When the Arctic Melts

In the middle of the night in the middle of the summer in the middle of the Greenland ice sheet, I woke to find myself with a blinding headache. An anxious person living in anxious times, I've had plenty of headaches, but this one felt different, as if someone had taken a mallet to my sinuses. I'd flown up to the ice the previous afternoon, to a research station owned and operated by the National Science Foundation. The station, called Summit, sits ten thousand five hundred and thirty feet above sea level. The first person I'd met upon arriving was the resident doctor, who warned me and a few other newcomers to expect to experience altitude sickness. In most cases, he said, this would produce only passing, hangover-like symptoms; on occasion, though, it could result in brain swelling and death. Belatedly, I realized that I'd neglected to ask how to tell the difference.

N.S.F. Summit Station—according to the agency's many rules, this is how visiting journalists are required to refer to the place—was erected in the late nineteen-eighties. Initially, it was occupied only in the summer; now a small crew remains through the winter, when, at Summit's latitude—seventy-two degrees north—the sun never clears the horizon. The station's main structure is known as the Big House. It resembles a double-wide trailer and teeters almost thirty feet above the ice, on metal pilings. Arrayed around it are a weather station, also elevated on pilings; a couple of very chilly outhouses; several tanks of jet fuel; and an emergency shelter that's shaped like a watermelon and called the Tomato. Some of the station's residents used to sleep in tents, but a few years ago a polar bear showed up, so the tents have been replaced by metal sheds.

The Greenland ice sheet has the shape of a dome, with Summit resting at the very top. The ice dome is so immense that it's hard to picture, even if you've flown across it. It extends over more than six hundred and fifty thousand square miles—an area roughly the size of Alaska—and in the middle it is two miles tall. It is massive enough to depress the Earth's crust and to exert a significant gravitational pull on the oceans. If all of Greenland's ice were cut into one-inch cubes and these were piled one on top of another, the stack would reach Alpha Centauri. If it melted—a rather more plausible scenario—global sea levels would rise by twenty feet.

Until relatively recently, it was thought that Summit would be, if not unaffected by climate change, at least untroubled by it. Such is the ice sheet's bulk that at its center it creates its own weather. But in the past few decades Greenland has changed in ways that have stunned scientists who spend their lives studying it. Since the nineteen-seventies, it has shed some six trillion tons of ice, and the rate of loss has been accelerating. Crevasses are appearing at higher elevations, glaciers are moving at non-glacial speeds, and large parts of the ice sheet appear to be twisting, like a writhing beast. In July, 2012, surface melt was detected at Summit for the first time since modern measurements began. In 2019, the station experienced melt in mid-June and then again in late July. On August 14, 2021, it rained, an event so remarkable that it made news around the world. ("For the First Time on Record, Rain Fell at the Summit of Greenland," ran the headline in the Sydney Morning Herald.) There was late-season melt at Summit in September, 2022, and more melt in June, 2023.

The story of climate change is generally told in terms of human action, and for good reason. The almost two trillion tons of CO2 that people have pumped into the atmosphere have changed the planet in ways that every day become more apparent. Last year, average global temperatures set a new record, and by a wide margin. Canada experienced record wildfires; the Caribbean saw record ocean temperatures, which devastated its coral reefs; and Libya was hit with record rainfall, which led to a dam collapse that killed more than five thousand people. This year's global temperatures will almost certainly surpass last year's. Among the many climate-related disasters of 2024 so far have been a heat wave in Mecca that killed thirteen hundred pilgrims during the hajj and Hurricane Helene, which caused at least twenty billion dollars' worth of damage. How people—or governments and corporations, run by people—respond to the mounting losses will have repercussions that will last, for all intents and purposes, forever. As no less an authority than the United Nations' Intergovernmental Panel on Climate Change put it, upon releasing its latest scientific assessment, "The future is in our hands."

