Severe storm prediction model could aid first responders

DEKALB, Ill. — What if first responders and emergency management personnel could know about severe storms, like tornadoes and hail storms, weeks in advance?

A professor of meteorology at Northern Illinois University is developing a model that could do just that and recently received funding from the National Science Foundation.

“I can’t say a tornado is going to run through Rochelle, Illinois, four weeks from now. But we can say the potential for a tornado outbreak in northern Illinois is high four weeks from now,” said Dr. Trent Ford, Illinois state climatologist.

“That puts emergency managers on alert to know that around these four or five days, four weeks from now, we are looking at a setup that could be conducive to having multiple tornadoes break out. That is beneficial to make sure we have resources available when we see those sorts of severe weather,” he said.

Ford was talking about the work of Victor Gensini, professor of meteorology at NIU and deputy director of NIU’s Center for Research Computing and Data.

In March, the National Science Foundation awarded Gensini a $475,000 grant over three years to extend research into development of those long-range forecasts that could predict severe storms, like tornadoes and hailstorms.

In 2018, Gensini led a study to identify a method to predict the likelihood of hailstorms. His work “aims to identify atmospheric conditions that heighten the probability of severe U.S. weather events two weeks to two months before they occur,” according to NIU.

NIU said the forecasting project that received the NSF funding will involve “study of statistical associations between intra-seasonal atmospheric modes that can lead to severe weather and creation of conceptual models to explain the physical mechanisms. Artificial intelligence and machine learning will be used to help recognize these statistical relationships.”

“At the end of the day, we’re trying to push the envelope of what’s possible for severe weather predictions,” Gensini said, in the NIU news release announcing the grant.

Ford said the NIU work points to how far weather forecasting and modeling has come in recent years.

“The models and forecasting have advanced a tremendous amount over the last 15 years. When we look at the retrospective, after we get through a week of temperatures, and say, ‘What did it look like?’ We are seeing very low errors on those temperatures. To be able to forecast a temperature within just a couple of degrees, seven days out, is amazing, honestly, and we take it for granted,” he said.