study addresses a gap in the current research on tornado resilience at universities and may support these institutions in developing a culture of tornado resilience among international students.

U.S. emergency services have been responding to increasingly severe and complex disaster events. Failures in organizational culture, however, can result in operational collapse, misdirected or misaligned responders, delayed or deficient decision-making, or other response issues that negatively affect response operations. When considering how to improve emergency response, more focus is needed on organizational culture and how leadership and decision-making can influence organizational performance. To explore this issue, we used organizational theory to develop a valid and reliable survey designed for use in fire services. This included two pilot studies with eight focus groups with firefighters, factor analyses, and reliability analyses. To date, this instrument has been deployed to three separate fire departments, garnering 1,633 responses, with more departments scheduled to participate. The instrument measures 20 concepts related to leadership, decision-making, and organizational culture. Results suggest that there are strong negative perceptions across the three departments centering on communication with senior leadership, perceived cohesiveness of senior leadership, perceived unfair processes and procedures, a lack of strong and shared values, and a lack of dedication to career development. Future research will explore the relationships between organizational effectiveness, commitment, satisfaction, and trust in leadership and improved organizational leadership, effectiveness, and operations.

Standards for emergency power supplies at nursing homes across the country should consider the area’s extreme temperatures and power failure risk. Hurricanes Irma and Michael knocked out power for 6.5 million and 325,000 Floridians, respectively. Before full power restoration, temperatures in South and Central Florida amid Hurricane Irma damages exceeded 90 degrees, and temperatures reached 88 degrees in Florida’s panhandle following Hurricane Michael. Without moderately cooled environments, the risk for adverse heat-related health outcomes increases, particularly for the elderly. This research, initiated with funding from the Natural Hazards Center Quick Response Grant Program, assessed the role of governance in reducing exposure to heat stress in Florida nursing homes. Community-friendly summaries of emergency power plans from facilities affected by Hurricanes and Matthew were utilized to identify compliance dates and the extent of emergency power supplied at each facility. Records from the Florida Agency for Health Care Administration were assessed to establish a timeline of compliance for facilities across the state. Hundreds of facilities had delays implementing required improvements, regardless of their past hurricane experience. The opportunity to expand beyond the required emergency power supplies to include other life-sustaining equipment was missed for many facilities. Deadlines and extensions set to address public fears did not consider the availability of generators, technicians, and inspectors. In order for reforms to be comprehensive and reduce compliance delays, requirements and industry-specific resources should be carefully considered and planned in advance of legislative action.

This study investigates the spatial and temporal patterns of the April 27, 2011, tornado outbreak in Central Alabama. The 2011 Super Outbreak was the largest, most costly, and one of the deadliest tornado outbreaks ever recorded in the United States. In this study, we examine 29 documented tornado tracks (889 total data points) in Central Alabama and reveal findings related to complex topography and its effects on tornado intensity. The case study on the Tuscaloosa-Birmingham tornado shows that heavy damages occurred at relatively lower elevations and, in contrast to suggestions from other studies that tornadoes divert from their original path in hilly terrain, we found little or no diversion based on the tornado track. Additionally, the temporal trends of this particular outbreak are consistent with evidence from other studies that show a small peak in nocturnal tornado activity in the southeastern United States. The findings from this research may be useful to the emergency management community, planners, and the weather enterprise because they indicate the variation of vulnerability over space and time in the context of natural hazards.

Earthquake alerts on estimated fatalities and economic losses from the U.S. Geological Survey (USGS) Prompt Assessment of Global Earthquakes for Response (PAGER) system are being used by an increasing variety of end-users for situational awareness and response planning, including both key national and international response agencies. With an increasing user community comes an unprecedented expectation that PAGER will function correctly immediately and continuously. However, initial alerts by PAGER within minutes following an earthquake include several large uncertainties, namely initial estimates on location and depth, which can affect the severity of estimated shaking, as well as loss model uncertainties. Historically, PAGER alerts were updated only when improved estimates of earthquake shaking were available from the ShakeMap (via strong motion recordings, finite fault, or “Did You Feel It?” reports), but these updates did not incorporate any reports on damage/losses. We introduce a Bayesian framework for incorporating early reports on fatalities within approximately the first 24 hours to update PAGER’s overall estimates of fatalities and alert levels. The framework accounts for uncertainties associated with the early fatality reports, as well as model-related uncertainties, in order to improve the overall impact forecast. The updating appears to improve the loss estimate and alert level to the correct level within the first 24 hours, even when the initial estimation from PAGER is assumed to be off by two orders of magnitude. Such updating of loss estimates is critical for agencies such as the U.S. Agency for International Development and the Office of U.S. Foreign Disaster Assistance in devising timely and effective response strategies.

This study evaluates the existing standard design snow load for buildings and performs cause analysis for a gymnasium collapse due to snowfall. Frequency analysis was used to estimate the 100-year snowfall (design snowfall), and snow load maps for dry snow and wet snow were derived considering their respective unit weights. In Korea, dry snowfall generally occurs from November to January, while wet snowfall typically occurs from February to April. A comparison was performed between the current standard design snow load set by the Korean Ministry of Land, Infrastructure and Transport (MOLIT) and the proposed design snow load estimated in this study. Generally, the unit weights of snow for dry and wet snowfall are 100 kgf/m3 and 300 kgf/m3, respectively. At the time of the accident, the height of the snowfall was 0.5 m. As per the standard set by MOLIT for buildings, the design snow load of Gyeongju City, Gyeongsangbuk-do, was 50 kgf/m2. However, because the accident happened in February 2014, the actual snow load was expected to be 150 kgf/m2 from that of a wet snowfall. The results show that the newly developed snow load map is expected to be useful in the design of building structures against heavy and wet snow loading, and a proposed design snow load for building design that considers the unit weight of wet snow in the southeastern part of Korea where the highest snowfall occurs during spring season would be valuable.

The August 2014 tailings pond spill at the Buenavista del Cobre mine in Cananea, Mexico brought life in the Sonora River region to a virtual standstill. The only exception was the water trucks delivering clean water daily to thousands of people in the seven municipalities along the river. As a follow up to a Natural Hazards Center Quick Response Grant, the research team went to the area at approximately 8 and 26 months after the spill. They interviewed over 100 residents about the spill’s impact on their livelihoods and where they sought information for how to respond to and recover from the spill. A quarter of people felt that there were no good information sources, while another quarter thought that media sources were the most reliable. More than 80 percent of interviewees said that the people they had specifically talked to were good sources of information. A quarter of interviewees thought the government was the least reliable source of information. It was most commonly reported that interviewees talked about the spill with friends, although respondents reported speaking with a broader set of people (e.g., family, coworkers) at 26 months, compared to 8 months after the disaster. Overall, there was a general distrust of information regarding the spill, particularly from official sources. People tended to trust the information they received from close relationships more.

Evacuation is frequently used by emergency managers and other officials as part of an overall approach to reducing the morbidity and mortality associated with hurricane landfall. In this study, the evacuation shelter capacity of the Houston-Galveston metropolitan statistical area (MSA) was spatially assessed, and shelter deficits in the region were estimated. These data provide essential information needed to eliminate shelter deficits and ensure successful evacuation during future storms. Global Moran’s I and Getis-Ord Gi* statistics were used to assess regional spatial auto-correlation and clustering of evacuation shelters. Shelter deficits were estimated in four ways: aggregate deficit for the Houston-Galveston MSA, evacuation zip-zone, county, and distance or radii of evacuation zip-zone. Evacuation shelters were disproportionately distributed in the region, with lower capacity shelters clustered closer to evacuation zip-zones (50 miles from the coastal zip-zone), and higher capacity shelters clustered farther away from the zones (120 miles from the coastal zip-zone). The aggregate shelter deficit for the Houston-Galveston MSA was 353,713 persons. To reduce morbidity and mortality associated with future hurricanes in the Houston-Galveston MSA, authorities should consider the development and implementation of policies that would improve the evacuation shelter capacity of the region. Eliminating shelter deficits, which has been done successfully in the state of Florida, is an essential element of protecting the public from hurricane impacts.

Repeated exposure to hurricanes, tropical storms, and other natural disasters impacts the mental and physical health of populations living along the highly vulnerable U.S. Gulf Coast. In this study, the self-rated physical and mental health of Gulf Coast residents was estimated, and factors associated with differences in self-rated health were identified. The 12-item Short Form Health Survey (SF-12) was administered online to a sample of 3,030 Gulf Coast residents in December 2017. Responses were scored using the algorithm recommended by Quality Metric to calculate mental component summary scores (MCS) and physical component summary scores (PCS). Multivariate linear regression models were fitted to identify predictors of self-rated physical and mental health among residents. Residents of U.S. Gulf Coast states have poorer self-rated physical and mental health compared to the U.S. population. Women and respondents reporting higher perception of flood risk had worse self-rated mental health, while older adults (for example, 65 + years versus 15 to 24 years), mobile home residents, and respondents with higher perception of surge risk had worse self-rated physical health. Findings indicate that residents of U.S. Gulf Coast states have poorer self-rated health compared to national standards, which may have practical implications for hurricane-associated physical and mental health services planning and delivery. 

Post-disaster housing recovery is an important indicator of individual recovery after disasters. Without the reestablishment of housing, normal activities cannot be carried out, and all dimensions of recovery are delayed. While many studies have shown that social vulnerabilities are critical predictors of housing recovery, only a few studies have used longitudinal data. A 2014 study by Peacock et al. is one such exception that analyzed the trajectories of assessed housing values from pre-disaster to post-disaster periods. The current study used temporary housing residency as the criterion variable and aimed to demonstrate how pre-disaster social vulnerability variables affected length of temporary housing residency after the 2011 Great East Japan Earthquake (GEJE). This study used two sets of population recovery data. The first was from the 2015 Natori City Life Recovery Population Survey (N=1,695, out of 2,331 households, with a return rate of 72.7 percent). The second was the entire record of temporary housing residency in Natori City, Japan (N=3,088 households). These data were integrated, and survival analyses were conducted. These analyses clarified relationships between the Life Recovery Survey social vulnerability variables and the length of stay in temporary housing. The results showed that vulnerable households, such as those with smaller family sizes, those living in rental housing dwellings (before GEJE), and unskilled laborers stayed a significantly longer time in temporary housing. These results imply the necessity of pre- and immediate post-disaster housing recovery planning for these identified vulnerable populations.