But, like so many stories that get told, this one doesn't tell the whole story. The future depends on how humanity reacts to global warming, and it also depends on how the Earth does. Owing to advances in everything from satellite altimetry to deep-sea drilling, a great deal has been learned in the past few decades about the planet's history. Much of the new science suggests that the climate is, all on its own, unstable, prone to dramatic and sometimes sudden shifts.

The history of Greenland is a case in point. During what's known as the Last Glacial Maximum, some twenty thousand years ago, an ice sheet stretched more or less continuously from Greenland across Ellesmere and Baffin Islands and down over Canada and much of the northern United States. So much water was tied up in the ice that sea levels were four hundred feet lower than they are today, and it was possible to walk not just from Siberia to Alaska but also from Australia to Tasmania and from England to France. When the ice began to recede, around fifteen thousand years ago, large swaths of the world experienced catastrophic flooding. During one particularly sodden period, known as meltwater pulse 1A, sea levels rose by more than a foot a decade.

Most scientists believe that ice ages—there have been at least ten of them over the past two and half million years—are initiated and terminated by periodic shifts in the Earth's orbit, caused by, among other factors, the tug of Jupiter and Saturn. But orbital shifts produce only slight changes in the amount of sunlight that reaches different parts of the globe at different times of the year. Such slight variations are insufficient to explain the growth and subsequent retreat of massive ice sheets. Rather, it seems, the orbital shifts act like a trigger, setting off other processes—feedbacks—that greatly amplify their effect. One relatively straightforward feedback features albedo, from the Latin word for "whiteness." Ice and especially snow have a high albedo. They reflect lots of sunlight back to space. Thus, as an ice sheet grows, the planet absorbs less energy. This has a cooling effect, which encourages the buildup of more snow and ice, which results in more reflectivity, and so on. Start to melt an ice sheet and the same cycle spins in reverse.

Today, feedbacks are, to put it mildly, a growing concern. A report published last year by more than two hundred researchers from around the world noted that many of the systems that determine the climate exhibit nonlinear behavior. Such systems may "shift to a very different state, often abruptly or irreversibly, as a result of self-sustaining feedbacks." The researchers identified two dozen potential "tipping systems," among them the Greenland ice sheet.

At a certain point, the report warned, feedbacks could become so powerful that, even if CO2 emissions were cut dramatically and temperatures stabilized, the ice sheet would continue to shrink, possibly until it collapsed. The "best estimate" of when this critical threshold will be reached is when average global temperatures rise 1.5 degrees Celsius—roughly three degrees Fahrenheit—above preindustrial levels. Even after that line is crossed, it will take many centuries for the changes set in motion to play out. Still, as a practical matter, there will be no going back. When it comes to tipping systems, the future is in our hands until it isn't.

Days at Summit begin with a staff meeting held in a heated tent that's outfitted with a treadmill, weights, and yoga mats. In front of the treadmill, people have taped scenes from more temperate climes—ones with trees and flowers. On my first morning at the station—I still had a headache, but no fatal brain swelling—the station's supervisor opened the session with a request for volunteers to lug a pair of propane tanks up the stairs to the Big House. Summit's cook announced that the latest shipment of food was short on lettuce. Someone pointed out that there were problems with the flags on the station's runway, which is made entirely of snow. A fourth person promised to clean the outhouses. After the meeting, I got to launch the daily weather balloon, which was about three feet tall and dangled a cartridge of electronic instruments. The balloon, filled with helium, flew out of my hands. I tried to follow it as it sailed over the ice, but I soon lost sight of it.

The view from Summit in all directions is pretty much the same: white. The Norwegian explorer Fridtjof Nansen, who, in the eighteen-eighties, led the first team to cross Greenland on skis, recalled the ice sheet's monotony—an "interminable flat desert of snow." There was, he complained, "no break or change in our horizon, no object to rest the eye upon, and no point by which to direct the course." Especially when it's cloudy, the ice, free of shadows, appears as one enormous blank page.

