VMASC Researchers’ Study Looks at Region’s Vulnerabilities Following a Major Storm

May 30, 2013

Rafael Diaz

A massive survey that reached more than 7,000 homes in Hampton Roads, done as part of a study by two researchers at Old Dominion University's Virginia Modeling, Analysis and Simulation Center (VMASC), is proving to be an invaluable tool to help the region better understand residents' perceptions of risk, and how they would respond to a major storm event.

It provides the most accurate forecast to date of a giant storm on vulnerable populations, localized in every part of Hampton Roads.

The work of Joshua Behr and Rafael Diaz, research associate professors at VMASC, is being presented around the country by Virginia's Department of Emergency Management (VDEM) to showcase the best practices that could come as a result of the research.

"This study has just exploded. The potential importance of this information is immense," said Behr, who comes to VMASC from a social science background.

Funded through a six-figure donation by the Perry Family Foundation, the initial 2011-12 survey was the largest of its kind ever done in Virginia. It focused on the 24 communities that comprise the greater Hampton Roads area, running from northeastern North Carolina to the edge of the Capital Region.

Behr and Diaz developed a multidimensional metric of community and social vulnerabilities that explains the propensity of residents to either shelter in place or evacuate the region, as well as identifies the health and housing needs of those residents following the storm.

"Through our studies of hurricanes in the United States - as well as catastrophic events around the world including mud slides in Bolivia, cyclones in Sri Lanka and the Fukushima nuclear disaster in Japan - we began the process of conceptualizing the different types of vulnerabilities that may be evident in Hampton Roads when a large hurricane hits," Diaz said.

The behavior of residents is estimated through factors beyond the physical location of a resident relative to the storm path, to include financial and social networks, medical fragility, past storm experience, trust in government and other psychosocial aspects of vulnerability.

"Through the extensive process of interviewing households following Hurricane Irene, we were able to confirm, measure, and document over a dozen dimensions of vulnerability present in Hampton Roads. We identify particular geographies and population segments that are hypervulnerable, meaning there is a synergistic dynamic at play when an area or population is experiencing several vulnerabilities at once," Behr said.

That information was combined with existing flood maps, road networks, and information about first responders and health providers such as primary care offices.

This has allowed the researchers to produce data that reveal how the range of vulnerabilities varies across the region. Of particular interest to them was gaining a better understanding of medical vulnerabilities within the region. For this, their models placed "medically fragile residents and their caregivers in precarious situations when faced with the choice of where to shelter or to evacuate," Behr said.

Citizens with mobility impairments, the need for specialized transport, and sensory or cognitive impairments are particularly vulnerable, Behr said. "They may not have the means or ability to effectively get out of harm's way, and then recover, both financially and health-wise, to a state of normalcy that may approximate the pre-event quality of life."

With the survey data in place, that information was intersected with a simulation of a major storm event in Hampton Roads, which the researchers call "Sandtrina" - after Hurricane Katrina, the 2005 storm that devastated New Orleans, and last fall's Superstorm Sandy, which blasted the U.S. northeast coast.

Katrina was chosen because of the close similarity of Hampton Roads to New Orleans, Behr's hometown, in terms of topography, demographics, economy and industry.

"Katrina and Sandy can be viewed as laboratories that yielded insights we could apply to Virginia," Diaz said. "By adapting the breadth of wind field and storm surge of Superstorm Sandy, and modeling that type of scenario to this region, we have learned quite a bit without actually having to directly experience such a storm."

The intersection of the data streams can help predict how many people in each of greater Hampton Roads' 24 communities might be expected to seek shelter or evacuate, but also the characteristics and needs of the affected citizens.

Combining the information gleaned from mapping behavior vulnerabilities and modeling real-world storms could prove to be of real benefit to agencies charged with responding to such an event. "This research may pay huge dividend in terms of training and refining state emergency preparation efforts and, ultimately, saving lives and property," Behr said.

As an example, the Federal Emergency Management Agency (FEMA) could use the information to help plan for immediate post-event housing.

An extension of the survey and the modeling of the storm scenarios has been supported by VDEM and FEMA with $400,000 through the Urban Area Security Initiative.

"We've listened to emergency planners from all over the region, who expressed a desire for more quality localized data," Behr said. "In the past, regional estimates of displaced populations, housing damage and dollar impacts stemming from a fictitious storm scenario were so overwhelmingly large that they had the potential to lose their practical connection to local planning efforts."

The new study by Behr and Diaz is attracting attention. VDEM presented some conceptual findings at the Interagency Coordinating Committee on Hurricanes conference in Atlanta, and department officials later took it to the National Hurricane Conference in New Orleans.

Most recently, the ODU researchers gave the keynote presentation at the Hampton Roads Regional Catastrophic Planning Team's State Disaster Housing Workshop in late April at Christopher Newport University. In attendance was a wide array of planners and emergency officials from every level of government.

"Through our presentation, I believe we were able to communicate the power to be had from leveraging a collaborative engineering, health and social science team. These multidisciplinary efforts are very much the hallmark of Old Dominion University's research approach," said Diaz.

Behr added: "That's what we wanted to do, to be a resource and demonstrate the practical application of modeling and simulation. And not just for here. The approach we have developed and the instruments we have validated here in the region are replicable in any other region, and may prove to be of use to emergency planners elsewhere."