Research on the formation of twisters has taken a positive turn: Scientists have discovered some of the key differences between storm systems that spawn tornadoes and those that don’t.
Two factors seem particularly crucial: whether a storm has high wind shear, or change in wind speed and direction with height, and high “helicity,” a measure of how much the storm tends to rotate. A team of meteorologists, led by Andrew Mercer of Mississippi State University, reports the finding in an upcoming Monthly Weather Review.
“Now we can say, Yes, this is what a tornadic outbreak looks like, generally speaking, versus a nontornadic outbreak,” Mercer says.
Weather forecasters at the federal Storm Prediction Center in Norman, Okla., are already using the new information to help guide their predictions. One day, Mercer hopes, meteorologists might be able to look at an approaching storm system and determine whether it will produce tornadoes.
Until now, most research on tornado prediction has focused on storms known to have spawned twisters, and trying to identify why. Plenty of wicked-looking storm systems never create actual tornadoes — although they can cause harm through high winds, hail or other hazards.
Several years ago, a team based at the University of Oklahoma, where Mercer was a graduate student, decided to study a broader range of storms. In a series of papers, the researchers scrutinized large-scale storms, such as low-pressure systems some 1,000 kilometers across, to see whether those ended up producing tornado outbreaks. In a 2009 paper analyzing 50 tornadic and 50 nontornadic storms, the scientists reported that it might be possible to tell the differences between the two 24 hours before tornadoes formed.
In the new study, Mercer and his colleagues did a deeper statistical analysis of the same storms. The researchers identified four basic patterns observed in each type of storm, then compared those with computer simulations to see whether tornado-spawning tempests could be predicted.
One meteorological parameter, a measure of the storms’ instability, didn’t differ much between tornadic and nontornadic storms, says Mercer. But wind shear and helicity did. In other words, a storm can look unstable and yet not spawn tornadoes if it doesn’t have particular patterns of wind shear and helicity.
Meteorologists across the country launch weather balloons twice daily, which provide the necessary data on winds. But translating those observations into useful predictions remains a challenge, says Chuck Doswell, a semi-retired meteorologist at the Cooperative Institute for Mesoscale Meteorological Studies at the University of Oklahoma in Norman.
False alarms are a particular problem, as the new technique tends to flag too many storm systems as tornadic. “There are times when it looks like it’s going to be more significant than it is going to be,” Doswell says.
In 2006, Doswell published a paper cataloging 6,000 tornadic and nontornadic storms. He, Mercer and others are now looking to expand the detailed analysis from the current 100 storms to all 6,000. Studying more storms should help the scientists improve their ability to predict tornadoes.
“We’re trying to go after that portion of the severe weather occurrences that we think we have a reasonable chance of being able to forecast,” Doswell says.