Post-disaster recovery is an unequal process—some segments of a community recover quickly, whereas others lag behind because of pre-existing social inequalities. Social vulnerability acknowledges that social structures shape disaster vulnerabilities and recovery outcomes. Public housing residents are among the socially vulnerable groups that rebound slowly in the aftermath of disasters because of their housing tenure, limited access to recovery funding resources, and limited voices in recovery planning. We examined the recovery of public housing units in Lumberton, North Carolina, following Hurricane Matthew flooding in 2016 to understand how funding sources, plans, and policies affect recovery outcomes. This research is part of a larger, interdisciplinary, recovery-based field study conducted by the Center of Excellence on Risk-Based Community Resilience Planning with collaborators from the National Institute of Standards and Technology. Lumberton is a racially and socioeconomically diverse community, with 729 public housing units. Many of these units were impacted by the flooding, which displaced tenants. Two years after the floods, 182 of 267 damaged public housing units are either not repaired or abandoned. Using field notes, recovery resources, household dislocation, and in-depth interviews with local authorities and residents, we investigate the impact of funding resources and policies on the recovery outcome of this community. Findings show that limited financial resources, changes in the state recovery priorities, dependence on governmental support, and bureaucratic funding allocation processes resulted in delays in the recovery of public housing developments.

This study conducted a risk assessment and risk classification for heavy rain damage in the region, and then developed a prediction function for heavy rain damage by risk class. A risk index of heavy rain damage using PSR and DPSIR models was developed for the risk assessment, and risk classes (identified as colored zones) were obtained according to the index. Multiple regression analysis, principal component regression analysis, and artificial neural networks (ANN) were applied to develop the prediction function of heavy rain damage. In order to evaluate the prediction performance of the prediction function, we divided heavy rain damage data into a learning section, from 2005 to 2012, and an evaluation section, from 2013 to 2016. The ANN using the DPSIR model showed the best prediction performance, with an NRMSE of 8.65 percent. Therefore, the ANN model using the DPSIR was selected as the prediction function. Successful prediction of heavy rain damage based on such a prediction function could be very helpful for disaster preparedness and management.

The aim of this study is to develop a heavy rain damage prediction function that can be helpful in preparing for and recovering from localized heavy rain disasters. We collected heavy rain damage as the dependent variable and hourly rainfall data as an independent variable. In addition, we classified damage data into four categories according to damage scale and obtained various rainfall variables with different durations. We applied regression models, machine learning (decision tree, random forest, support vector machine), and deep learning (deep neural network) techniques to develop a categorical heavy rain damage prediction function. To verify, we used accuracy and the F1-score, which are classification performance evaluation indicators. We then selected the model with the highest prediction performance as the final model. The best prediction performance model was the heavy rain damage prediction function using a deep neural network with a deep learning technique (accuracy 75 percent; F1-score 38 percent). These outcomes could be useful for disaster preparedness, recovery, and management.

The advisory and warning information being used in the Korea Meteorological Administration (KMA) is the Heavy Rain Report. This advisory information is given to the public when the maximum rainfall exceeds 60 mm over a three-hour-duration. However, this information may have different meaning according to the region. In this study, we consider the regional characteristics of damage and rainfall. To do this, we define damage and hourly rainfall data using the Hazard-Triggering Rainfall. One to 24-hour-duration rainfall variables in a region is obtained and the variables are reduced to one principal component (PC) for analysis. Then, the frequency of the PC score is obtained in each class interval, and the probability that damage will occur is estimated for each class. The Hazard-Triggering Rainfall is defined as the PC score, which has the damage probability of 50 percent. We divide the data into groups based on a learning time period (2005 to 2013) and evaluation time period (2014 to 2017). Ten local governments were selected for testing the performance of the Hazard-Triggering Rainfall using a F1-score. As a result, the F1-score for the report of KMA was 46 percent and the Hazard-Triggering Rainfall was 57 percent. Therefore, Hazard-Triggering Rainfall may be more suitable for the information in this case study. We also consider that the regional characteristics of damage and rainfall may need to include more properties of the region in future studies.

This poster will provide a summary of articles that were published in a special issue of Southern Economic Journal. As a review, Lim and Skidmore focus on twenty years of cross-county NFIP data and provide some of the first evidence that NFIP has reduced flood-related fatalities. Ahmadiani et al. examine trends in flood insurance and mitigation activities and Sims and Null employ a stochastic dynamic programming approach to assess the usefulness of downscaled climatological and hydrological forecasts in guiding flood mitigation investments. Frimpong et al. examine factors affecting homeowner acceptance of “buyout” for high risk properties. Bakkensen et al. identify empirically the impact on property sale prices following a cluster of hurricanes. Koning et al. integrate an adaptive agent-based modeling and hedonic estimation approach to measure the effect of flood insurance uptake on property value. In the concluding article, Miller et al. suggests a way forward for the NFIP. The authors argue that the two initial goals of NFIP were to (1) shift risk away from the tax payer and onto the those residents who choose to locate within a flood risk area, and (2) provide universally available and affordable flood insurance options can be achieved by adopting risk-based pricing in tandem with subsidization of low-income households.     

Droughts are often responsible for greater damages than other natural disasters. The U.S. National Oceanic and Atmospheric Administration has identified drought events as some of the most serious natural disasters ever observed and has noted that rising temperatures and precipitation amounts due to climate change could increase damage from droughts, as well as their direct and indirect impacts. The present study aims to develop estimation equations of drought damage for each continent by classifying the global drought damage data in Africa, the Americas, Asia, Europe, and Oceania. This study analyzed data provided by the Centre for Research on the Epidemiology of Disasters through the emergency events database EM-DAT. From 1900 to 2018, 316 droughts occurred in Africa, 161 in the Americas, 173 in Asia, 45 in Europe, and 30 in Oceania, respectively, for a total of 725 drought events. Many casualties were attributed to these droughts, including 867,131 in Africa, 77 in the Americas, 9,663,400 in Asia, 1,200,002 in Europe, and 684 in Oceania, for a total of 11,731,294 deaths. Based on a classification by continents, this study develops an estimation equation considering drought damage and casualties.

Stormwater management infrastructure is designed to mitigate flooding and influences the flood exposure of an area (e.g., a watershed and/or neighborhood) by managing runoff and streamflow. Cities or counties invest in stormwater management infrastructure through comprehensive plans and capital improvement plans that are based on the consideration of the life-cycle of infrastructure, land use/land cover change or increasing rates of urbanization, and local climate change impacts. A growing number of studies have revealed the unfair distribution of stormwater infrastructure across neighborhoods based on the racial/ethnic composition, suggesting discrimination in the provision of infrastructure. However, current research is limited by the types of infrastructure considered. To fill the gap, this study examines the distribution of stormwater management infrastructure across diverse neighborhoods in the city of Houston, Texas, with an environmental justice framework. Specifically, we examine the structure of different types of infrastructure, including roadside open ditches and buried grey infrastructure. Then, we use bivariate correlations and regression to analyze the distribution of infrastructure across neighborhoods (block groups). We hypothesize that socially vulnerable neighborhoods have more open ditch systems and less grey infrastructure for stormwater management when controlling for density. Preliminary results support the hypothesis and unequal distribution of stormwater management infrastructure in Houston. This study provides empirical evidence in environmental justice research to describe the provision of stormwater infrastructure as an environmental good to reduce the risk of flooding that is potentially distributed unequally across neighborhoods based on race/ethnicity and income.

Communities in Colorado are increasingly experiencing disruptions from fire, flood, drought, and extreme heat. With this rise in hazardous events, there is a pressing need for communities to increase their resilience. An interdisciplinary team from the Cooperative Institute for Research in Environmental Sciences Education and Outreach Program is developing and implementing an innovative, action-oriented youth engagement project that targets rural Colorado students, teachers, and communities. Our engagement model empowers youth to envision community resilience through immersive scenario-based role-play based on a solid understanding of the relevant science; learn about natural hazards through engaging, Colorado-focused lessons; initiate conversations about hazard preparedness and responses from within communities; and develop and implement student-led resilience action projects. The project team is developing instructional materials for middle and high school students: four lesson plans focused on different hazards (fire, flood, drought, extreme heat), four complementary scenario-based role-play games with a focus on youth empowerment, and a teacher workshop based on these materials. Each school implementation follows a sequence in which the lesson plan activities are conducted first, followed by a scenario-based role-play game and reflection. Building on their experience with the game, students develop resilience strategies for their communities and present those at a community Resilience Expo. Over the course of the three-year program, the project activities will train and support 140 teachers, engage over 400 students, and foster 11 Resilience Expo events across Colorado, from primarily rural communities. The instructional units and the games will be used in classrooms with over 600 students.

During the disaster recovery process, it is challenging to continuously assess community health and well-being. Near real-time data of the fitness activity level throughout a community can be a valuable source for assessing community recovery and for building capacity for community resilience. This exploratory study used cycling and running activity records from a personal fitness monitoring app as a proxy for community health and wellbeing. This large-scale data enabled us to assess the post-Hurricane Harvey recovery of the fitness activity level throughout Harris County, Texas. The data revealed geospatial patterns of fitness activity level recovery, including how long it took for an affected region's fitness activity level to return to the pre-Harvey level. This study highlighted the need for more in-depth studies on the validity and limitations of such data in assessing community health and wellbeing while demonstrating its value in supporting a near real-time, scalable method to monitor the recovery of affected communities. 

In the past decade, overwhelming amounts of building damage data have become available after major earthquakes, ranging from engineering models, crowdsourced estimates, and remote sensing maps. However, this data has highly variable quality, making it difficult for stakeholders to trust and use it as a basis for crucial recovery funding decisions. This project aims to reduce the mental effort of sifting through post-disaster damage data by creating a data fusion framework that stakeholders can use to combine multiple sources of damage data into a single estimate of damage. Using building damage data collected after the 2015 Nepal earthquake, we show that the data fusion framework results in a more accurate damage map than a single damage data source. Post-disaster, stakeholders can use damage maps from this data fusion framework for post-disaster needs assessments, where they often face a trade-off between accuracy and data collection speed. 

As the frequency and intensity of natural disasters increases, the threat of property damage also rises, thereby exacerbating the total cost of damage, the share of homeowners who seek disaster assistance, and the amount of assistance requested. Limited funds for disaster assistance, combined with rising demand for such funds, dictates a broader understanding of the long-term mobility decisions of homeowners affected by disasters. Funded by a Natural Hazards Center Quick Response Grant, the authors conducted 15 interviews with homeowners in Lumberton, North Carolina, who received disaster assistance from the Federal Emergency Management Agency for property elevation, reconstruction, or acquisition following Hurricane Matthew. The unique timeline of the distribution of disaster assistance, which occurred between Hurricanes Matthew (2016) and Florence (2018), presented an opportunity to position homeowners’ mobility decisions within the context of recurrent disasters. Therefore, the interviews addressed both homeowners’ decision-making processes when applying for disaster assistance and changes to homeowners’ perceptions of their decisions following Hurricane Florence. The authors find that wealthier homeowners are more likely to choose acquisition than either elevation or reconstruction, while socially vulnerable homeowners expressed concerns about the affordability of relocation. Age and the duration of homeownership also played a significant role in homeowners’ decision-making processes: a majority of the interviewees, many of whom are elderly, had lived in their homes for several decades and expressed no desire to relocate. Research outcomes can inform disaster recovery policy of factors that shape the mobility decisions of homeowners affected by natural disasters and changes in the perceptions of those decisions in the wake of increasingly frequent recurrent disasters. 

