On 20 May 2013, at 2:56 p.m., a tornado touched down in central Oklahoma. Over the next 40 minutes, it ripped through the towns of Newcastle, Moore, and south Oklahoma City. The storm destroyed dozens of houses and cars, two farms, two elementary schools, a strip mall, and several other buildings as it killed 24 people and injured hundreds.
Climate change is known to affect many types of extreme weather, such as hurricanes, droughts, and floods. But until recently, few studies have addressed whether it will affect tornado outbreaks like the one that decimated central Oklahoma. Matthew Woods, a recent graduate of the University of Illinois at Urbana-Champaign, aimed to fill that gap with his recent research in atmospheric sciences and meteorology.
“Climate change certainly raises the ceiling for future tornadoes, in terms of strength,” Woods said. Using a modeling framework called pseudo–global warming methodology, he predicted that the frequency of warm-season tornadoes will decrease slightly in the United States, but those that do occur may be stronger. Meanwhile, the cool season is likely to see both more frequent and more intense tornadoes. Woods will share his results at AGU’s Fall Meeting 2021 during a poster session on 13 December.
Studying the Past to Predict the Future
Tornadoes are very localized, which makes them difficult for climatologists to study. Climate models are intended to describe widespread effects, and most consider only points spaced tens or hundreds of kilometers apart, whereas most tornadoes are around 50–100 meters across. “Climate models, they don’t explicitly resolve or capture storms because the storms fall between the grid points,” Woods said.
Instead, Woods turned to pseudo–global warming methodology, in which researchers model historical weather events over small geographic areas using conditions that mimic what future weather conditions could look like. Woods modeled two historical events: the 20 May 2013 Oklahoma tornado outbreak and a 10 February 2013 outbreak that took place along the U.S. Gulf Coast. Both events involved extremely strong tornadoes capable of inflicting extreme damage.
Woods adjusted conditions such as temperature, humidity, wind, and soil moisture to account for changes that are predicted to occur by the 2090s according to five different climate models, then simulated what the Oklahoma and Gulf Coast tornado outbreaks would look like in each hypothetical future scenario. Simulating the spring Oklahoma outbreak suggested that tornado occurrence will decrease by about 15% during the warm season, but tornadoes that do occur are likely to be nearly 100% stronger. Meanwhile, modeling the winter Gulf Coast outbreak told him that the probability of tornado occurrence in the cool season will increase by around 400%, with substantial increases in storm intensity.
Mateusz Taszarek, a severe thunderstorm researcher from Adam Mickiewicz University in Poland, said that pseudo–global warming methodology complements two approaches that researchers have previously used to study how severe storms (which have the potential to cause tornadoes) will change under future climate conditions. He and other researchers have examined changes in severe storm frequency over the past few decades and have also assessed the potential for future atmospheric conditions to support severe storms.
Results have recapitulated Woods’s findings, but Taszarek thinks the locations Woods examined are as important as the seasons. Previous research predicted that tornado frequency will decrease in the Great Plains but increase along the Gulf Coast. Taszarek added that strong winds brought on by the jet stream are likely to further increase tornado frequency along the Gulf Coast during the winter months, the season when the outbreak Woods modeled occurred.
Meteorologist Pieter Groenemeijer from the European Severe Storms Laboratory agrees that Woods’s approach is a valuable complement to previous work, but he and Taszarek stressed the need to examine more events. “I think this is actually probably the thing that is most important—that they would simulate more of these historical outbreaks and see if they get a consistent signal,” Groenemeijer said.
Protecting Vulnerable Populations
Taszarek is particularly interested in Woods’s model of the Gulf Coast tornado outbreak because it took place in a highly populated region with variable landscapes where it’s difficult to see storms approaching. In addition to performing more studies like Woods’s, he thinks scientists need to research communication strategies that will help the public appreciate the risks posed by storms. “I think there’s a lot of social science that needs to be done,” he said.
Woods also worries about the potential for future storms to affect population centers. “The risk in the future is going to likely increase, not only because of climate change, but just because of population growth,” he said. “We’re making the target bigger.”