Poster Session Abstracts

In Texas, household evacuation is a key risk countermeasure by which local officials and populations can minimize loss of life due to hurricanes and tropical storms. Efficient and timely evacuations of hurricane threatened areas remain a significant concern in which repetitively examining households’ decisions to evacuate (or not) is of critical importance. Various factors may influence the complex process of decision-making.

This study explores factors that shaped evacuation decisions ahead of Hurricane Harvey on August 25, 2017. The data used consists of a random sample of nearly 900 households residing in eight counties along the Texas Coastal Bend area. A weighted correlation matrix lists indications of possible evacuation predictors. Logistic regression models explore possible factors that may influence the decision to evacuate.

Several factors are consistent with some previous findings that sought to understand household-level evacuation decision-making. The respondents' decision to evacuate is significant and positively correlated to several forms of risk perceptions and mandatory evacuation orders. Also, the decision to evacuate is significant and negatively associated with voluntary evacuation orders and to be living in a single-family home at the time of the event. These results may indicate the importance of issuing a mandatory evacuation by officials. The results of this study can help better understand whether residents evacuated and presents statistical models to predict evacuation choices. Future research can compare household evacuation decisions over time, in different regions, and for different natural disasters.

Psycholinguistics is a field in behaviorism science that George Miller established to study the psychological impact of languages on the human mind. Specific research and application of psycholinguistics in emergency communication are limited, where it is often purely analyzed under language barriers. The main objective of this study is to develop new knowledge about psycholinguistic in emergency communication through highlighting some of the communication gaps that are usually overlooked in emergency planning, and provide some recommendations to improve the overall emergency communication systems by reconsidering the way we look at language as an important psychosocial factor that impacts vulnerable communities. Previous research studies in psychology, linguistics, and emergency communication were critically analyzed. A qualitative methodology involving semi-structured interviews with subjects from Gainesville, Florida, who speak English as a second language, was chosen to provide a flexible approach to broadly explore the phenomenon being studied. We studied the psychological/linguistic dimension within local emergency communication in English and compared it to nine other languages. This study provided insights into one main research question: how can different languages influence our understanding of emergency notification? Five main themes were found; gaps in direct translation, variations in emotional impact, variations in grammatical language structure, fusion attitudes, and lack of technical terminology. Our data spotted considerable variations in the way people perceive local communication based on their mother language, which can significantly influence people’s behavior during emergencies and possibly put them at higher risks.

California, Oregon, and Washington are the first three states to implement the U.S. Geological Survey (USGS) operated ShakeAlert earthquake early warning (EEW) system that detects significant earthquakes. Data provided by ShakeAlert are used by distribution partners to deliver alerts that could provide seconds for automated system actions and mobile alerts that encourage people to take protective actions before shaking arrives. To understand how schools—which play a critical role in preparing and keeping children safe—in earthquake-prone regions can use EEW effectively, we are conducting a mixed methods study of the willingness and ability of K-12 schools in Alaska, California, Oregon, and Washington to integrate EEW into existing drills and emergency plans. In early 2020, we interviewed school district administrators, emergency managers, teachers, parents, and students in Anchorage, Alaska (n=88), and Ridgecrest and Trona, California (n=30) about their recent earthquake experiences, gaps in preparedness, and perceptions of EEW. Results from our qualitative analysis found that respondents had a limited understanding of EEW or how it might function in schools. While many people believed that early warning could help students mentally prepare for an earthquake, they also noted barriers, including limited funding, disruption in classrooms, and drill fatigue. These findings are being used to guide the second phase of our study—an online survey of school superintendents that will be administered in fall 2021. Together our two-part study will assist the USGS, its partners, and emergency managers in making evidence-informed decisions about incorporating ShakeAlert in schools. 

Heavy rainfall from Hurricane María triggered more than 71,000 landslides across Puerto Rico in 2017. In the aftermath of the disaster, Puerto Rican officials expressed a need for educational materials about landslide hazards to help guide residents, emergency managers, and planners in reducing risk. After extensive collaboration, the Landslide Guide for Residents of Puerto Rico was released to the public in Spanish and English in 2020. The process of collaborative risk communication—used to develop and distribute the guide and a set of related products—is described in a new publication, “Principles of Collaborative Risk Communication: Reducing Landslide Losses in Puerto Rico” in the Journal of Emergency Management. The nine principles, from cultural competence to reciprocity, may inform future interdisciplinary risk communication efforts among government agencies, universities, and local organizations.

The core team behind the Landslide Guide includes students, faculty, and scientists from the University of Puerto Rico Mayagüez, the Natural Hazards Center, and the U.S. Geological Survey. A broader network of practitioners and scientists in Puerto Rico contributed time, expertise, and content to the development and dissemination of the Landslide Guide, which was just the first product in a suite of dual-language landslide risk communication tools for Puerto Rico. Derivative products were co-developed in response to specific stakeholder requests and include a short animation, radio script, story map, presentation slides, and educational activities. Project presentations have been viewed more than 15,000 times collectively. The guide, related products, and presentations are available for download at hazards.colorado.edu/puertorico.

Household power outage experience varies with household situations. While there are methods to anticipate where the greatest direct impacts of a hazard may occur, these methods were not intended to anticipate the secondary impacts of hazards such as power outages. To understand how households experience power outages in different geographical locations, 896 participants were surveyed across three cities in the United States: Detroit, Miami, and Phoenix. Participants were recruited through Amazon’s Mechanical Turk service to complete the survey hosted in Qualtrics. It was hypothesized that people from non-white minorities and lower socioeconomic status experience more frequent and longer power outages and are more likely to have experienced greater economic impact (such as disposing of perishable food) while simultaneously less likely to receive help during their longest outage. Regarding racial status, the study found that non-whites in Phoenix and Detroit were more likely to experience longer outages than whites. Income was not a strong factor with having to dispose of food because of a power outage or for failure to receive assistance during the longest reported outage. This research demonstrated that social vulnerability indicators were not reliable predictors of more frequent and longer-duration power outages. Further assessments in varying geographical and political contexts are necessary to increase understanding of how households experience power outages. This research is necessary as the likelihood of power outages is expected to increase, given the increasing trend of recent reported electrical disturbances coupled with increased demand from a growing population.

Wind Tunnel facilities across the world are currently able to fabricate wind boundary layer profiles using first and second-order statistics primarily through finding the mean and turbulence that is essentially identical to winds experienced in nature. This is made possible with roughness elements within the aforementioned wind tunnels that allow for fabricated turbulence to be experienced at the end of the tunnel. Due to the manual and time-consuming nature of the turbulence causing mechanisms to be rearranged to change the roughness gradient, there is little known in regards to what effect higher-order turbulence has on the structure due to the expense of constant trial and error being too large with potentially months of data collection. However, this is no longer the case for the Powell Laboratory at the University of Florida, where there is a wind tunnel with a terraformer mechanism that can readjust the roughness gradient present every 30 to 60 seconds. This will allow for the months-long process to be shortened incredibly to a handful of weeks. The majority of the statistical analysis to determine whether or not the first and second-order boundary conditions would intersect perfectly will be done through Matlab programming. There are no findings as of right now because the research is still very much a work in progress.   

Oil spills caused by natural hazards can have devastating effects on both marine life and human life. These natural hazards can cause the pipe flowlines to buckle. This buckling can occur under the presence of hydrodynamic forces from sea waves or hurricanes. The frictional variability between clays and flowlines in the seabed of offshore environments is not yet fully understood. Often, engineers overestimate the necessary embedment depth of these flowlines in the seabed in order to avoid any disturbances or buckling caused by these hydrodynamic forces. By better understanding the irregularity of frictional resistance in the seabed and creating more practical pipe-soil designs, more cost-effective options can be explored. Therefore, for this research project, a pipe model was simulated in a laboratory to test the effect of cyclic loading on the axial resistance between a pipe and clay from the Gulf of Mexico. This model consisted of a pipe being displaced by a motor-powered load at relatively slow speeds as it was pushed and pulled in the longitudinal direction of the pipe. The data for the cyclic load tests will be processed using Python in order to access the data more efficiently and to use it for plotting and calculations. These plots are essential in understanding the relationship between the displacement of the pipe and the coefficient of friction. 

Following two consecutive earthquakes that occurred in Nepal in 2015, a number of humanitarian aid projects, supported by international non-governmental organizations (NGO}, have been implemented in severely impacted districts to assist in the recovery efforts. Despite the goal of investing in disaster reduction to build community resilience, many recovery projects have yet to strengthen the capabilities of these communities. This is due to a lack of consideration of local contexts while implementing project activities. This study analyses the experience of a flood-proof housing project in a rural village in southwestern Nepal.

Semi-structured interviews with key informants in the locality and diary methods (field-based and remote) were conducted from 2019 to 2021. The best practices are drawn out to illustrate how to strengthen community resilience by incorporating local knowledge and engaging vulnerable households. Using a participatory approach to reconstruction and retrofit, this project strategically develops a multi-organizational collaboration among international and local partners. The study reveals that trust and transparency are the key factors to enhancing community resilience and adaptative capacities. It concludes that multifarious and highly context-specific aspects are considered to be a barrier between communities, local governments, and NGOs. Once these barriers have been removed the results in attaining actively local disaster preparedness, fostering local vulnerability reduction, strengthening adaptation, and community resilience are improved.

The COVID-19 pandemic poses risks to the mental health of college students and faculty at institutions of higher education (IHEs). In the United States, students have already reported stress, anxiety, depression, and suicidal thoughts. Faculty mental health has important impacts on student wellbeing through faculty-student relationships and influences on students’ academic continuity. Thus, research is needed to identify areas where students and faculty need support in recovering from the COVID-19 pandemic. This study addressed this question by evaluating the impacts of the pandemic on student and faculty/academic advisor mental health. Data were collected from undergraduate college students (N=40; 85% female; 65% White/Caucasian) and faculty/academic advisors (N=13; 100% female; 84.6% White/Caucasian) at two- and four-year IHEs in the United States. An online survey was administered via REDCap between November 2020 and February 2021. COVID-19 stressors, stress, depression, and anxiety were assessed using a COVID-19 Stressors questionnaire, the Perceived Stress Scale, the Patient Health Questionnaire-8, and the Generalized Anxiety Disorder Screener-7, respectively. Data were analyzed using descriptives and chi square analyses. Students and faculty/academic advisors reported an average of 7 and 6 COVID-19 stressors out of a possible 14, respectively. Significantly more students reported moderate-high stress than low stress (90.0% vs. 10.0%; 2=30.4, p<0.001). Significantly fewer faculty/academic advisors reported moderate-high depression than minimal-mild depression (15.4% vs. 84.6%; 2=6.2, p=0.01). No other comparisons in student and faculty distribution across stress, depression, and anxiety severity levels were significantly different. Implications for research and effective IHE support will be discussed.

The global COVID-19 (SARS-COV-2) pandemic has had a significant impact on multiple domains of daily functioning in communities, families, and couples. Research on couple dynamics amid mass stress events using dyadic data is still limited. As such, the current study employs a cross-lagged panel design with 535 cisgender, heterosexual couples to explore longitudinal paths between pandemic related loss, attachment behaviors, trauma coping self-efficacy, and anxiety symptoms. Data were collected at three time points between April and October 2020. Findings demonstrated the potentially mobilizing nature of loss. Higher levels of reported loss in the 3 months leading up to the first wave of data collection associated with higher levels of partner attachment behaviors at the next wave. Additionally, higher levels of wave 1 loss for females also associated with higher trauma-coping self-efficacy and lower anxiety at the next wave for men. Higher levels of attachment behaviors generally associated with improved trauma coping self-efficacy at the next wave. Higher levels of trauma coping self-efficacy appeared to offset later reports of loss for men and associated with lower anxiety at the next wave within men and women. Finally, higher anxiety tended to associate with higher levels of reported loss at following waves and degraded trauma coping self-efficacy at the next wave within men and women. An investigation of indirect effects revealed that higher levels of trauma coping self-efficacy at wave 1 associated indirectly with lower levels of anxiety at wave 3. Implications for clinical practice and further research are discussed. 

Growing out of the progressive mainstreaming process of radical and feminist disaster literature, ‘protection’ and ‘inclusion’ recently became the new buzz words of the disaster law and policy field. The most recent resolution of the International Federation of the Red Cross and Red Crescent Societies on disaster law recognizes the importance of integrating protection and inclusion of vulnerable groups into relevant disaster laws and policies. In parallel, UNWomen recently launched the Women’s Resilience to Disasters Knowledge Hub, which includes a policy tracker intended at looking at the United Nations Member States’ progress in achieving inclusive disaster risk reduction strategies. Using the pressure and release framework and the concept of intersectionality to recall the radical aspects of these notions and their theoretical implications for disaster risk reduction, this poster will critically review whether and how the concepts of inclusivity and protection have so far been addressed in specific disaster laws and policies. It will demonstrate that the way these concepts have so far been approached in disaster laws and policies remains for the moment quite technocratic–characterized by a failure to engage with the broader transnational systemic forces playing key roles in the creation and preservation of risk and vulnerability, and to address the intersectional dimension of vulnerability and risk, therefore resulting in ineffective if not discriminatory recommendations.