Many countries, including Taiwan, face new emergency management challenges because of an aging population. In the last decade, many senior citizens’ welfare institutions in Taiwan have experienced disaster events such as flooding. Understanding the emergency behavior of the institutions is a foundation for risk reduction. This study aims to examine the flood response characteristics of the institutions to provide a scientific foundation for policymaking. Data about the flood preparedness and response behaviors of senior citizens’ welfare institutions in Taiwan was collected in late 2018. The survey collected about 360 responses from 1,100 facilities. Approximately 25 percent of the institutions are located in flood-prone areas. The emergency plan qualities and emergency exercise frequencies seem to be adequate, but the real capabilities still need to be verified. The results also show that the institutions tend not to adopt precautionary evacuation because of operational continuity and cost considerations. This study suggests that the government should strengthen the institutional efforts for disaster preparedness and reduce the obstacles to evacuation. This will make the institutions more capable to conduct sheltering-in-place or off-site evacuation under continuity of care and reduce the risk of senior citizens’ welfare institutions in the future.

Heavy rainfall associated with Hurricane Maria triggered over 40,000 landslides on the island of Puerto Rico. In the aftermath of the disaster, Puerto Rican officials expressed a desire for educational materials related to landslide hazards. To address this need, the U.S. Geological Survey and the Natural Hazards Center at the University of Colorado Boulder have partnered with geologists and students from the University of Puerto Rico Mayagüez to develop landslide outreach materials. We aim to increase awareness of landslide threats and inform risk reduction strategies by creating a range of products tailored to the Puerto Rican context that emergency managers and mitigation practitioners can use to communicate landslide risk to residents of Puerto Rico.

This poster will feature a mock-up of a Spanish-language graphic-based educational booklet on landslides that reflects the local geologic and cultural setting of Puerto Rico. Printed pamphlets are a commonly used form of educational materials for landslide hazards, and we examined a range of existing pamphlets to inform the development of the educational materials for Puerto Rico. Informal interviews and existing risk communication literature targeted to communities on the island suggest that such pamphlets are useful but not sufficient to adequately communicate hazard risk in Puerto Rican communities. Thus, the booklet we are developing will be the first step in a broader effort to facilitate discussions with disaster risk communication stakeholders and will be expanded into multiple modes (e.g., animated videos, flyers, and social media posts) to be distributed in community-based settings. The project will use ongoing monitoring and evaluation to update the educational materials based on user input and to inform future risk communication efforts. 

Governments in Taiwan have developed sheltering strategies in their local emergency management plan to build disaster sheltering capacity. However, shelter preparedness has never been systematically examined. Without proper evaluation, stakeholders cannot effectively improve sheltering preparedness before an emergency. Therefore, this research aims to develop and implement an evaluation framework of sheltering preparedness in study areas to identify the sheltering preparedness characteristics in Taiwan. This research adopts cluster sampling and evaluates 68 shelters in the two township areas of Taichung and Tainan City. Researchers conducted site surveys and interviews to collect hazard, building, facility, emergency functions, and operational characteristics of the samples. The findings indicate that shelter designation is related to the authority of the township office. Most of the shelters are civic activity centers that host less than 100 people. Conversely, only a few schools are designated as shelters. The location and structure are verified in all the shelters to ensure safety after a disaster. However, staffing planning (quantity and training) for shelter operation is unrealistic. Governments require additional efforts toward improving current sheltering preparedness to increase capabilities before the next extreme event.

Evaluating and modeling evacuations requires input from both engineers and behavioral scientists. This project leverages an interdisciplinary team to analyze the decisions made by people in the path of an impending hurricane. Data on decisions, information seeking, and uncertainty levels were taken from a post-Hurricane Matthew household mail survey. This data was used to examine several questions, including the order in which evacuation-related decisions are made, which is modeled using multinomial random parameter modeling. The outcome is an explanatory model for the decisions that households tend to make first when faced with hurricane evacuation. The data also examine the role of certainty in the decision to evacuate or stay and the factors associated with certainty levels about living in an evacuation zone; evacuation preparedness time; evacuation transport mode, route, destination; and hurricane characteristics. In addition, households’ access to and use of information sources, both from media and personal networks are analyzed to predict perceptions regarding information and uncertainty. The evacuation decision and its antecedents between household members are also explored and compared. The Interdisciplinary Research in Hazards and Disasters (Hazards SEES) project will culminate with the integration of behavioral decisions at the household level into a high-fidelity agent-based simulator that considers different disruptions to a road network and measures travel times and accessibility impacts.

Disaster research is a relatively young field that is only two or three scholarly generations old. As such, many remember firsthand the voices and perspectives of senior scholars and those who conducted research that is foundational to our collective understanding of disasters. As these scholars leave the field, they often leave behind valuable information in the form of project notes, draft manuscripts, datasets, and other supplemental materials collected over time. These materials may be invaluable to current and future generations of researchers addressing related problems from new perspectives. The increasing interdisciplinarity of the field, combined with renewed interest in disaster research by outside parties (e.g., those interested in the impacts of climate change), necessitates both a secure physical home and organizational mechanisms to keep these data collections safe and ensure their availability to all who may wish to use them. Preservation of these materials will allow for the advancement of disaster science, cultivation of a deeper scholarly understanding of previously studied topics, and a broadened public awareness of various disaster-related topics. This poster will describe and depict the acquisition processes, contents, and uses of scholarly papers of William A. Anderson, T. Joseph Scanlon, and Enrico L. Quarantelli, including a comparative content analysis of each collection. 

Older and/or disabled people have been known to suffer more serious damages during disasters. The 2014 Great East Japan Earthquake in Tatsuki demonstrated that the root causes of proportionately greater damages are the siloed approaches taken by social services and disaster management organizations and the lack of coordination between regular and disaster response operations. One solution is to involve social workers, who regularly make plans for everyday community needs, and ask them to simultaneously prepare disaster care plans. This paper reports a project that links regular social services and local responses during disasters to persons with disabilities (PWD). In 2016, a three-year project was launched in Beppu City, Japan that led to a standard operating procedure (SOP) for assessment, informal human resources matching, and disaster response simulation during disaster drills. At the end of the project, a quasi-experimental, propensity-score-matched impact evaluation demonstrated a significant increase in disaster risk reduction literacy scores only among the experimental group PWDs. In 2018, Hyogo prefecture initiated the Beppu-model SOP utilization grant program. Based on the preliminary results in Harima township, one of the two initial municipalities, Hyogo prefecture decided to expand the grant program to all Hyogo municipalities. As a result, 37 out of 41 local governments have applied for the project as of 2019. This poster concludes with a preliminary impact evaluation of the Beppu Model SOP implementation/utilization project in Hyogo. 

This year is the thirtieth anniversary of the Hazard Reduction and Recovery Center (HRRC) in the College of Architecture at Texas A&M University. This center is situated in urban planning and landscape architecture and draws upon this scholarship and practice to support research, education, and outreach around hazard and disaster concerns. The HRRC includes the expertise of planners, landscape architects, sociologists, geographers, architects, policy analysts, economists, and engineers. The mission of the HRRC is to increase our understanding of the nature and impact of natural and technological hazards upon humans and the physical and built environment in which they live and to increase our knowledge regarding hazard mitigation, preparedness, response, and recovery. The mission is also to enlarge the hazard research community through graduate student training, faculty development, and educational endeavors, as well as to disseminate research findings to the research community and to practitioners so they can use this information to mitigate, prepare for, respond to, and recover from disasters. The HRRC provides assistance and consultation to those state, national, and international agencies responsible for hazard analysis, emergency preparedness and response, disaster recovery, and hazard mitigation. This poster provides highlights of the HRRC over the past 30 years and recent developments.

Although post-disaster housing is part of community recovery, its modeling is still in its infancy. This research aims to provide a spatial regression model for predicting households’ recovery decisions using publicly available data. For this purpose, a Bayesian hierarchical geostatistical model with spatial random effects was developed. To calibrate the model, household data collected from Staten Island, New York in the aftermath of Hurricane Sandy was used. The model revealed that on the scale of census tracts, residents with higher income or larger household size were significantly less likely to reconstruct. In contrast, odds of reconstruction rose with increase of long-term residents. The model outputs were employed to develop a propensity score for each tract to predict the probability of reconstruction/repair in each census tract. The propensity scores developed in this study can serve as a decision-support tool to tailor recovery policies.

Post-earthquake damage assessment can be significantly expedited when machine learning (ML) algorithms are used. Currently, there are few rapid quantifiable methods to determine if buildings are safe for reoccupation after extreme events. However, recent advances in remote sensing, computing technologies, and data science have prompted the development of ML methods that can assess and quantify the conditions of structures in near-real time. This study introduces a framework known as the human-machine collaboration (H-MC) for structural health monitoring (SHM). The H-MC is a model in which humans co-work with artificial intelligence to complete specific tasks. The H-MC for SHM uses the speed of machine learning techniques and expertise of structural engineers for rapid post-earthquake damage assessment. In this framework, response data from undamaged structures are used to develop an ML model. Concurrently, analytical models are developed by the domain experts. During an earthquake, ML tools rapidly generate damage-sensitive features, and using the simplified analytical model of the structure developed by a domain expert, rapid damage detection is occurs within seconds after an event. The effectiveness of the framework is shown using data from real instrumented structures.

Tropical storms, the main driver of storm surge in many regions, can also generate heavy rainfall. The resulting combination of riverine, pluvial, and coastal flooding can cause significant losses, particularly in densely populated coastal environments. Hewanorra International Airport (HIA), located in a low-lying area of Saint Lucia, is the only international airport on the island, and is susceptible to flooding caused by heavy rainfall, high river water levels, and storm surge. The airport has been shut down in several extreme events (e.g., Hurricane Tomas in 2010) because of runway and terminal flooding. The closure of HIA, as well as flooding in surrounding urban areas, has led to substantial economic losses. This study set up and calibrated a two-dimensional sub-grid hydrodynamic model (LISFLOOD-FP) to simulate inundation extents caused by compound flooding. The model is forced with rainfall observations and estimated storm surges. A lumped rainfall-runoff model (HYMOD) is calibrated to generate upstream flow boundary conditions. LISFLOOD-FP simulations are validated using aerial videos and soil moisture maps, as well as available satellite images from Sentinel-1 and Pléiades sensors. Sensitivity analysis is conducted to account for the uncertainties of input data by applying design rainfall/runoff events and extreme sea levels with different return periods. Results show that representing the compounding effects of multiple flood factors can lead to more accurate inundation extents. Rainfall is the major contributor to flooding in this area. 

