Meteorologists worry about weather alert fatigue
It’s hard to make a tornado. Even on May 6, when forecasts suggested a day of widespread storms with the right conditions to spawn multiple strong, long-track tornadoes, it’s never guaranteed.
“To get this to happen, it requires an unbelievably delicate mix of air and water, which is constantly moving, constantly changing,” said Rick Smith, warning coordination meteorologist for the National Weather Service in Norman.
The wrong air temperature, humidity or wind direction — any little thing can kill a tornado’s chances of forming. On May 6, the threat of those long-track tornadoes didn’t pan out despite the accuracy of the overall forecast. It still turned out to be a deadly night.
“We’re working as hard as we can. People devote their lives to this,” Smith said. “We spent close to a week looking at what was coming up on May 6, worrying about it and trying to produce the best forecasts.”
To produce those forecasts, meteorologists spent time carefully picking out the words they would use, even organizing a conference call to decide
whether a forecast would include the attention-grabbing term “outbreak.”
Crafting the right message for the public
As Oklahoma yet again braced for a day of tornadic weather on Monday, the National Weather Service in Norman crafted a simplified message meant to cut through the noise and plainly tell people what to expect.
The message that morning was clear: Don’t focus on risk “categories.” Not everyone will see a storm. And if you see a storm, assume it’s dangerous.
This type of messaging was an intentional effort to speak more casually to the public.
“Social media does give us the ability to break outside of our traditional, very scripted formatted warning products and things that we do traditionally,” Smith said. “Don’t talk to them like a scientist with a master’s degree who’s studied this their whole life. Talk to them like it’s your aunt or grandma or best friend calling you on the phone.”
The National Weather Service and its parent organization, the National Oceanic and Atmospheric Administration (NOAA), have taken steps in recent years to understand how the general public responds to severe weather messaging. It’s a delicate balance to strike.
“We can have the perfect forecast, but if people don’t understand it, if they don’t trust it, then they’re probably not going to make decisions that best protect themselves,” said Makenzie Krocak, a research scientist with NOAA’s National Severe Storms Laboratory.
The study, which included focus groups, simulations and surveys, have lead to a simplification in how the National Weather Service issues bulletins.
A major factor in reviewing messaging has been the proliferation of weather information on the internet. It’s allowed the general public to see raw data and technical discussions even if they don’t have the training to understand it, and social media has spawned countless sources of analysis and forecasting that wasn’t available 30 years ago.
Ultimately, Smith said, the average Oklahoman might just settle on remembering “we’re going to get tornadoes.”
“The message gets really boiled down to something simple, which is understandable because most people don’t have the time or the need to really sit down and understand the probabilities and all the different scenarios,” Smith said. “It’s never a slam dunk, even on May 6 where every parameter that you can imagine seemed to be just sitting there, waiting for the right storm to come along.”
National Weather Service meteorologists also analyze their performance after a weather event. Smith described it like watching game film to find out what, if anything, can be done better next time.
Weather-warning fatigue is not well understood
Meteorologists also worry about weather fatigue. May 6 was just the latest in a string of severe weather events that have been packed into just a few short weeks.
And if the average Oklahoman expected to see a tornado on any of those days, they might have been disappointed. After the deadly Barnsdall tornado on May 6, some in the weather community began responding to others who complained that the forecast didn’t live up to their expectations.
Krocak, the social science researcher with NOAA, said there’s quite a bit of research on the impact of false alarms. Unfortunately, the results are mixed. Some studies show that false alarms influence behavior, while others show no link.
“So it’s something that forecasters really, really worry about,” she said. “One thing to remember is if we’re trying to reduce the number of false alarm warnings, that means that we’re going to reduce the number of warnings we issue. That probably means we’re going to miss more tornadoes.”
That tradeoff is a discussion that happens with any severe weather threat, she said.
“That’s certainly something that they’re thinking about because we don’t want to over-warn people. On the other hand, we really need to make sure that we get the message out before something bad happens,” Krocak said.