Cancer Alley is a 136,794 meters stretch of chemical and industrial plants along the Mississippi River between New Orleans, Louisiana, and Baton Rouge, Louisiana. Since 2005, the area has experienced more than two dozen hurricanes with major rainstorms in between. Cemeteries, although just as vulnerable to storms and cancer-causing chemicals as the local population and natural environment, are overlooked casualties of frequent hurricanes and plant siting. During hurricanes and annual flooding, cemeteries in South Louisiana sustain significant damage such as dislodged coffins, difficult to reintern remains, and burial records damaged or destroyed. African American cemeteries are vulnerable to climate change impacts such as flooding and are often inaccessible, undocumented, and rarely recognized as environmental justice concerns. Recently, environmental justice activists have mobilized to resist a Formosa plant's siting close to a historic black cemetery in St. James Parish. The authors hypothesized that the Formosa siting is not an isolated case but instead reflects a pattern of racialized multi-hazard exposure of African American people and cemeteries. They created a database of cemetery locations—many of which were previously unmapped—based on the race or ethnicity of those interred in two parishes. Then, they performed a spatial analysis comparing cemeteries' exposure to flood hazards and proximity to hazardous chemical sites based on racial makeup. Findings show that black cemeteries have more multi-hazard exposure than other cemeteries due to accessibility and flooding. Results indicate that racialized multi-hazard exposure of cemeteries should be an emerging concern of Gulf Coast disaster recovery planners and researchers.

This study investigates how different predictors influenced survivors’ emotional response during Hurricane Laura and COVID-19 pandemic. The Protective Action Decision Model (PADM) literature indicates that environmental/social cues, risk perceptions, and attitudinal variables shape emotional response during a hurricane. That said, this is the first PADM study that investigated a dual-threat disaster that has conflicting protective action recommendations. Therefore, in addition to the above variables, this study includes concerns and emotional reactions to COVID-19 in the models to investigate how emotional responses are formed in this dual-threat environment. Household survey data were collected during early 2021 using a disproportionate stratified sampling procedure to include households located in mandatory and voluntary evacuation areas across Texas and Louisiana affected by Hurricane Laura. Regressions analyses were used to identify significant predictors of optimistic, depressed, annoyed, nervous, fearful, relaxed, energetic, alert, and passive emotional responses during Hurricane Laura. The findings suggest that some hurricane social cues and attitude variables are significant predictors in Optimistic, depressed, annoyed, fearful, and alert models. Hurricane risk perceptions and COVID-19 emotional reactions are strong and significant predictors in most of the models. The findings of this study could be translated to other multi-hazard events around the world.

Over the past century, humans have rapidly expanded into the hurricane-prone areas of the United States. While there is much interest in estimating and projecting population trends for hurricane- and flood-prone areas, much is still unknown about the historical settlement patterns of these places. Through the use of future sea-level rise (SLR) zones produced by the National Oceanic and Atmospheric Administration (NOAA) and a novel property data set from the real estate company Zillow Group, Inc., that spans 150 million property records from pre-1900 to 2015, we examine the how, when and where of coastal settlement in the United States. Our results reveal broad trends of increasing structure density along the coast, particularly after 1950. Nationally, SLR zones are more densely developed and have expanded at greater rates than inland areas. These trends vary by region, decade, settlement size, and settlement type (e.g., beach towns, urban areas, size of sea-level rise zone). In this study, we demonstrate significant effects of hurricane events on the spatial structure of settlements, with areas affected by hurricanes having higher structure density. We contend that an understanding of these historical development patterns is crucial for formulating effective coastal environmental planning and policy.

This research is supported by the National Science Foundation (award number1924670) and the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health (P2CHD066613).

A growing trend over the past several years has been the formation and formalization of civilian volunteer rescue organizations. What once were emergent groups made up of spontaneous volunteers are now, in many cases, nonprofit organizations with a formalized structure in place. A large contributor to their formalization process has been the use of social media and social networking sites (SNS) to grow their organizations and create camaraderie and community amongst the members. This poster presentation uses nearly three years of ethnographic data, virtual ethnographic data, and social media data to examine how Cajun Navy groups, civilian rescue organizations, and disaster crowdsourcing organizations use social media and SNS to communicate, formalize, and grow. After preliminary analyses of our observations and social media data, we constructed a communication and social media-based typology that examines how the organizations communicate internally among members and externally to the general public. Within each section of the typology are examples of how civilian volunteer rescue organizations use social media and SNS to promote organizational sustainability, share information, and build community and camaraderie within the organization. 

This poster draws on data from 1,230 members of the Social Science Extreme Events Research (SSEER) network to examine the relationship between disasters studied and SSEER members’ level of involvement in the field. SSEER members self-identify as one of the following four categories:

1. Core Researcher: Strongly self-identifies as a hazards/disaster researcher, deeply committed to the field, and engaged in hazards/disaster research for a sustained amount of time.

2. Situational Researcher: Not previously trained or involved in the hazards field, but had the opportunity to study new phenomena based on a situational event.

3. Periodic Researcher: Not primarily engaged in hazards/disaster research but focuses on related topics inconsistently.

4. Emerging Researcher: Includes students and others new to the field and those still learning its history, theories, and methods.

Our analysis found that of 1,166 discrete disaster events, the top 10 most studied by SSEER researchers included: Hurricane Katrina, 2005; Hurricane Harvey, 2017; Hurricane Maria, 2017; Hurricane Sandy, 2012; Hurricane Irma, 2017; 9/11 Terrorist Attacks, 2001; Indian Ocean Earthquake and Tsunami, 2004; Fukushima/Great East Japan Earthquake and Tsunami, 2011; COVID-19, 2019; and the BP Deep Water Horizon Oil Spill, 2010.

Understanding the relationships between researcher type and disasters studied can provide context surrounding questions like: What types of disasters do researchers at various levels of involvement in the field typically study? And how does self-reported involvement in hazards and disasters research vary by number and type of disasters studied? To learn more, visit: https://converge.colorado.edu/research-networks/sseer

Earthquake scenarios are generally selected to serve a wide variety of local and regional needs ranging from testing a community’s ability to respond to earthquakes to developing proactive, targeted mitigation strategies for minimizing impending risk. Earthquake scenario efforts under the auspices of the Training and Communication for Earthquake Risk Assessment project highlight the potential value of such scenarios, but numerous challenges persist. We discuss some of the ongoing developments related to hazard and consequence-driven scenario generation for Quito, Ecuador, a city of 2 million people with high vulnerability to earthquake and volcanic activity.

As a part of this research effort, exposure and vulnerability models from the Global Earthquake Model Foundation, city-specific microzonation-based urban hazard models, and U.S. Geological Survey ShakeMaps are utilized to develop consequence-based seismic scenarios are developed for Quito. Scenarios are conditioned on the occurrence of a particular consequence such as loss, damage, or another metric. Additionally, social vulnerability indicators are used to examine impacts and consequences beyond risk, loss, and damage. More broadly, the combination of consequence-based scenarios and social vulnerability are examined to provide a more in-depth look at how a major earthquake can impact neighborhoods of varying social well-being to highlight how seismic scenarios can be selected and deployed to enable stakeholders to make critical decisions following an earthquake and strengthen a community's resilience.

Natural disasters ravage the world's cities, valleys, and shores every month. Having precise and efficient mechanisms for assessing infrastructure damage is essential to channel resources and minimize the loss of life. Using a dataset that includes labeled pre-and post-disaster satellite imagery, the xBD dataset, we train multiple convolutional neural networks to assess building damage on a per-building basis. To investigate how to classify building damage best, we present a highly interpretable deep-learning methodology that seeks to explicitly convey the most useful information required to train an accurate classification model. We also delve into which loss functions best optimize these models. Our findings include that ordinal-cross entropy loss is the most optimal loss function to use and that including the type of disaster that caused the damage in combination with a pre-and post-disaster image best predicts the level of damage caused. The highest accuracy percentage on the testing set that we achieve is 74.6%; the non-optimal nature of this is largely attributed to the limited discernibility between the major and minor damage categories. We also make progress in the realm of qualitative representations of which parts of the images the model uses to predict damage levels through gradient class-activation maps. Our research seeks to computationally contribute to aiding in this ongoing and growing humanitarian crisis heightened by climate change. Specifically, it advances more interpretable machine learning models, which were lacking in previous literature.

This study examines the effect of the COVID-19 pandemic on evacuation decision-making for those impacted by Hurricanes Laura and Sally. The COVID-19 threat exacerbates the difficulty of planning and preparing during the hurricane season, as measures like social distancing conflict with congregations that result from evacuations. The impacts of COVID-19 will likely be felt through the 2021 hurricane season, so it is important to understand people’s perceptions of safety and actual behaviors during hurricanes. Data were collected through an online survey (in English and Spanish) using Qualtrics and statistically analyzed. The majority of respondents stayed at home during these hurricanes; approximately a quarter of additional respondents identified that they evacuated somewhere else or to a motel. For evacuation decision-making, there was no statistically significant relationship between age, education level, gender, or religion and the decision to evacuate; however, a weak relationship existed between political affiliation and evacuation decisions, with democrats being more likely to evacuate. Additionally, about one-third of respondents viewed themselves as vulnerable to COVID-19 due to pre-existing health risks. For public shelter perceptions, all groups had a negative perception of the shelters regardless of whether they stayed at home or evacuated, with over half of respondents feeling that shelters during the COVID-19 pandemic were more at-risk than enduring a hurricane at their houses. These results can be utilized by local officials and emergency managers to create strategies that mitigate the compounding risks to the public from both COVID-19 and the impacts of hurricanes.

The experiences of small organizations, including small businesses and nonprofits, have partially defined our collective understanding of the COVID-19 pandemic. Perceived to be unprepared for the disaster that has befallen them, there has been a perpetual concern for their wellbeing, from the micro, individual level through to the macro, society level. Although small organizations did indeed lack formal business continuity plans, they were not without the capacity for adaptation. Rather than crumbling in the face of adversity, small organizations have engaged in intrapreneurship, a process of innovating on internal processes to apply existing capabilities and navigate the pandemic. In this, small organizations have worked through a process that has allowed them to define their purpose and realign their processes to meet the needs of their employees, customers, communities, and society. 

This research aims to identify and define what a testbed is, how hazards and disasters researchers use testbeds, and what testbeds exist for future research. To achieve this goal, our team conducted a systematic literature review on peer-reviewed publications that used testbeds or were ambiguous about whether they were presenting a case study or testbed. The literature review aimed to explain methods for testbed development, synthesize hazard, infrastructure, population, economic, and community resilience models examined using testbeds, and understand challenges commonly faced in this process. We also virtually administered an expert survey to 260 researchers, including those who have authored testbed publications and other experts identified by our team. The goal of the survey was to formulate a consensus definition of a testbed and learn any disciplinary-specific differences in defining and using testbeds. Findings from the literature review and expert survey, as well as the authors’ own experiences, were used to produce a step-by-step methodology for testbed development to provide guidance to researchers interested in using and developing testbeds in the future.

This poster presents findings-to-date of the literature review and expert survey, as well as our proposed method for testbed development demonstrated by an example of Onslow County, North Carolina. The Onslow Testbed models a real community subject to hurricane and flood hazards featuring buildings and transportation infrastructure, organizations, and population demographics. The example includes data sources, data accuracy checking, and how data were merged to develop the testbed.

Amphibious construction presents intriguing possibilities in the quest for sustainable, low-impact responses to the impending global climate change crisis. Forward-looking strategies are needed that are capable of providing adaptability to future flooding levels that are difficult to quantify in advance. Suitable new housing types are required for populated regions where sea level rise and heightened storm activity are expected to intensify flooding. Amphibious foundation systems allow a house to remain close to the ground with the appearance of an ordinary house but to rise with rising floodwater and float on the surface until the flood recedes, at which time it returns to its original position. This strategy has great potential to benefit vulnerable populations that currently face the difficult choice between leaving their communities or living in fear of the devastation and trauma that severe flooding can impose.

Amphibious retrofits to existing structures function entirely passively, in synchrony with natural cycles of flooding, allowing water to flow where it will rather than attempting to control it. Since the height to which amphibious structures rise is in response to the depth of the water, they enhance resilience by taking both changing sea levels and land subsidence in stride. Amphibious retrofitting is a particularly appropriate strategy for communities with a strong connection to place and respect for natural ecosystems. The poster features case studies of inexpensive prototypes implemented in Louisiana, Bangladesh, Ontario and Vietnam, and visionary projects designed for other flood-vulnerable locations worldwide.

The National Science Foundation-supported CONVERGE facility at the Natural Hazards Center has developed a series of free online modules to train hazards and disaster researchers and practitioners, with an emphasis on students and others new to the field. Since July 2019, CONVERGE has released seven training modules on the following topics: social vulnerability and disasters, disaster mental health, cultural competence, Institutional Review Board procedures, emotionally challenging research, gender-based violence in fieldwork, and broader ethical considerations. Each module features learning objectives, interactive case studies and sliders, and links to additional resources, such as standardized measures and datasets. Users who successfully complete the quiz at the end of the module receive a certificate, which is worth one contact hour of general management training through the International Association of Emergency Managers certification program. Annotated bibliographies that summarize the literature used to develop the modules are also available. These modules can be used as classroom assignments, and a list of sample activities is provided through the CONVERGE Assignment Bank. This poster highlights the training modules that have already been released while also showcasing forthcoming training modules, including collecting and sharing perishable data and reciprocity in field research. The poster also presents user background characteristics of the more than 3,000 module registrants and evaluation results from pre- and post-assessments of user knowledge, attitudes, and skills. These training modules are funded by the National Science Foundation, the Centers for Disease Control and Prevention, and the U.S. Geological Survey.

