
“You want to be semi in the path of a tornado, a couple of kilometers away,” explains Karen Kosiba (BS ’99), managing director of the FARM (flexible array of radars and mesonets) research facility at the University of Alabama in Huntsville. “There’s something called the hook, and the tornado forms at the bottom of that. So you want to be ahead of that and ready to move pretty quickly.”
Throughout her career, Kosiba, who studies the low-level wind structures that affect how tornadoes form and evolve, has witnessed more than 200 tornadoes across the country. During one close call, in El Reno, Oklahoma, in May 2013, she and her team watched darkness descend over the vast grasslands outside Oklahoma City as a massive supercell with winds up 296 miles per hour suddenly shifted direction.
Recognizing conditions “too complicated, too dark, too murky, too incomprehensible to us,” Kosiba and her team abandoned their research plans, fighting rush-hour traffic to leave the city as debris whirled across the interstate. In a frightening scene out of a Don DeLillo novel, some motorists headed the wrong way on the highway, electric transmission lines crashed to the ground, vehicles spun off the road, and eight people died, including three storm researchers.

Such cataclysmic scenes are rare due to sophisticated mobile radars, weather stations, and predictive tools, Kosiba says. By analyzing wind patterns not only close to the ground, where the strongest winds occur, “but also at 10, 20, 30 meters above that, we get a profile of what the winds are doing,” she says. Still, her research is not without occupational hazards.
Unlike hobbyist and media storm chasers who position themselves in high-visibility areas optimized for photographs, Kosiba’s teams—typically a driver, a navigator, and a radar operator—often drive through slanting rain and hail to situate themselves within a few kilometers of supercells.
Stationed in the cab of a radar-outfitted pickup called a Doppler on Wheels, they capture data on wind direction, strength, and distribution, as well as environmental surface conditions that influence tornado behavior, such as buildings and trees.
Her team’s work includes data from case studies of hundreds of tornadoes, has been published in science journals like Nature and PNAS, and is being used to inform hazard-prediction models and help architects and structural engineers design buildings that are more resilient to winds. It’s also improving forecasting and evacuation models to limit the loss of human life when powerful tornadoes, such as the one that occurred in El Reno, rip through communities.
“If you have better forecasting, even if it doesn’t directly save houses, it will save lives because people will have a better understanding of what’s heading toward them,” Kosiba says.
Kosiba, who as a child collected caterpillars, built balsa bridges, and stayed up late watching lightning storms, may never satisfy her insatiable curiosity. But her study of physics at Loyola University Chicago with professors such as Gordon Ramsey, PhD, offered her an applied scientific framework that set the stage for a thrilling career. “It’s funny, some of the things that come back to me. It’s been a while, but electromagnetics and radar are pretty big for me on a daily basis,” she says.