Workshop Abstracts
Dara Angelo, University of South Carolina
Impacts of Optimism on Evacuation Behavior
This research examines the relationship between life orientation (optimism or pessimism) and evacuation behavior. Life orientation has been shown to affect physical and mental health, organizational skills, decision-making, and stress management. Optimism is shown to be a strong predictor of the use of problem-focused coping strategies and better cognitive and emotional functioning. Both aspects could contribute to positive evacuation behaviors. According to this assumption, I hypothesize that optimists would be more likely to prepare for and evacuate during a hurricane.
There is a growing research interest in the link between optimism, preparedness, and resilience. However, much of the research is difficult to implement because it requires measuring the ways people cope with stress and involves following subjects during a stressful event. The goal of this research is to use data from the Hurricane Evacuation Study conducted by the Hazards and Vulnerability Research Institute to fill a gap in the link between optimism and resilience. The study consists of surveys sent to residents of eight coastal counties of South Carolina asking questions about evacuation intent and behavior. The survey also includes the ten questions from the Life Orientation Test used to determine the respondents’ level of optimism. The preliminary findings suggest no correlation between life orientation and evacuation intent or prior experience among respondents. However, there was a positive correlation among the elderly.
Toward a Comprehensive Understanding of the Economic Costs of Disasters
Most observers agree that the economic costs resulting from natural disasters are increasing, but few people understand the nature of these costs and how they are estimated. Cost estimates are published in media reports after each event, but despite the opacity of the proprietary models from whence these estimates often come, they seem to become the "official" estimates.
The International Disaster Database in Brussels, the National Climate Data Center at the National Oceanic and Atmospheric Administration, and the Spatial Hazard Events and Losses Database (SHELDUS) at the University of South Carolina all report economic damages for major events. But there are no standards for these estimates, and inconsistencies exist for the same event across databases.
The Federal Emergency Management Agency's HAZUS model is perhaps the most comprehensive tool, but it is usually deployed to estimate the U.S. government's public assistance requirements. Congress has raised concern over the accuracy of these estimates. Moreover, the HAZUS data requirements are massive and this inhibits regular updating of the relevant cost information.
The insurance industry produces disaster cost estimates, but these are limited to physical damages to insured properties, not to all properties, and their models are proprietary and unavailable. Finally, some scholarly work has been done to estimate the economic losses from business interruption, but there is no comprehensive framework to guide the assessment of all economic costs that result from natural disasters. This paper begins to fill this void.
L.C. Chen, National Taiwan University
C.C. Chang, National Taiwan University
Y.L. Kuo, National Taiwan University
C.Z. Deng, National Taiwan University
H.C. Li, National Taiwan University
Resident Risk Perceptions, Preparedness, and Losses from Flood: The Case of Benhe Community in Kaohsiung City during Typhoon Fanapi
Located in a frequently flooded lowland area, Benhe community was the only place where a flood detention pond was built in 2004 in Kaohsiung City. The community has been developed intensively and the pond has become a popular recreation area for local residents. However, when Typhoon Fanapi brought record rainfall of 345.5 millimeters (13.6 inches) in three hours into the Kaohsiung area in mid-September 2010, Benhe residents suffered severely from flooding because the strong rainfall exceeded the capacity of pond's pumping station.
Based on interviews with the community leader and 10 households, we uncover three additional factors contributing to unprecedented losses. First, the detention pond has lowered the risk perception and creates a false impression that the area is protected from flooding. Second, the community flood alarm system malfunctioned because it was not well-maintained after the completion of the detention pond. The residents could not remove motor vehicles and install floodgates in time, even though they set up floodgates on the ground floor and at the entrance of basements after a serious event 10 years ago. Third, the flood depth was higher than the floodgates. Water damaged public utility equipment stored in basements and most motor vehicles.
To adapt to the extreme hazard event in the future, perceptions of unavoidable flood risks have to be reinforced continuously. Assisting the community to relocate public utility equipment to attics would be a useful mitigation. An effective warning system is also required. This case study implied that the residents' risk perception could be lowered after a few non-flood years. Future research is required on how various flood experiences influence risk perception and the relationship between risk perception and preparedness.