Analysis of satellite imagery for flood monitoring is a rapidly developing area of research due to recent advances in satellite data. Although static flood maps are readily constructed, few approaches for generating the time series of flood dynamics exist due to challenges associated with missing data, temporal mismatch between satellite image capture dates and flood events, and lack of in situ measurements. Machine learning (ML) approaches offer novel opportunities to address these issues and create dynamic flood maps, but additional research is needed to develop such workflows. This study explores the use of ML to create daily flood maps, specifically as applied to floodwaters in North Carolina following Hurricane Florence (2018) and Hurricane Matthew (2016). Specific objectives of this study were to 1) test the suitability of the ML algorithm Random Forest for dynamic flood extent modeling in low-lying, flat terrain, 2) develop a methodology in which watershed characteristics and antecedent rainfall coupled with machine learning techniques can be used to create dynamic flood maps, and 3) identify the best predictors of flood potential by interpreting feature importance in the model. Using the Random Forest algorithm, geophysical and socio-environmental predictors, and Planet Labs satellite imagery, daily maps of flood extent were produced with an overall accuracy of 0.91, F1 score of 0.72, and kappa coefficient of 0.67. Furthermore, an analysis of model predictor variable importance indicated that the hydrography, road network, height above nearest drainage, and landcover type are the most important characteristics in predicting landscape-scale flooding. 

The Federal Emergency Management Agency (FEMA) recently released three new resources to support more effective communication of evacuation and shelter-in-place orders. This poster presentation presents information on the recently released resources and their uses in effective disaster communication.

The research report Improving Public Messaging for Evacuation and Shelter-in-Place documented findings from over 120 peer-reviewed articles on public understanding and decision-making for evacuation and shelter-in-place actions. The report then presented data-driven recommendations for improving public messaging to inform the public about the risk and increase compliance with instructions to evacuate or shelter-in-place.

A complimentary slide library was released with the report containing presentation-ready versions of the research report’s findings and recommendations. Emergency managers can use these slides to inform community members about risk and educate planners about developing effective warnings that increase public compliance with the recommended protective actions.

The third resource is Shelter-in-Place (SIP) Pictogram Guidance. The pictograms provided clear, visual guidance to the public on shelter-in-place protective actions, classified by 10 hazard types and three building types. In addition, the SIP Pictogram Guidance includes recommended interior locations for specific hazards, additional actions for protection, and guidance on how long to stay sheltered-in-place. The pictograms are applicable for use in multiple communication channels, such as posters, websites, just-in-time social media posts, and Integrated Public Alert & Warning System Wireless Emergency Alerts.  

Proactive evacuation action plays a vital role in preventing/reducing human damages when facing imminent threats such as heavy typhoon/rainfall. The authors previously demonstrated that those who are equipped with well-developed disaster risk reduction literacy (DRRL), which consists of understanding the threats, being aware of the preparedness measures, and having the confidence to act quickly, initiated faster evacuation compared with those with low DRRL. Based on the finding, a series of disaster risk communication (DRC) was held in Onojo and Itoshima cities, Fukuoka, Japan with the aim of enhancing the residents' DRRL. One of the critical parts of the DRC is the interactive workshops in which the participants share the local information on dangerous spots, past disaster damages, and vulnerable community members to consider their possible behavioral responses at disaster times. This study attempts to evaluate whether the workshops contributed to their DRRL development. The data was obtained from the questionnaire survey conducted following the intensive DRC period (n=321). Propensity score analyses revealed that full workshop participation leads to improvement in understanding the threats, whereas it did not elicit significant improvement in being aware of the preparedness measures and having the confidence to act quickly. The efficacy of the DRC workshop and the key directions for designing future DRRL building methods are discussed to gain insights into effective disaster risk reduction practices.

Disaster video games have the potential to foster people’s participation in learning about disaster and disaster risk reduction (DRR). The findings from recent disaster video game research have revealed several implications for DRR education, alongside disaster and video game research at large. With calls for a DRR curriculum that positions learning as active, interactive, action-orientated, and with a connection to local experiences, ‘serious’ and mainstream disaster-related video games can be one tool to support such a DRR curriculum. Yet, for video games to be realized as powerful learning tools, more significant consideration is needed toward how people can learn from video games (video game-based learning) and how video games can be used to teach (video game-based pedagogy). In particular, such a video game pedagogy should be centered around social interaction, metagaming, and gameplay to further enhance people’s participation in the learning process. This poster highlights how an alternative epistemological approach, by partnering constructivist learning theory with video games, can lead toward active, interactive, and action-orientated DRR education.

The purpose of the study was to investigate the relationship between levels of emergency preparedness and family resilience of collegians. Data were collected from a convenience sample of 606 students at a large, mid-south, public university before the syndemics of COVID-19 and systemic racism in the United States. Respondents completed a Preparedness Assessment (PA) informed by the Federal Emergency Management Agency that asked about steps taken toward preparedness. Students also completed the recently developed Walsh Family Resilience Questionnaire that asks about how their family responds to stress based on her framework. The demographic characteristics of this sample largely mirrored national trends, particularly at the same type of institutions. A PA index was calculated by summing 33 preparedness actions ranging from possessing the required contents of a disaster supply kit like food and water to the possession of an evacuation plan or emergency preparedness education. An overall family resilience score was computed by combining the weighted factor score of each of the 32 items. A correlation analysis was conducted to examine the relationship between levels of preparedness and family resilience. Preparedness was found to be significantly, albeit weakly, correlated (r = .12, p =.01) with family resilience. Preparedness and resilience and the relationship between them are central to the work of disaster professionals addressing response capacities in their communities. Knowledge about inequities in preparedness within communities, including college students, would better inform response needs and targeted education intended to improve academic training, preparedness, and overcome inequities.

In the Arequipa region of Peru, numerous geoenvironmental hazards impact the daily life of small communities situated in the mountains, on the coastline, and in the hills in between. A lack of hazard susceptibility characterization in the region means that there is limited capability to predict and mitigate hazards, leaving small communities without the necessary tools to reduce their vulnerability to hazards. The objective of this study is to remotely characterize geoenvironmental hazards at 10 sites in the Arequipa region, including landslides, debris flows, flooding, erosion, liquefaction, rockfall, and seismic and volcanic hazards. First, we develop hazard inventories from aerial imagery and assign hazard ratings based on established criteria found in the literature. Next, we assign a confidence level to the hazard ratings and determine the field observations required to verify the hazard ratings. Then, we will create GIS-based models for automatic mapping of hazard identification and susceptibility and validate the models against our maps. These methodologies will decrease the dependency on time-consuming field investigations to characterize geoenvironmental hazards in remote sites, providing preliminary maps that improve upon current approaches in the literature. A diverse assortment of environments and hazards are included in this research; thus, the resulting framework is expected to be relevant for further hazards analysis in Peru and other countries with similar hazards and geomorphic and climatic settings. Community members, local governments, and hazard and disaster-focused organizations can implement these methodologies to prepare for hazardous events and develop mitigation strategies at a community level.  

COVID-19 is an inverted disaster. Infrastructure is intact, water is potable, power is running, yet the challenges to the maintenance of social order persist. Where family is often an asset in disaster preparation, survival, and recovery, the nature of COVID-19 requires social distance and isolation, preventing care work from retaining its traditional forms while simultaneously keeping families together through spatial confinement. COVID-19 also demands the spatial rupture as well as the emotional congealing of social kin families-those of first responders, co-workers, neighbors, and even athletic teams and clubs into new systems of care, as the support work once completed by familial kin is now shared within different identity communities. This working group consists of twelve separate projects focusing on child care, LGBTQIA+, disability, immigration, medicine, coping strategies, and sports during the pandemic. Each project operationalizes the family (broadly defined) as a key unit of analysis.

Introduction. Induced seismicity has increased in prevalence in the central United States following an increase in hydraulic fracturing and the reinjection of the resultant wastewater. Induced earthquakes (IE) can result in property damage, a decrease in property value, and detriments to personal wellness. At least 22 out of Oklahoma’s 77 counties are affected, though many residents cannot adequately prepare or address current levels of damage.  

Methods. In this study, we use a mixed-methods design to investigate the earthquake preparation actions of the residents of Cushing, Oklahoma, and Pawnee, Oklahoma, two areas with high rates of moderate magnitude IEs. 

Findings. Through the quantitative component of the study, we find no statistical link between prior experience with earthquakes and preparation efforts in the sample population. In the qualitative component, we find two major themes: factors influencing earthquake preparedness and homeowners’ perspectives on preparing for IE. Factors influencing preparedness include lack of earthquake experience, inaccurate knowledge and lack of communication, contradictory preparations, and coincident preparation for severe weather. Homeowners’ perspectives include resentment, choosing to live in an area of naturally low seismicity, and IE not being the residents’ responsibility. 

Discussion. The lack of association between IE experience and preparation suggests that future research is needed regarding the motivators of preparative actions in populations experiencing IE. Current earthquake preparedness guidelines need revision to accommodate the unique risks associated with induced seismicity. Recommendations should account for the infrastructure being addressed, the knowledge level of the target audience, and other characteristics of affected areas.

To prevent future property damage and loss of life, many communities that are vulnerable to flooding collaborate with their local and state governments to participate in property acquisition programs known as buyouts. Most research to date regarding the effectiveness of buyout programs has focused on their shortfalls. Cases that highlight successes tend to focus on large, “high capacity” buyout programs, which aren’t necessarily representative of experiences and needs of the diverse range of communities that currently experience flood risk in the United States. This research aims to more broadly explore innovative practices that communities have used to implement buyouts and the factors that have led to the success of those innovations. This work highlights the findings of semi-structured interviews conducted with individuals involved in buyout programs across the United States. Participants were drawn from buyout programs representing communities of varying sizes and capacities. The analysis is guided by both policy diffusion and internal determinants of policy innovation frameworks. The policy diffusion framework refers to the transference of ideas across jurisdictions over time and space. The internal determinants model of policy innovation assesses a community’s social, economic, and political context to conceptualize the obstacles and motivations for adopting certain policies. Results will showcase diverse, innovative practices in buyout programs across the United States, and more importantly, highlight the contextual factors that have contributed to those innovations. 

Hurricane Maria was a powerful category four hurricane that devastated the Commonwealth of Puerto Rico upon landfall on September 20, 2017. Several warnings, alerts, announcements, and broadcasts helped inform people of the impending hurricane. Yet, much remains to be known about the role and influence of these messages on Puerto Ricans ’ protective action and evacuation decision-making. To better understand how the communication system performed and how it can be improved in the future, in 2018, the National Institute of Standards and Technology (NIST) began a multi-year investigation into the emergency communication system and the response to this system by the community. In January 2021, NIST released a progress report of these efforts, as well as those concerning the wind environment and technical conditions associated with deaths and injuries and the performance of representative critical buildings. This poster highlights ongoing efforts to assess Hurricane Maria’s impacts on emergency communications and evacuation decision-making and summarizes the emergency communication investigation progress to date. The investigation aims to recommend improvements to systems and practices that would make communities in Puerto Rico and across the United States more resilient to impacts from hurricanes and other disasters.

Nearly three years after Hurricane Harvey struck Houston in 2017, thousands of Houston residents remain displaced or are still living in damaged homes that endanger their health. This report uses participatory action research to identify and analyze the barriers to recovery from the perspective of residents living in low-income Black and Brown neighborhoods in Northeast Houston. Community-based research methodology is used here as an intervention to the exclusion of disaster survivors and frontline communities from decision-making regarding disaster recovery processes and resource allocation. 

To expand their expertise beyond their own experience, five Hurricane Harvey survivors interviewed up to five friends or neighbors and participated in subsequent focus groups. The findings from these conversations are complemented by the experience of staff members of West Street Recovery, a grassroots disaster recovery organization that has navigated the recovery process alongside Northeast Houston residents, rebuilt homes, and advocated for improvements in disaster preparedness and recovery governance. 

Collectively this research group identified barriers to a just recovery, describing how existing inequities are reproduced and amplified within the stages of evacuation, provision of immediate needs, applications for aid, temporary housing, and home repair following Harvey. This paper then explores the dramatic interventions necessary to reorient the disaster recovery apparatus toward justice and catalyze structural transformations in the unjust class, race, and geographic systems that currently dominate recovery. 

The Incorporated Research Institutions of Seismology, in collaboration with the American Geosciences Institute’s Geoscience Online Learning Initiative, is developing a 3-hour asynchronous module to provide geoscience career information to first- and second-year undergraduate students using diversity, equity, and inclusion (DEI) best practices in its design. The module presents information on career opportunities in seismology, geodesy, near-surface geophysics, and natural hazards, and their impacts on society through careers in social science, emergency management, and technological innovation. We present content through multiple modalities, including videos, animations, readings, blogs, discussion, and short quiz assessments. The module supplements existing introductory geoscience courses introduces students to the plethora of geoscience fields and career opportunities, and illuminates how students can build a unique career that is authentic to their passions, skills, and social identities. The five-module sections introduce geoscience concepts, connect how geoscience impacts society, present diverse career pathways, highlight skills needed to be successful, and integrate students’ social identities with pursuing a geoscience career. Module developers intentionally include diverse representation, perspectives, and current DEI best practices to be inclusive of all potential students interested in geoscience. The module is in the late stages of development and was piloted to a small cohort of undergraduate students at Fort Valley State University, a historically black university outside of Atlanta, Georgia, in June 2021. We will factor in feedback and informal assessment of these materials before additional pilots commence in the fall of 2021. 

In recent years, the importance of Community-Based Disaster Response Organizations (CBDROs) has increased in the United States and Japan, with new roles and structures emerging in both countries. This research explores the roles of community-based organizations in disaster prevention and response in the two countries and compares and contrasts two localized examples of CBDROs: Seattle Emergency Hubs (SEH) in Seattle, Washington, and Sendai Bosai Leaders (SBL) in Sendai, Japan. In both cases, the roles of the CBDROs and officials are seen as complementary, with recognition that in a catastrophic event, residents cannot expect to be rescued by officials. In the case of Sendai, this is based on the experience of the 2011 Great East Japan Earthquake.

