Thursday, January 29, 2015
Blizzard to fizzle: Why the winter storm of the year wasn't
Posted: Tuesday, January 27, 2015
First, weather forecasters predicted that a monster 2 feet of snow would bury the Philadelphia region. That was Sunday.
By Monday afternoon, the prediction was for more of a thick blanket of up to 18 inches accompanied by wind gusts of 30 mph.
In the end, the predictions came true — just not in the tri-state region, which instead experienced a classic case of blizzard fizzle, forecasters said.
“It’s always a question we meteorologists ask ourselves, did we make the right call,” said Todd Miner, a meteorologist with AccuWeather Service in State College.
While parts of New York and New England states along the Atlantic coast were buried under a crippling winter storm that as predicted included blizzard conditions into Tuesday evening, at least a foot of snow, 70 mph wind gusts, thunder snow, coastal flooding and power outages, the Philadelphia and surrounding areas woke Tuesday to 4 inches or less of new snow.
The anticipated all-day storm had pretty much petered out in Bucks County by 8 a.m. Tuesday with Doylestown reporting 1½ inches and Penndel 3 inches. In Montgomery County, accumulations ranged from a half-inch to 2½ inches, according to AccuWeather.
Left in this winter weather’s wake: thousands of kids with a day off from school or a late opening, states of emergency in place throughout Pennsylvania and New Jersey, bare supermarket shelves, travel bans, airport delays, packed SEPTA trains running on a reduced schedule for the day, and embarrassed weather forecasters.
“My deepest apologies to many key decision makers and so many members of the general public,” Gary Szatkowski of the National Weather Service in Mount Holly, New Jersey, wrote in a Tweet. “You made a lot of tough decisions expecting us to get it right, and we didn’t. Once again, I’m sorry.”
So what happened?
Computer weather models are not as predictable as people might believe, forecasters said. All along it was anticipated the storm would have a sharp cutoff point as far as what parts of the tri-state area would be get a dusting versus getting walloped, forecasters said.
With Monday’s storm, the center of low pressure developed 90 miles farther east into the Atlantic Ocean, taking with it all the heavier bands of snow that were expected to develop down the New Jersey coast and eastern Pennsylvania, NWS meteorologist Kristin Kline said.
New model data questioning the track of the low pressure center and moisture wraparound started coming in around late Monday afternoon, Kline said. Before 10 p.m., as NWS meteorologists started watching the radar, they started seeing things that led them to believe the anticipated big snow totals weren’t coming for Philadelphia and New Jersey and they tried to put out revised snow total predictions, Kline said.
The problem with Monday’s storm system is that with only 12 to 18 hours out, the model data offered a wide range of options as far as where the storm was set up and intensify and where snow bands would develop. Usually in that time range the model data is consistent, but in this case it wasn’t, Kline said.
“There was a lot of uncertainty with this storm,” Kline added. “It was one of those things where someone was going to get pounded.”
AccuWeather meteorologist Todd Miner explained that the main weather system responsible for Monday’s storm — an Alberta Clipper — swept across the northern part of the United States until it hit the Atlantic Ocean, and that typically means a bad winter storm for New England states. The problem with storms like that one is forecasting its western boundary — the back-end so to speak — which makes the difference between a winter wallop or glancing blow, Miner said.
With this storm, meteorologists knew from the get-go that the far eastern Pennsylvania extending into New York region was on the cusp of the storm — the so-called “zone of uncertainty” — Miner said. As a collective unit, the decision was made to err on the side of caution when it came to suggesting storm impact, he added.
“We can’t make a Tuesday forecast on Wednesday. People want to know things ahead of time,” Miner said. “It takes a lot of time to analyze these models and figure out which one to go with.”
Another challenge for modern weather forecasting is the number of available computer models and data, Miner said. Until the mid-1990s, forecasters had only a few models that simulated atmospheric conditions. Now hundreds exist.
“When a fair bit of them call for a big storm in a metropolitan area — even if it’s just half of them saying it — what do you do?” Miner said. “Go for it and have nothing happen or vice versa.”
Sometimes old-fashioned conceptual models, which look at how the atmosphere typically behaves and what weather patterns usually evolve, are more accurate. In the case of the Alberta Clipper that preceded Monday’s storm, the old models would have leaned toward exactly the type of snow event Pennsylvania and New Jersey experienced, Miner said.
For now, the Mount Holly weather service bureau will re-evaluate its storm modeling to figure out how it can do a better job letting people know about uncertainty with the forecasts like the one Monday, Kline said.
As for the rest of the week, the National Weather Service in Mount Holly is predicting a quick-moving clipper system will pass through the Philadelphia area Thursday into Friday that could produce about an inch of snow, Kline said.