Jeffrey Czajkowski, University of Pennsylvania
Howard Kunreuther, University of Pennsylvania
Erwann Michel-Kerjan, University of Pennsylvania
Flood Risk, Mitigation, and Insurance in Texas
One of the many hurdles to privatizing flood insurance when the National Flood Insurance Program (NFIP) was established in 1968 was the inability of insurers to quantify the financial risk of insuring properties with large catastrophic loss potential from flooding. The development of flood risk models in recent years has refined insurers’ ability to quantify this financial impact.
Catastrophe models enable private insurers to set premiums for flood insurance that reflect risk if this coverage were offered by the private market. This paper specifies risk-based flood insurance premiums using a catastrophe model developed by CoreLogic for single-family residential structures with high exposure to flooding damage in three Texas counties. It uses data on 1.5 million individual single-family residential parcels.
One rationale for privatizing the NFIP is to encourage cost-effective mitigation measures by homeowners. We estimate the potential reduction in risk-based premiums from the implementation of mitigation efforts. Finally, we undertake cost-benefit analyses of these efforts to determine how economically efficient flood mitigation measures may be, accounting for different discount rates and time horizons.
Based on these analyses, we examine how well a flood insurance program would perform with premiums reflecting risk compared to the current NFIP program. For such a program to be politically feasible, grants in the form of insurance stamps must be provided to those requiring special treatment, such as low-income residents currently residing in flood-prone areas. The paper concludes by suggesting future research on the role of insurance to encourage residents in hazard-prone areas to adopt cost-effective mitigation measures.
Birnur Guven, Houston Advanced Research Center
Robert Stein, Rice University
Leonardo Dueñas-Osorio, Rice University
Devika Subramanian, Rice University
David Kahle, Rice University
The Severity of Perceived Risk in Predicting Evacuations from Hurricanes
The objective of this study is to show whether people’s risk perceptions of hurricane-related hazards can be summarized by a smaller number of multivariate risk measures. The study aims to understand the role and strength of such composite measures of risk in providing a viable means of predicting responses to severe weather events.
We draw on survey data with residential populations in the upper Gulf Coast areas of Texas after hurricanes Rita and Ike. Composite dimensions of perceived risk were determined using the statistical procedure known as multiple correspondence analysis. Behavioral responses to both hurricanes are modeled with both the marginal and composite measures of perceived risk.
We found people’s perceived risk of different hurricane-related hazards can be reduced to a single score that spans all risk types. Evacuation behavior is strongly dependent on whether one has a high risk of all types of hurricane related hazards regardless of the specific identification of the risk type.
The results suggest that people are less sensitive to risk type than they are to the general seriousness of the risks. Using this composite risk measure, emergency managers can be informed about the severity of public risk perceptions and might better craft their public directives in ways that minimize disruptive evacuations and achieve greater compliance with government directives.
Maka'ala Ka'aumoana, Hanalei Watershed Hui
Debbie Gowensmith, Hawaii Community Stewardship Network
Cheryl Anderson, University of Hawaii at Manoa
Sarah Henly-Shepard, University of Hawaii at Manoa
Gabrielle Schuerger, Impact Assessment, Inca
Edward W. Glazier, Impact Assessment, Inc.
John N. Kittinger, Impact Assessment, Inc.
Climate Change & Disaster Vulnerability—Building Socio-Ecological Resilience in Hawaii
Pacific Island communities are among the most vulnerable to natural hazards and predicted impacts associated with climate change. Located in the Pacific Ring of Fire and 2,500 miles from the nearest continental landmass, the Hawaiian archipelago is at heightened risk.
In Pacific cultures and communities, hazards have been mitigated in the past through risk management strategies that endowed communities with the adaptive capacity to weather disasters. These strategies include place-based knowledge systems, social institutions, and embedded learning processes for adaptive resource management passed from one generation to the next, maintaining linkages between healthy ecosystems and communities.