While a common view of the post-disaster landscape is one of chaos, panic, and violence, these perceptions are myths. In reality, people help one another, and masses of people, many of them volunteers, converge on the affected area to aid the survivors. These volunteers arrive for many reasons, including altruism, curiosity, or a personal need to participate in the disaster. This spirit of helping is not isolated to the first moments after an event; rather, it continues for months and possibly years into the recovery effort. Many of these individuals who travel from outside the affected area exist at the intersection of volunteer tourism and disaster tourism, in addition to their unique position as disaster volunteers. Given the number and influence of volunteers following disaster events, it is important to explore how these individuals understand their roles within the post-disaster landscape. This research investigates how volunteers situate themselves and how they construct their own identities during and after their volunteer service. The author uses autoethnography to analyze her own experiences as a disaster volunteer after five events (four within the United States and one international), as well as 10 in-depth interviews from former and current volunteers. Through this analysis, the author presents a better understanding of how disaster aid workers belong to a disaster site and construct their post-service identity in relation to place.

Hazard mitigation planning generally takes place at the municipal or county level and is rarely incorporated into comprehensive planning. This poster reports on the initial stages of a community workshop-based action research project to consider multi-hazard mitigation as a step in updating a local comprehensive plan. This work aims to: identify opportunities for more robust mitigation strategies than is typical in “worst-case” hazards planning processes (i.e., strategies that are appropriate for more than one hazard type and severity level, and for ongoing community development goals), and to articulate the nature of conflicts between mitigation and those goals. In order to capture meaningful robustness in the strategies they generate, community stakeholders considered scenarios of both gradual change (sea level rise) and sudden, more existentially threatening change (near-source megaquake tsunamis). The context for the project is the Washington coast, where there is a 10 to 15 percent chance of a magnitude 9.0 earthquake along the Cascadia Subduction Zone within the next 50 years. Coastal areas are preparing for shaking intensities, tsunamis, subsidence, and other hazards resembling the 2011 Great East Japan Earthquake and Tsunami. We found that values-driven, asset-based participatory community workshops in which stakeholders can consider both chronic/gradual and episodic/sudden hazards can aid in producing long-term, localized resiliency plans. Consideration of multiple low- to high-severity hazard scenarios is helpful for developing robust land use strategies for inclusion in comprehensive plans, while focusing only on severe (but low-probability) existential threats produces a more limited set of strategies.

Seismic intensity scales are based on three observations: geological changes, impacts on the built environment, and the response of humans who encounter various levels of ground motion. Historically, intensity scales have incorporated changes based on scientific and engineering research in the first two scale components. However, assumptions about human behavior not based on empirical social science were either carried forward or dropped from recent versions of these scales. Our research utilized the USGS “Did You Feel It?” database to explore human response to several large earthquakes around the globe, including events in Haiti, Pakistan, Nepal, China, Mexico, Japan, New Zealand, and the U.S. The study examined how people respond to large earthquakes, in developed vs. developing nations, in different cultures, and at different shaking levels. The objectives are to better understand human behavioral response in disasters and provide a basis for incorporating empirically grounded observations in seismic intensity scales. The study is a collaboration of Japanese, American, and Korean investigators in a cross-disciplinary pursuit of “Social Seismology.” Social scientists have long demanded a place at the table in natural hazards areas dominated by physical scientists and engineers. This multi-national and multi-disciplinary study provides a social science perspective on a tool long utilized to describe the effects of an earthquake, one that could and should be grounded in empirical research.

The occurrence of drought is difficult to classify in time and space compared to other natural disasters. In addition, drought causes damage to agriculture, life, and industry. Climate change has resulted in an increase in natural disasters globally, including rising temperatures, rising sea levels, extreme heavy rain, and megadrought. This study is intended to develop an estimation equation for the potential occurrence of drought by using global data on drought damage since 1900. In total, 725 droughts occurred in 146 countries from 1900 to 2018. In 1983, 32 damaging events were recorded, the highest number in one year during the observation period. In addition, 38 damaging droughts occurred in China, which experienced more damaging droughts than any other country. Examining drought damage by periods of approximately 30 years, 18 droughts occurred from 1900 to 1930, with 16 droughts from 1931 to 1960, 250 droughts from 1961 to 1990, and 441 droughts from 1991 to 2018, respectively. Compared to the early 1900s, damaging events increased by nearly 10 times through 1961, and by more than 20 times after 1991. This study aims to develop an estimation equation of potential drought globally by statistically analyzing drought damage data.

Emergency management and crisis response have long been acknowledged to be interdisciplinary fields. Recent events have especially highlighted the overlap between emergency management and emergency public health. This paper explores the ways disciplinary perspectives shape crisis response. Examining the cases of the Nepal earthquake on April 25, 2015 and the 2014-2016 Ebola epidemic in West Africa, this study examines the planning and implementation of crisis medical relief efforts initiated by a range of governmental and nongovernmental organizations. Through interviews and document analysis, the study captures the role that disciplines had in shaping the planning and implementation process. Interviewees reflected diverse disciplinary backgrounds and positions within the relief organization. Findings indicate that, despite engaging in similar strategic processes such as developing situational awareness and decision-making, disciplinary differences emerged in how people and organizations engaged in these processes. In particular, perspective (shaped by discipline, organization role, and background) influenced the kinds of issues people focused on and how they approached the decision-making process. While these differences presented some challenges for interpersonal interactions among relief workers, there is evidence to suggest that ultimately the existence of these interdisciplinary teams and the involvement of group members from different backgrounds benefited the relief planning and provision processes. This highlights the need for the integration of interprofessional training in professional settings and in education programs, particularly in the fields of emergency management and emergency public health.

In the past three decades, an increasing severity and frequency of disasters have been acknowledged. As a response to this growing trend of devastation, a greater dependence on post-disaster housing is emerging. This leads to concerns with regards to the housing effort in post-disaster settings. Hurricane Harvey’s devastation caused prolonged displacement for nearly 40,000 people, with 8,000 families occupying all vacant rental options/hotels and an additional 30,000 occupied emergency shelters. This number of displaced peoples resulted in a temporary housing dilemma that caused the government to spend as much as $250,000 for setup and maintenance for each trailer or manufactured home deployed. This study addresses the dual challenge behind excessive expenditure and disregard for community needs in post-disaster temporary housing allocation. The proposed approach adopts a simulated demand forecasting method and a supply chain optimization model which accounts for all rental, hotel, and temporary housing options. This combination creates a model that optimizes the supply chain and provides the most cost-efficient stocking quantity of temporary housing. As natural disasters continue to batter American coasts, the post-disaster housing dilemma will only continue to persist. This study addresses this challenge through a supply model with the potential to reduce the associated financial risk by nearly $200 million of expected losses in comparison to historical U.S. housing inventories. 

Increasing weather-related hazards have caused significant damages to U.S. housing stock, as exemplified by the 2018 billion-dollar hurricane season. Reconstruction of damaged residential housing is an essential component of housing recovery and long-term resilience of communities. Post-disaster housing reconstruction is a resource-driven process that is influenced by the availability and accessibility of financial capital and construction industry resources. Previous case studies have illustrated that housing reconstruction outcomes are uneven across different disaster-affected regions of the United States and may be attributable to unequal access to reconstruction resources. However, there is a paucity of quantitative studies exploring the strength of spatial relationships between resources and reconstruction outcomes across varying geographical regions of the United States. Although qualitative case studies have identified socioeconomic factors as a catalyst for post-disaster resource accessibility, there are a lack of quantitative studies that correlate socioeconomic characteristics with post-disaster housing reconstruction outcomes at the regional scale. Using geospatial regression analysis tools, this study develops two predictive models using Ordinary Least Square regression (OLS) and Geographically Weighted Regression (GWR) to explain housing reconstruction patterns and explore spatially varying local relationships between socioeconomic, construction industry, and federal government resourcing factors and housing reconstruction outcomes. OLS results show a positive relationship between resourcing variables and housing reconstruction outcomes, while GWR results reveal that the relationship between housing reconstruction outcomes and resourcing variables significantly varies across regions. The study will help policymakers identify regions vulnerable to resourcing crises for housing reconstruction after large-scale disasters and effectively allocate reconstruction resources.

Through the United States Army Corps of Engineers (USACE) Silver Jackets program and the USACE national nonstructural committee, an inter-agency partnership between state agencies, including the Kansas division of water resources and the Missouri State Emergency Management Agency, has begun to provide improved technical services to communities wrestling with a new era of flood risk management. The presenter will discuss how communities must diversify their approaches to comprehensive floodplain management and not rely on a single measure to manage risk. The poster will have case study specifics for several recent Silver Jackets projects in Kansas and Missouri, including a leveed community in Florence, Kansas that embraces nonstructural measures, and several communities using floodplain management plans, such as Ottawa, Kansas, and Independence, Missouri. The poster will describe the full set of measures used for flood risk management, putting an emphasis on nonstructural measures, including elevating structures, dry and wet floodproofing, use of emergency action plans, and the establishment of flood warning tools. Community planners dealing with the maintenance of levees, whether physical maintenance or levee accreditation, will benefit from learning about important next steps as communities deal with flood risks, from changing rainfall intensities to rising flood insurance premiums. Floodplain managers will learn the importance of not relying on one major feature, in case it should fail during the next major flood. Viewers will gain an understanding of sustainable flood mitigation using flood risk management plans or floodplain management planning, where nonstructural assessments enhance the effort.

An Enhanced Fujita-5 (EF-5) tornado struck Joplin, Missouri, on May 22, 2011, destroying 25 percent of the structures in Joplin and damaging an additional 50 percent. On the ground for 38 minutes, the tornado claimed 161 lives and injured over 1,150 people. Both spontaneous and formal memorials appeared in the days and weeks following the disaster. The Joplin Memorial Run (JMR), rebranded in 2012 as a memorial to the tornado victims, draws participants from across the country to run in the events and perform commemorative activities. The JMR offers a half marathon, 5K, team relays, and a children’s fun run, in addition to the Walk of Silence, a remembrance honoring those who perished in the tornado. Given that the JMR seeks to draw both runners and those wishing to pay tribute, this study examines the role of the memorial race in the long-term community recovery and specifically examines the runners’ motivations for participating in the race events. To answer these questions, we attended the 2018 JMR and conducted over 400 semi-structured interviews with race participants in addition to participant observations from the race day and the Walk of Silence. The research identified that while some people attended the JMR solely as runners or to pay tribute, most participants cited multiple motivations for their involvement such as a love of running, paying tribute to lost loved ones, supporting Joplin’s ongoing recovery, and strengthening community and familial bonds. Finally, we consider the ongoing role of the JMR in the recovery of Joplin.

Disaster trauma is best understood as a neurobiological “injury” that responds better to rapid interventions than to traditional talk-therapy models of care. Current approaches to Psychological First Aid (PFA) and Critical Incident Stress Management (CISM) have received critique validated by practitioners working with disaster survivors. As evidence-based research into emerging neurobiological and somatic interventions advances, an array of new methods to treat disaster trauma is available. A convergence of the treatment methods studied can be synthesized and taught to survivors to be self-administered, thereby increasing the utility and availability of this type of care. This poster will highlight research into over fifteen modalities and their applied benefit to disaster survivors as a convergent option for post-disaster care. 

