A few weeks ago in Arlington, Texas, the local forecasters predicted downpours on a day that the Texas Rangers were playing the New York Yankees. Thousands of fans stayed home.
Some clouds rolled in. A shower hit, but the game played on. The foul forecast did more than spoil a day: The Rangers estimated that it cost $250,000 in ticket and concession revenue.
Blown forecasts can wreck the bottom line not only for baseball teams but for vacation destinations, state highway associations and local restaurants as well. People have always complained about the weatherman, but these days, critics believe that bad forecasts are happening more often. And within the growing world of weather forecasters and meteorologists, a tempest is brewing over whether some are offering forecasts that science can’t back up.
Interest in the weather is, as meteorologists might say, at a record high. Forecasts can be retrieved from almost anywhere—local TV and radio, 24-hour cable networks such as The Weather Channel, digital channels such as NBC’s Weather Plus, weather Websites, and updates on mobile devices. As a result, competitive pressure to attract viewers has never been greater.
To set themselves apart, many weather outlets are increasingly offering longer-range forecasts and touting cases of extreme weather at or near the top of the newscast. Fox-owned WFLD Chicago, for one, introduced a primetime newscast last month called The Ten, featuring a 10-day weather trend cheekily titled “The 10 Day at 10:10 on the Ten.”
But a growing chorus in the meteorological community warns that stations are risking their credibility. Meteorologists say long-range forecasts are akin to “throwing darts” and “tossing a coin” and are virtually impossible to predict accurately. “There’s no one [in the meteorological community] who hasn’t raised the issue; it’s a well-known stress point for them,” says Dr. Anthony Socci, a senior science and communications fellow at the American Meteorological Society (AMS). “They’re selling a skill that isn’t there, and they’re not being candid with viewers.”
Socci blasted TV stations in his essay “Is Fear of Competition Among Local and National TV Stations Undermining the Credibility and Standards of Forecasting and Meteorology?,” which appears in the April issue of the Bulletin of the American Meteorological Society. Too often, stations offer probabilistic predictions—involving specific temperatures—in long-range forecasts as opposed to more-general trends. Another big transgression: predicting specific snowfall amounts two or three days before a storm. Such practices “exceed the limits of existing forecasting skill,” Socci wrote.
Intense competition is driving “endless rounds of one-upmanship, where science and alleged forecast skill become casualties and weapons in the insular battle among TV stations and networks all vying to lay claim to being 'first’ in delivering something new,” he writes, “even if not verifiably scientifically skillful, credible, or useful at the moment.”
The race to be first in the market can result in irresponsible behavior, says Louis Uccellini, director of the federal government’s National Weather Service, National Centers for Environmental Prediction (NCEP). He believes that stations equipped with their own Doppler radar are sometimes quick to sound the siren for severe-weather warnings that conflict with the more conservative data from the state and federal weather organizations. “They’re competing with the other stations as to who’s first with the story,” he says. “It’s a problematic area.”
The media is an “absolutely essential” partner in weather news, Uccellini adds, but forecasts offering such details as highly specific long-term predictions have questionable validity. “[Some stations] put more stock in techniques that we have questions about,” says Uccellini. “The techniques still need to be proven, as far as I’m concerned.”
But getting viewers to tune in at 6 and 11 p.m. gets harder with each passing ratings book, and stations are under mounting pressure to grab them with eye-catching weather news. With news representing an estimated 35%-40% of a typical station’s revenue (and weather news is some 30% of that), stations are fighting desperately to keep viewers tuned in.
Many find that long-range forecasts and breathless teases about extreme weather do well in luring eyeballs. By some estimates, two-thirds of stations around the country offer forecasts for seven days or more. The Weather Channel provides 10-day forecasts online that include high, low and average temperatures, as well as sun/cloud predictions and chance of precipitation. Accuweather.com, meanwhile, offers high and low temperatures for a full 15 days out.
Station execs say they’re simply meeting the public’s desire for more information. Years ago, when WFXT Boston switched its forecast from a five-day to a four-day, chief meteorologist Kevin Lemanowicz believed that the four-day chart was easier for viewers to read, and he felt more confident forecasting for one less day. But viewers felt they’d been shortchanged. “We got lots of calls, people saying, 'Where’s the five-day?” says Lemanowicz. WFXT reinstated the extra day soon after.
Meteorologists and station managers agree that there has never been so much general interest in the weather. In a recent Web-usage study that Frank N. Magid Associates conducted for B&C, 74% of respondents access local weather news online at least once a month, way more than those going online monthly for local news (50%) and traffic reports (26%). On Hearst-Argyle station sites in 2006, weather stories accounted for 16 of the top 20 most-read stories.
