Flooding is the biggest natural hazard worldwide in terms of both lives lost and property damage. In the U.S., the average cost of a flood event is $4.6 billion, and 21.8 million homes and businesses are situated in the pathway of a potential flood. Aside from the financial costs, there’s risk of loss of life and displacement, and that risk is growing as climate change increases the intensity and impact of each event.
Understanding how people prepare for and respond to flood events is important for managing risks associated with flooding and assigning insurance protections against it to manage the cost of rebuilding. This is one focus of a research grant recently awarded to a team of researchers led by Ethan Yang, Professor Civil and Environmental Engineering at Lehigh University by the National Science Foundation (NSF) and the Japan Science and Technology Agency (JST). The research, which will take place in the U.S. and Japan, will study flood-prone areas in each country, with a focus on human-centered data. Along with Yang, the other researchers on the U.S. team include David Casagrande, Professor Sociology and Anthropology, Lehigh University; Alka Sapat, School of Public Administration, Florida Atlantic University; and in Japan, the researchers are Tomohiro Tanaka, Kyoto University; Kensuke Otsuyama, University of Tokyo; and Kei Ishida, Kumamoto University.
The three-year, $1 million award (split evenly between both teams), will allow Yang and his team to study flood-prone areas in their respective countries. Because it’s rare that a federal funding agency like the NSF wants to spend its money on other countries’ research, Yang says, this is a great opportunity for trans-national collaboration: “If you want to do an international collaboration, both countries need to agree, OK, we're going to fund a team inside our own country, but you guys should collaborate with each other.” And that’s what’s happening here: funded equally by both nations, each team will pursue the same work with the same proposal. Such a two-country partnership can help build synergies, share capabilities, and make greater research strides than researchers in one nation alone.
This research will focus on two river basins, the Kuma River in the Kumamoto Prefecture in southwest Japan and the Passaic River in northern New Jersey. The two are comparable, with similar geographies in terms of flow length and drainage areas; both have long histories of fluvial (riverine) and pluvial (flash) flooding; each has had previous government investment in flood risk mitigation; and the researchers are familiar with each river.
The goal of the research, as the proposal says, is to enhance our understanding of comprehensive flood risk management. This will be done by developing a human-centered, multi-scale, multi-agent system-of-systems knowledge platform, or computer model, that can simultaneously analyze natural and human systems that can simulate impacts of and recovery from flooding hazards. Most of the research will take the form of data collection and computer modeling, with several information-sharing workshops throughout the duration of the grant.
The researchers will use the catastrophe modeling concept, a probabilistic computer modeling process that allows researchers to evaluate events based on data from various factors. Lehigh’s Catastrophe Modeling Center was founded in 2021 by an interdisciplinary team at Lehigh that included Paolo Bocchini, Civil and Environmental Engineering, and Brian Davison, Computer Science and Engineering, via an internal grant; it has recently been promoted to University Center status. The process is used by insurance companies to calculate risk; as with many climate-change-impacted natural disaster events, it is increasingly difficult to rely on past events because such events are becoming more intense. Catastrophe modeling allows for better prediction of risk and potential cost.
Catastrophe modeling is commonly used with flooding scenarios. Flood catastrophe modeling, for example, might use water depth, amount of rainfall, duration, and other factors to predict where a flood might occur and what areas will be affected by it. Other systems, like government policies or insurance criteria, can also be incorporated into the model so that all the systems are factored into a single model. (This is the system of systems approach mentioned in the proposal’s goal.) This allows researchers to see how a change in one system might cause changes throughout the entire ecosystem; for example, a change to a government policy will change a resident’s choice about moving away from a flood-prone area.
One novel aspect of this study is its use of human-centered data. As the JST says, “at the core of disaster resilience is the human factor, including the way human behavior and decision-making influences disaster vulnerabilities, mitigation, and response in complex and sometimes adverse ways.” To compile this data, researchers in both countries will co-develop surveys that will be sent to residents in both flood areas to uncover information such as flood-risk perceptions, flood planning actions, generational place attachment, income, and other data points. Minoritized and non-minoritized groups will be analyzed separately so that researchers can study their responses independent of each other. This is a novel element of the study, because minoritized groups, such as the elderly, poor, disabled or racially stigmatized groups, have fewer resources and less resiliency. Yang uses as an example two neighborhoods that have experienced a flood: the wealthy neighborhood might sustain $3 million in damage, while a poor community’s damages are $1 million. Comparing the absolute numbers here, he says, doesn’t indicate the severity or even the numbers of people affected. The $3 million of damages might be to one or two homes, while the $1 million might be to ten homes or a business. Gathering more nuanced information such as annual incomes can yield a totally different result and different pictures of the effects. “That’s basically the fundamental concept that we want to explore in our project,” Yang says. Doing this kind of survey also “gives us an opportunity to compare different flood related policies,” he says.
The award is rooted in the Sendai Framework, a 2015 United Nations initiative resulting from the Tohoku earthquake that led to the Fukushima nuclear accident. The Framework calls for countries to understand disaster risks, to govern in ways that reduce the risk, invest in strategies that create resilience, and prepare effectively for natural hazard events. All of this reflects, Yang says, Japan’s determination to learn from the past. “They established this framework in order to prevent or to mitigate the impact or the damage that happened from the next natural hazard, whether that's earthquake, tsunami, flood, hurricane, whatever,” he says, noting that Japan has a long history of learning from prior experience with disasters.
In the long run, that’s the goal with Yang’s research. The modeling approach should be adaptable to other countries and communities after the human-centered data specific to that location is gathered via survey. This work can help communities prepare for flooding and recover from them, creating greater resilience.