SBL participants are closely involved with conducting disaster drills, preparing supplies, and in case of a disaster, providing emergency supplies and managing evacuation centers housed in municipal buildings as well as community facilities. In contrast, SEH are based on the identification of a physical location (the Hub location) which are not public buildings. Moreover, the main role of the Hubs is to function as information conduits rather than evacuation centers.

With a linked history, volunteer-run CBDROs play significant roles in both countries, while their functions and relationships to official government agencies vary greatly, as do the ways they are nested within and networked with other neighborhood associations. Comparing the experiences of SEH and SBL provides insight into the successes of, and challenges faced by, localized CBDROs, resulting in lessons learned for practice.

Studies on hurricane evacuation behavior focus primarily on why individuals do not evacuate when they are instructed to, while few examine the reasons why individuals choose to evacuate under conditions that would not require them to, termed shadow evacuation. Such behaviour might delay the timely evacuation of more at-risk coastal residents. To better understand the reasons for shadow evacuation, this study analyzes data from a household evacuation behavioural survey administered between 2012-2013 to 479 households in Cameron, Hidalgo, and Willacy counties, Texas, known as the Rio Grande Valley. This area features a number of Colonia neighborhoods with high rates of shadow evacuation in previous hurricanes. The analyses describe and regress shadow evacuation intention by risk perceptions, information reliance, preparedness level, previous experience, and selected geographic, social, and demographic characteristics. Results of the preliminary analysis show that residents in Colonia neighborhoods had a slightly higher rate of shadow evacuation intention (OR = 1.28) than those from urban areas. The ORs varied in some risk areas along with the impacts of gender, race, and language preference. A follow-up regression analysis reveals that previous experience, risk perception, and preparedness level could effectively mitigate residents’ shadow evacuation behaviour. The findings of this study will help local authorities to better understand the causes of shadow evacuation so that hurricane evacuation planning, risk communication, and public education efforts to curb unnecessary evacuation can be improved.

The regional distribution and frequency of localized extreme rainfall are increasing due to climate change and geological effects. For these extreme events, river basins are the most vulnerable, and where 85% of flood all disasters occur in Korea. Specifically, more than 45% of river flood disasters occur in small stream basins as they are more vulnerable to floods than local and national rivers. In general, nonstructural measures, such as a flood warning system, can be an implementable management option since structural measures like flood restoration projects may not be within reach due to economic and environmental constraints. This study developed the Small Stream Flood Warning System to reduce flood damages and specifically reduce human casualties in small streams susceptible to flooding. In addition, the CCTV-based automatic discharge measurement technology is developed to measure real-time velocity, depth, and discharge in small streams where otherwise, direct field measurement is impossible because of safety issues. The system under development estimates channel depth and discharge using measured-based nomographs and rainfall data forecast by the McGill algorithm for rainfall nowcasting, using semi-lagrangian extrapolation. The table-type nomograph describes the relationships between rainfall, discharge, and depth and has been developed using field measurement data over four years, 2016-2020. This study evaluates the estimating accuracy of the methodology and is verified with data from field-measured discharges and depths. The results of the study demonstrate that the level of accuracy of estimating flow discharges and depth is confirmed when compared to field measured data. 

Emotions are often ignored within predominant Western science, premised on emotions as being irrational and unobjective. Increasingly, emotions are becoming situated more prominently within disaster literature. It is recognised that the influences of popular culture on people’s understanding of disasters are significant. LEGO© bricks are an example of this popular culture that is adored by hundreds of millions globally. Many regard LEGO© as a special toy for being able to elicit many positive emotions whenever played with as well as a tool for fostering education. Education is acknowledged as vitally important to establishing diverse outcomes for disaster risk reduction. The resulting emotional discourses highlighted the positive emotions of joy, interest, and excitement and the negative emotions of fear and hopelessness surrounding disasters. Significant discrepancies and accuracies emerged between the emotional discourses in the LEGO© content and the actual emotional behaviours of people in disasters. The use of LEGO© is subsequently promoted to improve disaster risk reduction outcomes. 

Socially vulnerable populations are at increased risk of poor health outcomes, including unintentional injuries- fall accidents, motor vehicle accidents, drug poisoning, drowning, and suffocation. This study aimed to estimate the burden of and temporal trends in mortality from unintentional injuries by race and ethnicity in the United States, from 1999 through 2019. National mortality data on unintentional injuries were obtained from the Centers for Disease Control and Prevention Wide-Ranging Online Data for Epidemiologic Research. A JoinPoint regression model was used to construct temporal trends in injury mortality overall and within racial/ethnic subgroups. An annual percentage change in rate was estimated to characterize the change in mortality with each discrete change in trend. An average annual percentage was also estimated to highlight the overall change in trend. Over the two decades of the study, 2,687,987 persons died from unintentional injuries in the United States. The majority were males (64.1%; age-adjusted mortality rate (AAMR)=56.1; 95% CI: 56.0 – 56.2), non-Hispanic whites (75.9%; AAMR= 44.1; 95% CI: 44.1- 44.2), and lived in the southern U.S. census region (40.5%; AAMR= 45.2; 95% CI: 45.1 - 45.3). Mortality trends in unintentional injuries had stabilized in non-Hispanic Whites since 2017 but had increased at an annual rate of 7.8% in non-Hispanic Blacks since 2012; 5.2% in Hispanics since 2014; and 1.7% in Asian Americans since 2009. Interventions aimed at improving injury mortality must consider the social factors that contribute toward poor health among vulnerable populations. 

This two-phase, qualitative study aims to explore the needs of stakeholders surrounding Intellectual Disabilities (ID) and the elderly with dementia in missing or emergency situations. This study conducted semi-structured interviews with two stakeholder groups: the first responder's group and the support groups, including families and caregivers for the elderly with dementia and ID. In phase 1, semi-structured interviews were conducted with the supporting group, including families living with the elderly with dementia and ID and social workers. In phase 2, a focus group interview (FGI) was conducted with first responders, including firefighters, police officers, emergency medical technicians who have experience helping the elderly with dementia or ID to extract the information needs for providing proper assistance in missing or emergency situations. Data were collected and analyzed using thematic analysis. Interviews with the support group show that they experienced heightened anxiety of missing and themes identified as contributing to anxiety included: inadequate supports from first responders in missing or emergency situations due to the lack of understanding of characteristics of ID and dementia. Results of FGI show that the current emergency response system in Korea is not sufficient to provide necessary information regarding the individual’s special needs and condition to interact in missing or emergency situations. Various ways of providing the information to the first responders in the emergency scene were identified. This study provides direct insight on the current system’s gap in getting assistance from first responders in missing and emergency situations for elderly with dementia and ID. 

Despite a growing literature on post-disaster housing reconstruction, our understanding of factors affecting relocation decisions in areas vulnerable to sea-level rise (SLR) remains limited. This article fills this gap by examining: (1) the factors that affect relocation in a disaster-stricken area that is vulnerable to SLR; and (2) the extent to which SLR risks contribute to relocation decisions. Based on the case of Monroe County affected by Hurricane Irma in 2017, we suggest that: (1) the factors that drive relocations are pre-disaster economic challenges, specifically a lack of affordable housing, low wages, and the cost of living, as well as the impact of the hurricane; (2) SLR-related risks have little/no direct influence on relocation decisions; instead, the relocated residents' focus is on the next storm and their short-term needs. The SLR problem is invisible to, less understood by, and has a political nature for many. 

The National Institute of Environmental Health Sciences (NIEHS) Disaster Research Response (DR2) program leads the nation in building research infrastructure, providing access to training, and supporting the formation of critical stakeholder partnerships that are tantamount to conducting timely, often complex, studies that respond to disasters and emerging threats to public health. The DR2 has recently updated and enhanced the collection of disaster research tools in a new, searchable, researcher-friendly portal. The tools portal features a search engine designed for multidisciplinary researchers in the United States and globally. Researchers can access tools and resources for their studies, many of which were specifically developed for understanding disaster-related events, several of which had not been previously available. The collection features more than 500 resources and tools, such as surveys and protocols, to aid in the timely design and implementation of health research after disasters and other public health emergencies. Useful details are provided for each tool to help researchers identify resources based upon various characteristics of interest, such as study population, mode and length of the study period, and languages available. Selected instruments are available in REDCap format for easy integration into popular data collection systems. To help researchers address the COVID-19 pandemic, the portal offers more than 125 resources, including survey instruments, Internal Review Board reviewed research protocols, common data elements, data dictionaries, and consent forms.

Oceanographic sensors play a critical role in our understanding of our world's oceans' physical, biological, and chemical interactions. Nearshore sensors are prone to damage or loss due to energy from breaking waves and storm conditions. Additionally, nearshore sensors deployed at a beach pose a potential danger or hindrance to beach patrons using the area for recreational activities. This work explores the potential to supplement in situ measurements using shore-based camera systems to measure characteristic wave geometry and velocity and expand the range of conditions observed. The camera systems use recent advancements in particle image velocimetry (PIV). This study seeks to assess the feasibility of field applications of PIV analysis to measure the velocity of vortical motion in breaking waves, specifically focused on the external vortex of a wave. Preliminary field results indicate that PIV is capable of capturing the external vortical motion of a breaking wave without artificial seeding of the fluid. Ongoing experiments use reference distances and Acoustic Doppler Velocimeters to measure the accuracy of the results obtained. The findings from this study will determine the methods that will allow the greatest accuracy of the camera systems and PIV to measure nearshore wave-breaking behavior. This study is part of a larger project that seeks to better understand the relationship between the internal and external vortical motion to alleviate the need for submerged instrumentation, enabling cameras to be placed along a shoreline to expand the range of conditions observed.    

This poster uses data from the Social Science Extreme Events (SSEER) network (N=1,230). It examines the characteristics of social scientists who study hurricanes, some of the most often studied events in the SSEER database. I use Hurricane Harvey as a case study because it stands as one of the most costly disasters in U.S. history and is one of the most often studied events in the SSEER database (about 12% of SSEER members); it, therefore, presents a special opportunity to explore my primary interest in emerging researchers (students and others new to the field), since they will be at the forefront of disaster preparedness and mitigation in the future. Furthermore, discovering similarities and differences between researchers of the same event can foster new understandings like: What knowledge, skills, and training create the pipeline that leads social science researchers to study extreme events? What role do popular disaster events like hurricanes play in the development of emerging researchers? And what kinds of researchers were most likely to study Hurricane Harvey?  

Stormwater infrastructure deters extensive localized flooding by controlling stormwater runoff during rainfall events. The capacity and condition of infrastructure elements in an area are determined by a local government’s decision on stormwater management, which influences the overall infrastructure performance and the extent of floods during and after rainfall events. Therefore, it is important to examine the distribution of stormwater infrastructure performance across a municipality to understand the distribution of localized flooding impacts. A growing number of recent studies have examined the uneven distribution of stormwater infrastructure capacity across neighborhoods within a municipality. However, such studies are limited by infrastructure types and measurement while considering the macro-scale design capacities of one or two types of infrastructure only at the neighborhood level.

Using a cross-sectional multilevel analysis of Houston, Texas, this study examines how the capacity and condition of stormwater infrastructure at the road segment level associate with socio-economic compositions of the adjacent parcels and neighborhoods. The results support the hypotheses based on environmental justice and hazard vulnerability literature that socially vulnerable people have insufficient access to public resources and capital investment. Socially vulnerable people live in areas with more roadside ditches and fewer storm sewers, less total volume of conveyance systems, and more deteriorated storm sewers when controlling for density and imperviousness. The findings have substantial implications when developing land use plans and capital improvement plans to reduce the impacts from localized flooding and make them more equitable.   

Weather risk communication often uses research techniques familiar to designers to create effective messaging for the public. Until recently, both designers and design researchers have not been engaged deeply in this problem space. This presentation focuses on a portion of a two-year research project carried out by the authors to support the National Hurricane Center (NHC). The objective of this project was to identify opportunities to better convey tropical storm risk information to the general user audience through their online presence. This part of the research project consisted of a two-day design workshop and a set of user interviews with the general public. The virtual workshop brought together design researchers and hurricane risk communication experts to discuss information needs of the public related to storms, constraints, and capacities of different user groups and to develop design concepts for how to improve the NHC’s digital presence. To evaluate these design concepts, we conducted a set of semi-structured interviews with 14 members of the general public who had varying levels of prior experience with hurricanes. The goal of the user sessions was to understand participants’ expectations and gain feedback on the design concepts, then transform that into implications about user behavior around localization. We used thematic analysis to evaluate the expectations and feedback of users and understand NHC’s place as a source for localized risk information for the general public. This poster highlights the insights we discovered from the interviews and corresponding recommendations for future design opportunities. 

As society advances, the number of disaster and crisis situations will too. Research is needed to improve human resilience in extreme contexts; however, challenges in studying such threats make theory development and testing particularly difficult. Recent research has employed social science fiction, an interpretive approach to analyzing media as a proxy case study, to study environments that cannot otherwise be evaluated in real-time (e.g., leadership during a zombie apocalypse).

As humans venture farther into space (e.g., a mission to Mars), one critical area of extreme context research is team resilience in long-duration space exploration missions. Understanding what teams do in space to prepare for, respond to, and recover from adversity will be critical to future missions. Therefore, the purpose of this study is to examine representations of team resilience in science fiction television series and develop a theory to inform team resilience in extreme contexts.

Ten television series, spanning air dates of 2001 to 2020, were selected based on the criteria of having a stable team comprised of three or more members whose primary operational setting is space and/or a particular spacecraft. Episodes in which the focal team of the series faced adversity were analyzed using a coding framework developed from team process taxonomies and recent models of team resilience. Resulting team resilience themes include adaptive coordination and decision-making, interpersonal dynamics as sources of strength and vulnerability, and member functional diversity and crisis roles. Implications for a theory of team resilience in extreme environments will be discussed.