In fact, the ice sheet is packed with information, like a giant encyclopedia. Among the first to recognize this was Ernst Sorge, a German glaciologist. "I'm looking at a landscape whose vast simplicity is nowhere to be surpassed on earth, and which yet conceals a thousand secrets," he wrote.

Sorge was part of a famous—infamous, really—expedition that set off from Copenhagen in the spring of 1930. The expedition's leader was another German scientist, Alfred Wegener, who's best known for having come up with the theory of continental drift. One of Wegener's goals was to establish a camp at a site dubbed Eismitte, or "ice middle," about a hundred miles south of where Summit now sits. Sorge and a colleague were supposed to overwinter at the site and take meteorological measurements. Owing to a series of unfortunate events, a third man, who was suffering from frostbite, ended up stuck at the camp as well and had to have his toes amputated with a penknife. Meanwhile, Wegener died as he was trying to fight his way back to the coast, eating his sled dogs along the way. His body is still buried somewhere in the ice.

From the surface, the camp at Eismitte looked like a snow fort with a round turret. Beneath the surface were chambers—a living room, an instrument room, and a storeroom—that had been dug out of the snow. Fascinated by the subglacial world, Sorge kept on digging until, at the far end of the camp, he had sunk a shaft more than fifty feet deep. Studying the walls of the shaft by lamplight, he discovered that he could tell the difference between snow that had fallen on Eismitte in the summer and snow that had fallen in the winter. By counting backward through the seasonal layers, he calculated that his shaft extended through twenty-one years' worth of accumulation.

In the decades that followed, researchers delved deeper and deeper, using increasingly sophisticated drills. The farther the drills descended, the denser the layers of old snow became, until they were compressed into ice. But even in the icy depths the difference between summer and winter precipitation could be discerned. This made it possible to date each layer back through the centuries.

Meanwhile, scientists found that they could tease out a wealth of data from every annual increment. By analyzing the ice with a mass spectrometer, they could calculate what the average temperature on Greenland had been in any given year. And by extracting the gases contained in tiny bubbles of trapped air they could reconstruct changes in the atmosphere.

In the nineteen-nineties, a team of American researchers working at Summit succeeded in drilling all the way from the top of the ice sheet to the bedrock. In the process, they pulled up thousands of long, skinny cylinders of ice—two miles' worth. In ice from fifteen hundred and two feet down, there was snow that fell when Nero was emperor; at twenty-three hundred and fifty feet, snow from the reign of Tutankhamun. At the very bottom was snow that fell before the start of the last glaciation.

Analysis of the core showed, in extraordinary detail, how temperatures in central Greenland had varied during the last ice age, which in the U.S. is called the Wisconsin. As would be expected, there was a steep drop in temperatures at the start of the Wisconsin, around a hundred thousand years ago, and a steep rise toward the end of it. But the analysis also revealed something disconcerting. In addition to the long-term oscillations, the ice recorded dozens of shorter, wilder swings. During the Wisconsin, Greenland was often unimaginably cold, with temperatures nearly thirty degrees lower than they are now. Then temperatures would shoot up, in some instances by as much as twenty degrees in a couple of decades, only to drop again, somewhat more gradually. Finally, about twelve thousand years ago, the roller coaster came to a halt. Temperatures settled down, and a time of relative climate tranquillity began. This is the period that includes all of recorded history, a coincidence that, presumably, is no coincidence.

Richard Alley, a glaciologist at Penn State and the author of a book about the ice-coring project, summed up the findings as follows: "For most of the last 100,000 years, a crazily jumping climate has been the rule, not the exception."

To work at Summit, scientists have to apply to the N.S.F., an independent agency of the U.S. government. The same goes for journalists. My trip to the station was arranged by the agency's Polar Media Program, which sent an "escort"—a Washington, D.C.-based press officer—to accompany me. So intent was my escort on not letting me out of her sight that it became something of a station joke. At one point, when we briefly became separated, I asked someone in the Big House to let her know that I had gone to another building.

