National Geographic Daily News
Photo of storm chaser Tim Samaras

Storm chaser Tim Samaras (here in Palmer Divide, Colorado) pioneered new ways of collecting tornado data.

Photograph by Carsten Peter, National Geographic

Samantha Larson

for National Geographic

Published October 24, 2013

It was a storm of epic proportions: The tornado that hit El Reno, Oklahoma, was as wide as Manhattan, spun off subvortices as fast as NASCAR drivers circle a track, and had some of the strongest winds ever measured. It took the lives of three highly experienced storm chasers. Yet it's going down in the record books as a mundane EF3.

The reason: The May 31 storm just didn't do enough damage to achieve a higher Enhanced Fujita (EF) rating. Now weather scientists are asking if the rating system needs an overhaul in the age of mobile Doppler radar and other sophisticated tracking techniques, and some are pressing for a new rating formula that would include measurements of maximum wind speeds.

Scale Based on Damage

In 1971, when Ted Fujita introduced the original Fujita (F) scale, it wasn't possible to measure a tornado's winds while they were happening. So he proposed creating after-the-event surveys of a tornado's path and correlating storm damage with the intensity of the wind that created it. Was the surface peeled off a home's roof? That's F1 damage, probably from 73- to 112-mile-an-hour winds. Or was the house totally leveled? That's at least an F4, with winds ranging from 207 to 260 miles an hour.

The Fujita Scale

Light Damage
0-72 MPH
Branches broken off trees; sign boards damaged
Moderate Damage
73-112 MPH
Surface peeled off roofs; moving cars blown off roads
Considerable Damage
113-157 MPH
Mobile homes demolished; cars lifted off ground
Severe Damage
158-206 MPH
Trains overturned; roofs torn off houses
Devastating Damage
207-260 MPH
Houses leveled; cars thrown, large missiles of debris
Incredible Damage
261-318 MPH
Houses swept away; bark stripped from trees

(Wind speeds are estimates)

Fujita's scale was an important tool for scientists, enabling them to keep a running database of tornado intensities. But it was far from a perfect system. "This was just a crude estimate," explains meteorologist Howie Bluestein of the University of Oklahoma. It did not, for example, account for key factors such as structural variations in different types of building.

This was just a crude estimate.

Seeing the need for a reassessment of the rating in the early 2000s, the Wind Science and Engineering Research Center at Texas Tech University brought together a forum of weather experts to give new definition to the scale. The forum added more "damage indicators" (DIs) to Fujita's system, choosing 28 objects, such as barns and hardwood trees, to include in surveys. Then six experts estimated the wind speeds for each DI, ranging from light damage to total demolition, and created a damage checklist to aid surveyors in their postmortem tornado tours. The Enhanced Fujita scale was born.

Photo of the El Reno Oklahoma tornado on May 31, 2013.
This storm spawned the El Reno tornado, which was rated an EF3 despite its very high wind speeds.
Photograph by Chris Machian, The World-Herald/AP

The meteorological community hailed the enhanced scale as an improvement, and the National Weather Service (NWS) began using it in 2007. But one central weakness remained: There was little hard data on how much damage a particular wind speed can cause. The correlation between wind speed and damage was based on "six very smart guesses, probably the best that we had. But they were still guesses," says meteorologist Josh Wurman, a member of the Fujita scale forum.

... very smart guesses, probably the best that we had. But they were still guesses.

The EF scale is especially inexact for the many tornadoes that occur over open country, as even strong winds wouldn't have much impact on, say, a wheat field. So using damage as a sole measure can be limiting.

Debate Over the El Reno Tornado

When NWS sent a team of four to do the El Reno damage survey, they unanimously found it to be an EF3. However, they also acknowledged that much of the tornado's path was over empty fields. So it wasn't a surprise to Tim Marshall, who was on the team, when he later heard that Doppler radar had measured winds in the EF5 range.

Meteorologist Jeff Snyder (of Bluestein's lab) and Wurman independently shared their radar data with the NWS; both sets of data showed winds near 300 miles per hour. The local NWS in Norman, Oklahoma, decided to incorporate that information and changed its EF rating from a 3 to a 5. This decision was widely supported: "If the new technology allows us to make wind measurements safely and remotely right near the ground, it seems ridiculous not to use those measurements," Bluestein says. "The damage surveys were very, very inexact and they were what you used when you had absolutely no other information."