The extent to which disaster management professionals can successfully frame risk, reduce panic, and communicate recommended protective actions may be limited amid an emerging public health crisis, such as the COVID-19 pandemic. One significant obstacle is that individuals have a finite number of experiences from which to estimate their respective levels of risk exposure, and the lack of consistent background knowledge creates vulnerability across subsets of a population. Powerful and creative communication methods that can capture the attention of the public have long been sought. Streaming online media used during the COVID-19 pandemic provides an opportunity to compare the efficacy of different media styles and formats. A total of 243 China-based COVID-19 knowledge-related messages, categorized into three styles—live-action screencasts, infographic or typographic videos, and anthropomorphic animations—were collected and analyzed together with 12,400 associated user comments. The results reveal that anthropomorphic animations have stronger associations with instructing information (e.g., physical-based track) than adjusting information (e.g., psychological-based track) when compared to the other media styles. Subsequent regression analyses show that the use of anthropomorphic animation and messaging directed towards adjusting information are two significant predictors of audiences’ positive reactions to videos. This finding implies that anthropomorphic animations could be a more effective communication method in bridging risk information and knowledge during a health crisis than other messaging formats.

The DAT/Artathon is an annual virtual workshop that brings together early-career disaster researchers and practitioners from around the world who are working to visualize and tell stories with risk and resilience data. Over the course of six 2-hour sessions in four weeks, participants share a skillset with their cohort and gain feedback on individual visualization projects on current topics in hazards research and practice. The first week is devoted to co-teaching data visualization skillsets in a sequence of 15-minute mini-lessons—topics range from user empathy to visual encoding. After the first week, participants work on their individual projects, and the remainder of the virtual meetings are devoted to providing and receiving feedback on each others' projects. The first DAT/Artathon (Summer 2020) brought together 12 researchers and practitioners from 6 different countries. Participants shared that the workshop helped them build new skills, expand their professional network, and finish a visualization project. This year, the workshop is scheduled for July 19 - August 12, 2021. To find out more about the Risk and Resilience DAT/Artathon, visit: https://datartathon.com. 

In Taiwan, most safety policies and regulations of Long-Term Care (LTC) institutions are mainly focused on fire safety, and a few address flood evacuation planning, whereas very few policies and guidelines are focused on earthquake evacuation planning. Taiwan is highly prone to earthquakes that can create catastrophic damage to infrastructure and society, with potentially severe impacts on LTC institutions with insufficient response capabilities. Because the laws governing disaster preparedness are not retroactive, most small existing nursing homes and elderly welfare institutions only meet the lowest safety standards due to limited operating costs. In addition, emergency response plans and drills are often regulated under the traditional evaluation mechanism. This study intends to enhance the preparedness and emergency response capacity of these vulnerable institutions concerning earthquakes. Firstly, it develops a model of earthquake risk for LTC institutions. Secondly, it analyses the spatial distribution of highly exposed and vulnerable institutions by utilizing Taiwan Earthquake Impact Research and Information Application Platform (TERIA). The TERIA platform will be used to simulate different earthquake scenarios and impacts to the LTC institutions. By allowing institutional management teams to visualize earthquake impacts, the project intends to promote disaster risk identification and risk communication. As a result, stakeholders will be able to analyze the vulnerabilities of LTC institutions using a participatory approach and develop specific and practical coping strategies to reduce localized earthquake impacts.

Earthquake and tsunami preparedness education in the local school system is an equitable way to reach the broadest representation of families from all socioeconomic conditions in a community. Students are potential change agents who hold promise for future community-wide preparedness. Today's students will become the community leaders, decision-makers, and taxpayers of tomorrow who will determine where and how economic resources will be allocated for addressing resilience. What is often missing in children’s disaster education is an explicit theoretical model for reaching specific learning outcomes. To further develop a theory for effective disaster education, our study examined middle school students’ knowledge and beliefs about earthquakes and tsunami through the lens of conceptual change theory. Four related constructs of conceptual change were examined, including students’ science knowledge, their preparedness knowledge, their ontological beliefs about the nature of the phenomena, and their epistemic beliefs about the nature of their knowledge of the phenomena. Study results indicate that all four conceptual change constructs contributed to prominent themes in students’ conceptions of earthquake and tsunami. Earthquakes and tsunamis are emotionally charged topics and students seek information about these hazards, often through online sources, which may foster naïve conceptions. Results indicate that geoscience instruction should be accompanied by instruction on preparedness actions that are locally relevant to students’ lives to reduce fear and anxiety and to develop students’ perceptions of themselves as active agents who can have a positive influence on the outcome of these events.

This study examines the risk communication strategies of local emergency management organizations during the 2019 Arkansas River Floods. Using semi-structured telephone interviews with local emergency management officials, the researchers gathered data on risk assessment, warning dissemination, and public response monitoring strategies adopted by the local emergency management organizations during the floods. The preliminary findings suggest that the emergency management organizations not only utilized multiple channels to disseminate flood risks to the public but also improvised their strategies to facilitate risk assessment and warning dissemination during the floods. Additionally, the findings indicate that while some organizations passively monitored public response to warnings, others actively engaged in two-way communication with the public and adjusted the original warnings as needed. This study further provides recommendations to facilitate risk communication and to overcome communication challenges during a disaster.

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

The Rhode Island Coastal Hazards Analysis, Modeling, and Prediction system (RI-CHAMP) is a high-resolution Hazard Consequence Modeling System designed to help state and local emergency managers (EMs) and facility operators anticipate the consequences of a major storm striking their community. Traditional tools available to EMs, such as vulnerability assessments, do not provide actionable data regarding specific local concerns, such as emergency vehicle access and potential communication disruptions. The development of high-resolution storm models can aid EMs in making informed storm preparedness measures at the local scale. This research approach captures critical infrastructure (CI) managers’ concerns about hurricanes and Nor’Easters across Rhode Island for use in an online dashboard viewer that integrates those concerns with wind, wave, and storm surge model outputs. The current work focuses on the Wastewater Systems and Maritime Transportation Systems CI sectors, expanding upon methods and research developed during the U.S. Department of Homeland Security-funded studies. Our approach utilizes web and mobile applications to simplify the data collection process and establish a workflow for CI managers to report and maintain their hazard concerns in an Infrastructure Assets Consequence (IACT) database. To date, the IACT database contains 92 facilities, 325 assets, and 558 hazard consequences. The online dashboard viewer that integrates numerical storm model outputs with the IACT database will be used in emergency operations centers to flag the potential flooding and wind impacts during a real-time storm event or for planning scenarios, thus informing local and statewide emergency response activities.  

Hazardous materials incidents not only result in physical impacts but can also have significant environmental and health implications for a community. Although several studies have concentrated on the importance of hazardous materials management and policies, only a few have examined hazardous materials incidents in a community. This study fills the gap in knowledge by analyzing hazardous materials transportation incidents in the state of Arkansas. Using secondary data and spatial analysis methods, this study examines the risks associated with hazardous materials transportation. Furthermore, it also identifies vulnerable areas for hazardous transportation incidents in Arkansas. The study findings indicate that hazardous material incidents in Arkansas are dispersed throughout the state, with a major concentration in two areas–the City of Little Rock and Pine Bluff. The findings of this study can help emergency managers and transportation planners to rethink their hazardous materials transportation routes and modify them to facilitate safe transportation of hazardous materials through their community.

The COVID-19 pandemic has continued to have enormous implications for emergency management in prisons. Incarcerated individuals often lack equitable access to resources to prepare for, respond to, and recover from disasters, yet this remains an overlooked issue largely due to limited information made accessible for the public. However, the COVID-19 pandemic’s far-reaching impact provided an opportunity to observe how state prison systems and individual prison units make decisions about certain resources, including personal protective equipment (PPE). For this paper, we collected and analyzed official announcements released by state prison agencies in the United States and news media documents to examine how PPE (masks, hand sanitizer, cleaning products) were distributed to incarcerated individuals and differences among state agencies. Our findings illustrate the inadequacies in current U.S. correctional systems to distribute resources to incarcerated people impacted by the COVID-19 pandemic. Conversely, the systems lead to increased inequality in disaster impacts to incarcerated individuals who are disproportionately representative of marginalized and oppressed communities. 

In each of the past six years (2015-2020), the United States experienced at least ten climate-related hazard events causing losses exceeding one billion dollars each. Climate-change-perpetuated increases in storm frequency and intensity, when combined with changes to economic, social, and environmental landscapes, are expected to generate even greater losses. Because losses are expected to outpace the growth of the U.S. economy, developing techniques to mitigate losses through improved understanding of built environment performance in windstorms is integral to the success and survival of communities. Building codes provide the first line of defense for residential buildings, which drive most losses. Ideally, regulations should represent evolving understandings of risk mitigation measures, though political and other pressures significantly influence the code amendment process. Thus, it is critical to in-situ validate the efficacy of codes whenever rare design-level events occur to authenticate or discredit specific amendments and code requirements. As a design-level event that was well-documented by the Structural Extreme Events Reconnaissance (StEER) network, Hurricane Laura provides an opportunity for such validation. In this poster, StEER data is utilized to construct samplings of post-Rita homes that will be forensically evaluated to (1) identify specific causes of failures, (2) determine if homes performed as intended based upon governing codes at the time of construction, and (3) verify if failure probabilities observed in practice are consistent with those predicted by established loss models like HAZUS. These findings will provide evidence-based suggestions for future construction requirements in Louisiana.  

More devastating disasters are occurring at a higher frequency. In the wake of catastrophic events, survivors need financial and non-financial support to recover. Unfortunately, insurance, personal savings, and governmental resources often fail to address all the needs of the most socially vulnerable populations. Nonprofit organizations step in to fill these “unmet needs” that remain. Long-Term Recovery Groups (LTRGs) are either a single organization or a network of organizations meant to provide a coordinated approach to determine and address the unmet needs of disaster survivors. They handle case management, construct and repair homes, pay medical expenses, pay utility expenses, coordinate childcare/education/work needs, among many other needs. Most of the research about nonprofits in disasters focuses on the response and relief phase of disasters - providing food, water, and emergency shelter - rather than reconstruction and recovery. Recovery is known as a time for resilience efforts to be undertaken. Yet, there is little research published on LTRGs or their contribution to community disaster resilience or to the resilience of marginalized populations. This research contributes to this knowledge gap in two ways. First, we define and describe LTRGs, providing one of the first detailed accounts of these organizations. This definition and description are drawn from a four-year qualitative study across seven communities affected by disasters. Next, we use what we learned in the qualitative research to work towards quantifying the extent of LTRGs across the United States by building a database on LTRGs with information on their organizational size, financials, and general operation.

During any natural disaster, such as a hurricane or a pandemic, people usually depend on the information shared with them by agencies. Nowadays, many agencies use social media platforms such as Twitter, Facebook, and Instagram to share important updates during any disaster period. The goal is to disseminate the information among the maximum number of users on social media. To achieve this goal, agencies often collaborate and share each other’s updates with their respective followers. This study explores the relationship among these agencies in Twitter, which is helpful to identify if the agencies have a structured network among themselves and how the network functions during any disaster period. For this study, we have chosen the Eastern and Central parts of Florida, which are often subjected to hurricanes and storms. To explore the relationship among agencies, we have used the social network analysis (SNA) method. In the first network, we have explored if the agencies have a well-defined relationship in Twitter, i.e., if they are following each other. We have estimated the network density and identified the most powerful agencies in the network depending on the degree of centrality and betweenness centrality matrices. To explore the information sharing among agencies during any disaster, we will develop a network based on the retweets and mentions, estimate the network density, and identify influential users using SNA. The same approach can be applied to any local or central level agency network.

Oaxaca is a state located in the south of Mexico on the Pacific coast. Due to its geographical location, climatic and socioeconomic conditions, it is one of the most vulnerable states to extreme events that cause disasters. Thirty out of 32 hazards classified as highly dangerous at the national level are present in Oaxaca.

The aim of this research was to analyze the patterns of disaster occurrence in Oaxaca from 1970 to 2013. We did a secondary analysis using the DesInventar records. Two thousand three hundred three events were identified between 1970 and 2013.

Oaxaca's patterns of disaster occurrence from 1970 to 2013 show that the highest percentage of these events is due to meteorological hazards with 74.3%, followed by geophysical hazards with 7.7%, and biological hazards with 7.7%. Disaster concentration areas were identified using spatial correlations. Regions from the coast, isthmus, and Papaloapan were the most affected. Likewise, the impact of disasters caused by various hazards has caused multiple losses during this period: 1,028 deaths, 804,372 victims, 32,796 injured, and 1,782,155 affected people were recorded.

Of the 2,303 disasters analyzed, 74.4% have been registered as of the year 2000, showing that the frequency of recording events and their consequences are increasing, making the need for disaster prevention and mitigation imminent.

The growth of the disaster research field has brought more attention to potential ethical concerns of disaster research studies. Disaster survivors, responders, and those who assist in cleanup and remedial efforts may be left with significant unmet needs and long-term physical and emotional challenges resulting from their experiences. From a human research protection perspective, primary attention has focused on the vulnerability of individuals and/or populations affected by a disaster who may serve as research participants. It is important for Institutional Review Boards (IRB) and investigators to collaboratively address how best to protect the welfare of individuals and communities affected by a disaster. A new approach is needed to systematically consider the various factors relevant to an assessment of human research protection issues following disasters. From our literature review, we discovered that few IRBs and investigators have experience in the development and review of disaster research protocols and are likely to reflexively resort to the standard approval criteria when presented with disaster research protocols. While the standard review criteria are effective for the review of most research, it does not take into consideration the unique circumstances that surround disasters, which could potentially put participants at risk of harm. This poster describes our review factors model that emphasizes contextual considerations to strengthen ethical research practices, as well as how the proper use of preparedness training and tools, and collaborative interactions with investigators, IRBs can effectively conduct the ethical review of research while being responsive to the need for implementing timely disaster-related research. 