Integration of these methods into community recovery programs could reduce suicide rates, stress related illness, unemployment, domestic violence, depression, PTSD, or other stress related conditions that compound in the months and years after a catastrophic event. These evidence-based trials reveal a high degree of effectiveness for alternative and somatic approaches in post-disaster treatment. 

Sparked by the acquittal of police officers accused of beating Rodney King, the 1992 Los Angeles Riot is often framed as a race riot. While this makes intuitive sense, as Rodney King was a black man who was badly beaten by white police officers, and because this sparked major conflict between blacks and Hispanics and Koreans and whites, the authors argue that class—in addition to race—offers a more compelling explanation for this behavior. The analysis uses a comprehensive dataset compiled from five sources to explore theories of Deprivation, Assembling Theory, and Emergent Norm Theory. Through this, the authors find that race and class are closely linked in the United States and offer a new explanation of the riot as a protest by people of low socioeconomic status against their disenfranchised position.

Disaster management should include Texas Freedom Colony cemeteries for their historical and cultural significance. With Hurricane Harvey still fresh in recent memory, archaeologists are questioning the best methods to help save and preserve cemeteries from natural disasters such as flooding and severe winds. The Sweet Rest Cemetery in Tamina, Texas is prone to flooding due to poor street infrastructure, lack of drainage, and sewer lines. This research applies tasseled cap transformation to measure the change in land brightness, vegetation, and wetness of Sweet Rest Cemetery using Landsat 7 ETM+ raster data files dated October 1999 and October 2017. The results illustrate a positive feedback loop between land development and the cemetery environment.

Hurricanes Harvey and Maria, Superstorm Sandy, and other natural hazards in recent years have cascaded into devastating and widespread social and technical disasters. Extreme weather events have caused disasters along failure pathways that are increasingly interconnected across supply chains, financial services, energy grids, civil infrastructure, and other socio-technical systems. This work develops a qualitative framework to characterize how system interdependencies, land-use patterns, and demographic factors combine to produce catastrophic disasters following Hurricane Florence. We evaluate how these system shocks triggered overflows at concentrated animal feeding operations, breaches of coal ash ponds, and failures in energy systems. We used event and decision trees to identify the causes of failures and to understand how the cascading disasters could have been avoided, in order to better prepare for future hazards. We identified daily management practices, entrenched regulations, and policies, as well as economic interests of key stakeholders that contributed to these disasters. We present recommendations to improve the resilience of integrated sociotechnical systems in the face of natural hazards in North Carolina.

In the face of climate change, plans need to anticipate future flood risk. We propose a monitoring framework to assess collective climate adaptation efforts by all plans that affect development in a community. The framework is driven by two questions: Are plans consistently reducing vulnerability to current and future flood risks? Are there short-term policies that exacerbate vulnerability in the future? Following the resilience scorecard developed by Berke and colleagues, the framework identifies land-use policies from a city network of plans and scores them “+1,” “0,” or “-1,” based on their effectiveness against three future flood risk scenarios—current, 2050, and 2100. Scores are then aggregated by neighborhoods to assess the spatial distribution of vulnerability. The resulting scoring matrix identifies adaptation gaps in short-term policies that induce vulnerability in future flood-risk scenarios (scored “-1”), policies that conflict with other adaptation efforts, and neighborhoods that reveal high vulnerability. These adaptation gaps indicate the need for reducing conflicts among policies, avoiding vulnerability-inducing policies, implementing additional adaptation policies, and monitoring policies over time. As a pilot case, we use the framework to assess flood adaptation in Dordrecht, Netherlands, a highly flood vulnerable city that is world-renowned for its adaptation efforts. Results suggest that while fairly robust in the current scenario, several Dordrecht neighborhoods reveal adaptation gaps in 2100. These gaps are primarily due to policies that increase housing intensity, which increase vulnerability in the 2100 scenario despite being effective against flood risks in the current scenario. This elicits the need to rethink present development policies or implement additional protection measures before 2100. 

Because earthquake preparedness is multi-faceted and circumstance-specific, effective communication strategies require sustained engagement with target audiences. Video games can potentially promote such engagement and provide tools for researching what motivates preparedness behavior. In this study, we use four versions of a custom-built video game in a two-by-two experiment to engage people aged 18 to 29 in the Portland, Oregon metropolitan area in preparing for the anticipated Cascadia Subduction Zone earthquake. Using a social cognitive framework, we assess pre- and post-play levels of self-efficacy, outcome expectations, and intent to act relative to a number of earthquake preparedness and response actions. In particular, we focus on whether these metrics are affected by levels of identification with the player’s avatar or resources available within the game. The game has a pre-, syn-, and post-earthquake phase. The player must take preparatory actions before the earthquake, avoid injury during the quake, and perform three response tasks after the earthquake. The high-identification condition allows the player to choose and name an avatar from among six characters of diverse appearances, while the low-identification condition involves a randomly assigned avatar named Player. In the high-resource condition, players have more money to buy supplies before the earthquake and greater strength to perform a rescue after the earthquake. Preliminary results show that self-efficacy and intent to act increase significantly for actions depicted in the game but not for other relevant actions, and that these effects are not significantly different across experimental conditions.

A key priority for the Earthquake Engineering Research Institute (EERI) is to advocate for comprehensive and realistic measures that reduce the harmful effects of earthquakes. EERI’s public policy and advocacy committee has worked diligently over the past two years to fulfill this aspect of leadership and has impacted federal, state, and local legislation. Using recent lessons learned from the successes and failures of seismic safety legislation, the public policy and advocacy committee is launching its Citizen Advocate Initiative to provide guidance and resources on how EERI members can be influential constituents at every level of government for earthquake hazard risk reduction. The Citizen Advocate Initiative seeks to empower EERI professional and academic members in numerous disciplines including, but not limited to, architects, planners, social scientists, seismologists, structural engineers, geotechnical engineers, and public policy practitioners, to take action with lawmakers in their communities as seismic safety experts. The goal of the Citizen Advocate Initiative is to create a toolkit that promotes various methods of involvement that reflect the diverse EERI membership. This includes speaking with and educating public officials and the general public, as well as increasing collaboration with like-minded organizations, in efforts to promote seismic safety legislation, among other objectives. This initiative is currently under development for a Fall 2019 launch and will obtain feedback from many multi-disciplinary practitioners and academics in seismic risk reduction and policy fields to bolster efforts as legislative seismic safety champions in the United States and internationally.

Transformative scenarios seek to serve as reflexive participatory tools through which actors generate a collective synthesis of a situation, think about what could happen, and explore what might be done and their role in the situation. The process of developing scenarios also affords an opportunity to cultivate cohesive relationships among participants that can serve as a foundation to promote community resilience to extreme events. As part of the research activities of the Environmental Protection Agency College/Community Underserved Partnership Program and the National Science Foundation Urban Resilience to Extremes Sustainability Research Network at the University of Puerto Rico Río Piedras, our team is integrating residents to the development of transformative sustainability transitions scenarios. This poster offers an overview of the study design, and of the proposals and strategies generated through scenario workshops with residents of nine communities in the municipality of San Juan. Our research highlights the importance of understanding and integrating the design and implementation of transition policies and programs. We seek to promote community resilience throughout the multi-level structures of governance within the city.

Communities are encroaching on wildland fire hazard regions as urban areas expand. Paradise, California was leveled in nearly six hours during the Camp Fire in November 2018. This poster will present critical infrastructure damage within the community and aims to monitor ongoing recovery efforts. This work, funded by a Natural Hazards Center Quick Response Grant and a National Science Foundation Rapid Response Research RAPID grant, examined the wildfire mitigation strategies implemented by schools and hospitals in Paradise. This study qualitatively assessed the effect of the disruption to the operation of these critical facilities on the community. Many structures on the hospital campus in Paradise were damaged. Some schools completely collapsed, but many only sustained isolated damage. The school district has reopened two of its eight physical facilities and is continuing to operate in temporary locations in surrounding communities. This has burdened the Paradise community and surrounding cities. Additionally, the heating of plastic pipes and depressurization of the water distribution system during the fire caused water contamination. Without direct access to non-contaminated potable water, schools and other facilities have struggled to reopen. The recovery efforts of the schools and hospitals are dependent upon one another and on the reconstruction of housing within Paradise. The results of interviews conducted with school and hospital administrators showed that balancing the return of critical facilities will be imperative for residents to return to Paradise. 

The post-disaster landscape abounds with memorials that facilitate psychological recovery and collective remembering. The commemoration is intrinsic to all stages of recovery, yet the memorials forged through the recovery process have received limited attention from disaster science scholars. Meanwhile, comprehensive research in the cultural geographic tradition provides a basis for understanding memorials and their functions. The present study draws from this literature to examine commemorations produced following a sample of U.S. based disasters. An array of memorial texts, including both tangible artifacts and intangible performances, are considered in the current study. Departing from previous work, we focus on post-disaster remembrances produced during the long-term recovery process rather than limiting analysis to only formally designated disaster memorials. The sample is defined by time rather than by subject. Discourse analysis investigating the meanings and motivations of the memorials is performed on data collected through photo documentation and photo elicitation. Findings reveal the power of these memorials to remake places, as well as disaster events; to construct differing levels of intimacy or distance with survivors and victims; and to unite or fracture communities depending on their treatment of survivors’ memories. These patterns enable further systematic exploration of post-disaster memory work and its implications for place attachment, social cohesion, and community resilience.

Hurricane Maria struck Puerto Rico on September 20, 2017, as a Category 4 hurricane. Hundreds of thousands of individuals fled in the aftermath of the storm and settled across the mainland United States. The literature discussing the reasons for displacement is well-established, but there is scant research on where individuals go when they are displaced and what aspects of their destinations influence their decisions to remain or return. The work that will be presented in this poster uses a proprietary data set aggregated from cellphone data to track the movement of populations between counties in the mainland of the United States and Puerto Rican municipalities from the date of the hurricane to six months after, at monthly intervals. The data tracks more than 400,000 individuals who left Puerto Rico for the mainland United States and more than 200,000 who subsequently returned to Puerto Rico. Using this data set, we examine the characteristics of counties on the mainland where disaster migrants settled to address two questions: Are disaster migrants displaced to areas with comparatively low-levels of vulnerability? Do patterns of settlement by post-disaster migrants resemble pre-disaster migration trends? Our initial findings suggest that previous migration patterns are highly significant in identifying initial post-disaster destinations, while their importance diminishes in explaining whether individuals remain or return. This suggests that social support is more valuable initially, but that economic factors become more important in the decision to stay. This shows that understanding pre-disaster migration patterns can predict post-disaster migration trends, with significant policy implications. 

