WASHINGTON — Citing the chaos in central Oklahoma last month when people tried to flee from dangerous storms in their cars, a weather expert from the University of Oklahoma said here Wednesday that research on storm warnings should include an examination of how people respond to them.
Kelvin Droegemeier, a professor of meteorology and a researcher at OU's National Weather Center, told a House subcommittee that lawmakers should be cautious in focusing on reaching the goal of providing an hour of lead time to people about potential tornadoes.
“We need to understand how we convey and formulate uncertainty in messaging to the public, how the public responds and comprehends warning information,” Droegemeier told a Science, Space and Technology subcommittee.
“And I think that we have to even ask ourselves whether the whole current watch and warning system really needs to be rethought from the ground up starting with people, because the people are the ones who are affected ultimately.”
The researcher's comments came during a hearing on legislation by Rep. Jim Bridenstine, R-Tulsa, that would boost funding for weather research and make it a higher priority for the federal government. The bill creates a program to “develop and extend accurate tornado forecasts and warnings beyond one hour in order to reduce loss of life, injury, and damage to the economy.”
Droegemeier said it was important to look at the lead time.
“But I think we want to be careful about not focusing entirely on that magical number because ultimately the question is ... what are people going to do with that one hour?”
On May 31 in central Oklahoma, he said, thousands of people fled their homes and “put themselves in harm's way because they had a very large amount of lead time without really knowing what to do with it ... And so I would argue that our focus really should be on the goal of zero deaths.”
A preliminary study at OU regarding the public perception of tornado warning accuracy showed that, out of 3,000 people surveyed, 25 percent said they would take no protective action for a hypothetical storm in the light intensity category, Droegemeier reported in written testimony to the subcommittee.
Seven percent said they would take no action even after the warning level rose to “significant” and “severe, devastating and incredible events.” And 18 percent for those categories said they would drive away from the warned area.
Kathryn Sullivan, acting director of the National Oceanic and Atmospheric Administration, and Bridenstine engaged in a lengthy exchange over whether the government should provide as much lead time as possible.
Sullivan told Bridenstine that she wouldn't necessarily set an hour of lead time as a goal, saying “there's a genuine question about how humans respond to impending risks.”
“So the government should make decisions about how much lead time we give people because we know better than they know how to protect themselves?” Bridenstine said.
“No,” Sullivan said. “We should understand how to communicate the information we have so it's effective for the people who have those decisions to take.”
“So let's say we have an hour lead time,” Bridenstine said. “Would you suggest we should withhold that information because an hour is too much and people aren't smart enough to take cover?”
“I'm not saying withhold it,” Sullivan said. “I'm simply saying that the challenge of communicating forecast information effectively to decision-makers is a genuine question that needs to be approached thoughtfully.”