"Don't worry," he replied. "I'll tell her you wandered off into the snow."

At the time of my stay, in mid-July, roughly half of the station's forty-odd residents were contract employees who were there to keep the place running, an immense logistical challenge. Summit's (more or less) constant below-freezing temperatures make basic operations like supplying drinking water both complicated and energy-intensive: the water has to be melted from snow, the pipes that circulate it have to be heated, and the fuel that provides the energy has to be flown in. (Thanks to a long-standing arrangement, supplies and also people are transported to Summit by the New York Air National Guard's 109th Airlift Wing, which operates a fleet of ski-equipped cargo planes out of an old U.S. Air Force base on Greenland's west coast.) Several people I met at Summit had spent the previous winter working on Antarctica, either at McMurdo or South Pole Station, and were planning to return there for the next austral summer. They said that the challenge of the work was part of its draw, as was the space-station-like camaraderie.

Another largish contingent was installing a network of radio antennas. These, it was hoped, could be used to detect ultra-high-energy neutrinos. Potential sources of such neutrinos, I learned, include gamma-ray bursts, pulsars, and clusters of galaxies known as flat-spectrum radio quasars.

"We cannot detect the neutrinos directly," Felix Schlüter, a German astrophysicist, explained to me one day, when I sat down with him in the Big House. "We can only detect them when they're interacting with the ice and producing other ps." The following afternoon, I (and my escort) set out with Schlüter on snowmobiles to visit some antennas that had recently been installed about three miles from the station. During the ride, I pulled my neck gaiter over my mouth, which turned out to be a mistake: the moisture from my breath caused it to freeze to my face. When we got to the spot, it was maybe ten degrees Fahrenheit, and a brisk wind was blowing. There was some scaffolding and green flags, but most of the critical equipment had been buried hundreds of feet down. We were deep enough into Nansen's interminable desert that the Big House, the Tomato, and the outhouses had all sunk out of view.

Every summer, the N.S.F. flies a group of high-school students up to Summit as part of a program whose stated goal is to "inspire the next generation" of polar researchers. Zoe Courville, a snow scientist with the Cold Regions Research and Engineering Laboratory, in New Hampshire, was at the station to prepare for the annual visit. One morning, I watched her and an energetic young technician named Caleigh Warner dig a snow pit behind the Big House. The idea was to use it, à la Sorge, to show the kids the difference between summer and winter snow. When Courville and Warner were done, the pit was about seven feet deep and accessible only by a set of snow stairs. I clambered down.

For the first few layers, the seasonal differences were clear even to a novice like me. Summer snow is coarse and grainy; winter snow, smooth and dense. Courville ran a bare finger down one wall of the pit. A couple of feet below the surface, she paused. "That's probably last summer's melt," she said. "When you look at the grains, they're rounded and fused together." I ran a gloved finger along the same spot. The melt layer was thin and brittle. I broke off a piece that resembled a shard of glass.

"When I first started coming up here, twenty years ago, we had models that were predicting what the climate would be like in Greenland," Courville said. "And we're starting to be outside of even the most dire predictions in terms of temperature increase."

"I try to be optimistic about things," Courville told me at another point. "I don't know that it's all doom and gloom. But from what I've experienced here in Greenland, at the center of the ice sheet, we're approaching the point of no return."

When the students finally arrived—they were a day late, owing to various problems with the ski-equipped planes—they were buzzing with excitement. They took turns descending into the pit in heavy boots and parkas that they had borrowed for the trip. A visiting scientist from the University of California, Irvine, brought out some cylinders of ancient ice to show them. The kids were invited to smash the cylinders with a mallet, an activity they took to with gusto. Ziplock bags were passed around so that everyone could take a chunk as a keepsake. The gesture struck me as curious, since the ice was destined to melt as soon as the students took off again for the coast. But maybe, I thought, that was the point.

Today, the spot where the Summit core was drilled is preserved almost like a shrine. The metal casing that the drillers left behind still sticks up out of the snow, even though the borehole beneath it has collapsed. The casing is wrapped in a custom-made jacket, and attached to it is a red flag that shudders in the near-constant wind.

