Hell hath no fury like weather watchers scorned—and if that wasn’t already apparent to most meteorologists it was certainly made clear recently when snow-expectant New Yorkers prepared to get dumped on but merely got dusted.
The Big City hunkered down on January 26, awaiting a snowstorm the National Weather Service had described as “crippling and potentially historic.” The dire forecast (and possibly painful memories of Buffalo in November) prompted Governor Andrew Cuomo to essentially shut down the city, closing the subway for the first time ever. Bus service, commercial flights, PATH trains, and highways were also shut down and nearby state leaders took similar measures.
Imagine the chagrin, then, when parts of the stranded region awoke to find less than a foot of snow covered their forcibly parked cars and quiet streets. No matter that many areas did indeed get the predicted amount of white stuff, the question perpetually asked by news outlets was how could forecasters get it so wrong?
There are a few answers to that—the inability to anticipate the western edge of the storm, differences in weather models, and the general difficulty in forecasting snowstorms. There’s also the valid point that for much of the forecast area, snow speculations weren’t wrong. In fact, according to National Weather Service Director Louis Uccellini, the real problem with the forecast wasn’t the variables, but that most of us don’t understand them.
“It is incumbent on us to communicate forecast uncertainty,” Uccellini said of NWS forecasters. “We need to make the uncertainties clear.”
How to best do that is a topic that has long occupied meteorologists. The World Meteorological Organization has issued guidelines on communicating uncertainty, stating that improved understanding of the changeability in forecasts would enhance decision making and boost audience confidence. The National Research Council has also issued recommendations, specifically for NWS.
The suggestions range from simple ideas such as using color in weather maps and considering how warnings are worded to more technical concepts such as moving from deterministic to probabilistic modeling. In the NWS case, though, years of incorporating this research haven’t created enough change.
“Despite efforts made to educate NWS forecasters on how to interpret and convey this information, they are severely constrained in doing so when adhering to the currently inflexible state of the official product suite,” writes Steve Tracton for the Washington Post’s Capital Weather Gang blog. “Not to mention, users need to be educated on how to understand this information.”
It’s not really clear if a better understanding of uncertainties would have caused officials to act any differently. Many have stated that they would rather be over-prepared for a nonevent than underprepared for a deadly storm. And while that attitude is heartening, it can be a fine a line between unnecessary readiness and warning fatigue.
“It's not whether the city should have prepared so much, it's how people respond,” Irwin Redlener, Director of the National Center for Disaster Preparedness, told the Associated Press. “We don't want the population to get so cynical that they're not heeding the warnings.”
In that sense, then, apologies from forecasters aren’t helpful, because they deflect attention from their audience’s responsibility to be sure they understand the information that they’re using to make decisions. Still, as University of Georgia atmospheric scientist Marshall Shepherd points out, in today’s information environment, it will fall on forecasters to get the point across.
“There is more risk and nuance in weather forecasts than the public is interested in consuming so it is a challenge to craft a message that gets attention, is not ‘hype,’ yet has actionable information,” he writes for Weather Underground. “We must continue to have the discussion about how to communicate uncertainty and risk effectively.”