Tornadoes, wildfires, tropical cyclones and sea level rise are all on the list of dangers made worse by climate change. Answering questions about how severe these disasters will be, how soon and how frequently they will occur, and what areas are likely to sustain the most harm is of vital importance. Responding to this need, Columbia climate scientists at the National Center for Disaster Preparedness (NCDP) have led the U.S. Natural Hazards Climate Change Projections project, which can provide officials, relief organizations, city developers and families with the information they need to plan for the worst.
This initiative brought together several public and academic researchers to develop a novel, interactive dataset to track and predict the occurrence of climate change–fueled extreme events on a county level, through the middle and end of the century. The project builds off of the Natural Hazards Index v2.0, a present-day look at 14 different hazard types.
The team identified significant future climate change–driven hazards, including an escalating wildfire threat to San Diego and Washington State’s Yakima County. As high as the risk is now, the models point to it getting worse. The Dakotas, which currently do not have a high incidence of wildfires, should prepare for an increase. Louisiana will see fewer wildfires in the future, but underlying data suggests that’s because there’s going to be more precipitation, and more precipitation could mean more flooding.
These are just a few of the predictions modeled as part of the project.
“We show what the baseline hazard is and then what the percent change is going to be in the mid- and end-century,” said Jonathan Sury, senior staff associate at NCDP and the lead researcher on the project. “How much worse are things going to get? For example, when you look at the raw numbers for wildfire, it’s not much, but when you interpret the data, even if you go from 0.2% to 0.4%, you’re doubling the risk. And if you don’t have enough firefighting resources such as fire trucks, or if you don’t have enough homes that have been prepared and adapted to the potential for wildfire, the outcomes could be catastrophic.”
These new datasets are comprised of maps, graphs and an introductory narrative that helps the user learn about the hazard, how the hazard’s profile may change under the influence of climate change and additional information about the dataset itself, which may be of particular interest to academics and researchers.
The project identifies another notable and alarming red flag: an increased tornado risk for the East Coast.
“We knew trends were moving eastward for tornadoes, and now we can actually show the big shift and the increase in the number of potential tornadoes from Tornado Alley over out east and then up on the Eastern Seaboard,” said Sury. “Our previous tornado dataset was just looking at historical data. Our extreme heat layer was looking at trends over the past 40 years. This is looking at how things are changing under different climate conditions and different time periods between now and the future.”
“While the challenges of climate change and disasters may seem overwhelming, it is also important to note that we have more knowledge at our disposal than ever before,” said Jeffrey Schlegelmilch, NCDP director and professor of professional practice of climate at the Columbia Climate School. “By working across sectors and engaging partnerships like this, we can provide data that is empirically rigorous and immediately relevant to stakeholders outside of academia. This helps to foster better decisions, better investments and better resilience for our communities.”
Recognizing the risk and value of predictive data, the U.S. Federal Emergency Management Administration was in the process of putting together a tool to track how future climate change factors could intensify natural disasters. The Trump administration has put those plans on indefinite hold while also deleting many governmental agency tracking websites.
“We are in a time where the hazards we face are increasingly impacting the lives and livelihoods of our communities,” said Schlegelmilch. “At the same time, we are seeing datasets being taken down, right when we need them most.”
The disappearing dataset policy might pose challenges. However, according to Sury, the new predictive map aggregates existing datasets, incorporates the best, most currently available data, and leverages the most advanced technological methods to produce novel research that predicts— over time—the way climate change will intensify and escalate the risk of the four types of hazards: tornadoes, tropical cyclones, wildfires and sea level rise.
“Because the very climate these hazards are emerging from is changing, we can’t rely solely on past experiences and historical data to guide our understanding of risk”
Traditionally, extreme event preparedness science relied on historical event data to anticipate and model future trends. This project shifts the paradigm.
“Because the very climate these hazards are emerging from is changing, we can’t rely solely on past experiences and historical data to guide our understanding of risk, and how we make investments to help reduce those risks,” said Schlegelmilch.
The new map and underlying dataset, which were released on April 21, is the third hazards data product based off a beta version of the Natural Hazards Index, first released in 2016. This tool was developed to supplement the Preparedness Wizard, a clickable resource for building household emergency plans. In 2023, Sury and his team developed the U.S. Natural Hazards Index v2.0 to update and better visualize natural hazard data for 14 hazard types in the U.S. and Puerto Rico.
In this latest dataset, the research team built on deep existing research and analysis, tracking the trajectory of sea level rise and deriving the tropical cyclone data from a combination of existing, ongoing and new research. Here, too, predictions show escalating risk.
“Tropical cyclones cause enormous damage in the United States and risks are expected to grow with climate change,” said Simona Meiler, postdoctoral researcher in weather and climate risk science at Stanford University, who co-developed and supplied the modeled dataset to NCDP.
“The climate system’s response to anthropogenic forcing is not homogeneous and remains uncertain; changes in near-future hurricanes may vary geographically,” said Chia-Ying Lee, associate research professor at Columbia’s Lamont-Doherty Earth Observatory. “We need to look at changes in regional risk, which this map provides. In this dataset, it is assumed that storm frequency increases with global warming.”
“By making this data available in an accessible format, we aim to support not just climate scientists and risk modelers, but also urban planners, emergency managers and researchers in other fields, like economics or human mobility. The goal is to make future tropical cyclone risks more visible and actionable, particularly in the most vulnerable regions,” said Meiler.
As daunting as these predictions are, researchers underscore that knowledge is society’s best possible defense. The team has made the data open-source and available to everyone to position the research to do the most good.
“These will be most helpful for anyone with a long-term interest in the future,” said Schlegelmilch. “For individuals, it may involve looking at the kinds of emergencies that you may face, which can impact investments you make to your home, or what kinds of hazards you plan for in your own personal preparedness.”
“Investors will want to understand exposure to hazards, whether for physical assets or even economies exposed to increasing hazards. This in turn can help to incentivize resilience investments to offset these risks. Emergency managers and city planners can use these to help look into how hazards may change, which will impact disaster plans specifically and development plans more broadly,” Schlegelmilch said, adding that “these are just a few of the many, many potential use cases across the country.”