Communities of color are often overlooked or exploited by media after disasters. Due to their socioeconomic conditions, a history of disinvestment, declining populations, or unincorporated statuses, members of these communities lack the resources needed to develop sophisticated messaging to make their voices heard by media and during planning processes. Communities of color may find that their concerns and experiences during disasters have been marginalized during recovery planning and funding allocation. Some may characterize these African American communities as voiceless or invisible to the media. The paper presented in this poster uses communicative and advocacy planning as theoretical principles to argue the importance of giving these communities a voice in a world that continues to silence them. This study, conducted by the Texas Freedom Colonies Project and its collaborators, highlights one method used to provide a voice to these voiceless communities. Jones Future Academy is located in Sunnyside—a freedom colony in Houston, Texas. This small high school served as the site for this case study. The researchers interviewed 200 high school students (grades 9 through 12), asking questions related to their experiences with Hurricane Harvey. The results indicated that from a child's perspective, marginalized communities are not equally represented across media streams and that there is a need to address this issue.

The contributions of women and girls, visible especially at the community level, remain largely isolated from disaster risk reduction (DRR) decision-making, whether in government, private sector, or multi-stakeholder. Although achievement has been made at the policy level, links between gender-sensitive risk assessments and gender dimensions of resilience building remain tenuous in reality. The increase in the magnitude and frequency of disasters, both nationally and globally, creates an urgent need for women’s contributions, especially at the grassroots level, to be acknowledged and adequately addressed at all levels of disaster management. There is also an inadequate focus on individual and institutional capacity and tool development for mainstream gender and DRR at the policy and program levels. The poster presentation will focus the roles, contributions, and knowledge of women, which is the vital entry point in combating natural disasters, particularly in terms of risk reduction and management. With investments in their capacities, women can perform multifunctional roles as first responders, participants, managers, decision-makers, and leaders in the field of risk reduction and resilience building.

Sea level rise, as one of the most widespread and important climate change factors, has become a pressing threat to transportation infrastructures, especially in coastal regions. Exposure to flooding would not only reduce road capacity and cause closure and damage, but it would also affect the broad community’s accessibility to essential services. While many studies have evaluated the potential impacts of sea level rise on transportation infrastructure, few studies have investigated the spatial patterns and the associated social and physical factors to such vulnerability. Using Honolulu as a case study, this project adopted a normalized Hansen accessibility index to explore the contributing factors and spatial distribution of transportation vulnerability. The results show that vulnerability is spatially distributed unequally. It reveals that the residents in the northern part of the island near Kahuku, the eastern part near Hawaii Kai, and the central part near Honolulu Harbor may have experienced more impacts from tidal flooding than others. These communities also have a low level of accessibility to various opportunities, even without coastal flooding. In general, accessibility reduction increases with the percentage of roads affected in the traffic analysis zone (TAZ); the total length of affected roads; the percentage of residents in educational, health, and social services industries; and the percentage of white populations. Accessibility is negatively related to the percentage of residents in public administration, information, and retail industries. The findings indicate the important role that infrastructure protection could play in reducing vulnerability, and that certain economic sectors should be given special attention to adapt.

Communication plays an important role in helping people prepare for, respond to, and recover from all kinds of public health threats, whether natural, accidental, or intentional. People can make informed decisions about the actions needed to stay safe when information is accessible and understandable. Effective emergency risk communication plans integrate the access and functional needs of people in a community. Involving community partners and trusted leaders who support populations with the greatest risk is critical. The creation of integrated community-based networks to facilitate effective communications should be completed well in advance of a disaster. The recently updated workbook from the Centers for Disease Control and Prevention, Public Health Preparedness Workbook and Toolkit - Integrating Partners in Risk Communication Strategies, is structured around a flexible, functions-based framework that provides tools to organize plans for a broad group of individuals with disabilities and others with access and functional needs. This poster will explore the workbook contents and provide an overview of access and functional needs, and highlight planning considerations and resources. Further, step-by-step partner outreach guidance and noteworthy practices and strategies from the field will be discussed. Tools, templates, and worksheets will be reviewed that are useful for documenting and identifying ongoing improvement opportunities and institutionalizing best practices. This poster will solidify action steps that can help jurisdictions more effectively carry out their risk communication planning and effectively meet the needs of the whole population, including individuals with disabilities and others with access and functional needs.

Climate change has been linked to rising temperatures, precipitation, and sea levels. However, many factors associated with drought, such as its precursors, causes, scope, and damage extent remain uncertain. Droughts are responsible for direct damages to agriculture, industry, and livelihoods, and preparing for these disasters is difficult. The present study considers factors associated with damages, such as fatalities, injuries, damages, and loss of homes, and develops an estimation equation of drought damage costs based on drought damage data from 146 countries from 1900 to 2018. This study analyzed data provided by the Centre for Research on the Epidemiology of Disasters (CRED), through the EM-DAT international disaster database. In total, 11.7 million people experienced damages due to droughts from 1900 to 2018, including 2.7 million people who were injured or lost their homes. Total damage costs amounted to more than $169 million. Furthermore, 3 million fatalities were attributed to drought globally in 1928 alone. The highest damage costs in a single year were recorded in 2015, at more than $25 billion. 

Over 300,000 households in Puerto Rico were damaged by Hurricane Maria in September 2017. Studies have shown that housing conditions are representative of social and economic well-being. Trickle-down housing theory suggests that socioeconomically vulnerable populations are more likely to reside in structures that are less resilient to the forces of extreme events, thus receiving higher levels of damage. This study will explore the connections between preexisting social and structural vulnerabilities, as well as their relationship with housing damage in the aftermath of Hurricane Maria. It is expected that in Puerto Rico, socially vulnerable areas suffered disproportionately. A rich set of data will be applied to explore vulnerability and housing damage in Puerto Rico. Text analytics will be used to derive high-quality information and patterns from the data, including Twitter posts, open-source interviews, and news headlines. A geospatial dataset of island-wide household damage will be spatially joined with census data to extract statistically relevant trends. These diverse methods of analysis will be coordinated to determine the relationship between existing vulnerabilities and housing damage. The future findings of this study may indicate that targeted mitigation efforts should be enacted to reinforce household resiliency in socioeconomically vulnerable areas. Furthermore, housing recovery programs may need to be reworked to focus on the highly impacted vulnerable populations to avoid the persistence or enhancement of preexisting vulnerabilities in the wake of a disaster.

Hurricane Maria devastated Puerto Rico in September 2017, leaving the island with overwhelming reconstruction needs but limited resources. In such situations, some households, whether by choice or necessity, rely on informal reconstruction, such as building without the oversight of contractors or failing to adhere to building codes. In many instances, this is made possible through social resources in the form of social capital, as homeowners may seek the expertise of family members with construction knowledge or local organizations with leftover material. While social capital has been studied extensively in the post-disaster context, it is still poorly understood in the context of informal housing reconstruction. This study investigates the relationship between household social capital, broken further into bonding, bridging, and linking capital, and informal housing reconstruction. Data collection included door-to-door surveys (N=305) in two rural municipalities in Puerto Rico. Survey questions gathered social capital indicators, such as how often respondents interact with people from different communities or the local government, and questions aiming to capture the percentage of reconstruction on their homes that could be considered informal. Data analysis includes logistic regression, and results show a significant relationship between bridging capital and evidence of informal reconstruction. Theoretical contributions include a refined image of the social resources that homeowners rely on when the official process becomes inaccessible or undesirable, toward a holistic understanding of post-disaster reconstruction. Practical contributions include implications for planning and post-disaster networks, as we identify key players that are effective in distributing resources and promoting access to reconstruction.

Although Puerto Rico is suffering under draconian austerity, measures have been imposed by the financial oversight and management board as they restructure what is now an unpayable $74 billion debt. Immediately after Maria an estimated 2,000 people departed daily, however, the trope of the “emptying island” is not a new phenomenon. While Hurricane Maria might be considered exceptional by Caribbean standards, massive hurricanes are not new. I argue that these are two crisis triggers that have led to a nation-changing critical juncture that will have enduring legacies for the island. The alternatives in Puerto Rico’s history are stark. In the first scenario, the land, energy, food, and water are attentively and democratically managed by the Puerto Rican people in an exercise of collective sovereignty. In the second scenario, which some have referred to as “Puertopia,” a small elite group purchases the island out from underneath its residents, partially through the privatization of public goods and services. I predict that attitudes in Puerto Rico are being shaped by Hurricane Maria, and the apparent mishandled response and these attitudes are inculcating a stronger Puerto Rican national identity.

Aizawl is one of the most risk-prone cities in India. Located in the northeastern part of the country, this ridge-top capital of the Mizoram state lies in the Indo-Burman mountain ranges close to several earthquake faults. Approximately 350,000 people in the area live on steep slopes. Natural and human-induced landslides occur each year, causing large losses of life and property. An earthquake could generate hundreds of landslides all at once. The Mizoram Department of Disaster Management was aware of the risk but unsure of the path to mitigation until GeoHazards International (GHI) began to assist through a Munich-funded project. GHI worked with several line departments to study and develop a detailed scenario of a major earthquake in Aizawl and detailed recommendations to make the city more resilient. The transdisciplinary approach of GHI brought together international, national, and local experts from practice and research. We found that the event would cause 18,000 building collapses, trigger more than 1,100 landslides, kill more than 25,000 people, and damage critical utilities, such as communications, electrical and water systems, and roads connecting Aizawl to the rest of the country. Without proper preparation, Aizawl would be an isolated city struggling to rescue its own without external help. After dissemination of the scenario facilitated buy-in from local decision makers, GHI helped Aizawl develop a long-term plan to reduce landslide risk and is currently helping the municipality achieve its plan milestones. It is a mitigation race against the next earthquake, and GHI is working to ensure that Aizawl is prepared.

Effective communication is paramount in any risk mitigation or disaster response campaign. In order to most effectively convey information about natural and technological hazards to the public, it is important to know in advance how the public perceives different sources of information so that the most effective method and medium can be used to communicate with the public. This is especially pertinent in contexts where risk communication has become politicized, such as with hazards influenced by climate change. This work will present the findings of a household survey sent to residents in target Colorado and Oklahoma communities that have experienced varying levels of human-induced earthquakes. These earthquakes are induced by the injection of briny wastewater, a byproduct of hydraulic fracturing, deep underground. Because of the connection to energy production, the causes of earthquakes and how to prevent them are often a contested issue in the public sphere. The survey seeks to understand what factors, such as sentiment toward oil and gas, exposure to earthquakes, education level, and geographic location, relate to perceptions of sources, such as how trustworthy and legitimate they are regarding information on the induced earthquakes. This work presents results from an ongoing analysis of survey responses. The results of this work will help inform future risk communication strategies and bring together knowledge from engineering and social science disciplines to understand how to better communicate risk to the public.

A breakthrough in disaster risk reduction requires an approach that views disasters not as abnormal events, but as a regular part of a community’s evolution and that recognizes disaster risk management as inextricably interwoven with daily life. A novel computational modeling framework implements this approach to understand escalating regional natural disaster risks and possible interventions. The system-wide framework considers the perspectives and interactions among multiple stakeholders (government, primary insurers, homeowners), multiple interventions (home strengthening, insurance, land use planning), and secondary factors that affect risk.

