Why are snow forecasts so hard? And why do most people feel forecasters are almost always wrong?
Baltimore is right along the Chesapeake Bay, which generally funnels in warmer air. At the same time, a storm system from the west will channel cold air (the cold part usually comes from Canada). These meet somewhere in Maryland, and sometimes in a tight gradient (sharp difference over a short distance).
Where exactly this gradient, and how tight it is, is typically the big question in snow forecasts. These forecasts tend to fluctuate based on the latest prediction of where this gradient will be.
An important factor that makes many think the forecasts are often wrong is misuse and misunderstanding of forecasts published by others. For example, an app or website may claim 4-7 inches of snow will fall next Thursday, which is 6 days away.
But that has a 80% of being wrong, since almost nothing works out exactly the way we think it will from 6 days out.
So knowing that and taking those types of forecasts with a grain of salt is important. And don't just take rumors as fact.
Also, know that forecasts rely on computer model guidance. These models take atmospheric and surface (on the ground) conditions, and with formulas, make a forecast. There are several different models, and here are a few:
GFS - this is the American model, and it can predict things up to 384 hours away (that is 16 days).
GEM - this is the Canadian model, which does best in cold air situations (which makes sense since it comes from Canada).
ECMWF - this is the European model, which is often very accurate.
Closer in to the 'event', the NAM model performs at a more HD level, getting more granular information.
All of this combined leads to a forecast; but these things can wobble over time as it gets closer to the forecasted date.
Getting back to Baltimore forecasts, we sit right on the rain/snow line (same as the gradient).
Some experts weighed in on the difficulties facing forecasters.
“You’ve got the mountains; you’ve got the Bay; you’ve got the Atlantic — all things that can come into play,” says NBC Washington Meteorologist Veronica Johnson".
“In a snowstorm, the hardest part of our forecast is the I-95 corridor, and it’s that way for Washington; it’s that way for Baltimore; it’s that way for Philadelphia; it’s that way for New York; it’s that way for Boston.
“West of I-95, you’re almost always going to get snow. East of I-95, mostly you’re going to have some kind of a rain-snow mix if not all rain. Then right along I-95, where all of those major cities are, you’re going to get some kind of a mixture. And that’s why predicting the snowfall is so hard in this area.”
"Snow forecasting is arguably the most difficult, most complicated thing for forecasters to predict," James Hoke, director of the National Oceanic and Atmospheric Administration's Hydrometeorological Prediction Center, says.
"It's like a giant math problem that's constantly changing," says Ralph Roskies, scientific director for the Pittsburgh Supercomputing Center. "And the computer does the arithmetic."
"There are many areas over the oceans where we don't get frequent observations," says Kevin McCarthy, deputy director of the Hydrometeorological Prediction Center.
"I would say that for most meteorologists, forecasting snowfall amounts is the hardest thing we do," said Spectrum News Chief Meteorologist J.D. Rudd, based in Wisconsin. "Forecasting snow amounts is very, very tough. The slightest change in a storm's track or intensity can be the difference between three and nine inches of snow."
"There are the social 'media-rologists' that immediately post models many days out and start internet rumors," Spectrum News Chief Meteorologist Eric Elwell said. "These people typically don't understand the nuances and biases of the models, especially beyond a couple of days."
Sentry Page Protection
Please Wait...
