Figuring out when a tornado will strike can help threatened communities brace for the worst, but according to Matt Parker, associate professor of marine, earth, and atmospheric sciences, the current accuracy of forecasts and warnings can lead to a “wolf-cry” syndrome, working against meteorological credibility.
Parker is working to reduce this margin of error to better understand how tornadoes tick to improve forecasts.
“The real goal is to issue warnings with increased lead-times and without a lot of false alarms,” Parker said. “What we don’t understand is why three-quarters of all rotating thunderstorms don’t succeed in making tornadoes.”
Parker and 100 other researchers from 40 different universities participated in project VORTEX2 to collect data from 50 thunderstorms that contained all the conditions that induce tornadoes. The project, the largest study in its field conducted, extended between the summers of 2009 and 2010 and Parker traveled to Colorado to monitor potential storms.
According to Parker rotating thunderstorms, called supercells, can contain the ingredients of tornadoes. Strong vertical wind shear, the change in wind speed and direction over a short distance in height, serves as a condition that causes storm rotation. Parker also cites storms with warm downdrafts of air and strong upward motions induce twisters.
But even with all the ingredients, Parker said the combination isn’t enough for an assertive forecast.
“We are not very good at separating the hits from the misses,” Parker said. “When you have a lot of false alarms, the public isn’t as responsive to the warnings and they may not always take the necessary steps of action. But that’s not anyone’s fault. The forecasters are operating that the state of our understanding. It’s just that our understanding has a lot of gaps.”
According to Parker, the information of the 50 storm systems will keep his research team busy for the next five to ten years.
“To take teams out into the field and to operate the equipment is extremely expensive, and we don’t have the funding to do that all the time,” Parker said. “Right now, the way to do that is just for students and researchers to slog down and work before we can draw conclusions.”
The conclusions may keep tornado-prone regions in the future safe, Parker said, despite the time it will take to filter the data.
Currently, a tornado forecast’s best prediction is one to two days, according to Parker. Tornado watches only extend one to two hours before a tornado may form, while the average tornado warning, which is issued when the threat is immediate, is between 13 to 15 minutes before. Parker said this is good, but could be better.
“If you’re alone, 13 minutes may be enough to run down to your basement, but if you run a nursing home or a school, 13 minutes is not a lot of time at all.”
Before this lead-time in forecasts can improve, Parker and his colleagues have quite a bit of work ahead. But for him, it isn’t drudgery.
“Since I was a kid I was interested in severe weather,” Parker said. “Going back to fifth or sixth grade, I just loved storms. Fortunately, I’ve had enough things in a row to work out in my career that I am able to pursue severe weather and how it works.”