Photo of El Reno Oklahoma storm damage.
Debris hangs from a tree on June 1, 2013, the day after the tornado in El Reno.
Photograph by Justin Sullivan, Getty Images

But incorporating radar data into the EF scale may not be that simple. While the EF scale is intended to describe tornado-strength winds as they occur for three seconds ten meters off the ground, radar data describes instantaneous winds, and usually at higher altitudes. "It's not an apples to apples comparison," Snyder explains. "We don't know how winds change with height inside of a tornado." The fact that radar data are available for a very small fraction of the tornadoes that occur creates further inconsistencies.

Because of these complications, the NWS decided to stick to strict damage analysis to make its tornado ratings, at least until there is a standardized way to incorporate radar measurements. When the Norman office submitted their El Reno report to headquarters, the EF5 grade was rejected, and the storm was again classified as an EF3. "While the wind measurements from the mobile radars are considered reliable, NWS policy for determining ratings is based on surveys of ground damage," a NWS statement read.

Even EF3 tornadoes can kill people.

The back-grading of the El Reno storm led to some heated controversy within the meteorological world. Meteorologist Jim LaDue of the NWS is trying to coordinate an effort to incorporate radar measurements into the EF scale. He explains that a primary reason to rate a storm in the first place is to "figure out what meteorological conditions accompanied strong and violent tornadoes." In order to do this, "we need as accurate a truth as possible of how strong these tornadoes were. Not just what the damage was, but what were the maximum wind speeds," he says.

The debate also has been fueled by the tragic fact that this particular tornado was the first to cause the death of storm chasers, including Tim Samaras, one of the field's greatest heroes. "When people are killed, it's almost as if the loss is not being respected if it happened in less than the worst event," Wurman says. "But even EF3 tornadoes can kill people."

3 comments
gordon Couger
gordon Couger

How many people a tornado kill depends a great deal on how much warning they have. I took pictures after the F4 that hit Vernon and Wichita Falls Texas on April 10th 1979. The storms clipped the north edge of Vernon killing 11. As I recall 9 of the deaths in Vernon were on the highway in cars. Reports have the storm wiping out a good part of town in Vernon I have family there and was in Vernon the next day and the major damage was limited to north west edge of town, including the Canton Cafe, International Harvester dealership and little more.

Whitaker Falls had strip plowed thought it a mile wide and 7 miles long.  With better of the 42 fatalistic there 25 were vehicle related. One couple I know caught out in a pickup truck got out and laid flat on the ground. She had a broken leg, he had a broken arm and both their back look like they had been hit by shrapnel but they lived thought it.

The true killer tornado's are the one's that hit in the middle of the night in place unprepared for them.  See: http://www.tornadoproject.com/toptens/toptens.htm Top Ten killer Tornadoes all before 1953 Only one of them is in tornado country the Woodward storm on April 9, 1947. I have read the logs of W5YJ one of the Amateur radio station that handled relief for that storm.

de W5RED


wendy layten
wendy layten

Wow, I'm speechless. Amazed and dumfounded.

El Gabilon
El Gabilon

So, under the Fujita Scale a tornado that strikes an area where no damage is done such as a grassland, the tornado did not exist since it did not damage to trees, houses, buildings, etc. Rather than saying....well, err, uh huh, we really don't know how to measure the winds of a tornado, we will create a phony system and use that.  Is that science?  If the Fujita Scale is used to determine what amount of federal funds will be used to help the people who were hit by this disaster it will be lower than what it should be in the El Reno, Oaklahoma disaster.   But Bubba we think..........

How to Feed Our Growing Planet

  • Feed the World

    Feed the World

    National Geographic explores how we can feed the growing population without overwhelming the planet in our food series.

See blogs, stories, photos, and news »

The Innovators Project

  • Brave Sage of Timbuktu

    Brave Sage of Timbuktu

    Abdel Kader Haidara had made it his life's work to document Mali's illustrious past. When the jihadists came, he led the rescue operation to save 350,000 manuscripts.

See more innovators »

Phenomena

  • How Sloths Save Their Energy

    How Sloths Save Their Energy

    They effectively "tape" their internal organs to their ribs and hips to prevent pressure on the lungs. By Ed Yong.

See more posts »

Latest News Video

See more videos »

See Us on Google Glass

Shop Our Space Collection

  • Be the First to Own <i>Cosmos: A Spacetime Odyssey</i>

    Be the First to Own Cosmos: A Spacetime Odyssey

    The updated companion book to Carl Sagan's Cosmos, featuring a new forward by Neil deGrasse Tyson is now available. Proceeds support our mission programs, which protect species, habitats, and cultures.

Shop Now »