During the summer of 2012, fires exploded across the drought-stricken Colorado Front Range—a heavily populated area where the Great Plains meets the Rockies. One evening in early June, lightning struck a tree in the foothills west of Fort Collins. It ignited a fire that burned quietly for a few days and then rocketed downslope, fueled by a windstorm and bone-dry trees, dead from a mountain pine beetle infestation, and engulfed 30 square miles of forest in a single day.
“This is the fire we always worried we might have,” Larimer County Sheriff Justin Smith had said at a news conference that night. The High Park fire grew to 136 square miles—four times the size of Manhattan. It was, at the time, the second-largest fire recorded in Colorado history.
Jason Box, a glaciologist who grew up in Colorado, watched the disaster play out on television in the departure area at LaGuardia Airport in New York. “People were glued to the screens,” he says. Box, then a professor at the Ohio State University who now works for the Geological Survey of Denmark and Greenland, was waiting for a flight that would take him to Greenland for the 2012 field season to study the dynamics and melting of the Greenland ice sheet. He suddenly had a thought: Could soot from the wildfires melt Greenland’s ice sheet?
Scientists have known for years that soot reduces the ability of snow and ice to reflect solar radiation back into space. They’ve found tiny black particles in the Arctic snow and ice that have come from the burning of fossil fuels, agricultural fields, trees, and grasslands thousands of miles away. Pure white snow is highly reflective—it has an albedo of 0.9, meaning it returns 90% of the solar energy that hits it. But snow that’s darker—say, if it is covered with soot—absorbs the sun’s energy, warming, melting and becoming even darker. It then absorbs more energy, launching a positive feedback cycle that causes local—and even regional—warming.
If this cycle were to happen on a large scale in Greenland, it could spell trouble for the ice sheet, which holds 8% of the Earth’s freshwater and is suspended frozen atop the bedrock. If the ice sheet melted entirely, global sea levels would rise 23 feet. Yet even one foot—a plausible scenario that could play out within the next 35 years—would be enough to inundate millions of homes and send the cost of coastal damage from erosion, storm surges, and salt water encroachment soaring. Combined with the recent news that the West Antarctic ice sheet is already collapsing—which itself could release enough water to raise sea levels 13 feet—our descendants are likely looking at a very watery future.
In Greenland that summer, Box tried to collect snow samples that would allow him to test his hypothesis, but launching a new project on the fly proved impossible. “I underestimated in the end how hard it would be to get those samples,” he says. “It was pretty discouraging.” But because any black carbon from the wildfires would get buried in subsequent snowfalls, he knew he had time. Or so he thought.
A Record Melt
Box has carefully tracked Greenland’s albedo for two decades. Year after year, he has plotted the ice sheet’s evolution. Between 2000 and 2011, when the sun was at its strongest in July, the ice sheet’s albedo dropped from 0.74 to 0.66. Box credited the decline to a trio of factors: more warm air from the south, fewer clouds—which reflect solar radiation—and less new snow. Normally, the higher altitude areas of the ice sheet, about 50% of its area, don’t melt at all. But Box proposed in a paper published in July 2012 that with another decade of similar warming it would be reasonable to expect “100% melt area over the ice sheet.”
And then, almost as soon as he’d suggested it, it happened, surprising everyone. Over a few days in July 2012, almost all Greenland’s ice sheet lost its luster. The melt area expanded, growing from 40% to 97% of the ice sheet. Even the very cold areas at high elevations melted to some degree. The albedo also hit record lows, but was the drop due to the melting or to something else, like black carbon?
The event coincided with a dome of warm air parked over the ice sheet, but Box still wondered if the wildfires might somehow be involved. After all, 2012 had become a banner year for wildfires. Peat fires in northwestern Canada left six-mile-long scars in the tundra, and wildfires in Russia had charred taiga and forests in remote parts of eastern and central Siberia. By the end of the fire season, in Russia alone, some 17,000 wildfires had consumed more than 74 million acres—more than eclipsing the near-record 9.3 million acres burned in the U.S. that same year.
