text-only page produced automatically by LIFT Text
Transcoder Skip all navigation and go to page contentSkip top navigation and go to directorate navigationSkip top navigation and go to page navigation
National Science Foundation
design element
News From the Field
For the News Media
Special Reports
Research Overviews
NSF-Wide Investments
Speeches & Lectures
NSF Current Newsletter
Multimedia Gallery
News Archive
Press Releases
Media Advisories
News Tips
Press Statements
Speech Archives
Frontiers Archives

Tracking Tornadoes, Nature's Most Powerful Winds

February 1997

Late in the afternoon on the lonely roads of the Texas panhandle, an armada of vans, trucks and airplanes bristling with scientific instruments races toward a rendezvous with the weather. It is June 2, 1995, and a two-year storm-chasing project, VORTEX, is about to yield one of its biggest successes.

As the storm experts converge on the little Texas town of Dimmit, one of their vehicles, a truck topped by a radar with a six-foot diameter antenna, called DOW, trains its electromagnetic beam on a violent twister stripping away highway asphalt and tossing aside automobiles as if they were tumbleweeds. DOW gets within two miles of the tornado and records detailed images, including spiraling bands of clouds and concentric rings of debris.

Six days later the crew gets another chance: ELDORA, a radar aircraft, tracks an even stronger, half-mile-wide tornado near Kellerville, Texas. Flying at 1,000 feet, scientists on board watch the twister rip up dirt and asphalt just four miles away as their radar charts unprecedented cross sections of the entire storm, including a tube-like vortex reaching high into the thundercloud.

These and other storm chases are more than just adventures in bravery. They are serious scientific experiments made possible by the use of mobile Doppler radars. And, they are cutting through the thick veil of clouds that has masked the origins of nature's most powerful winds.

Thanks to the new mobile radars, storm chasers today have detailed evidence of the inner workings of tornadoes and the thunderstorms that spawn them. The radar imagery spurs computer simulations, map-room hypotheses and even more chases. One of the goals: to solve a mystery that has stumped meteorologists for decades--why some, but not all, severe thunderstorms give rise to deadly twisters.

Such a breakthrough was one of the hopes for VORTEX, the Verification of the Origins of Rotation in Tornadoes Experiment, funded in part by NSF. The chase phase partnered scientists from the National Oceanic and Atmospheric Administration's National Severe Storms Laboratory (NSSL), the National Center for Atmospheric Research (NCAR) and universities to record the genesis of tornadoes and other aspects of thunderstorms.

Meteorologists already know that the most powerful--and most destructive--thunderstorms are the best organized. These thunderstorms organize themselves into one or two updrafts five to 10 miles across, lifting air to 50,000 feet at speeds of up to 150 mph. The wind environment causes the updraft tower to rotate counterclockwise when viewed from above. As the tower tilts, precipitation falls outside its edges, producing evaporation-cooled downdrafts that wrap themselves around the updraft. Somehow, sometimes, rotation near the ground develops into a tornado.

"The big issue that's still unresolved with tornadic storms is at the lower levels," says Stephan P. Nelson, director of NSF's Mesoscale Dynamic Meteorology Program, which funds DOW, ELDORA, VORTEX and other weather research. "We have a good, general understanding of the structure of storms and how they form, and good information on the initiation of rotation in mid to upper levels of the storms. But what finally causes a tornado to touch down on the ground is still a mystery."

Developing Doppler Radar

In the last 25 years, Doppler images have confirmed important mid-level storm features, such as the initiation of the highly organized rotation in severe thunderstorms at or above 14,000 feet in altitude. With this rotating central updraft, called a mesocyclone, a thunderstorm has a 20 percent chance of producing tornadoes.

Until about 10 years ago, the critical low-level circulations were largely out of reach of Doppler radars, because the radars were too big to haul on a storm chase. Yet they are the best tools for observing whirlwinds.

Like conventional radar, Doppler radar emits a beam of electromagnetic waves and records the response bounced back from airborne water, ice or debris. Doppler radar can also discern movement by detecting a shift in frequency--the Doppler effect--caused by movement towards or away from the radar.

In 1987, University of Oklahoma meteorologist Howie Bluestein and his team began chasing tornadoes with a breakthrough portable Doppler radar designed by engineers at Los Alamos National Laboratory. Bluestein's crew caught up with a ferocious storm near Red Rock, Okla., in April 1991, clocking a record 286-mph wind. Bluestein is now using a radar engineered at the University of Massachusetts. It gathers data with high enough resolution to discern the extreme winds inside a tornado.

DOW, the Doppler-on-Wheels, was built in 1995 by Josh Wurman and Jerry Straka of the University of Oklahoma with support of NCAR and NSSL. It has a slightly wider beam than the University of Massachusetts radar and can penetrate further into the dense water and ice of a thunderstorm, although with slightly less detail. Next spring, Wurman will use two more powerful DOWs for 3-D imaging.