The multi-stakeholder disaster risk management framework will include seven interacting mathematical models representing simulations of damage, losses, and ways to strengthen homes; decision-making by each stakeholder type including oligopolistic competition among insurers; and the changing building inventory and regional economy. It expands on an existing framework developed by the investigators, through the integration of six tasks: (1) Identify a broad range of government interventions, including risk-focused and risk-influential actions; (2) Develop an empirically-based homeowner decision model that considers the process by which homeowners initiate risk-focused and risk-influential decisions; (3) Develop a dynamic model of the insurance market; (4) Develop a dynamic building inventory model; (5) Develop a dynamic regional economic model that interacts with the building inventory; and (6) Conduct a full-scale case study for hurricanes in North Carolina.

This project promises an improved understanding of regional disaster risk management by uniting the conceptualization of disasters as part of the normal life of a community with the power of dynamic engineering modeling of risk.

This interdisciplinary research project combines sociological, transportation engineering, and systems engineering approaches to measure evacuation performance from two perspectives: transportation agencies and individual households. Current research focuses on three areas. First, household evacuation demands are estimated by a two-stage model using survey data from households in the Hampton Roads area (Virginia). It estimates evacuation perception at the first stage, which is found to have a significant impact on estimating evacuation demands at the second stage. In a second effort, household evacuation route choice is estimated for traffic simulation. It is found to depend on route characteristics (e.g., expected travel time, willingness to use the recommended route, and destination type) and socio-demographic factors (e.g., dwelling type and the number of children in the household). Third, in terms of assessing evacuation strategies, two normative performance measurement models are formulated. In the first, the formulation is informed by the traffic simulation model that uses the demand estimated from the two-stage model using the Virginia data. In the second, the data and information obtained from households in Florida who experienced Hurricane Irma are used directly.

Safer housing is a goal of post-disaster reconstruction programs, and while there have been studies of community recovery and resilience following disasters, little work has focused on the houses rebuilt after disasters. To make housing safer in future disasters, houses must be designed and constructed using safe engineering principles, and they must be safely maintained over their lifetimes. We posit that the latter is influenced by household perceptions of the safety of their homes, which can be influenced by the training programs implemented during post-disaster reconstruction. In this research, we studied post-disaster housing programs in 12 communities in the Philippines following the 2013 Typhoon Haiyan. Our interdisciplinary research surveyed 538 households to determine their perceptions of safety and safe housing design in future typhoons and earthquakes, assessed the expected safety of the houses in future typhoons and earthquakes using performance-based engineering methods based upon collected drawings, measurements, and material samples of the housing types, and compared the survey and performance-based assessment results to identify where household perceptions of safety align and conflict with engineering assessments of safety. This poster presents the findings from two communities, hypothesizing reasons for conflicts between engineering assessments and household perceptions of safety based upon case knowledge. Future work will recommend communication strategies for post-disaster reconstruction programs that can address these conflicts, converging local and engineering knowledge.

Risks associated with flooding often persist after the water recedes. Although these risks are sometimes sensed by residents (seen, smelled, touched, or heard), they are rarely documented or measured. Over time, these risks may become “masked” by other factors. Furthermore, environmental testing, often used to assess post-disaster environmental impacts, tends to ignore ecological complexity and social history, perpetuating an incomplete understanding of one’s environment and risk. As such, the development of new knowledge systems is needed to unmask these risks so they can be incorporated into recovery efforts. Engaging disaster survivors in citizen science may serve to fulfill this need by increasing both social awareness and scientific knowledge of risk. We collected survey, soil, and interview data in Lumberton, NC following Hurricane Florence to examine (1) the presence of risk due to fecal contamination of soil and (2) the impact of citizen science engagement on recovery knowledge systems. Results indicate that citizen science served as an effective approach for facilitating discussions about environmental risk while contributing to a science-based dialogue about E. coli presence and persistence in residential soils post flooding. The data serves as an important proof of concept for future studies that aim to deploy citizen science after disaster.

Natural hazards, such as earthquakes and tsunamis, can have adverse impacts on infrastructures and populations globally. In Wellington, New Zealand, awareness of these hazards is high, but preparation is low. Using the theory of planned behavior, we conducted an online survey with a large community sample (N=604) to identify predictors of intentions to prepare for natural hazards. Results indicate that attitudes and perceived behavioral control were the only positive predictors of intentions to prepare for natural hazards—in particular, experiential attitudes (perceptions of the experience of preparing as positive), instrumental attitudes (perceptions of the outcomes of preparing as positive), and self-efficacy—with intention, predicting information-seeking behavior. A secondary goal of the study was to examine possible framing effects. Natural hazards or natural disasters are used inconsistently, and findings in other areas, such as climate change communication, demonstrate that relatively minor changes in the framing of target issues can impact intentions and actions. Our findings offer preliminary support for a similar framing effect for natural hazards/disasters, both in intention formation and in the association between intentions and behavior. The findings of this research have important implications for public information campaigns and interventions aimed at increasing preparedness for natural hazards. 

A mixed methods study was conducted to analyze how coastal management programs across the United States implement federal consistency review by assessing enforceable policies by type and category within a hazard and environmental context. A qualitative content analysis was completed on 35 federally-approved lists that included federal activities programs to review for federal consistency. Case studies of 13 programs explored the structure of the respective state and federal consistency programs. Six states were adopted local plans and regulations as part of their state enforceable policies framework. Ten states were found to not use local policies but did utilize the adoption and implementation of local coastal programs or local program delegation. A qualitative content analysis was conducted on 82 coastal program documents (across 23 coastal programs) to document the type and scale of policies used for federal consistency review implementation and the extent that programs use hazards- and environmental protection-focused enforceable policies. Results indicate that federal consistency reviews vary widely across coastal management areas by federally-approved review lists and the content and structure of enforceable policies. This study adds to the literature on the intersections of hazard mitigation, local land use planning, coastal management, and federal consistency. The descriptive analysis shows how coastal management can implement federal consistency reviews to increase community resilience, and the role local land use planning has in this process. Further research is needed to better understand these interconnections and how they are implemented. 

Tropical cyclones are a considerable threat to the people, economy, and environment of Mauritius, with intense cyclones having an approximate return interval between eight to 15 years. The strength of cyclones is expected to increase with climate change, and recent hydro-meteorological disasters have already been attributed by various sources to climate change. However, the digitalized instrumental record available is relatively short, and there is little known about community experience and memory of cyclones, either from a current or historical standpoint. Critically, while community awareness of environmental risk is shown to be present in Mauritius and key to the capacity to respond, the levels and distribution of this knowledge are almost entirely unknown. In order to adequately understand community response to cyclones, this research suggests that it is essential to first understand the longue durée, given that the prerequisites and conditions for any so-called “natural” disaster trace their origins far into the distance in both time and space. This poster presents the final findings of the authors’ PhD research, which deployed a combination of interviews and archive research to illustrate the past experience and impact of tropical cyclones in Mauritius and the long-term patterns of social and cultural learning from disasters. The results present compelling evidence of cycles and trajectories of vulnerability in Mauritius, which are currently under-researched and accounted for in the disaster risk reduction policy.

With increasing global temperatures and drier climates due to global warming, wildfires are becoming more frequent and deadlier. In July 2018, Mati, Greece, was devastated by a wildfire that destroyed the town. There were at least 91 fatalities and 186 people were seriously injured. An interdisciplinary agent-based wildfire evacuation model was created to couple humans, wildfire, and the Mati town transportation network to understand how heterogeneous decision-making affects an individual or a household life safety. There are four model components: the population distribution model, the wildfire propagation model, evacuation shelters, and the transportation network model. The agents are individual people or a household as one decision-making unit. The agents have the following evacuation behaviors: evacuation on foot, evacuation by car, and sheltering in place (stay and defend). The critical variables in the evacuation models are walking/driving speed variations, household decision-making and preparation time (milling time), and warning diffusion. These diverse decisions affect life safety measured by mortality. 

Recurrent flooding has caused repetitive loss of life and property damage in the United States. Following the Great Flood of 1993 in the Midwest, federal disaster mitigation policies began shifting in focus from structural measures to non-structural measures (e.g., land use planning, property buyout, and acquisition). However, such non-structural mitigation approaches are primarily voluntary. Local communities, rather than the federal government, need to undertake initiatives to reduce the impact of hazards within their jurisdictions. To facilitate disaster mitigation policy adoption at the community level, it is important to understand factors that influence local government policy adoption with respect to the property buyout program. The objective of this research is to identify barriers and enablers of disaster mitigation policy adoption, and to provide theoretical and practical implications for subnational governments seeking to increase future policy action to reduce community risks and enhance resilience. Data comes from three sources: (1) property buyout program data from 1990 to 2018 was obtained from the Federal Emergency Management Agency (FEMA); (2) socioeconomic, sociodemographic American Community Survey data was obtained from the U.S. Census Bureau; and (3) floodplain data was collected from the FEMA Q3 Flood Data database, in addition to flood event data from a summary of disaster declarations and FEMA’s grants dataset.

The analysis uses logistic regression to explore the relationship between the adoption of the buyout program and socioeconomic factors. The potential impact factors at the county level include total population, population density, gender, race, age, education, median household income, median housing value, poverty level, structure type, floodplain percentage, and flood-related event frequency.

Hurricane Sandy devastated the New York City (NYC) shoreline in 2012, destroying many homes. NYC has been pushing for and investing in green infrastructure (GI) as part of the rebuilding process. More than 4,500 GI assets have been built or are under construction, and an additional 5,500 assets are slated for construction in 2019. However, information about GI is neither widely known nor easily understood by investors, researchers, or members of the public. To promote understanding of GI, it is important to effectively explain NYC’s GI processes and existing outcomes. Information visualization is the process of representing data in a meaningful and visual way so that users can better understand the data. This study focuses on visualizing GI projects in the five NYC boroughs. The purpose is to first create information visualization of GI and then assist investors in tracking project progress, help researchers access GI data, promote public understanding of GI, and improve public participation. Data used in the study comes from NYC Open Data. There are 5,675 rows and 32 columns, including project status, geolocation, and tree species. The analysis uses R software to create different visual analysis diagrams to explore the locations of GI in NYC, the percentage of projects in the five boroughs, and how much stormwater is absorbed by GI. The study also explores various effective ways to present information visually.