In some island communities many of these practices have been preserved. For example, the community of Hanalei, Kaua‘I, successfully weathered the aftermath of Hurricane Iniki through use of local and indigenous traditional skills, knowledge sets, and social networks. Long-time residents viewed material preparedness and shared knowledge as part of their traditional customs and responsibility. As a result the community remained resilient, allowing for social and ecological recovery.
But major social and ecological transformations have occurred since Iniki, leading long-time residents to question whether the traditional knowledge and practices endure. We conducted interdisciplinary, community-based, participatory research to assess the vulnerability and adaptive capacity of the community to natural disasters and climate change, and to identify the elements that comprise socioecological resilience at the community level in order to engage in long-term resiliency planning and policy implementation.
Kelly Klima, Carnegie Mellon University
M. Granger Morgan, Carnegie Mellon University
Iris Grossmann, Carnegie Mellon University
Kerry Emanuel, Massachusetts Institute of Technology
Does It Make Sense to Modify Tropical Cyclones? A Decision-Analytic Assessment
Dramatic increases in damages caused by tropical cyclones (TCs) and improved understanding of TC physics have led the Department of Homeland Security to fund research on intentional hurricane modification. We present a decision analytic assessment of whether it is cost effective to attempt to lower the wind speed of TCs approaching South Florida by reducing sea surface temperatures with wind-wave pumps. Using historical data on hurricanes approaching South Florida, we develop probabilities of how storms might evolve. The effects of modification are estimated using a modern TC model. The FEMA HAZUS-MH MR3 damage model and census data on the value of property at risk are used to estimate expected economic losses. We compare wind damages after storm modification with damages after implementing hardening strategies protecting buildings. We find that if it were feasible and properly implemented, modification could reduce net losses from an intense storm more than hardening structures. However, hardening provides "fail safe" protection for average storms that might not be achieved if the only option were modification. The effect of natural variability is larger than that of either strategy. Damage from storm surge is modest in the scenario studied, but might be abated by modification.
Kelly Klima, Carnegie Mellon University
Wändi Bruine de Bruin, Carnegie Mellon University
M. Granger Morgan, Carnegie Mellon University
Iris Grossmann, Carnegie Mellon University
Public Perceptions of Hurricane Modification
If hurricane modification were a feasible strategy for reducing hurricane damages, residents of hurricane-prone areas would need to make more informed decisions about whether or not to support its implementation. Here, we examine Florida residents’ perceptions of hurricane modification techniques to alter path and wind speed. We conducted an initial set of semi-structured mental model interviews, followed by a survey study with a larger sample. We found that participants questioned the feasibility of hurricane modification, expressing negative responses to the idea of tampering with nature. Hurricane modification scenarios evoked anger at scientists, but responses were less negative when hurricane modification involved reducing wind speed without changing direction. Although interviewees recognized the natural variation inherent in hurricane forecasts, they did not mention it in the context of hurricane modification. Recognizing such uncertainty should reduce attributions of blame after a failed hurricane modification attempt. In the follow-up survey, we unexpectedly found that participants who recognized uncertainty expressed more anger at scientists attempting hurricane modification, as damages were increasingly worse than forecasted. If the efficacy of techniques can be increased, people may be willing to support hurricane modification. However, such an effort would need to be combined with open and honest communications to members of the general public.
Scott Knowles, Drexel University
James Kendra, University of Delaware
The Emergency Managers: A Profession on the Edge
Recently, a discussion on the International Association of Emergency Managers listserv detailed the projected areas of expertise for a pandemic flu planner. The skills list ran to about 50 areas of expertise, from public health quarantine to epidemiological monitoring. What profession could possibly expect its members to be expert in so many specialized knowledge areas, and also have the political competency to practice these skills in the midst of a terrifying disease outbreak? Emergency management.
According to the prevailing framework of emergency management in the United States, emergency managers are ready for all hazards, and every phase of a disaster— mitigation, preparedness, response, and recovery. The reality, though, is that no profession in American history—except, perhaps, for military officers—has assumed so much proficiency and so much knowledge in its practitioners. In the absence of realistic disaster policies and in the retreat of insurance companies from the realm of disaster expertise, emergency management has been a profession on the ascent since the 1990s. The mountain of risk it is climbing, though, is a treacherous one—as Hurricane Katrina demonstrated.