While these ground-based radars effectively show features of the tornadoes, ELDORA, or Electra Doppler Radar, reveals the nature of the broader thunderstorm. The aircraft has a new Doppler radar developed jointly by NCAR and French scientists. Two spinning tail antennae scan overlapping helixes, yielding dual-Doppler imagery capable of mapping 3-D wind fields.

ELDORA debuted in 1993 but has since been upgraded to scan small features. With it, Bluestein, UCLA tornado expert Roger Wakimoto, and others have reaped highly detailed cross-sections of thunderstorms. Observations of extreme updrafts, microbursts and turbulence at the edge of outflowing air are among ELDORA's contributions to thunderstorm knowledge.

"The radar images show details of flow structures on a scale we've never seen," says Morris Weisman, a thunderstorm computer modeler at NCAR.

The most sophisticated thunderstorm simulations are run on state-of-the-art supercomputers at NCAR, the University of Illinois Supercomputing Center and other similar facilities. They can barely depict large tornadoes, Weisman says.

Nevertheless, current hypotheses about low-level rotation and tornado formation are derived in part from computer simulations.

Building Hypotheses

In one hypothesis, upper-level rotation in the storm eventually reaches the ground to form the tornado. In another, existing rotation at the ground stretches upward and accelerates like a spinning ice skater drawing in her arms.

Still other hypotheses hold that the thunderstorm's cooled downdraft has a spin that gets pulled up into the bottom of its updraft, initiating rotation near the ground.

DOW showed that the Dimmit, Texas, tornado had a relatively clear eye that penetrated to within 300 feet of the ground and contained a strong downdraft blowing at 50 mph. Wurman speculates that debris was centrifuged out to the walls of the vortex and was forced out by the downdraft in the center.

DOW also showed that the tornado's strongest winds were about 700 feet up, as expected. Surprisingly, the VORTEX chasers did not observe temperature contrasts on the ground. Scientists had thought these contrasts would be a mechanism for promoting the low-level rotation leading to tornado formation. Despite these puzzling results, Wakimoto says ELDORA imagery from one VORTEX storm shows a broad downdraft developing in the mesocyclone--an event modelers showed would be initiated by low-level rotation.

Solving Mysteries

Such continuing mysteries do not daunt Weisman. "VORTEX observations should show if our results are reasonable or just a fantasy model," he says. Meanwhile, as modelers are busy simulating events, testing hypotheses and settling controversies, the mobile radars still have plenty of data to gather, even after tornado-chasing season is over.

This winter, for instance, ELDORA has been participating in an international project studying North Atlantic storms. And last September, Wurman demonstrated DOW's capabilities in studying a much bigger prey--Hurricane Fran.

Wurman's team can operate DOW safely a mile or more away from a tornado, but the hurricane chase forced them to sit right in the middle of Fran. Gusts neared 100 mph, but a new eight-foot radar dish withstood the 12-hour onslaught, providing provocative imagery of the inner workings of a hurricane.

With the help of these pioneering tornado hunters and their radars, hurricane experts may soon find themselves solving some mysteries of their own. As we increase our understanding of hurricanes and tornadoes, we are better able to predict their behavior, and to get out of their way.



If you should happen to find a pair of ruby slippers in a cornfield in Kansas, call meteorologist John Snow, dean of the College of Geosciences at the University of Oklahoma.

Snow and his meteorology students run the Tornado Debris Project, a three-year, NSF-funded program to collect and trace items picked up by tornadoes. The researchers map where the debris falls to the ground and trace it to its origin.

"Every tornado that strikes an urban area that has identifiable debris has produced a plume, quite independent of intensity," Snow says. He hopes eventually to model the debris plumes in computer thunderstorm simulations.

Snow says they have received hundreds of calls from people who found canceled checks, photographs, cassettes, floppy discs and scraps of clothing in their yards and fields. He cautions, however, that not every tornado leaves a trail. Tornadoes in sparsely populated areas often loft natural debris that cannot be traced.

So far, the team has traced items from 25 storms in three years. The results have disproved earlier theories that only the strongest tornadoes loft debris significant distances. The research may ultimately have applications in forecasting hazardous material fallout from tornadoes.

Two canceled checks are among the best traveled of the debris Snow's group has tallied--perhaps because they are easily picked up from torn-up attics and are easily traced. One check flew 230 miles in a Nebraska tornado. Another was picked up in Wisconsin and carried 125 miles away, across Lake Michigan. A much heavier item, a man's jacket, was carried 20 miles in another storm.

The debris tends to be found dry and in good condition. Snow says this indicates it is carried in very strong updrafts that don't have time to condense large water droplets. The debris may be borne aloft for several hours--and miles--after the tornado dissipates. So, if you find debris that might have fallen out of the sky, call the Tornado Debris hot line at 1-800-3-DEBRIS. You never know what might have taken flight in tornado season.


Return to February 1997 Frontiers home page   Other Contents of This Issue
Visit Other Frontiers Issues page   Other Frontiers Issues
Visit Other NSF Publications page   Other NSF Publications
Visit Office of Legislative and Public Affairs page   Office of Legislative and Public Affairs


Email this pagePrint this page
Back to Top of page