Why the forecasts were wrong, and what they got right
There's some hand-wringing over the forecasting models used. Here's the basic idea, from CNN meteorologist Brandon Miller:
"Meteorologists largely depend on three major forecast models: the NAM (North American Mesoscale), the GFS (Global Forecast System), and the ECMWF (European Center for Medium Range Forecasting). Meteorologists will look at all of these and, using their own expertise, local knowledge, etc., formulate a forecast. In this case, the NAM and the ECMWF both showed 2 feet of snow or more for New York City, while the GFS (which has just been upgraded this winter) showed a more conservative 6 to 12 inches.
"The National Weather Service forecasters in New York certainly went all in with the NAM and ECMWF forecasts, and all but ignored the GFS, without providing much room for uncertainty that should come with the GFS showing a vastly different solution." (In 2012, the ECMWF "was the media-darling model for properly forecasting Superstorm Sandy to a 't' while the GFS did not," Miller says. "Now it is the opposite.")
Still, "the forecast wasn't too far off if you look at it as a whole," Miller says. "The heaviest snow fell across much of Connecticut, Rhode Island, Long Island and Massachusetts. Winds have gusted up to hurricane strength, which was also in the forecast." Predicted storm surge flooding also came to fruition, he said.
"But of course the headlines are going to come from New York City and New Jersey, where a forecast of more than 2 feet turned out to only be about half that," Miller said. There were 2 feet of snow elsewhere, and "a miss of only 30 to 40 miles in a forecast that was first given 48 hours in advance is not that bad from a strictly forecasting perspective."
But from a "practical perspective," it's a "big miss, and has major ramifications with business disruption, wasted resources and tax dollars, etc.," Miller said.
CNN did note the discrepancy among forecasts Monday, even as government officials were announcing closures.
"I just got the brand new models in just a minute ago. And one model says for New York City 2 inches -- not 2 feet, 2 inches. The other model I looked at said 27 inches," CNN meteorologist Chad Myers noted. "I hate it when models don't agree to that extent."
It's not an exact science, and it's important for people to know that, Miller says. "As meteorologists we must convey the uncertainty associated with these forecasts."