Weather forecasters say the storm that will soon hit the state is still a tricky one to forecast, even hours before it arrives.
The National Weather Service said in a forecast discussion posted Wednesday morning that it was “very challenging” to predict what exactly will happen late Wednesday afternoon and night into early Thursday morning.
The forecasters said the various supercomputer models they use to assess future weather were unusually divergent and “struggling” with the forecast, particularly unusual given how soon the snowfall is expected to begin.
They said there was “tremendous uncertainty” about the path of the storm, but they were following a “middle ground” and predicting the greatest threat for heavy snow south of the Massachusetts Turnpike to near the Cape Cod Canal.
The latest snowfall total forecast map, issued by the weather service Wednesday afternoon, showed only about 4 to 8 inches falling in Boston. The weather service had predicted more than a foot in Boston as recently as Tuesday.
Carl Erickson, a senior meteorologist at Accuweather.com, said the supercomputer models were “not in very good agreement” in the run-up to the storm, but by Tuesday “more of a consensus was being reached.”
Still, there’s uncertainty, he said, especially on the edge of the storm where there will be a very sharp cutoff from where snow will be falling quite heavily and where “there’s hardly anything going on.”
Another factor that has made forecasting the storm difficult is that it had more than the usual number of moving pieces. There were “several pieces of energy, low pressures, to keep track of,” he said.
“Not only at the surface, but several thousand feet up in the upper levels, there’s pieces of energy,” he said. One is over the upper Ohio Valley and one is over the North Atlantic, “all just kind of pinwheeling around helping to draw in all that North Atlantic moisture.”
“You had a fresh, cold air mass in place, and you also drew all this moisture in from the Atlantic to meet up with that cold air,” he said. “Those are the two big ingredients that you need.”
Steven Decker, associate teaching professor at Rutgers University and director of the undergraduate meteorology program, said another factor that might have led to the difficult forecast is thunderstorms in the Southeast caused by the same weather system that is now rolling toward Massachusetts.
Thunderstorms are a wild card in snow forecasting, he said, because they can cause atmospheric changes that can throw off weather models, he said.
If there are enough thunderstorms and they’re strong enough, they can change the jet stream, change the storm’s track, and change how much moisture it’s carrying.
“If the weather model doesn’t fully capture all of its effects, that can lead to errors” later in the model’s timeframe, he said, noting that the storms hit two days ago in Tennessee and Alabama, and a day ago along the southeastern coast. The models should improve going forward, now that actual data from those storms can be incorporated into the models, he said.
Yet another factor in predicting a spring snowstorm is the warmth of the earth and the higher angle of the sun, meteorologists said.
“The extra question you have to deal with is how much of that snow is going to stick,” Decker said.
Solar radiation making its way even through cloudy skies can, if the snow is only light, keep the ground warm enough to prevent snow from accumulating, Erickson said.
If it’s snowing heavily enough, the snow will accumulate anyway, but if it’s only light to moderate, “you have to take the time of year into effect,” he said.
Shortly before noon, Erickson said, the snow was falling very heavily in New York City and inching its way northward. It’s expected to arrive in Eastern Massachusetts sometime Wednesday afternoon.Martin Finucane can be reached at firstname.lastname@example.org.