Box was aching to learn whether black carbon had contributed to Greenland’s epic 2012 melt, but he couldn’t scrounge together the funding he needed for an expedition. When he tried to apply for fast-track funding from the National Science Foundation that year, a program manager told Box that the organization had already supported another group doing similar work. Box wasn’t deterred. “I wasn’t going to let that stop me.”
“Conventional funding wasn’t an option anymore,” adds Box, who instead rolled out a crowdfunded campaign he called “Dark Snow.” He launched a website, a Facebook page, and a new Twitter feed. He enlisted the support of Peter Sinclair, a climate activist and multimedia blogger, to produce videos posted to the Dark Snow YouTube channel. “He’s totally indie and an important amplifier in our messaging,” Box says. The news media seized the project’s catchy title and DIY vibe—and Box’s spunk—publishing over 30 articles about Dark Snow. “The crowdfunding isn’t just about financing our work,” Box says in a Dark Snow video, “it’s about connecting people to the science.”
Dark Snow’s goal was to link a specific set of fires to a specific melting event. “We saw an unprecedented melting in Greenland in 2012, beyond what can be explained by climate warming alone. Our hypothesis is that it was the light-absorbing impurities seen in the record-setting forest fires,” says McKenzie Skiles, a snow hydrologist and PhD candidate at the University of California, Los Angeles and manager of the snow optics lab at NASA’s Jet Propulsion Laboratory in California who joined Dark Snow to analyze the snow samples.
One group Dark Snow was up against was the NSF-funded team led by Chris Polashenski, a geophysicist, and Zoe Courville, a research mechanical engineer, at the U.S. Army Cold Regions Research and Engineering Laboratory, in Hanover, New Hampshire. The team secured $2.3 million from NSF and NASA for the project, the Sunlight Absorption on the Greenland ice sheet Experiment, known as SAGE.
In late 2012, Box and Polashenski independently pulled a NASA image taken by the CALIPSO satellite in July showing a smoky plume drifting over Greenland’s ice sheet. It added to their arsenal of evidence, but it wasn’t enough proof. “It set the stage that the big melt event happened because the wildfires in Siberia emitted black carbon and caused more sunlight to be absorbed and caused the ice sheet to melt,” Polashenski says. If it turned out to be true, he says, “It would be a good story.”
Raining Black Carbon
The idea that black carbon could boost the melt rate of snow or ice wasn’t new in 2012, nor was the idea that soot from wildfires in North America and Siberia could settle on Greenland. In 2007, Joe McConnell, a hydrologist at the Desert Research Institute in Reno, Nevada, had measured black carbon levels in ice cores pulled from west central Greenland that dated back to 1788. Between 1850 and 1910, black carbon concentrations grew seven-fold and then fell, charting the path of coal burning during early industrialization. Major wildfires also showed up on the record as peaks of vanillic acid, which is produced when lignin from conifer trees is burned.
Black carbon is a powerful contributor to global warming, second only to carbon dioxide. But unlike carbon dioxide, it doesn’t stay aloft for long. Within a week or so of its emission, it tumbles from the sky, on its own or in rain or snow, sometimes landing on snow, glaciers, or sea ice.
Although black carbon emissions have been in decline since the industrial revolution, scientists have renewed their interest in it lately because even small amounts deposited on Arctic snow may play a key role in a warming world.
Open biomass burning, including wildfires, accounted for more than a third of global black carbon emissions in 2000. In a recent study of black carbon emissions, Andreas Stohl, an atmospheric transport modeler at the Norwegian Institute for Air Research, in Kjeller, Norway found that during the summer months—when most fields and forests burn and when ice-albedo feedback is strongest—biomass burning is the largest source of black carbon at two study sites in Greenland, Station Nord (elevation 90 feet) and at Summit (elevation 9,624 feet). “The most sensitive period is right around the summer solstice, in June, when there is more solar energy coming into the Arctic than any other place on Earth,” says Mark Flanner, a climate modeler at the University of Michigan in Ann Arbor.