The National Science Foundation-funded Social Science Extreme Events Research (SSEER) network was formed, in part, to respond to the need for more specific information about the status and expertise of the social science hazards and disaster research workforce. SSEER is open to all social and behavioral scientists and those in allied disciplines who study the human, economic, policy, and health dimensions of disasters. This poster summarizes the findings from the 2018, 2019, and 2020 SSEER annual census reports. Specifically, it focuses on key variables, including the number of researchers in SSEER and its annual membership growth from 648 researchers in 2018, 949 in 2019, to 1,230 in 2020. The poster also highlights the disciplinary background of researchers and their geographic location by featuring the SSEER interactive web map as well as associated data publications. These annual census reports have implications for training, mentoring, and workforce development initiatives geared toward ensuring that a diverse next generation of social science researchers is prepared to study the root causes and social consequences of disasters.

More information on SSEER, including copies of the annual census reports, is available at: https://converge.colorado.edu/research-networks/sseer.

Floodplain management is beginning to experience a paradigm shift due to a greater understanding of the ecosystem services provided by floodplains and predicted flooding impacts from climate change. Local governments have the most authority in managing their floodplains and must be equipped for this paradigm shift. A comparative case study methodology of three counties provides context into how local floodplain programs are implemented. The research questions included:

• What are the key elements to an effective local floodplain program?

• What are the major barriers that impede local governments in implementing floodplain regulations and complying with the National Flood Insurance Program (NFIP)?

• How can local governments improve their programs in an attempt to overcome those barriers?

While components of strong organizational capacity were found to be key elements of an effective local program, they do not necessarily correlate to compliance with the NFIP. Low organizational capacity was found to be a major barrier to implementation and compliance, but high levels of organizational capacity are not necessarily required. The floodplain management components of the NFIP itself were found to be a major barrier as well. By taking advantage of education and training opportunities, including floodplain manager certification, local governments can begin to overcome barriers. The results are intended to assist local officials in improving these programs in the face of climate change. The results may also prove useful to both state- and national-level floodplain management staff, helping them fine-tune training, outreach, and compliance programs directed to local government staff.

Wearing a mask has long been recognized as a contagion source-containment approach rather than a protective action against infection from respiratory diseases. At the start of the COVID-19 pandemic outbreak, the World Health Organization recommended that only the sick and healthcare workers should wear a medical mask. This advice was subsequently revised after global health experts found that wearing masks could slow down the spread while also protecting people from new infections. Subsequently, the Chinese government made wearing a mask mandatory. Inconsistency in reporting the efficacy of mask use resulted in considerable confusion in households. This was further exacerbated by grave shortages in mask supply, especially in rural locales. Within this context, the study seeks to understand the dilemma that rural households in China faced when assessing their decisions concerning mask use. It reports on data collected from 492 Chinese households from nine randomly selected villages in Ya’an, Sichuan province. Findings suggest that these socio-economically backward communities, with inadequate medical support, were more concerned with the facilitating attributes of wearing a mask (i.e., effectiveness concerns and perceived social expectations) than the inhibiting attributes of mask use (i.e., expense concerns and life-convenience concerns). Regression analyses demonstrate that the former was influenced by authorities, expected consequences of infections, and educational achievement, while the latter was influenced by information from public media. An accurate understanding of risk exposure was effective in alleviating household worries. These findings can be used to improve risk communication accuracy and help to mitigate the effects of conflicting information.

The COVID-19 pandemic has proved a unique challenge to manage during hurricane evacuations as social distancing needs directly conflict with the mobility and congregation typically seen during an evacuation. This study, conducted during 2020 and 2021, examines how the compounding hazards of the COVID-19 pandemic and hurricanes affect peoples’ risk perceptions, preparedness planning, and evacuation decisions while considering the role of social determinants of health in disasters. Two surveys were disseminated virtually, one each during the 2020 and 2021 hurricane seasons, using convenience sampling to residents in Puerto Rico and the U.S. Virgin Islands (PRVI). Ultimately, the COVID-19 pandemic has had a profound effect on how people make their decisions to evacuate. Over half of the sample considered themselves vulnerable to COVID-19 due to existing health concerns. When asked about their evacuation actions, a significant number of individuals who would have previously evacuated to a shelter would choose not to during the pandemic. Additionally, individuals were shown to have a negative perception of public shelter options, with many people viewing public shelters as overtly risky due to COVID-19 concerns and would instead choose to shelter-in-place despite a recommendation to evacuate to a safer location or a government-operated shelter. Data from the 2021 hurricane season, although still preliminary, have shown that vaccination status would not affect evacuation decisions for 70% of the sample. This study will help apply knowledge for future pandemic disaster planning by revealing improvements for public messaging that will ultimately save lives and determining planning and logistical requirements. 

The National Research Council of Canada (NRC) recently developed a National Guide for Wildland-Urban Interface (WUI) Fires to help communities better resist WUI fires. NRC hopes to upgrade the Guide’s recommendations into a standard that can be adopted in provincial building codes. Its recommendations resemble the International WUI Code, California Building Code, FireSmart Canada, and Firewise USA. It offers a spectrum of design and retrofit options, from structural (reliance on noncombustible building material) to vegetation management (keeping brush trimmed and far from the building) or a balance of the two. To help Canadian communities judge the business case for the Guide, NRC contracted with the Institute for Catastrophic Loss Reduction (ICLR) for a benefit-cost analysis of the Guide, which they call an impact analysis. ICLR, in turn, contracted with SPA Risk LLC to design and carry out the analysis. We found that following the Guide can cost under CAD $5 per square foot and that benefits can exceed 30 times the cost. Benefits include reduced damage to building and contents (about half), casualties (about 1/3rd), and smaller quantities from other sources. We also estimated jobs saved and accounted for demand surge and climate change. For the impact analysis, see the report and overview.   

Non-profit organizations and other civil society organizations are an important part of disaster resilience, but it is still unclear how and what these organizations learn from disaster experiences. For example, will organizations apply the lessons they have learned during the COVID-19 pandemic to future disaster preparedness? This study explores how congregations learn from disasters and examines the nature of cross-sector collaboration during a crisis. Congregations and faith-based organizations behave in a similar fashion to other non-profit and community-focused organizations. As a result, they can expand the academic and practical knowledge of how civil society organizations learn during a crisis. 

Using a natural experiment, the study evaluates if congregations increased their risk assessment and disaster planning, disaster networking, and collaboration activities from 2019 to 2020. Initial results support two themes. First, while awareness of pandemics has grown, congregations have narrowly applied the lessons they have learned. Disaster learning is both temporally and spatially immediate. Second, congregations turned to other houses of worship for help and relied less on government sources or other disaster-related organizations. These results could have substantial implications for community resilience, especially given the role congregations and similar civil society organizations play in disaster readiness and resiliency.

Texas is one of 13 states in the United States that does not provide system-wide temperature regulation (air conditioning and heating) in their state prison systems. Without temperature regulation, prison agencies develop policies to mitigate the impact of extreme heat on individual incarcerated persons. Yet each year, the Texas Department of Criminal Justice reports that staff and incarcerated persons have fallen ill or even died from complications from extreme heat in Texas prisons. This study includes an analysis of 345 surveys from incarcerated persons in Texas about their experience with heat mitigation policies as well as the impact of extreme heat on their lives and personal health between 2018 and 2020. Findings demonstrate how the structure of prisons and the characteristics of imprisonment present significant challenges to incarcerated persons accessing resources to mitigate the impacts of heat. Additionally, surveys collected in 2020 reveal how the COVID19 pandemic, which has disproportionately impacted incarcerated populations in Texas, created additional challenges to individually focused heat mitigation strategies in Texas prisons. 

Climate change causes weather extremes to rise in frequency and severity, which could have detrimental effects on human life, property, and livelihood activity. There is uncertainty about the influences of anthropogenic climate change on the occurrence and severity of small-scale, sudden onset weather phenomena such as hailstorms and subsequent loss and damage. However, several studies indicate a connection between hailstorm activity and hailstorm-related loss and damage. Severe hailstorm events had observed in Bangladesh in recent years, which are, in fact, rapid-onset disasters but low exposure in terms of giving government response and media consideration. Hence this study examines potential impacts and management strategies for loss and damage resulting from hailstorm events among smallholder farmers in Phulbari Upazila of Kurigram District of Bangladesh. Firstly, the direct and long-term economic and non-economic loss and damage caused by the hailstorm on human well-being and livelihood were assessed. Then, the study evaluated the current adaptation, coping, management, and response strategies at the institutional and community level in the context of such extreme events. Finally, a regulatory framework and implementation approaches had suggested achieving the country's resilience against disaster and climate change-induced loss and damage. Participatory Vulnerability Analysis, Key Informant Interviews, and Sample Surveys gathered the primary data for the study. In addition, secondary data were collected through analysis of literature, published and unpublished scientific articles and media reports, and other sources. This research outcome will help countries develop a guideline to address climate change and disaster-related loss and damage.

Disaster preparedness efforts regarding the location and prepositioning of relief supplies are becoming much more crucial than ever before. Location decisions are particularly challenging, as these require information before the disaster, i.e., the potential disaster site and the extent of the damage. After catastrophic events, local supplies are mostly destroyed in the affected region, posing unique challenges to distributing relief supplies. As a result, local supplies may not be available for the population in need. This research studies the location of points of distribution (PODs) where prepositioning of relief can be readily available and close to the potential disaster site. Given a fixed distribution center where relief supplies are stored, the proposed model considers identifying the districts' shapes and the location of the POD's inside the district such that it minimizes the total social costs. The social costs are the summation of the private or logistics costs and the externalities of the distribution of relief in deprivation costs. The people in need capture these deprivation costs based on the time spent without the relief supplies. The results provide disaster responders guidelines at the planning stage to locate planning points, allocate resources better, and provide alternative distribution strategies of relief.

Addressing urban and rural flooding is a major planning problem. As urbanization increases, new grey infrastructure is constructed in both urban and rural settings. This may lead to increased flooding on public and private property and result in increased property damage. Green infrastructure is an emerging concept for sustainable development in urban communities and government institutions alike. There is considerable literature illustrating the benefits of green infrastructure, but few studies discuss the barriers that arise during planning and implementation. There is a paucity of research identifying the sufficient level of investment needed to create and implement green infrastructure spatial plans. The research conducted here focuses on identifying barriers to green infrastructure use in three case studies containing varying scalar development units in the state of Hawaii, on identifying the criteria needed to design a strategic green infrastructure spatial plan that complements or replaces grey infrastructure, and on how best to ensure that social and environmental justice and social equity are addressed when including green infrastructure in both urban and rural communities. A mixed-methods research approach will be used in this study. Several perspectives will be applied to develop criteria for each case study and strategies to overcome the identified barriers. In addition, several focus group interviews for each case study will be conducted to gather opinions on the spatial plan concept and on the identified barriers in each one. It is predicted that each case study will have a correspondingly unique spatial plan. 

Flooding is the leading cause of disasters worldwide, and these events significantly impact affected communities. Floods can harm the environment and affect residents' physical and mental health; however, flood risk management decision-making is often driven by assessments of property damage. Engineers and planners lack the tools and methods to consider the potential health impacts of floods in risk assessments. This can lead to over-investment in flood risk management in areas with high property values and under-investment in more vulnerable areas. The purposes of this study are to identify physical and mental health impacts associated with flood events and to develop quantitative relationships between flood intensity, population characteristics, and incidence of these outcomes that can be incorporated into models used in flood risk management. To this end, we conduct a systematic review of natural hazards and epidemiology literature which empirically assess various health outcomes (specifically: post-traumatic stress disorder, anxiety, depression, physical injury, infectious disease, exacerbation of chronic illness) following flooding events. Full text analysis is ongoing, and once complete, we will conduct a meta-analysis of eligible studies and calculate pooled odds ratio estimates for each outcome of interest. We will also extract information related to flood magnitude, allowing us to control for this factor in our analysis and begin developing basic predictive models of health impacts from floods of varying magnitudes. The results of this study will inform efforts by the U.S. Army Corps of Engineers to improve the quantification of the benefits of flood risk management project alternatives.

Food banks and food pantries play a significant role in providing food assistance to the millions of food-insecure households in the United States. A large body of research has now addressed the spatial variation in food insecurity and access to food retailers, but far less attention has been given to the geographies of food pantries. Access to these services is an important strategy in managing food insecurity before, during, and after any extreme events. Access to these critical services is shaped by their locations and transportation infrastructure, which are commonly disrupted during extreme events in many areas. This study examined how extended disruptions to road networks due to flooding and closures during extreme events, including hurricanes, may affect access to food pantries. Using graph theory and road network analysis, this study quantified changes in access to food pantries (in terms of travel distance) due to roadway flooding and closures. Results of geographic analyses were compared to responses to a survey of food pantries in Southeast Texas following Hurricane Harvey to understand how critical infrastructure disruptions alter access to food for already vulnerable populations. Understanding the locations susceptible to disruptions in access to food aid can provide valuable insights on how these services are sited and how to respond during extreme events, thereby supporting food security for vulnerable populations. 