As members of a nascent profession, emergency managers face interwoven challenges in institutionalizing their claims of special expertise in handling dangers. Ownership of terms such as risk, danger, hazard, disaster, vulnerability, pandemic, terrorism, and resilience among others is claimed by multiple specialists. The basic sciences underlying disaster still lack
precision in many areas, making practice a continuing foray into uncertainty.
This presentation will examine the challenges of trying to build a profession at the borders of knowledge in numerous areas, managing ambiguity and doubt as much as risk and danger. Drawing on the history of science and technology, the presentation will explore the relationship of science and practice in advancing a profession in a social and policy environment that itself is demanding, shifting, and transient.
Michelle Lueck, Colorado State University
Lori Peek, Colorado State University
Craig Trumbo, Colorado State University
Holly Marlatt,Colorado State University
Hurricane Risk Perception and Evacuation Decision Making Among Persons with Disabilities
Persons with disabilities are more likely to be poor and to live in low-quality housing, which increases their exposure to hazard impacts. Also, a limited body of research shows that when disasters strike, these individuals could have a more difficult time taking recommended protective actions, evacuating, or withstanding the event. This presentation builds on prior work by examining the relationship between disability status (i.e., physical, cognitive, and/or sensory) and evacuation decision-making processes and behaviors. In this presentation, we report preliminary findings of the first wave of data from a three-year panel survey of a spatially random sample of more than 600 individuals living within 10 miles of the U.S. Gulf or Atlantic Coasts. Over 10 percent of our sample reported that they, or someone in their household, were living with at least one form of disability; and over 20 percent reported being concerned about the evacuation of a family member with a disability not living in their household. Our analysis will compare the responses among those with and without disabilities in terms of evacuation behavioral intention and reported barriers to evacuation. This paper will provide valuable insight into this vulnerable population's response to hurricane risk and evacuation planning.
David McEntire, University of North Texas
Abdul-Akeem Sadiq, Indiana University-Purdue University Indianapolis
Kailash Gupta, University of North Texas
Unidentified Bodies and Mass Fatality Management in Haiti: A Case Study of the January 2010 Earthquake with a Cross-Cultural Comparison
Past disasters such as the terrorist attacks of 9/11, the Indian Ocean tsunami, Hurricane Katrina, the Sichuan earthquake, the Haiti earthquake, and the recent Japan earthquake and tsunami underscore the need for a thorough understanding of mass fatality management (MFM). These disasters resulted in a significant number of deaths, overwhelming the resources of local, state, and federal governments. For instance, the 2004 Indian Ocean tsunami and the terrorist attacks of September 11, 2001, in the United States killed about 275,950 and 2,792 people respectively. These cases strained the response systems of the affected nations.
One of the most significant potential problems after such major disasters is how to effectively deal with unidentified bodies. This study uses the January 2010 earthquake in Haiti as a case study to better understand this neglected component of MFM. In addition, this study compares findings in Haiti to other disasters in India, Bangladesh, and Sri Lanka, and concludes with a discussion of implications for research and practice.
Richard Stuart Olson, Florida International University
Juan Pablo Sarmiento Prieto, Florida International University
Gabriela Hoberman, Florida International University
Disaster Risk Reduction, Public Accountability, and the Role of the Media:
Concepts, Cases, and Conclusions
This paper argues for the utility of—and offers, a relatively narrow definition for—accountability. We use that definition in combination with an applied analytic framework for accountability in the specific policy domain of disaster risk reduction. That framework then informs an analysis of the post-impact roles that media played in two major 2010 events—the January 12 Haiti earthquake and the February 27 Chile quake, supplemented by an analysis of media coverage in two “mirror countries” that share roughly equivalent disaster risk profiles, Jamaica and Peru. The more specific research questions focused on media attention spans, other emergency or disaster relevant media roles, and “zones of silence” in event coverage, particularly about pre-event accountability for effective DRR.