“At high altitudes you see a much stronger influence from further south, whereas close to the North Pole or over the Arctic Ocean—near the surface—you see almost only influences from the highest latitude sources that exist,” says Stohl, who has studied the Arctic black carbon question for many years, but hasn’t looked into 2012 sources yet. “North American fires are normally stronger and inject the material higher into the atmosphere than Russian fires.” In other words, North American wildfires, which typically happen farther south than Siberian conflagrations, could be responsible for a large amount of melting in Greenland, which is higher in altitude than much of the rest of the Arctic.
Into the Cold
In May 2013, Polashenski’s three-person crew set out on snow machines from Summit Station, 10,551 feet above sea level, along a traverse route northward and then west towards Thule, covering 50 to 90 miles a day. They towed sleds, heavily laden with equipment and food, including over 80 pounds of chocolate bars. They soon fell into a routine, setting out at midday and arriving at the next station at 2 p.m., where they would dig a snow pit to sample the last two years of snow accumulation and any carbon trapped within.
Each pit was a just over six feet square and just as deep, dug by hand in temperatures far below freezing. “You’re digging and digging and digging and sweating in that hole,” he says. “But then you get concerned about contaminating it. So, you climb out, put on clean gear, get back in and scrape another six inches off the sides.” Only then would Polashenski begin cutting Rubik’s cube-sized snow samples from the wall, one every inch, from top to bottom. They would finish by 2 a.m., eat, and fall into their sleeping bags. By mid-June, they had dug 35 snow pits, taken 1,100 snow samples, and made 1,190 albedo measurements.
The Dark Snow team arrived in Greenland near the end of June, and Polashenski’s group had already left. Box and the team flew by helicopter to a location called Saddle at 8,200 feet elevation, where melting is infrequent but had occurred in 2012. “We needed a hard number so we could estimate how much of the melting we could pin on black carbon,” Box says.
Box chose the last site—just east of Ilulissat, about mid-way up the western coast—because images showed it held the darkest surface. “You could walk for kilometers and kilometers and all you saw was this darkish surface,” says Marek Stibal, a microbiologist with the Geological Survey of Denmark and Greenland and a member of the team.
On their first day on the ice sheet, the wind mercilessly whipped tiny snow particles at their faces. “It was like getting sand blasted,” says Skiles, the PhD candidate. Box drilled down into the ice sheet and found the 2012 summer layer about six feet down. Skiles busied herself digging snow pits to collect samples at various depths so they could measure changes in the black carbon concentration. They visited three sites in all, at varying elevations, drilling seven cores and collecting hundreds of samples.
As Box’s team prepared to close the 2013 season and leave Greenland, the air was tinged with smoke. Russia was having another record year for wildfires. “I could see the haze in the air,” Skiles says. But for whatever reason, Greenland refused to repeat the previous year’s melt.
Pessimism or Possibility?
At a lecture at Dartmouth College in December, Polashenski’s enthusiasm for the black carbon hypothesis was beginning to fade. His preliminary analysis of the snow samples suggested the levels of the pollutant were too low for black carbon to be the driver of the widespread melting seen in 2012. “It’s hard to conceive that this black carbon layer made a big enough change to cause melt all through the ice sheet—it isn’t enough,” he says. In all his calculations and estimates, black carbon only seemed to account for a small fraction of the melt. By working with samples collected during the summer of 2012, he says, “We’re studying the perfect storm. We know black carbon can get to Greenland and we know that it reduces albedo, but the question is by how much and does it matter?”
Another contributor to the melting could be algae. During a visit in 2010, Stibal thought he spotted dust on the ice, but when he peered at it beneath a field microscope, he saw algae as well. They were producing a dark pigment, possibly to protect themselves from DNA-damaging UV radiation. But in the process, they were darkening the surface of the ice sheet, causing it to absorb more energy—and possibly boosting their own ice-darkening growth. The algae, Stibal says, may be contributing to the snow-ice-albedo feedback. The microbes could also be feeding off deposited black carbon—literally—by munching on phosphorous and other nutrients that arrive with dust and smoke plumes.