There is a need for planners to understand recovery processes to effectively provide communities with needed tools. The American Planning Association (APA) and the Hazard Reduction and Recovery Center (HRRC) at Texas A&M University are collaborating to develop a recovery guidebook for planners and allied professionals. This poster presents background research that will be used to provide evidence-based guidance to these professionals. We use qualitative interviews and online surveys to address three research questions: (1) What are planners’ (and allied professionals’) perceptions of the disaster recovery process and their roles in community recovery? (2) What do they need to know about disaster recovery to better support their communities? (3)How would they like to learn about or train in disaster recovery? A systematic literature review on community recovery was carried out, and an electronic library and annotated bibliography were made publicly available on the APA website. The systematic literature review helped frame the interview and survey questions. Interviews targeted professionals who had experienced or supported community recovery previously and examined these planners’ and disaster specialists’ concerns, perceptions about, and best practices for disaster recovery. Interviews were coded and analyzed using qualitative research methods. Online surveys were conducted with a sample of APA members to primarily examine planners’ concerns and perceptions toward community recovery. Results from these data will be presented in this poster.

The intentions of spontaneous, unaffiliated disaster volunteers are often seen as being at odds with those of large, organized nonprofits and government agencies. This dynamic largely exists because of the wide variety of goals, degrees of professional expertise, and organizational memory that arise following disaster events. In the United States, spontaneous volunteers learn the challenges of institutional disaster recovery very quickly when their efforts are not aligned with those of the state. This study aims to understand to what extent volunteers perceive that they have formal institutional access in state-managed disaster recovery processes. Participant observation, semi-structured interviews, and targeted written survey responses across two sites in the Florida Panhandle immediately following Hurricane Michael generated perspectives of how different varieties of actors working in the same context perceive themselves, their local organizational ecosystem, and develop or perpetuate institutional incongruousness and barriers to communication and coordination. Findings suggest a lack of cognizance of the full positive impacts of volunteer engagement on the part of spontaneously unaffiliated volunteers. This lack of cognizance is perpetuated by a perceived burden associated with navigating and reporting to state agencies. State agents are also found to perpetuate this disconnect through their lack of inclusive, comprehensive communication strategies. Attempts at mediation and coordination have been made, though common uncoordinated points have been identified. 

Resilience has become a buzzword in recent years, especially in the area of disaster risk reduction. Over the last 20 years, the social dimensions of resilience have become increasingly apparent and emphasized. Although there has been much debate and discussion, there is little consensus about what social resilience actually entails, and how it transfers from theory into practice and policy. This research investigates the ways in which the New Zealand government has responded to multiple adverse events (droughts, floods, major earthquakes, animal disease outbreaks) in the Hurunui District over the last decade, and how those responses impacted the community’s resilience. Perspectives were sought from three broad groups: farm households, local and regional governments and organizations, and the national government and national organizations. The lived experiences of people from each of these groups are explored—with a focus on well-being outcomes for those affected—and analyzed within a multi-capital framework. The intent is to understand how governance and public policy impact social and human capital following adverse events. The research is interdisciplinary and draws upon social-ecological system literature, key social science concepts, and the sustainable livelihoods approach in order to illustrate how the concept of social resilience can be used and applied as an analytical tool in a New Zealand context. It forms part of the Resilience to Nature’s Challenges National Science Challenge, a government-funded research program aimed at generating a transformative approach to disaster risk reduction in New Zealand.

Over the past several decades, the number of disasters has increased worldwide, causing extensive damages in communities. The planning field is facing the challenge of building resilient communities. Recently, a growing number of studies reorientated disaster research toward social infrastructure factors, such as social capital. Despite studies on progress toward building resilient communities, there is more to investigate regarding the quantitative approach to household recovery. Social capital has been categorized as bonding, linking, and bridging. At the household level, bonding capital includes close ties with family, friends, and other emotional support groups; bridging capital includes vertical ties with higher institutionalized power or authority gradients, and linking capital includes weak ties with social groups. During post-disaster recovery, the most commonly available social capital is bonding capital, while linking capital provides opportunities, such as government aid, and bridging capital offers information that serves as a resource for recovery. This research conducts a quantitative analysis of the relationship between social capital and household recovery. Through random sample surveys of 1,795 households in New York City after Hurricane Sandy in 2012, we collected information on households’ social capital and recovery status in 2016. We empirically tested the correlations between social capital and self-reported recovery outcomes. Preliminary results show that after a disaster, social capital can be a predictor of post-disaster recovery at the household level. This research provides evidence that building social capital could potentially enhance community resilience, achieving better recovery outcomes.

This poster describes thesis research that examined recovery from the 2008 Wenchuan earthquake, noting the specific capabilities of the Communist Party of China, relative to other nations. The first finding indicates that China experienced a rapid infrastructure recovery, rebuilding Wenchuan between 2008 and 2011 with few internally displaced people. By contrast, recovery in New Orleans has lasted for years following Hurricane Katrina, and many residents still remain internally displaced around the country. Japan has not yet fully recovered from the tsunami and Fukushima-Daiichi nuclear meltdown, and they are also experiencing significant displacement of former residents.

The second finding indicates that the 19 most wealthy provinces of China, as measured by annual fiscal avenue, were tasked by the Party to supply affected regions with 1% of their gross income to fund the recovery. By contrast, New Orleans received financial support from federal and state governments, which required no cooperation with other states. In Japan, the federal government was the sole source of funding for the recovery, while other prefectures supplied manpower and expertise.

Third, China used the recovery to engage in new forms of urban development and to improve land use planning. In comparison, the United States used the recovery as an opportunity to create new forms of urban development and Japan saw the recovery process as an opportunity to engage in rational land use planning. Finally, this research supports the conclusion that every recovery is different. The cultural, economic, and political context in which a disaster happens will differentially structure the recovery.

Coordination among agencies in response to extreme events could not only reduce the adverse impacts of disasters on the general public, but it could also enhance public organizations’ resilience to external disturbances. In the scholarship of emergency management, coordination means building a common operating picture among different organizations that have a sufficient level of shared information and can participate in emergency operations at different locations simultaneously. Organizations understand each other’s tasks, capabilities, and constraints, and thus could support and complement functions when an emergency occurs. However, building a common operating picture requires public agencies’ time and efforts to engage in emergency management activities before extreme events. Current emergency management practices fall short of engaging key stakeholders, while little is known about what motivates non-emergency management agencies’ participation in the coordination activities. This study attempts to fill the gap by exploring both the external and internal factors that are associated with U.S. public agencies’ engagement in coordination activities in response to extreme events. By using a 2016 national survey, the study examines one particular public transit agency’s engagement in emergency management coordination activities. Findings suggest that political influence could facilitate participation in emergency management activities. Organizational vulnerability to hazards, per se, does not translate into responsive actions automatically. Risk perception mediates the effects of public agencies’ hazard experiences on their responsive actions. The findings provide actionable, evidence-based policy recommendations for local emergency managers for increasing participation of broader stakeholders in emergency planning and preparedness.

The U.S. government has recognized the need to enhance disaster resilience at federal, state, and local levels. The United States mainly adopts local voluntary disaster resilience strategies, and the development of national disaster resilience is largely a bottom-up process. Varied local communities can serve as “policy laboratories” to experiment with innovative approaches to disaster resilience. If a policy is deemed to be successful, it can then be learned and adopted by other local governments. However, local governments may lack the technical expertise or financial resources to enhance disaster resilience without support from states and the federal government. More importantly, they may lack the will and ability to promote their experiences on a large scale. Therefore, it is pertinent to explore whether disaster resilience can be developed by local communities themselves, or if a strong federal presence is necessary. Using data from two Federal Emergency Management Agency hazard mitigation assistance grant programs—the Hazard Mitigation Grant Program (HMGP) and the Pre-Disaster Mitigation Program (PDM)—this study attempts to summarize the potential drivers of local hazard mitigation policy in the United States, including how federal and state policies influence local governments to adopt incentive disaster resilience policies and whether local mitigation plans prepared in response to federal and state mitigation policies are more or less effective than other types of actions. The aim of this study is to investigate the mechanisms and influencing factors of disaster resilience policy adoption from an intergovernmental perspective.

When does group loyalty break down in an emergency? The Station nightclub fire that occurred in Rhode Island in 2003 has been modeled repeatedly because a large and extensive dataset is available. However, models of the fire have exhibited dramatic differences compared to the dataset. It was assumed that people noticed a problem immediately or that they acted in a rational manner. This agent-based model is built around the idea of prevailing group loyalty as the norm because the evacuation was not rational, and the results of that evacuation pointed to group ties as an important factor. The model is built around individual and easily navigable priorities, and those priorities are in line with affiliative behavior patterns unless under extreme stress from environmental hazards. The degree of extreme stress required for the abandonment of the group as a priority varies by level of group intimacy. Results support this model as more accurate than models that do not incorporate group loyalty.

Catastrophic events can render landscapes uninhabitable, causing communities to implement buyouts to reduce future exposure to hazards. Buyouts can result from environmental disasters (e.g., severe floods) or technological disasters (e.g., the release of toxic chemicals). Buyouts permanently remove people from hazardous landscapes. Following the relocation of people, the built environment is dismantled, leaving open space. Open space often bears the imprint of the previous land use in the form of decaying slab foundations, street signs, and power lines. Along with these physical scars, the emotional trauma of the initial disaster is amplified by relocation and loss of sense of place. For some communities disbanded because of flooding, residents maintained a connectedness to their former home and neighbors through commemoration. Commemoration varies from officially commissioned to spontaneously developed events. Research on the commemoration of buyout sites generated by anthropogenic forces is less robust. This research focuses on the experiences of four communities awarded federal buyouts: Times Beach, Missouri, and Ponca City, Oklahoma, because of technological disasters; and San Antonio, Texas, and Wakenda, Missouri, because of flooding. The author uses critical discourse analysis to interrogate the narratives of formal and informal commemorative objects at these buyout sites. This research explores how a sense of place and belonging is (re)constructed through commemoration and the ways in which the cause of the buyout is depicted at sites of floods and technological disasters.

The Federal Emergency Management Agency (FEMA) Hazus program is responsible for providing state-of-the-art methodologies for estimating losses from floods, hurricanes, earthquakes, and tsunamis. There is a significant gap in our ability to quantify and communicate flood risk across the United States, driven by the lack of continuous and sufficiently detailed probabilistic flood depth data. While authoritative agencies, such as the U.S. Geological Survey and the National Oceanic and Atmospheric Administration, have taken responsibility for the measurement and dissemination of earthquake and hurricane hazard data, efforts to collect, measure, and store flood hazard information remain distributed across siloed groups. FEMA Hazus and Risk MAP programs are collaborating with hydrology experts from the Pacific Northwest National Laboratory (PNNL) to fill this gap in nationwide flood hazard data by completing a feasibility study for the development of national probabilistic (multiple return periods) depth grids using an extreme flood event model developed by PNNL using two-dimensional hydrodynamics. To establish the accuracy of this approach, flood loss estimates derived from PNNL depth grids in Minot, North Dakota, will be compared with estimates derived from depth grids produced by Hazus and FEMA engineering studies. If sufficiently accurate depth grids and flood loss estimates can be produced using PNNL and Hazus methodologies for Minot, this approach can begin to be expanded nationwide using best practice guidance from this feasibility study.