Science 20 February 2015:
Vol. 347 no. 6224 pp. 818-821
Vol. 347 no. 6224 pp. 818-821
Weather experts were on high alert in December 2009. A region of high pressure had settled over Greenland, forming a roadblock in the path of the circumpolar jet stream. Thwarted, the jet stream meandered, bending southward into a large loop and shunting cold Arctic air toward the center of the United States. Meteorologists were familiar with this “Greenland block.” It was a climate pattern historically favorable for winter storms—and as if on cue, record snowfalls blanketed the eastern United States over the next few months.
Against the backdrop of “Snowmageddon” and other powerful winter storms that have blasted the United States, Europe, and Asia in the past few years, a different kind of tempest has been swirling within the Arctic science community. Its core is a flurry of recent research proposing that such extreme weather events in the midlatitudes are linked through the atmosphere with the effects of rapid climate change in the Arctic, such as dwindling sea ice. The idea has galvanized the public and even caught the attention of the White House. But some Arctic researchers say the data don't support it or that the jury is at least still out. Even some of its proponents agree that the media hype is premature.
Now, scientists are starting to tackle the issue in earnest. Atmospheric links between the poles and midlatitudes are becoming a marquee topic at Arctic-related conferences. Last December alone it drew researchers to a workshop in Barcelona, Spain, and a dedicated session at the annual American Geophysical Union meeting in San Francisco, California. “A lot of scientists have worked individually, in isolation, pursuing their own ideas,” says Judah Cohen, a climate scientist at the Massachusetts Institute of Technology in Cambridge—one of several researchers who hope that a coordinated effort will measure the reach of the north.
THE IDEA THAT THE ARCTIC could have a regional, even hemisphere-scale impact on the atmosphere represents a paradigm shift for climate scientists. “When I was in grad school, [the thinking was that] the tropics are everything,” Cohen says. “The tropics, the ocean—that was it, there was no other player.”
The Arctic Ocean, after all, is small and cold; the tropical Pacific Ocean is vast and, of course, warmer. To influence the atmosphere on a global scale, very large amounts of heat and moisture must flow from ocean to air. That heat engine is huge in the Pacific and powers the El Niño–Southern Oscillation (ENSO), the planet's dominant interannual climate pattern. ENSO also has a helpful supply chain: Even as heat escapes from the eastern tropical Pacific to the atmosphere, currents flowing in from the western tropical Pacific provide a fresh supply.
But there is a new kind of engine forming in the Arctic. The rapid loss of sea ice in the Arctic Ocean due to global warming—the area of summer ice has shrunk more than 11% per decade since 1979—has created an expanse of dark open water newly available to absorb the sun's energy. This extra energy input and the corresponding flux of moisture and heat to the Arctic atmosphere are helping drive a strong local positive feedback to global warming, called Arctic amplification. As a result, surface temperatures are rising twice as fast in the Arctic as at lower latitudes.
But is the loss of sea ice influencing more than just local warming? In 2012, climatologists Jennifer Francis of Rutgers University, New Brunswick, in New Jersey and Stephen Vavrus of the University of Wisconsin, Madison, proposed that Arctic amplification driven by sea ice loss could significantly affect midlatitude weather by slowing the jet stream (Science, 18 April 2014, p. 250). The mechanism they suggested was this: A faster warming Arctic means a reduced temperature gradient between the Arctic and midlatitudes, which in turn weakens west-to-east winds. The result is a slowdown of the jet stream, allowing it to meander more and form elongated waves (called Rossby or planetary waves) that jut to the north or south. Among other effects, Francis and Vavrus proposed, the deep waves in the jet stream could enable a winter storm to push farther south and then linger in one location, possibly dumping record amounts of snow.
A handful of newer studies support one key part of this hypothesis: that large-amplitude Rossby waves in the jet stream are indeed linked, statistically, to extreme weather in the midlatitudes. For example, Vladimir Petoukhov of the Potsdam Institute for Climate Impact Research in Germany and colleagues in 2013 correlated summer heat waves in Europe with instances of slow-moving, high-amplitude Rossby waves. And last year, James Screen, a climate modeler at the University of Exeter in the United Kingdom, and Ian Simmonds of the University of Melbourne in Australia co-authored a paper that reported a strong statistical correlation between amplified, slow Rossby waves and months from 1979 to 2012 with extreme weather events. More waviness, they found, made the western United States more susceptible to heat waves and the eastern United States to extreme cold.
But other parts of the hypothesis remain controversial—particularly the part of the chain that links these changes to Arctic amplification. Some evidence supports Francis and Vavrus's idea that a decreasing temperature gradient between the midlatitudes and the Arctic might slow the jet stream and make it wavier.
