After a slow start this year, tornado season made a deadly debut Wednesday night. An estimated dozen tornadoes touched down in north Texas. The damage included at least six people killed and whole houses that were suddenly yanked off their foundations to join the airborne debris.
The twisters hit three Texas counties southwest of Dallas. Witness accounts suggest the largest funnel was close to a mile wide, and scientists estimate its wind speeds reached 210 miles an hour or more, according to Kurt Van Speybroeck, a meteorologist with the National Weather Service based in Fort Worth.
The biggest tornado landed in Granbury, a town of 8,000, destroying close to 40 buildings and mobile homes. Six people were killed there, and at least 100 suffered minor to severe injuries.
The twister was classified as an EF4 using the Enhanced Fujita Scale, which depicts the strength of tornadoes by measuring their destructive force on structures like barns, strip malls and automobile showrooms. The strongest twister is an EF5, of which only eight have been documented in the United States since 2000.
By measuring damage, meteorologists estimate the twister's wind speed in three-second gusts. The Granbury funnel had potential winds of 210 to 261 mph, speeds that exceed the strongest hurricanes, which are measured by sustained wind speed.
About 1 percent of all tornadoes in the country are rated EF4, according to Greg Carbin, a meteorologist with the National Oceanic and Atmospheric Administration's Storm Prediction Center. There have been 79 of them in the last 13 years, out of 16,638 tornadoes total.
"They're fairly rare," Van Speybroeck said.
Can warning time be expanded?
Unlike with hurricanes, which can be tracked from a long way off, the art of tornado prediction is still in its infancy. It is a maddeningly tricky task, and people often have as little as 10 minutes to prepare before destruction hits.
According to a report from NOAA, residents of Joplin, Mo., had about 30 minutes after the initial siren alert to seek shelter before a tornado touched down May 22, 2011, killing 158. It was the deadliest tornado in 58 years.
This week, a Colorado-based field project began to collect data that might help meteorologists develop a better model of when and where storms and twisters are likely to strike. They are aiming to develop precise warnings six to 24 hours in advance.
As part of the project, called the Mesoscale Predictability Experiment, or MPEX, a team of scientists will fly in a small jet over the central U.S. region commonly known as "Tornado Alley," dropping parachute-borne weather devices to gain a more detailed picture of what triggers thunderstorms and the tornadoes that often result.
Gaps in the current system
The current average lead time for tornado warnings is only 13 minutes, according to NOAA. Part of the challenge is that the National Weather Service's network of 102 weather balloons is spread too far apart to pick up the small-scale data that might indicate conditions leading to storms.
Current satellite imagery technology is also unable to collect the detailed, quantifiable information on temperatures and pressures at different altitudes that could help computer models produce accurate predictions.
"A lot of these features that we know have some effect on the triggering of thunderstorms ... are basically going unobserved by the operational network of instruments," said Jeff Trapp, a meteorology professor at Purdue University's Department of Earth, Atmospheric and Planetary Sciences.
"Even severe storms [at a] 100-kilometer scale are also going to be missed in the current observational network," said Chungu Lu, a meteorologist with the National Science Foundation who is overseeing the project.
"MPEX scientists are trying to capture the large-scale, upper-level atmospheric disturbances, like pressure changes, and use this information in the forecast model to better predict the initiation of storms," Lu added.
The project involves a large team of scientists, representing the National Center for Atmospheric Research; Colorado State University; the University at Albany at the State University of New York; Purdue; the University of Wisconsin, Milwaukee; and NOAA's National Severe Storms Laboratory.
Looking 'upstream' in Tornado Alley
MPEX data collection and modeling will take place over the next month, during the height of tornado season. Early in the morning, before commercial air traffic picks up, scientists will lift off from an airport in Broomfield, Colo., in a Gulfstream V jet, funded by the National Science Foundation, cruising about 7 miles above the earth over the southern portion of Tornado Alley.
The goal is to get a more complete portrait of conditions "upstream" of the thunderstorms, which usually form later in the day. At regular intervals during their six-hour, 3,000-mile trips, researchers will drop small instruments attached to parachutes out of the plane. These instruments, called minisondes, will collect temperature, moisture and wind data every quarter of a second as they fall to the earth. The jet itself is also outfitted with data-collection technology.
During the afternoon, other researchers will release weather balloons near burgeoning storms to learn how the surrounding atmosphere is affected.
"We know that once thunderstorms exist that they disturb their environment; this can make the rest of the atmosphere more chaotic," Trapp said. "We don't know the magnitude or the extent of this disturbance."
The scientists will then experiment with different ways to assimilate this data into computer models to learn which prediction methods are most accurate. The hope is to develop an early warning system similar to what meteorologists currently use for hurricane projections, Trapp said, or a "cone of probability of where a storm will form and what its path will be" in the crucial hours before the storm hits.
If successful, the MPEX (pronounced "em-pex") could save lives as severe storms become an increasingly persistent threat to Americans. According to Munich Re, a global reinsurance agency, severe thunderstorm losses in the United States have "increased significantly" since the 1980s. Last year, thunderstorms and tornadoes resulted in around $7 billion in insured losses (ClimateWire, Jan. 4).