The Summit core—officially the Greenland Ice Sheet Project Two, or GISP2, core—filled in a key chapter in climate history and, at the same time, opened up a huge gap. Apparently, there was some great force missing from the textbooks—one that was capable of yanking temperatures around like a yo-yo. By now, evidence of the crazy swings seen in the Greenland ice has shown up in many other parts of the world—in a lake bed in the Balkans, for example, and in a cave in southern New Mexico. (In more temperate regions, the magnitude of the swings was lower.)

Scientists are still struggling to make sense of the data. The best theory is that the wild swings were set off by some daisy chain of feedbacks involving the ice, the air, and—most important—the oceans.

The great wheelworks of the climate, the oceans transport fantastic amounts of energy—a quadrillion watts' worth—from the sun-drenched tropics toward the sun-starved poles. One particularly important loop in this system is the Atlantic Meridional Overturning Circulation, or AMOC (pronounced "ay-mock"). The loop might be said to begin in the North Atlantic, where the surface waters are especially cold and salty. The combination of low temperature and high salinity makes the water unusually dense, so it sinks. Warmer water from the south rushes in behind it; as this water cools, it sinks, drawing still more water north, and so on. Oceanographers measure currents in units called sverdrups. One sverdrup equals a million cubic metres per second. When the AMOC is operating at full strength, the water circulates to the tune of twenty sverdrups, a hundred times the flow of the Amazon River.

"A full AMOC collapse would be a massive, planetary-scale disaster" is how Stefan Rahmstorf, an oceanographer at the University of Potsdam, in Germany, recently put it. "We really want to prevent this from happening."

Greenland, the world's largest island, is a Danish territory. Though eighty per cent of the island is covered in ice, there are slender ice-free strips along the coast, and people have inhabited these areas, on and off, for nearly five thousand years. Today, most Greenlanders are of Inuit descent and speak both Danish and Greenlandic. About a third of the island's fifty-six thousand residents live in the capital, Nuuk; the rest live in towns and villages that hug the fjords.

Kangerlussuaq, which has a population of about five hundred, sits at the end of a particularly long fjord on Greenland's west coast. The town exists largely because of its runway, which was built by the U.S. Air Force during the Second World War and is now used by Air Greenland as well as by the New York Air National Guard. It has a grocery store, a restaurant overlooking the runway, and a recreation center open only to Guard members and their invited guests. After my stay at Summit, Kangerlussuaq struck me as positively cosmopolitan.

In Kanger, as it is often called, I had arranged to meet Marco Tedesco, a climate scientist at Columbia who studies ice dynamics. When I caught up with him, he was fuming over a rental car. He'd been under the impression that he'd reserved an S.U.V. with off-road clearance; instead, he'd been handed the keys to an ancient Honda. Would the car get stuck in glacial silt, which sometimes acts like quicksand? From Tedesco, I learned a new word in Greenlandic: immaqa, meaning "maybe."

Tedesco, who grew up near Naples, is tall and lanky, with a shaved head and a collection of tattoos that he has acquired in various places for various reasons. On his right arm is a shower of snowflakes; one is twelve-pointed, which, he told me, is a design very rarely found in nature and which he chose in memory of his mother. On his left arm, the assortment includes a water droplet that he got in Hawaii during a low period—"I felt like a drop in the ocean"—and on his chest are Chinese characters that he translated as "big truth." Tedesco had brought along a former graduate student of his, Paolo Colosio, who's now a postdoc at the University of Brescia. When I told them that my husband taught Dante, they both began reciting the opening canto of the Inferno: Nel mezzo del cammin di nostra vita / mi ritrovai per una selva oscura / ché la diritta via era smarrita.

We went to have dinner at the restaurant by the runway. The weather along the coast had been bad, and the place was crowded with people whose flights had been cancelled. (I later heard that Air Greenland is sometimes referred to as Immaqa Airline.)