Yet climate dynamicist Elizabeth Barnes of Colorado State University, Fort Collins, reported in 2013 that she could find no uptick in the instances of Rossby waves in the last couple of decades, the period encompassing extreme sea ice loss and Arctic warming. And in a 2013 study, Screen and Simmonds found no statistically significant changes in these waves from 1979 to 2011, despite a declining north-south temperature gradient.
Some theorists have proposed another mechanism by which dwindling sea ice could shape midlatitude weather: by forcing changes in a multiyear atmospheric cycle called the Arctic Oscillation. In a positive Arctic Oscillation phase, the jet stream is strong and less wavy, and cold air stays in the Arctic. In a negative phase, the jet stream is weaker and wavier, and the cold air spills southward. Some analyses suggest that in the last few decades the Arctic Oscillation has increasingly been in a negative phase, which might drive not just storminess but also longer term cooling in the midlatitudes.
But evidence that the retreat of sea ice has influenced the Arctic Oscillation is scarce. “Since any Arctic effect [on the jet stream] is buried in large random climate [variations], and the record is short,” such a link is simply impossible to prove at the moment, says James Overland, an oceanographer at the National Oceanic and Atmospheric Administration's Pacific Marine Environmental Laboratory in Seattle, Washington.
FRANCIS ACKNOWLEDGES the lingering uncertainty about how the changes in the Arctic could drive bouts of extreme weather farther south. But, she says, as the mystery attracts more researchers, they are finding clues. Last year, climate scientist Masato Mori of the University of Tokyo and colleagues modeled how sea ice loss in the Barents and Kara seas north of Russia could increase blocking events—and the resulting severe winters—over Eurasia. And Baek-Min Kim of the Korea Polar Research Institute in Incheon and colleagues showed in a model how changes in sea ice loss in the Barents and Kara seas could set off a chain of events that produced a negative phase of the Arctic Oscillation.
Another paper published last year, this one by Kazutoshi Sato of the Japan Agency for Marine-Earth Science and Technology in Yokosuka and colleagues, suggests that ice loss in the Barents Sea and cold Eurasian winters are actually part of an even bigger climate pattern originating in the North Atlantic Gulf Stream. Recent observations show that the current is pushing warm water farther to the north. That shift, Sato and his colleagues found, may produce planetary waves that both warm the Barents Sea and cause a cold anomaly over Eurasia.
“That's a pretty robust linkage there,” Francis says. “It's not going to happen every single year, but it's pretty convincing that there are real physical mechanisms disrupting the jet stream.”
IT'S TOO EARLY TO TAKE SIDES about whether the Arctic influence is real, many researchers say. Last December's Barcelona workshop grappled with that “two camps” misconception, Screen says. “I don't like to frame it like that,” he says. “The majority [of people] are somewhere in the middle.” In fact, a few are in an entirely different camp: blaming changes in Eurasian snow cover, rather than the retreat of Arctic ice, for the midlatitude weather extremes (see sidebar, p. 821).
“We have records of sea ice only going back 30 years, with dramatic losses in the last decade or so,” Screen says. A statistical analysis of that brief period of observations, he says, is unlikely to settle the debate. Instead, scientists need to turn to other sources of knowledge, such as models and theory, “to better understand what we might be seeing in the observations.”
Overland, for his part, calls this the “preconsensus period.” He says meetings such as the workshop in Barcelona (which both he and Screen attended) have shown encouraging signs of progress. For one, researchers plagued by inconsistent definitions of terms such as waviness are finally getting a chance to straighten things out. And he felt that the Barcelona workshop did come to at least one consensus: that if there are links between Arctic amplification and midlatitude weather, they're amplifications of already existing and known climate patterns—Greenland blocking and a region of persistent high pressure called the Siberian high—so there's no need to build brand-new physics into the models.
SORTING OUT THOSE “known knowns” is exactly what researchers should be doing, Barnes says. “In my mind, we've put the cart before the horse a bit” by focusing on weather extremes, she says. “We're missing the theory, the dynamics, the mechanistic understanding.” She suggests that the community go back to basics to determine how Arctic warming could influence the jet stream dynamics. “It's certainly not sexy like telling you that we're going to have more extreme cold events—but in terms of the science, it's the way forward.”
One key to success, she says, is teasing out the effects of the Arctic on the jet stream from those of other atmospheric powerhouses, such as the chaotic atmospheric flow at midlatitudes and the ENSO pattern in the tropics. Indeed, researchers need to look beyond the potential impacts of Arctic change, Overland says, and aim for a better picture of atmospheric dynamics as a whole.
“We're just getting into that time where people get interested,” Barnes says. “If you look over the literature in the last 10 years, there's a lot of focus over how tropical warming will influence the jet stream, but not a lot on how Arctic warming will influence it. As a community our eyes were turned to the south, and now we're looking north. That, to me, is what's exciting.”