In October 2016, Hurricane Matthew made national headlines as it battered hundreds of communities along the southeast coast of the United States – the names of which rarely make those headlines. One of them was Lumberton, North Carolina, a racially diverse city of 20,000 people, many of whom live in poverty.

In the weeks after this small coastal city withstood devastating flooding of the Lumber River, it hosted some unusual out-of-town visitors: researchers from Colorado State University and several partnering institutions, intent on documenting the community’s damage and recovery from the flood.

The researchers have since returned to Lumberton several times, carefully tracking the city’s long road to recovery. It’s all part of an unprecedented, federally funded effort to help communities like Lumberton prepare for and respond to disasters, including hurricanes, floods and fires.

A recent study documenting the flooding of Lumberton is authored by scientists in the Center for Risk-Based Community Resilience Planning, a 12-institution research center headquartered at CSU, established in 2015 by a $20 million grant from the National Institute of Standards and Technology. The center’s goal is to develop tools to enable worldwide research in a new area called resilience science, in which a community’s resilience to hazards can be quantified and ultimately improved.

The Center for Risk-Based Community Resilience Planning is one of three designated National Institute of Standards and Technology “Centers of Excellence.” Co-led by civil engineering faculty John van de Lindt and Bruce Ellingwood at CSU, the center comprises close to 100 experts from institutions including NIST, Texas A&M University and the University of Kansas. The Lumberton report is co-edited by van de Lindt, Walter Peacock of Texas A&M, and Judith Mitrani-Reiser of NIST.

In their new study, the team provides baseline measurements and observations of Lumberton’s response to and recovery from Hurricane Matthew-related flooding. In coming months, they will build on these findings with follow-up field campaigns, with the goal of extrapolating what they learn to a generic resilience model that can be applied to any community.

Florence follow-up

In September 2018, when Lumberton also bore the effects of Hurricane Florence, the center deployed a team to assess damage from that storm, too. That effort was led by center member Elaina Sutley, assistant professor at the University of Kansas, who completed her Ph.D. work at CSU in 2015.

“It is important for us to go back several times, since recovery is a continuous process that involves many actors and many decisions, and we want to be able to capture these details in our models,” Sutley said.

Comparing impacts from the two storms, they observed post-Matthew mitigation strategies and safety procedures that may have lessened Florence’s impacts. These included residents evacuating to safe areas sooner; homeowners having routed duct work from low-level crawlspaces to attics; city crews deploying pumps to move water away from structures; and members of the National Guard and volunteers shoring up the levee along the Lumber River that was breached during Matthew.

Even so, the researchers say Florence was a stark reminder that recovery, particularly for flood-prone communities, is a work in progress – often moving two steps forward and one step backward.

Defining ‘resilience’

NIST is the national organization dedicated to measurement science, and defining “community resilience” is a difficult, nonlinear measurement problem, van de Lindt said. Most importantly, the center’s approach is multidisciplinary, drawing on the practices of engineers and social scientists working together seamlessly.

For example, in the Lumberton study, the engineers and social scientists jointly wrote survey questions for households impacted by flooding, merging standard questionnaires from their respective fields. The surveys tracked things like structural damage, high-water lines and loss of electricity, while also assessing behavioral outcomes, like whether a family chose to stay or leave during the storm – and when they returned.

“Technical field investigations typically have engineers and social scientists working independently,” said van de Lindt, who co-led the first field team that visited Lumberton shortly after Hurricane Matthew. “But that’s not how our center works. There are no teams without both engineers and social scientists, and we all work together and learn from each other.”

Race, poverty predictors of slower recovery

The researchers’ major objectives in Lumberton included understanding how public schools cope with and respond to disruptions; seeing how private-sector businesses respond; improving damage assessment tools; collecting data on individual households’ disruption and damage; modeling long-term housing and recovery; and understanding how recovery of these systems is interdependent.

In interviewing residents and city officials, assessing damage, and documenting road and school closures, the researchers quickly confirmed that race and socioeconomic status are key predictors of longer recovery from disasters. This knowledge will help the researchers  develop models for other communities with different socioeconomic makeups.

“What we start to see in our report is that if you are Hispanic or black, your probability of dislocating from your house, at the same level of damage as a white family, is different,” van de Lindt said. These findings were primarily due to the fact that minority households were more likely to reside in houses located in Lumberton’s major flood zones. This finding aligns with a core social science observation on the intersection of physical and social vulnerability, resulting in exacerbated impacts.

Physical damage is only one part of disaster aftermath. “It’s really the human toll,” van de Lindt said. “When people are poor, or there are other demographics at play, you can see how different recovery is for them.”

Future goals

Unlike other disaster studies that focus on physical damage and then just make recommendations on how to rebuild, the center is looking at a broader set of conditions —structural, infrastructural, social, economic, and overall quality of life – over several years in a single community to better understand what makes for successful recovery and resilience, according to study co-editor Mitrani-Reiser, director of NIST’s Disaster and Failure Studies Program. “With that data and insight, we can recommend how communities across the United States can improve their resilience to natural hazards,” Mitrani-Reiser said.

The center will use what they learned from Lumberton to create a resilience model that works across different communities dealing with different circumstances. This work will culminate in a computational tool called IN-CORE that can define relative strengths and weaknesses within communities. These assessments will allow identification of better policies and decision levers – essentially, making a business case for community resilience.

Ultimately, such a tool could help predict, for example, the effects of a tornado or hurricane in terms of deaths, injuries, dollar losses, and number of years to rebuild. The center plans to release interim products throughout their five-year grant so communities can start benefiting from the researchers’ work as soon as possible.

The researchers will return to Lumberton this year.