Deep thinking on snowy spells

Scientists study North Atlantic weather system for clues to patterns of miserable winters

By Beth Daley
Globe Staff / February 13, 2011

E-mail this article

Invalid E-mail address
Invalid E-mail address

Sending your article

Your article has been sent.

Text size +

Piles of snow walled off Boston during a snowy stretch in the mid-1940s. Students reveled during the mid-’60s when school snow days were called with joyous regularity. During two years in the early 1990s, forecasters took to measuring drifts with yardsticks.

So is it possible this miserable winter is the start of another snowmageddon cycle?

Scientists don’t know but they are studying clues. They know that snowy seasons are often related to changes in atmospheric pressure over the North Atlantic. Researchers are now trying to understand what causes those shifts, called the North Atlantic Oscillation, in hopes of one day predicting what decades, or maybe even years, will potentially be snowier than others — information critical to communities, businesses, and the public.

Scientists may never be able to predict confidently exact seasonal snowfall because there are so many forces affecting local weather. But they hope to one day be able to say there are strong odds that a particular stretch of time will be snowier than average.

“If we could do this, cities might want to start planning how much salt they are going to buy in September,’’ said Louis W. Uccellini, director of the National Weather Service’s National Centers for Environmental Prediction. “It would be very useful.’’

The North Atlantic Oscillation swings between what scientists call positive and negative phases. During a positive phase, low pressure over the Arctic allows strong air currents known as the jet stream to flow smoothly from west to east, keeping cold air at higher latitudes. But in a negative phase, higher pressure over the Arctic can divert the jet stream southward, allowing colder air to move into the Northeast, bringing with it the potential for especially snowy winters.

“Imagine putting a large rock into a river, which forces the water to flow around it,’’ said Michael J. Iacono, a senior staff scientist for the Lexington-based Atmospheric and Environmental Research, a Verisk Analytics company. In this analogy, the rock is the zone of higher pressure; the water flow corresponds to the cold air pushed south — and the weather it would likely bring.

The oscillation’s impact is felt in more ways than snowfall: The precipitation, winds, and temperature shifts that follow its shifts can affect agriculture, fish stocks, and human health up and down the East Coast and in Europe. In 2004, Uccellini co-wrote a book called Northeast Snowstorms, in which he reviewed more than 130 years of snowfall data and concluded there was no statistically significant pattern of seasonal snowfall. He did, however, find a strong relationship between years with lots of snow and the negative phase of the oscillation.

“The only signal that was clear with snowstorms was a negative’’ North Atlantic Oscillation, he said.

The North Atlantic Oscillation this year has mostly been in a negative phase. But this winter’s misery — Boston has had 71.2 inches this season, 45.3 inches above normal — is also probably due to a better understood climate feature: La Niña, which periodically causes abnormally cool conditions in the tropical Pacific. This probably forced precipitation-laden storms north, head-on into the cold air forced south by the negative North Atlantic Oscillation.

Researchers knew it would be a La Niña winter. Pacific Ocean monitoring is so advanced scientists can predict the climate pattern months in advance.

But the North Atlantic Oscillation can be forecast only one to two weeks in advance. Unlike La Niña, which involves more slowly evolving climate processes in the ocean, the oscillation is an atmospheric phenomenon that is far more chaotic.

“It is very, very challenging to predict,’’ said James Hurrell, senior research scientist and chief of climate modeling at the National Center for Atmospheric Research in Colorado, who studies the North Atlantic Oscillation.

The link between the North Atlantic Oscillation and snowfall is intriguing because the oscillation can go into decades-long periods where it is dominated by a positive or negative phase. From the mid-50s until the blizzard of ’78, for example, it was in a negative phase more often than not, accompanied by some of the snowiest years of the past century. But in the 1979-1980 winter, a generally positive North Atlantic Oscillation took over until fairly recently. While some researchers believe there are discernable decade patterns to the oscillation, others say a good measurement record dates back only to 1950.

“So it’s tough to get a good handle until you see it over many more years,’’ said Mike Halpert, deputy director of the federal Climate Prediction Center. “There may be a 20- to 30-year cycle, but we can’t be sure yet.’’

Maddeningly, even if the oscillation has long-term phases, it can shift back and forth during a season. Much of the snow this winter occurred during a negative phase, but the wicked winter mix of ice, sleet, and snow early this month occurred when the North Atlantic Oscillation shifted more positive.

Still, Hurrell said there are advances being made in the study of sea surface temperatures that may give researchers clues about which way the North Atlantic Oscillation will “nudge’’ over time.

“There is an exciting prospect that there may be a very modest level of predictability,’’ he said.

That probably won’t mean an airtight seasonal forecast — a multitude of ocean circulation and other weather and climate patterns will make that difficult. Layered on top of the North Atlantic Oscillation effect is the possible impact of climate change on these natural patterns: For example, one theory today suggests melting Arctic sea ice is setting off a chain reaction that is having the unexpected effect of colder East Coast winters.

“Even if we got it all right, we still wouldn’t know exactly where and when the snow would fall — there is too much natural variability,’’ Halpert said.

But if scientists can one day calculate the probability that a stretch will be dominated by a negative or positive North Atlantic Oscillation, it could help people get ready for high energy use or potential spring floods caused by melting snow.

“It would be good to let people prepare,’’ said Uccellini .

Beth Daley can be reached at