Earthquake-induced soil liquefaction is a geotechnical engineering phenomenon that poses extreme danger to humanity and engineering construction both above and below ground. Extreme loss of life and property can ensue when this weakening of soil-structure interaction occurs. For this reason, methods must be identified, created, and employed to estimate the probability of liquefaction occurrence. Currently, Finite Element (FE) analysis performed utilizing well-calibrated constitutive models of soil is a reasonable and well-known method for accomplishing this. These models include parameters that are both mathematical and physical soil properties. However, these inputs are often imprecisely or completely unknown. For this project, Sensitivity Analysis was employed to identify the trivial parameters in the PM4S and model developed at the University of California, Davis (UC Davis). Upon completion, the probabilistic data-matching method, Bayesian Inference, was performed to estimate the range of individual parameter values that allowed the simulation model to best emulate specific experimental data for a sample of Ottawa F-65 sand from the Center for Geotechnical Modeling facility at UC Davis. Finally, Forward Uncertainty Propagation quantified the liquefaction risk using these calibrated input parameters. However, as few computational programs exist that could have completed all three of these separate processes, the new, state-of-the-art quoFEM application, developed at the University of California, Berkeley’s SimCenter, was used. Utilizing this tool, we performed an entire Uncertainty Analysis on the Ottawa F-65 sand sample that underwent excitation to train a FE model of a column of sand for widespread application in earthquake engineering and disaster resilience studies.  

Conventional wisdom holds that pre-event planning is a key factor in effective disaster response. In assessing the response to the COVID-19 pandemic, it is necessary to evaluate the extent to which emergency management agencies were prepared to respond to a pandemic, particularly given the unusual nature of this event (i.e., scope, scale, and length of response). While emergency management agencies at every level of government have been involved in the COVID-19 response, state-level governments have taken on a prominent and atypical leadership role. This study assesses the extent and role that emergency management agencies planned for in a pandemic scenario. Understanding the extent to which state-level emergency management agencies planned for an event like the COVID-19 pandemic and what they anticipated their roles to be provides insight for future revisions in pandemic planning. This study addresses two related research questions: RQ 1: To what extent did state-level emergency management agencies account for a pandemic in emergency management response plans before COVID-19? RQ 2: What was the planned role of state-level emergency management agencies in response to a pandemic? An analysis of state-level emergency management plans found that, although all states with available emergency management response plans included pandemics, there was significant variation in the extent of the inclusion and the role prescribed for emergency management. Public health and emergency management response plans were congruent with respect to the planned role of emergency management. 

Background: The response to the COVID-19 pandemic has placed significant strain on many professionals. While there has been substantial research on the physical and mental health impacts of pandemic response on patient-facing healthcare staff, there has been considerably less research investigating the wellbeing of the public health workforce.

Objective: Quick Response Research was conducted to assess the impact of the pandemic on public health workers and identify strengths, weaknesses, opportunities, and threats to the mental health of the public health workforce.

Methods: Key informant interviews were conducted with a subset of respondents to a cross-sectional survey. A semi-structured interview guide was used addressing the following topics: professional role during the COVID-19 pandemic; the impact of individual- and organizational-level considerations on mental health; and organizational strengths, weaknesses, opportunities, and threats associated with the pandemic.

Results: Participants included 24 epidemiologists, public health nurses, programmatic staff, evaluators, data scientists, and case investigators working in state and local public health. Half had six or more years of experience in public health. Five themes were identified through inductive coding of transcripts: 1) importance of teamwork and camaraderie, 2) potential for growth of the field of public health, 3) considerations for adaptive work environments, 4) ongoing communications challenges, and 5) constrained hiring capacity and burnout.

Conclusion: After more than a year of public health emergency response to the COVID-19 pandemic, there are detrimental and supportive factors that must be addressed to ensure both immediate and long-term mental health among public health workers.

Background: The utilization of Incident Command Systems (ICS) and Emergency Operations Centers (EOCs) in public health emergency and disaster response is a cornerstone of capabilities prescribed under the Center for Disease Control and Prevention’s Public Health Emergency Preparedness (PHEP) Capabilities.

Objective: To assess the use of ICS and EOCs by public health agencies during the COVID-19 pandemic response.

Methods: A content analysis of online materials available from 62 health departments funded by the Centers for Disease Control and Prevention for PHEP was conducted. Data was collected and inductively coded by two trained students with an initial agreement of 76.13% (κ=0.34).

Results: Bivariate associations between evidence of ICS utilization in COVID-19 response were assessed using relative risk regression. Fifty-eight percent (36 of 62) of PHEP grantees had accessible information regarding ICS use in COVID-19 public health response. PHEP grantees with online information about PHEP, Hospital Preparedness Program linkages, public health seats at state or jurisdictional EOCs, exercises, and ICS generally were more likely to have COVID-specific ICS information. Jurisdictions with a public health specific EOC were no more likely to provide COVID-specific ICS information online.

Conclusion: COVID-19 has placed significant strain on the operational response capacity of public health agencies. Given that this is the most significant event in which PHEP Capabilities have been tested, both in terms of geographic spread and duration, assessment of preparedness and response capacities such as the utilization of EOCs and ICS is critical for recognizing strengths and addressing shortcomings to improve future PHEP operations.

Resilience is a concept that is receiving increasing attention, is embraced across different fields, and is assessed at different levels. While its operationalization varies, resilience is often understood as the ability to rebound from a disturbance, return to function after disruption, and in some cases “build back better." Resilience is typically a positive attribute to strive for, and there is a growing emphasis on fostering resilience at all levels. While disasters and other crises are devastating for the impacted communities, communities often persevere and display resilience, and these experiences have the potential to serve as important learning experiences for the communities being impacted, as well as other communities. This research explores resilience in Wilmington, North Carolina, a city impacted by Hurricane Florence in 2018 and by COVID-19, by analyzing locally published newspaper articles that were generated in the aftermath of both events. Wilmington is characterized as having an increasingly sophisticated recovery and resilience community organization network. The analysis focuses on identifying themes that showcase how the community exhibited resilience. Specifically, this work explores what public discourse reveals about resilience in a community impacted by Hurricane Florence, a natural hazard, and COVID-19, a public health crisis. Overall, analysis reveals a strong short-term commitment to resilience but a lack of long-term planning and future visioning. This presentation will highlight how this community thrives in the wake of disaster and crises and provide insight for ways to further build resilience in the community.

Gender is an important social determinant of vulnerability to hazards. In disaster studies, the conventional, Western idea of gender is often perceived as the categorical distinction between men and women, which is hierarchical and in binary opposition. Mainstream disaster theories, limited to experiences of [cis]women in disasters, are yet to capture the complex contextual realities beyond the West. This has led to a ‘calculated blindness’ towards gender issues in disaster policies and actions. The broad categories of ‘vulnerable groups’ further overlook the diversity within one particular group. These ‘categories’ are formed through the process of othering those who do not embody the norm, defining their ‘abnormality’ and ‘faults,’ and pushing them to the margins of the social structure. The Hijra community, believed to be the oldest ethnic transgender group in India, faces several problems in their daily lives concerning accessibility to health services, public programs, education, verbal, physical and sexual abuse, economic challenges, and social discrimination. Apart from accounts of the 2004 Indian Ocean tsunami, Indian literature on experiences of transgender persons in disasters is very limited. Thus, this proposed research sets the agenda to reconstruct the idea of disasters from a queer perspective through the Hijra community. It aims to explore the experiences, capacities, and challenges of transgender groups in disasters and the role of intersectionality in addressing gender concerns. It intends to contribute to existing disaster discourses, underlining the need for gender-inclusive policies for effective disaster risk governance and advocate for profound academic research.

Catastrophic wildfires pose an increasing threat to communities throughout the world as climate change continues to accelerate. The United States is woefully underprepared and understaffed to meet the challenges of preventing and mitigating these destructive wildfires. In response to the need for trained and experienced wildfire personnel, universities throughout the nation are developing creative ways to help educate and train the next generation of wildland firefighters. Through opportunities such as trainings, certifications, and hands-on experience, universities are helping to not only train a much needed workforce but are also helping to bridge the gap between scientists and land managers. Some of these programs, like Clemson University's Fire Tigers, also assist in educating the public on wildfire prevention strategies currently in place, such as using prescribed fire.

Fire departments are often the first responders to natural or technological hazard events, and 80% of departments in the United States are fully volunteer or a combination of paid and volunteer firefighters. The combination approach is often seen as a solution to declining volunteerism without investing in a fully paid service. Organizational commitment in the fire service is usually high, and volunteer departments tend to have higher commitment than paid departments. Furthermore, research has shown a correlation between transformational leadership and organizational commitment.

For this study, data on firefighters' leadership and organizational commitment of firefighters was drawn from a pilot questionnaire about diversity and inclusion commissioned by the International Association of Fire Chiefs and developed by researchers at Oklahoma State University. The survey was deployed to six fire departments in 2019, some of which are combination departments that have both paid and volunteer firefighters. Factors for measuring organizational commitment and leadership in the existing survey data were analyzed for validity and considered for employment status and organizational structure. The volunteer firefighters within combination departments had a distinctly different perception of leadership and organizational commitment than both the paid members of combination departments and the volunteers within fully volunteer departments.

Further research is needed to validate these preliminary findings and to identify the roots of these perceived differences. It is clear that the combination department is not a simple solution for retention, and there is potential that volunteers may have less organizational commitment once their department pays some firefighters.

Fears of climate change and the escalating impacts of environmental damage are growing, and recent papers in the area of Sustainable Human-Computer Interaction (HCI) have called for urgent, non-linear solutions to these problems. Speculative design and related approaches, including design fictions, have been taken up to navigate these "wicked problems" that define contemporary nature/society relations. Unlike traditional design aimed at solving problems as they have been construed under the status quo, speculative design is about imagining how the status quo might be different and provoking discussion about alternative ways of being. To evaluate the potential and current adoption of speculative design and related practices by the HCI community, we surveyed speculative design projects published in Association for Computing Machinery venues between 2008 and 2021, assessing fundamental questions such as who is involved in the process, how sustainability is framed, and how speculation is used. Our evaluation of this body of work yielded mixed results; we find both promising trends as well as notable and problematic limitations in how the HCI community is taking up speculative practice in this domain. We build upon this evaluation to offer four provocations to designers seeking to use speculative practice to support sustainability goals.

Natural disasters, like storm surges, are often unpredictable in their timing and cause substantial loss of life and property. Climate change coupled with sea-level rise has increased the intensity and frequency of these storms, making them deadlier every year. Because of this, it is vital to accurately predict the impact that such weather events have on coastal populations. This type of predictive modeling is pivotal and a primary step in ensuring community resilience. Current physics-based mathematical models used to predict storm surges associated with hurricanes take a considerable amount of time to compute. Combining process-based and neural network-based models to imitate the behavior of these storms will help expedite the classification of the storm. This study provides a proof of concept and focuses on creating high-fidelity data of previous and potential storms in the Chesapeake Bay area using GeoClaw, a 2-D shallow water solver. This high-fidelity data is used as a training set for a neural network using TensorFlow. The goal of this project is to use the storm surge data from the GeoClaw simulations to train a neural network until it can reproduce the databases of the simulations developed. As this research has a short timeline of only 10 weeks, the model produced will serve as high-compression storage for simulations until it can be trained with more data. It is anticipated that other researchers will be able to utilize this coupling of models to quickly categorize storms and recognize the effects before they occur.  

An understanding of community resilience and how hazard events impact communities can lead to better policies and decisions designed to achieve community resilience goals. Communities achieve resilience when individual and household needs are equitably met through functioning physical and social systems. Social institutions, such as education and health care, help communities meet different needs. This poster presents a social institution resilience theory that can guide the integration of social institutions within computational modeling environments. The proposed theoretical framework combines the concepts of needs theory, community capitals, disaster phases, and social vulnerability. The research methodology utilized a systematic literature review and three case studies to answer the question of what community capitals facilitate or constrain a social institution’s ability to equitably meet community needs during different disaster phases. Findings suggest that the significance of community capitals depends on disaster phases. For example, a response model that predicts if a hospital can meet basic survival needs, such as emergency room services, may be valid with components that measure natural and built capitals. Meanwhile, a recovery model that predicts how long after a disaster it would take before a school district will be able to meet higher needs, such as students' need for belonging, would not be valid unless it included measures of social, human, political, and financial capital. The proposed social institution resilience theory will help researchers understand the complexity of and the capitals that impact community resilience.

The community timeline of disaster prevention action plans has not been understood by residents. The walking and cooking events have been held along the timeline disaster prevention action plan. As a result of a questionary survey, we could confirm the effect led to the awareness of disaster for the participants, but the effect for non-participants couldn’t appear. The number of participants needs to grow to foster the awareness of disaster of participants and non-participants. However, the problem is that the number of participants grew sluggishly. On the other hand, the following fields are increasing due to the aging there is a movement to make effective use of them. According to past research, residents have a habit of going to the fields two to three days before a typhoon arrives. By combining a walking event and a farm, residents go to the fields and secure ingredients. Residents acknowledge the profit by evacuating the shelter and raising awareness of evacuation behavior by preparing meals using the harvested vegetable at evacuation shelters. Furthermore, it is expected that the fieldwork can contribute to the fostering of local communities. Through the events of cultivation farms, we verified the effects of unconscious disaster prevention action on disaster awareness. Taking this research as an example, I would like to change the structure of the local community where the consciousness of the residents who cannot take evacuation action is latent when damage is considered due to a typhoon or heavy rain in Japan.   