Nicholas Pinter, Southern Illinois University
Anthropogenic Forcing and Lack Thereof: 2011 Mississippi River Flood
Flooding in 2011 began in the Ohio and Mississippi rivers confluence area. The crest then moved down the Lower Mississippi over about two weeks, causing widespread damage. Press coverage during the 2011 flood speculated about man-made contributions to the flood. These discussions, and the 2011 flood, can be evaluated based on extensive research that has tested for anthropogenic flood forcing.
Various researchers have assessed flows and basin precipitation for climate-driven trends. Our own 2008 research tested for peak discharge trends using a dataset of over eight million historical values. Statistically significant flow trends (all increases) were only found the Upper Mississippi. Any systematic climate change or land use shifts downstream were counterbalanced by reservoir construction on the Mississippi tributaries. Climate models predict future increases, but suggestions that 2011 discharges reflect past climate change (or urbanization, etc.) are difficult to justify.
Flood stages reflect both discharge and local conditions. “The water is supplied by nature, but its height is increased by man,” wrote engineer Charles Ellet in 1852. We tested flood stages at the same stations as above. Significant increases and decreases were identified throughout the Mississippi system, including stage increases up to 10 times greater than can be attributed to discharge alone. Local engineering—levees, navigational structures, and channel straightening—dominate stage changes. Stage peaks during the 2011 flood confirmed these historic trends. The Mississippi and many other U.S. rivers have been heavily modified, and these historic modifications have major influences on modern floods.
Ramani Sankaran, St. Peter’s University, Chennai
Didactic Simulation for Disaster Management
Didactic Simulation is an exercise in which participants compete as a group against other groups, as well as against each other. This exercise does not need a meeting leader. The principal features are rapid involvement, immediate application of new knowledge, peer consensus, simultaneous learning and teaching at the same time, and exchange of experience.
The simulation session is administered in three segments. First is the introductory remarks session. The participants then work on each item in small groups in the second session. The concluding session deals with substantive questions and comparison of solutions arrived at by various teams. Participants who find that their scores are consistently and substantially below these of the group may wish to review some areas.
The author has designed, developed, and tested an abridged version of the Didactic Simulation Game for hurricane management. The modules deal with objectives and priorities of the hurricane management system, criteria for effectiveness of disaster management, disaster supply chain management, and the effective handling of the post-disaster rehabilitation project.
The game stimulated discussion on the various aspects of hurricane disaster management and created an exchange of ideas on the advantages and disadvantages of various strategies. The scores allotted by the game designer are based on research from various disaster-related journals and reports.
Jeannette Sutton, University of Colorado Colorado Springs,
Carter T. Butts, University of California, Irvine
Emma S. Spiro, University of California, Irvine
Matt Greczek, University of Colorado Colorado Springs
Sean Fitzhugh, University of California, Irvine
Nicole Pierski, University of California, Irvine
Project HEROIC: Understanding Informal Online Communication in Disasters and Other Extreme Events
While often taken for granted, informal information exchanges—rumor, gossip, and casual conversation—can become an important "soft infrastructure" for decentralized resource mobilization and response during extreme events. Recent developments in social media technology and mobile devices have transformed informal communication, allowing individuals to reach a larger number of others across greater distances than ever before.
There is a growing awareness of the potential value of informal online communication networks for emergency warnings, alerts, and related applications in the disaster and emergency management community. Despite this recognition, however, much remains unknown about the dynamics of online conversation and the underlying networks that structure it.
Our research addresses this gap, employing a longitudinal, multi-hazard, multi-event study of online communication to model the dynamics of informal information exchange in and immediately following emergency situations. Our project consists of several linked activities. First, we are undertaking a baseline-controlled longitudinal study of the use of online communication over a multi-year period. Second, we are employing systematic content analysis of micro-blogging posts and related information (e.g., blog posts, web pages, and so on) to understand the way in which members of the public sort, interpret, and promulgate information during emergency events. Finally, we are developing statistical models for the joint evolution of content and network structure within informal online communication channels in response to emergencies and other exogenous events.