Tedesco told me that he had become interested in the Greenland ice sheet about twenty years ago. At the time, he was working for NASA, thinking about how to improve the detection of snowmelt via satellite. After a while, he decided that he needed to see the place for himself. "I wanted to look at things more completely," he said. Since 2010, he has visited Greenland fourteen times. On one visit, he launched a radio-controlled boat into a meltwater lake and, from a safe distance, watched the lake drain. On another, he installed sensors on the bottom of an empty meltwater lake and, from a not so safe distance, waited for it to fill.

On a trip last year, Tedesco brought along a drone to measure albedo at the edge of the ice sheet. Melt along the edge is exposing more rock and soil; since these are darker than ice, they absorb more sunlight, fostering more melt. But even where plenty of ice remains the reflectivity of the surface is dropping.

"The surface is darkening from an energy point of view," Tedesco said. "It's basically like exposing a wound and then putting some salt in it."

Beyond its runway, Kangerlussuaq has one main attraction: a twenty-mile dirt road that leads away from the coast, toward the ice sheet. The road, improbably enough, was laid for Volkswagen, in the late nineties. As the story goes, the carmaker had a cold-weather test facility erected on the ice which included a track and a dorm for workers. But, after a few years, the whole scheme was abandoned. The road is now maintained by the municipality of Qeqqata, which encompasses Kangerlussuaq and is the size of Ohio.

"I'm always very emotional when I drive this road," Tedesco said the next morning, as we headed out. "It's my adopted land." As he drove, he described to me a scheme of his own—never realized—to establish a museum of Arctic smells. Archived fragrances might include the herbal scent of the tundra and the perfectly blank smell of the ice. It was Colosio's first visit to Greenland, and Tedesco warned him that the place had a mystical draw.

"You're going to want to keep coming back," he said. "You're going to be under the spell."

The old VW road runs almost due east, through a flat, sandy valley flanked by glacially smoothed hills. The area's native trees are all low and shrubby, but a few miles out of Kanger we came to a grove of introduced pines. The pines seemed to be thriving in the warming climate, and people had decorated some of them with Christmas ornaments. We passed an Arctic hare—very white and surprisingly large—and then a family of reindeer.

After about an hour, we reached a spot where, across the valley, a tongue of ice spilled over a ridge. Tedesco identified the tongue as belonging to the Russell Glacier. (In addition to the ice sheet, which is essentially one enormous glacier, Greenland also has thousands of smaller, peripheral glaciers.) We stopped to take a better look.

When Tedesco first travelled the VW road, Russell ended in a dramatic wall of ice. Now the wall is gone, and the glacier looks deflated—more like an ice doormat. Tedesco compared visiting Russell to calling on a friend with a terminal illness. "You have to have the strength to say goodbye," he said. "You see this and you say, 'Oh, man, it's happening really fast.' "

The VW road originally ran from Kangerlussuaq all the way to the ice sheet. Thanks to melt, it no longer gets there. Instead, it gives out a half mile short, at a huge pile of dirt and jumbled rock—a moraine in the making. We parked near an old bulldozer that seemed to be rusting into the ground. Tedesco and Colosio strapped on backpacks filled with equipment, and we began hiking over the rubble.

It was cloudy and a relatively balmy forty degrees. (The average annual temperature in Kangerlussuaq is around twenty-four degrees Fahrenheit, compared with about minus twenty degrees at Summit.) "Speriamo che non piova," Colosio remarked. (Let's hope it doesn't rain.) We reached the edge of the ice sheet, which was so thin that we could walk right onto it, as you would step onto a curb. There was meltwater everywhere, collecting in puddles and running in rivulets. In some places, the rivulets had merged to form streams that had to be jumped across.

"If we come back in a few days, we'll have to bring bathing suits," Tedesco joked, laying down his backpack.

