LIVE VIDEO: Live Newscast    Watch
 

Scientists a step closer to predicting tornadoes

10:35 PM, Feb 14, 2012   |    comments
Screenshot from TornadoVideos.net
  • Share
  • Print
  • - A A A +

For decades, meteorologists have been able to forecast the severity of hurricane seasons several months ahead of time. Yet forecasting the likelihood of a bad tornado season has proved a far greater challenge.

Now, new research from scientists at Columbia University's International Research Institute for Climate and Society could eventually lead to the first seasonal tornado outlooks.

"Understanding how climate shapes tornado activity makes forecasts and projections possible, and allows us to look into the past and understand what happened," said Michael Tippett, lead author of a study in February's journal of Geophysical Research Letters.

The need for such data is reinforced by the still-fresh memory of 550 Americans killed by tornadoes last year - coupled with an unusually violent January for twisters.

In the new study, Tippett and his team looked at 30 years of past climate data. They used computer models to determine that the two weather factors most tied to active tornado months and seasons were heavy rain from thunderstorms and extreme wind shear (wind blowing from different directions at different layers of the atmosphere).

"If, in March, we can predict average thunderstorm rainfall and wind shear for April, then we can infer April tornado activity," Tippett says.

The method worked for each month except for September and October, and it worked best in June.

This is the first time a forecast of up to a month in advance has been demonstrated, he says.

"A connection between La Niña and spring tornado activity is often mentioned," Tippett says, "but such a connection really has not been demonstrated in the historical data and hasn't been shown to provide a basis for a skillful tornado activity forecast.

"Our work bridges the gap between what the current technology is capable of forecasting (large-scale monthly averages of rainfall and winds) and tornado activity, which the current technology cannot capture," he notes.

The research isn't ready for prime time yet, however, so no official forecast will be made for the upcoming season using these methods.

"This is a useful first step," says Harold Brooks of the National Oceanic and Atmospheric Administration, who was not involved in the study. He says it will be helpful to know, for example, that sometime in the last week of April, conditions will be favorable for lots of tornadoes in the eastern USA.

With greater lead time, a state emergency planner "could be better prepared with generators and supplies," Brooks says.

By Doyle Rice, USA TODAY

Most Watched Videos