There has long been a rift in the cumulative understanding of unsaturated soil behavior, rendering certain predictions about when, where, and how landslides occur relatively unreliable. It is often assumed in geotechnical analysis that the soil at a site is either fully dry or fully saturated. This assumption is inherently faulty as soil exists most commonly in an unsaturated state between these states. Researchers have only lately begun to study the properties of unsaturated soils using experimental and numerical methods. Experimental methods demand relatively many resources to test a small number of samples, whereas validated numerical simulations allow for testing of a wide range of variables that would be infeasible to experimentally test in a laboratory. In this numerical study, the Shan-Chen Lattice Boltzmann Method is used to model three-dimensional soil samples at different densities and inject water into the models until fully saturated. To do this, high-performance computers are used to conduct high-resolution simulations to capture the suction forces between the pockets of water and the soil grains. The collected data is then used to plot the Soil Water Retention Curve (SWRC) at the various porosities. Comparing and analyzing the SWRCs extrapolated from the data will give insight into the correlation between porosity and the resulting SWRC. Future analysis of the SWRCs will reveal information about the relationship between porosity and structural properties of soil such as effective stress, shear strength, permeability, aggregate stability, and others for potential geotechnical applications–especially in the area of terrestrial natural hazards.  

The use of images of children is a common branding tool used in various non-governmental organizations (NGOs) communication materials, reporting on the experience of children and their vulnerability in facing disaster, and communicating narratives of the catastrophe. The poster explores the iconography of children in reports and policy briefs in the field of disaster produced by international organizations and NGOs. It critically analyses the influence of the dominant iconography of children in shaping the discourse of disaster vulnerability and capacity, its narratives, and priorities. It more particularly shows that children are intentionally placed as a central figure of the representation of disaster in its humanitarian dimension. Images of children are often socially constructed to present them as vulnerable and resilient victims simultaneously in one scene. The iconography of children in disasters creates a narrative of vulnerable children with urgent needs for protection and lack of capacity. However, they are symbolized to become agents of change through education. This narrative shapes the focus of disaster risk management practices toward certain activities such as raising awareness and capacity building for children instead of creating a space where children can proactively and equally participate in the process of disaster risk management like adults. This multi-dimension iconography portrays the tension of different representations of disaster ‘victims’ images, and outcomes of NGOs work in empowering the marginalised.

Population growth, urbanization, and land consumption determine the vulnerability to extreme weather events and pose significant challenges to policymakers and urban planners. While the built environment plays a key role in these developments, detailed and consistently enumerated spatial data characterizing the built environment from a multi-faceted perspective is sparse, even in so-called data-rich countries. We try to fill this gap by integrating industry- and user-generated large-scale geospatial data sources to produce a set of gridded surfaces describing the physical, property-related, socioeconomic, and evolutionary (temporal) dimensions of the built environment in the United States, consistently enumerated at a fine spatial granularity of 250 meters. These built environment characteristics, such as built-up area, property value estimates, locally-owned properties, or building age, provide unique new opportunities to describe the built urban landscape from new perspectives. These data products represent the foundation of new ways to better understand the urban form and urban structure at a fine scale, providing unprecedented possibilities to carry out thorough assessments of the exposure and vulnerability of the built environment to natural hazards.

This research is supported by the National Science Foundation (award number1924670) and the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health (P2CHD066613). The use of Zillow’s ZTRAX data is acknowledged.

Non-profit organizations perform a crucial role in disaster recovery. Their horizontal and adaptive structures serve as intermediaries between the government and residents in the community. Non-profits are also better suited to promote social cognition, trust, and reciprocity, which often results in the creation of social capital. Despite non-profits' crucial role in disaster recovery, the literature has lacked insight as to how non-profits deal internally with the recovery process. We examined how health and non-health-focused non-profits resolve competing priorities when extreme weather and/or the recovery process happen simultaneously with a public health disaster? We conducted a telephone survey of 78 health and non-health sector registered non-profits in Puerto Rico and 30 key informant interviews in the context of three disasters in Puerto Rico, namely, the 2017 Hurricanes Irma and Maria, the 2020 Southwest Puerto Rico Earthquake sequence, and the 2020 COVID-19 pandemic. The results show that health and non-health-focused non-profits in Puerto Rico used Hurricane Maria and earthquake experiences to become more efficient at coping with and managing the effects of multiple disasters. Our study provides insights into the challenges faced by organizations as they adjust their mission and goals, program and activities, service provision, and community engagement. This study will help planners, emergency managers, and other stakeholders better engage with non-profits during recovery from multiple disaster events. This study gets the support of funds from the Centers for Disease Control and Prevention and the National Science Foundation received through the Natural Hazards Center in Boulder, Colorado.  

This research explores how service providers from community-based organizations (CBOs) navigate disaster recovery and immigration policies to help the Mexican immigrant community in Houston, Texas with their long-term housing recovery after Hurricane Harvey. I conducted semi-structured interviews, content analysis, and ethnographic observations with service providers from CBOs in Houston that serve this population with post-disaster housing. I argue that the disaster recovery system is an inherently anti-immigrant racial structure, passively and actively limiting the access to resources for the Mexican immigrant community. I found that to challenge this anti-immigrant racial structure of disaster recovery, service providers assist the community through providing direct assistance to immigrants excluded from other disaster recovery programs, collaborating with other organizations to combine limited resources, helping the community navigate anti-immigrant bureaucracy, and building trust by embedding themselves in the victimized community. However, findings also show that these organizations face significant challenges in conducting their work. The findings from this project will be of value to policymakers and stakeholders involved in post-disaster recovery work.

Light Detection and Ranging (LiDAR) point clouds can be colorized with 360° photo imagery. Colorized point clouds increase their situational awareness and data potential. Potential data uses are object classification, which aids in further understanding of multiple situations, from identifying blue tarps in a hurricane-affected area to assessing disaster damage such as landslides. Colorized LiDAR point clouds methods already exist, but these methods are limited to certain hardware and software configurations. With the current hardware and software, the University of Washington Natural Hazards Reconnaissance Facility (RAPID) possesses, an efficient and accurate workflow for point cloud colorization will be developed. A new method and workflow must be developed to colorize point clouds with local assets. Local assets such as Applied Streetview imagery and Phoenix miniRanger LiDAR scans are recorded on a mobile vehicle platform. Applied Streetview imagery is processed through Pix4D to generate point clouds. The LiDAR scans are processed through Phoenix Lidar Systems software. Both point clouds will be merged and colorized using Cloud Compare or Quick Terrain Modeler software. This study seeks to properly document how to colorize LiDAR imagery with current RAPID hardware and software. Colorized LiDAR data will be instrumental in analyzing future situations in natural hazard reconnaissance with a mobile platform. Natural hazard reconnaissance with a mobile platform capable of colorized LiDAR data will require further analysis to realize the full potential of the data.   

COVID-19 has had a tremendous impact worldwide and has exposed the massive vulnerabilities of specific underrepresented populations. Among these populations, Black Americans have experienced serious illness and death, and recent research suggests that systemic vulnerabilities are to blame. Since preventative measures to fight against COVID-19 largely depend on an individual’s ability to social distance and the devastating relationship between COVID-19 and Black Americans, this project recognizes that the ability to take part in preventative measures is not equitably distributed.

The objective of this research is to explore the relationship between COVID-19 deaths of Black Americans, social distancing measures, population density, and political affiliations to explore the extent to which the Black experience with COVID-19 may relate to their ability to social distance. The results will help decision-makers better enable the social distancing practice for vulnerable groups.

Data for this research were collected from three sources: social distancing data from Safegraph, sociodemographic data from the American Community Survey (U.S. Census Bureau), and political party identification from the 2020 presidential election results from Dave Leip's Atlas 2020 U.S. Presidential Vote Data. The analysis used linear regression to explore the relationship between COVID-19 deaths of Black Americans, social distancing patterns, political affiliations, and other key factors in 30 counties/cities during seven workdays in July 2020. The findings show that deaths of Black people caused by COVID-19 are significantly associated with social distancing and political affiliations.

Repetitive non-extreme flooding has disrupted the functioning of settlements, and studies suggest that sea-level rise will affect more people considerably sooner than previously imagined due to stronger heavy rainfall, storms, and rising groundwater combined with higher tides. To date, managed retreat, as a climate change adaptation strategy, has largely been reactive, which means measures have been taken after damage and disruption. In contrast, the pre-emptive managed retreat is not a widely accepted and practiced adaptation strategy. There is a need to shift managed retreat from an ex post (reactive) to an ex ante (proactive) strategy, helping coastal communities effectively prepare and adapt to uncertain futures.

To facilitate ex ante managed retreat, it is important to understand how the existing cases developed collaboration frameworks, coordinated governance arrangements, and increased community engagement by utilizing information technologies and different visual expressions. This research is to compare the similarities and differences of ex ante managed retreat implementation experiences among different cases. The results will assist stakeholders in informing and enhancing the decision-making processes for the agencies and the communities they serve in order to make proactive investments in long-term adaptation and resilience.

Data of ex ante managed retreat cases for this research come from three sources: the Georgetown Climate Center, the Managed Retreat Conference, and the Waterfront Alliance. The findings of this research recommend that communities should develop local jurisdiction-wide, community-driven adaptation strategies and empower stakeholders to take proactive action toward flood risk reduction and prevention in participatory planning processes.

Moving businesses and residents out of hazardous areas is an important mitigation priority. At the same time, disaster recovery spending is increasing as the frequency and nature of hazards intensifies. Research has suggested that disaster assistance, given its focus on infrastructure replacement, may encourage development in the same hazardous area or prevent recipients from moving. This study examined residential and business mobility of the recipients of post-disaster loans in Galveston County after Hurricane Ike in 2008 and after Hurricane Harvey in 2017. The study explored whether loans encourage or discourage residential and business movement out of hazardous areas, as well as the factors that influence location decisions. This research found that 19% of single-family residences receiving assistance changed ownership after Hurricane Ike and 13% changed ownership after Hurricane Harvey. For businesses, 9% and 24%, respectively, moved to a new location. Location decisions were not only motivated by economic and personal reasons, but also by disaster risk. Disaster assistance had an effect on whether businesses and residents chose to remain at their current location or mitigate against a future hazard, indicating a need to further integrate mitigation and recovery policy.

Where people go between evacuation and recovery remains an understudied aspect of disaster research. Whether experiencing multiple displacements, permanent displacement, or undergoing recovery in a damaged dwelling, the spatial and temporal dimensions of disaster displacement can have direct impacts on the recovery experience of survivors. Pulling from focus group data gathered in 2017 from Hurricane Sandy survivors in New Jersey, this qualitative research focuses on the experiences of those who recovered in-situ or within their damaged dwelling following the storm and the various ways this non-displacement impacted their recovery. A content analysis following a grounded theory approach produced the emergent themes of the in-situ experience, including a lack of suitable shelter, an exposure to secondary hazards, and an inability to achieve satisfactory emotional recovery. This study contributes to the growing body of literature surrounding recovery experiences, and it introduces valuable insights into the challenges that survivors face while recovering in-situ.

Ground shaking and liquefaction can have highly detrimental consequences on the built environment and on communities in regions of high seismicity such as California. Characterizing the near-surface soil layers and their properties allows us for direct evaluation of the amplification of ground shaking as waves travel from the bedrock to the surface. A soil model can also help infer the liquefaction potential and the ground deformation caused by horizontal spreading and vertical settlement in the area. This study focuses on the city of Alameda to illustrate the benefits of modeling near-surface soil characteristics. Cone Penetration Test data provided by the U.S. Geological Survey, local geotechnical borehole reports from construction and utility projects in Alameda, and previously developed sediment velocity models are used to estimate the soil profile and develop a random field that characterizes the soil behavior near the surface. A seismic site response analysis is conducted to improve our understanding of the seismic hazard and the risk local communities face.   

Knowing where wildfires are most likely to occur is vital for determining why wildfires occur where they do and for developing locally relevant wildfire mitigation strategies. We determined the extent to which significant hot spots of wildfire activity in Washington State occurred between 1970 and 2020. We then mapped spatiotemporal variation in wildfire locations and acres burned using various spatial statistical methods. For the purposes of this study, only wildfires that burned at least 100 acres were considered. We detected significant hot spots and cold spots in the state, with our results finding all hot spots are located east of the crest of the Cascade Range. We then determined the extent of spatial variation in wildfire occurrences and acres burned over time, revealing that the geographic area wherein most of the state’s acres burned has shrunk considerably since 1970 even though total acres burned has drastically increased. This concentration of large wildfire activity is increasing the risk of wildfire in the north-central portion of the state considerably. Our results showcase the advantages of using spatial statistical methods and the importance of both spatial and temporal considerations in the context of hazard mitigation plans, which must grapple with the ever-increasing complexities of climate-driven changes in hazard risk.

Communities in flood disaster-prone cities, such as Houston, can be under great threat from flood hazards. Especially in emergency situations, planning optimal reaction strategies must depend on the identification of socially vulnerable areas and communities in disaster events during both disaster management and recovery in terms of the evacuation of socially vulnerable groups and the accessibility to emergency facilities. The existing disaster information systems remain temporally imprecise, spatially vague, and do not consider social vulnerability and decision-making capabilities. Motivated by this observation, in this poster, we propose an interactive and collaborative spatial decision support system based on advanced cyberinfrastructure, WebGIS, and citizen science to improve situational awareness in disaster management. Our platform included in the system considers spatial and social vulnerability priorities to enhance knowledge elicitation and sharing among disaster responders and communities.

Based on identified emergent events from social media using a weakly supervised learning system and social sensing and surveys, we are able to model flooding risk perception and present identified spatial hotspots of socially vulnerable locations with a custom-designed online mapping application. By systematically integrating spatial analysis on identified socially vulnerable communities and evaluation on transportation accessibility for socially vulnerable communities, the proposed decision support system can be used to support optimal reaction strategies planning by visualizing communities which should be considered with high priority during both disaster management and recovery in flood events and further support disaster resilience planning.