Partly Cloudy
18°FPartly CloudyFull Forecast

After storm deaths, search on for perfect warning

Published: Sunday, June 9, 2013 5:30 a.m. CST
Caption
(AP photo)
A man stands on top of his car May 31 as it is flooded on South May Avenue near SW 25th in Oklahoma City.

OKLAHOMA CITY – The average tornado warning in the 1980s went out only after a twister was on the ground. Continuing advances in storm prediction have since enabled forecasters to warn people before a funnel cloud is upon them, giving them precious time to seek shelter.

Our better understanding of the nature of tornadoes such as those that have ripped through Oklahoma several times over the past month, killing dozens of people and injuring hundreds more, has still saved countless lives. And our further understanding of such violent weather patterns will no doubt save even more.

Some experts, though, acknowledge there is debate about whether there can be too much advance warning of a tornado strike, and if this could lead people to take foolish risks such as trying to dart across town to pick up a loved one or taking to the open road to try to outrun a violent storm.

“There’s a great philosophical discussion about what constitutes the ideal lead time,” said Greg Carbin, a warning coordination meteorologist for the National Weather Service’s Storm Prediction Center in Norman, Okla. “The more lead time the better, but the flip side of that is that accuracy and certainty in our predictions usually decrease with lead time.”

Over the last five years, residents in the U.S. have been given an average lead time of 13 minutes between the issuance of a tornado warning and a confirmed tornado on the ground. That’s a 17-minute increase from the 1980s, when tornado warnings were typically issued four minutes after a funnel had been spotted, said Lans Rothfusv, who is deputy chief of the warning research and development division at the National Severe Storms Laboratory in Norman.

Rothfusz credited the change to major advancements in technology, especially the introduction of the Doppler radar on the forecasting scene in the early 1990s.

So, is there a “sweet spot” or an ideal amount of warning time?

“While we strive for the longest warning time possible, the answer to that is: it depends,” Rothfusv said. “It depends on the individual needs or the group needs.”

Those in charge of hospitals, nursing homes or even stadiums filled with people, for example, would need significantly more time to prepare than a family that simply needs to get into the basement.

And while Rothfusv and other weather scientists will continue to work to increase warning times, they say finding the ideal amount of warning time must also involve the input of social scientists and experts in human behavior.

“For years we’ve been looking at this from the physical science side – better radars and better science, but we’re at this interface where nature and humans intersect,” he said. “What we’re getting to is realizing we need to understand better how people respond to warnings. Maybe we could give people one hour. We need to know what the response may be.”

When the widest tornado on record touched down May 31 about 30 miles west of Oklahoma City – an EF5 storm packing 295 mph winds at the edge of a metro area with 1.2 million residents – forecasters had already given residents nearly a half-hour to prepare for the threat.

With the additional time, and with the May 20 EF5 tornado that killed 24 people in Moore, another Oklahoma City suburb, still fresh in their minds, many panicked residents opted to flee their homes, and interstates and roadways became gridlocked with people trying to outrun the approaching storm. Many were encouraged by a local television meteorologist who warned viewers that if they couldn’t get underground, they should leave the relative safety of their homes and drive south.

“They were saying that if you weren’t below ground, you weren’t going to survive it, (and) I was still shell-shocked from May 20,” said Terri Black, a teacher’s assistant from Moore who left her home and ended up trapped in traffic during the May 31 storm. “That was very instrumental in my decision (to flee).”

Twenty people died in the May 31 tornado and subsequent flooding, including several who were caught in their cars along a crowded interstate.

For the tornado that hit Moore, forecasters gave 16 minutes’ notice before the twister touched down at Newcastle and moved northeast.

For those who live in Tornado Alley, the issuance of a tornado warning is an attention-grabber. It means that a tornado has either been spotted or that weather radars show low-level rotation in a storm suggesting a tornado is likely, and it usually prompts the sounding of emergency sirens in populated areas.

With hurricanes, Carbin said early warnings of up to 36 hours before landfall are often needed so residents of cities and towns in the path of the storm have time to evacuate, but tornadoes are a completely different, and highly unpredictable, weather phenomenon.

“With respect to tornadoes, nobody’s talking evacuation as far as I know,” Carbin said. “That’s not possible given the short time scales with which tornadoes develop.”

While the average lead time nationally is 13 minutes, Oklahomans and others in the Great Plains often receive more notice because the storms take longer to develop, so the telltale signs of a rotating storm appear earlier in the process. In areas that have smaller tornadoes — mountainous states in the West and those along the Eastern seaboard — twisters develop more quickly and cut into lead times.

The Storm Prediction Center had said for days prior to the May 20 and May 31 outbreaks that storms would likely develop. And when the days arrived, tornado watches — which mean conditions are ideal for tornadoes to develop — had been issued hours before the massive thunderstorms began to develop the rotation that spawned tornadoes. On May 31, the tornado watch area was a huge swath of central Oklahoma, from the Kansas state line to the Red River border with Texas. A tornado watch on May 20 spread across multiple states, making it virtually impossible for someone in the heart of the watch area to try and drive away from the storm’s projected path.

“That’s the nature of watches. An excess of 30,000 square miles is the average area for a watch,” Carbin said. “I think evacuating from tornado watches is a completely unrealistic method of going about this. They’re too large, and there’s a likelihood of not being able to get out of the way in time.”

___

Sean Murphy can be reached at www.twitter.com/apseanmurphy

Previous Page|1|2|3|Next Page

Get breaking and town-specific news sent to your phone. Sign up for text alerts from the Northwest Herald.

Reader Poll

How much money have you spent shopping on Black Friday?
$50 or less
$51-$100
$101-$200
$201-$500
More than $500