Despite the numerous confounders, Box remains confident. Using hypothetical data, a model he developed shows that even small increases in black carbon could double the initial reduction in albedo over time. The changes start slowly at first, but ratchet up by the end of the melt season. In other words, under certain conditions, the black carbon might, in fact, have been able to push the ice sheet past a threshold and explain the 2012 melt.
In February, Skiles analyzed the first Saddle core. The black carbon levels in the 2013 snow were low, but as she inched back through time, deeper into the core, the concentration jumped several times through the summer and peaked at 15 parts per billion. The Dark Snow team included the graph in a 2014 fundraising video posted to Facebook in mid-May. “It doesn’t seem to be just one event, but it seems to be several different fire deposition episodes,” Box says.
Box’s optimism is further buoyed by a study recently published in the Proceedings of the National Academy of Science. Using ice cores that date back to 1750 and surface samples from 2012 taken at four study sites in Greenland, including Summit Station, Kaitlin Keegan, a graduate student at Dartmouth College, looked for links between black carbon from wildfires and widespread melting of the Greenland ice sheet. The 262-year record shows four black carbon peaks at depths that correspond to 1868, 1889, 1908, and 2012, and that the ice sheet melted extensively in 1889 and again in 2012. The other two years—1868 and 1908—didn’t see significant melting because temperatures at the surface of the ice sheet were too cool in 1868 and the black carbon fell too late in the year in 1908 to trigger widespread melting.
“They’re validating the hypothesis that wildfire soot is important in the story of the 2012 Greenland melt,” Box says. “It’s satisfying to hear it from an independent study.” Still, he’s not certain Keegan and her colleagues, who include McConnell, the Desert Research Institute hydrologist, have found the smoking gun. For one, Box questions the warmer temperatures they saw in 1889. The Danish government has been monitoring air temperatures in Greenland since 1873, “and there isn’t evidence of a warm anomaly in 1889,” he says.
Keegan and her colleagues predict that with rising temperatures and an increase in the frequency of forest fires, widespread melting should occur on average every five years by 2100. Wildfires in the Western U.S. have become larger and more frequent over the last 30 years, and the length of the fire season expanded by 78 days between 1987 and 2003.
Warmer spring temperatures are drying soils earlier in the season and spurring wildfires later in the year.
“When the trees are well hydrated and the grasses are wet, they’ll still burn, but they won’t burn as enthusiastically,” says Don Falk, a fire ecologist at the Laboratory of Tree-Ring Research at the University of Arizona in Tucson. “High temperatures, wind, and low humidity create the conditions where fires can really just crank,” he says.
Whether fire conditions in the American West will lead to further melting in Greenland still hasn’t been proven, though. Field expeditions in Greenland are expensive and cold and can be frustrating for the scientists, who are often plagued with equipment malfunctions and poor weather. But the research that both Box and Polashenski are doing could help us decipher the future of Greenland’s glaciers in a warming world.
Both teams are back in Greenland this spring and summer to collect more samples and make more measurements. Polashenski’s group was unable to reach the area between Summit and NEEM this year, and instead focused on getting a better understanding of the distribution of black carbon across Greenland in 2012, along the Thule peninsula. Box is focusing on the biology of the ice sheet and how dark-colored microbes alter its albedo. His team plans to camp on the ice sheet for two and a half months, using a drone to measure the ice sheet’s reflectance. “By bringing the biology and the glaciology together, some innovation can occur and we can learn something new,” Box says.
By turning to crowd funding and opening up his field notebooks to the public, Box is giving us all the opportunity to learn something new. There’s a benefit to such open science, but also a risk. In the process, Box is broadcasting his progress to other scientists and risks being scooped. He says his team will publish as quickly as possible—but he’ll continue to use social media to promote their work. “We’ve been putting all our cards on the table, and that may help other people publish before we do,” Box says. “But we’re also telling people about the science first.” That openness is a key part of the project for him, because in the long run, connecting with the public on climate change could have more of an impact than any journal publication.