The unparalleled interest stems from a number of factors. With Hurricane Katrina, the 2004 tsunami, wildfires, and, most recently, spring snowstorms and nor’easters, devastating weather is frequently a lead news story. Global warming, as evidenced by Al Gore’s Oscar-winning documentary An Inconvenient Truth, is also at the top of people’s minds.
Moreover, as the Weather Channel turns 25 this month, there’s an entire generation that grew up with 24/7 weather news at their fingertips. “There’s absolutely more interest in weather than there’s ever been,” says Mike Steinberg, senior VP at AccuWeather. “People have recognized the impact that weather has on personal activity, as well as business. It impacts just about every person and every business every day.”
WHAT’S THE FORECAST FOR NEXT MONTH?
Some say the long-range forecasts are driven by consultants and station managers who put the almighty Nielsen numbers ahead of accuracy and responsibility. Paul Deanno, chief meteorologist at WTVJ Miami, says he’s a “huge opponent” of any forecast that goes beyond seven days but acknowledges the strain meteorologists are under: “There can be immense pressure from management to balance what’s scientifically accurate and what’s promotable.”
To be sure, many in the industry have nary a qualm with long-range forecasts. In Chicago, WFLD chief meteorologist Amy Freeze insists that science supports her 10-day forecast. At the 26 Hearst-Argyle stations, the typical forecast is around five days. But, for marketing purposes, many stations occupying a seven on the dial go with seven days. “With technology today, you can kind of go out to seven days, and it’s a good indicator of trends,” says Terry Casey, weather product manager for Hearst-Argyle. “The timing might be off a bit, but it’s a really good indicator of what the weather will be.”
At WCVB Boston, President/General Manager Bill Fine employs seven certified meteorologists, giving the station one of the most accredited weather departments in the country. WCVB offers a five-day forecast, then stretches to a seven-day outlook at the end of the program. Though acknowledging that the latter makes the meteorologists uncomfortable, he says it’s what viewers want. “People like to plan,” he says. “Everyone wants to see the weekend weather no matter what day of the week it is.”
At AccuWeather, the 15-day forecast is, in fact, a point of pride. “We pioneered the extension of the forecasting horizon,” says Lee Rainey, VP of marketing, “pushing it further and further out.”
As technology advances to give increasingly precise updates, the pressure at the station level mounts. “[The weather industry] is, in many ways, pretending to know things that we don’t know,” says Raymond Ban, executive VP of meteorology science and strategy at Weather Channel. “We’ve worked our way into a corner, and we’re eventually going to have to work our way out. It’s going to take the entire enterprise to change.”
If the weather department is consistently wrong, credibility qualms can bleed into the brand. It is possible, after all, that viewers could doubt the business reporter, the anchor and the station overall—and switch channels. “All we have in the news department is our credibility,” says Dan Forman, senior VP/station manager at WNBC New York. “Whether it’s sports, business or weather, it has to be backed by facts. You don’t play with that.”
Blown forecasts are far from victimless crimes. Communities often take cues from the stations’ weather reports, whether it’s local government ordering plows in advance of a blizzard, utility companies stocking up on energy, or families canceling travel or dinner plans. In Denver this past winter, stations were forced to explain why there was no snow after they predicted 12-18 inches, causing canceled flights and legislative meetings.
Eighteen months ago, Baltimore braced for a massive snowstorm, causing the Maryland State Highway Administration (MHSA) to order 2,500 trucks for salting and plowing. As it turned out, the accumulation amounted to a dusting. Fully deploying state highway crews costs some $60,000 per hour. The tab for a storm that never comes can run into the hundreds of thousands of dollars, says a spokesperson.
Maryland’s Highway Administration has learned not to rely on local broadcasters when extreme weather is lurking. Says the spokesperson, “The [stations] are going to play it up; that’s their job.”
Several years ago, a group of restaurateurs in Providence, R.I., banded together and kept track of the local stations’ snow forecasts versus the actual snowfall. They presented their findings to station managers, along with estimates of how much the forecasts were costing the economy (as high as $10 million-$20 million some days, they said) based on overzealous snow predictions. Said organizer Bob Burke, “Not even Nome, Alaska, gets as much snow as you’re predicting [for Providence].” (He says the predictions fell more in line with typical amounts after the restaurateurs approached the stations, and they were not forced to give their “Biggest Flake” award.)
Some are calling for an industry-wide move to deliver more-responsible weather news, but no consensus has emerged. In the meantime, education may be the best antidote. A former broadcast meteorologist, Marisa Ferger, teaches aspiring forecasters at Penn State, which features one of the nation’s top weather-broadcasting programs. She’s adamant about instructing students on the pressure to sell sexy weather stories, such as the ubiquitous “storm of the century” warnings.
“I tell students they’ll have to do things they’re not comfortable with,” says Ferger. “It’s going to happen, and they have to be prepared for it.”