It’s rare for a winter storm of this magnitude to stalk the nation’s most populous metropolitan area. But that’s what happened this weekend.
As of Friday, forecast models still couldn’t agree on whether the Northeast snowstorm would turn into a major snowstorm for the Mid-Atlantic and Northeast. It wasn’t until Saturday morning that the models really started to come together on a potentially historic scale. And even then, there was much more uncertainty than usual.
Since then, meteorologists have tightened the forecast, increasing snowfall amounts, wind speeds and overall threat levels to ensure people know the severity of what will happen over the next 48 hours.
The end result is likely to be fewer deaths and injuries than would have occurred had authorities not taken stock of the situation, declared a state of emergency, and taken other incremental steps, but in reality, this prediction left much to be desired in terms of ex ante accuracy.
Forecasters were also having a difficult time. Social science research shows that people tend to stick with the first storm forecast they hear, even after the weather changes. This can sometimes be deadly, as in the case of Hurricane Ian in 2022, when some Florida residents did not flee areas that were particularly threatened by the change in the storm’s path.
In this case, the first forecast people in the Northeast heard was that the storm was likely to pass through their region, but included a warning that they should keep an eye on the latest forecast in case conditions changed.
And oh, the details have certainly changed.
In many ways, the forecast confusion was comparable to another familiar situation. The hurricane rapidly gained strength as it approached land, catching people who might have been prepared for a weaker storm by surprise. The storm itself was hurricane-like, with winds up to 134 miles per hour and more than two feet of snow in some places, blowing from Delmarva to Boston.
Rapidly changing forecasts present unique challenges for forecasters, weather reporters, and public officials alike. Instead of taking days to familiarize themselves with the complexity of the forecast and prepare the public, governors and mayors had just a couple of days to communicate the growing severity of the event.
They appear to have succeeded in this, as airlines canceled thousands of flights ahead of the winter onslaught, travel bans were issued to cities and states before the worst of the situation, and other measures were taken to ensure public safety during the event.
Given the scenario of rapidly changing forecasts, a textbook ferocious storm crashing down Interstate 95, and increasingly urgent public safety messages, things could have been much worse.
Winter, and even hurricane history, has seen many failures in the response of cities like New York, with former Mayor Michael Bloomberg famously dismissing the threat of Superstorm Sandy after it was no longer officially called a hurricane.
A blizzard can make or break a mayor’s career, and New York City’s new mayor, Zoran Mamdani, tried to get ahead of it and clearly communicate the serious threat it poses to public safety. The storm was also an early test for the new governors of Virginia and New Jersey.
There are many sound scientific reasons for the persistence of forecast uncertainty in this situation, including the opposition of computer models to predicting complex weather patterns.
But given all the tools available to today’s forecasters, including the advent of AI-powered computer models, one might hope that this last-minute storm forecast would prove to be rare rather than common.
Because, in retrospect, there was much good fortune in the nation’s preparation for this storm, but forecasters were in the uncomfortable position of rapidly increasing warnings with little time to prepare.
