Tornado Trackers

IEEE members work to improve the science behind predicting tornadoes

8 March 2010
tech Photo: Jeff Snyder

Tornadoes are nature’s most violent storms, capable of quickly wreaking devastation and death. Most developed countries have radar to detect tornado formation, but scientists still don’t understand all the factors that cause the storms. Because they pop up and dissipate so quickly, predicting where and when a bad one will hit more than 13 minutes ahead of time has not been possible.

To improve forecasting and increase warning times, the largest-ever tornado-measurement program, called the Verification of the Origins of Rotation in Tornadoes Experiment–2 (VORTEX2), is being conducted in the United States, which has more tornadoes than any other country—about 1200 per year. Under the right conditions, the storms form almost anywhere. They have turned up in Africa, Asia, Australia, Europe, and South America.

VORTEX2, launched last year, is designed to measure tornadoes’ characteristics during their formation and life cycle. More than 100 scientists and students from 16 U.S. universities were deployed in 40 vehicles outfitted with mobile radar systems, weather instruments, and weather balloons and their launchers. They headed for Tornado Alley, which lies in the Central Plains, roughly from Texas to the Dakotas. It’s the most active area for the storms, which surface from May to October.

The teams were on the road last year during the most active period, 10 May to 15 June, and hope to head out again this year around the same time.

Scientists first began researching the tornado life cycle in the mid-1990s, during the original VORTEX study. The results enabled the U.S. National Weather Service to provide severe-weather warnings with 13 minutes of average lead time instead of 5 and reduced the false-alarm rate to about 65 percent, a 10 percent drop. The National Science Foundation and the National Oceanic and Atmospheric Administration are funding VORTEX2.

Several IEEE members are on the VORTEX2 team, including Senior Member Stephen Frasier and Graduate Student Member Vijay Venkatesh, both from the University of Massachusetts, Amherst, and Member Peisang Tsai, from the National Center for Atmospheric Research, in Boulder, Colo.

Frasier, an electrical and computer engineering professor, is director of the university’s Microwave Remote Sensing Laboratory. Venkatesh is an EE grad student who works in the lab, and Tsai was a research fellow at the lab until July.

“The lab’s bread and butter is designing, building, and operating radar systems and analyzing their data,” Frasier says. The UMass laboratory has been collaborating with other meteorological organizations since 1993 to study tornadoes and other severe storms. Researchers there built two mobile polarimetric Doppler weather radar systems: a 9.4-gigahertz X-band microwave radar and a 95-GHz W-band millimeter-wave radar.

The two Doppler systems send into a storm differently polarized signals that reflect off water droplets and debris and are then analyzed. The W-band radar offers the finest resolution of any mobile Doppler radar system in the country because it produces a very narrow beam (0.18 degrees wide). At a 10-kilometer range, the spatial resolution is roughly 30 meters, about five times as fine as most other radars, Frasier says.

The truck-mounted Doppler radars “see” inside the storms and determine what’s going on in terms of wind speed, wind shear, raindrop density, and other conditions. They map the wind field from the very lowest to the very highest tornado levels and document the structure of multiple, sub-tornado-scale vortices, believed to cause much of the local damage.

“We can deduce tornado properties by looking at the two different polarizations, similar to looking through the lenses of polarized sunglasses,” Frasier says. “The polarization helps us discriminate among rain, hail, and debris picked up off the ground.”

The radar signature is so distinctive that computers can be programmed to recognize it, electronically shouting “Tornado!” so warnings that are more timely and precise can be issued.

Last May, the group drove the lab’s pair of radar-equipped trucks the 2600 km from Massachusetts to meet up with the rest of the VORTEX2 team at the University of Oklahoma School of Meteorology, in Norman, the country’s leading severe storm research center. They then set out to get their samples. “Last year’s trip was a great educational experience, because it brought together teams from different disciplines,” Venkatesh says. “It was a privilege to be involved.”

Although chasing storms may look scary, Tsai says it’s quite safe: “You’re chasing with the most knowledgeable meteorologists, who know the safe side of the storm to be on. The biggest concern is driving safely.”

The data collected is analyzed by researchers and graduate students at the University of Oklahoma. It’s too early for preliminary findings to emerge from last year’s work, because most of the data is still being examined. “The data will be used to determine the life cycle characteristics of tornadoes so that we better understand how they form, why they form, and how they evolve,” Frasier says. “Ultimately, the goal is to do a better job of predicting when they’re going to happen, to help save lives and mitigate damage.”

At press time, the UMass group was awaiting word on funding for this year’s twister hunt. If the funding comes through, the team plans to test out a new solid-state version of its X-band radar—it relies on a solid-state power amplifier in place of the magnetron tube used last year.

Learn More