Tedesco's goal for the expedition was to repeat the albedo measurements that he had made last year, to see how conditions had changed. Once again, he'd brought along his drone. It was equipped with two sets of sensors—one to measure incoming radiation, from the sun, the other to measure outgoing radiation, reflected off the ice. To calibrate the sensors, he laid out a plastic sheet checkered in black and white, like a signal flag. It was apparently a lot more high-tech than it looked. "That little square cost me two thousand bucks," he said.

While Tedesco and Colosio tinkered with the drone, I wandered around. At Summit, it's always very white because there's always—or nearly always—fresh snow, and everyone wears goggles or sunglasses to prevent snow blindness. At the ice sheet's ragged edge, whatever snow had fallen during the winter had, by mid-July, long melted away, and there was only ice, which came in many shades, all of them gray. The ice was speckled with bits of dust, which glaciologists call cryoconite, and pocked with cryoconite holes, which form because dust absorbs sunlight more efficiently than ice does. The surface was changing so quickly that I could watch as neighboring holes merged to form pools. I also came upon a much larger hole, maybe twenty feet across and perfectly round, that went straight down. It had, I figured, once been a moulin, which is a shaft tunnelled out by a river of meltwater. When they're full, moulins are spectacularly beautiful and equally dangerous. This one was empty and dingy, with blackened edges. It looked like some kind of side entrance to the underworld.

In 2015, when world leaders agreed in Paris to try to limit the global temperature increase to 1.5 degrees Celsius, it was hoped that this would minimize the damage from feedbacks and prevent the world from crossing dangerous thresholds. Already, by some measures, temperatures have crept above that limit. According to the research group Berkeley Earth, over the past year they have averaged 1.66 degrees above preindustrial levels, and according to Copernicus, part of the European Union's space program, they have averaged 1.64 degrees above those levels. "The 1.5-degree limit is deader than a doornail," the former NASA climate scientist James Hansen, who is sometimes referred to as the father of global warming, has said. The world might need to spend decades at 1.5 degrees before triggering the tipping points associated with this temperature, but that is slim comfort, as in coming decades temperatures will almost certainly continue to rise.

There is, in principle, nothing wrong with a warmer world, or even one whose climate bounces around. At many points in history, the Earth has been much hotter than it is today. During a period known as the Cretaceous Thermal Maximum, for instance, around ninety million years ago, breadfruit trees grew on Greenland, and a rain forest thrived on Antarctica. During the Wisconsin, as temperatures yo-yoed up and down, people were living in Africa, Europe, Asia, Australia, and quite possibly North and South America. Some of them managed to get by, or we wouldn't be here.

But the society we have now was built for the climate we have now, or at least a close approximation of it. Alter the world by, say, drowning Dhaka or Shanghai and all sorts of knock-on effects follow. Last year's report on global tipping points—the one produced by more than two hundred researchers—predicted that "escalating climate change" will increase "the risk of violent conflict" and that this, in turn, will "undermine societies' ability to cooperate," leading to yet more climate change.

The roughly six trillion tons of ice that Greenland has lost translates to enough water to cover the Eastern Seaboard to a depth of eighteen feet. Roughly half the loss has come from surface melt, the other half from an increase in the discharge of icebergs.

From Kangerlussuaq, I flew to the town of Ilulissat, which is sometimes called the "iceberg capital of the world." (The town's name, in fact, means "icebergs.") Some hundred and fifty miles north of Kanger, Ilulissat sits on Disko Bay, at the mouth of another very long fjord. Icebergs break off into the fjord and float along until they hit an underwater sill just south of town. The bigger icebergs get stuck on the sill, and other icebergs pile up behind them, in a great glacial traffic jam. A few years ago, the Greenlandic government opened a museum not far from the ice jam, and on my first day in town I went to talk to its director, Karl Sandgreen.

From the outside, the Icefjord Centre looks like a cross between a milking barn and a concert hall, with lots of metal beams and a roof that meets in a swale instead of a peak. At the entrance, visitors are instructed to remove their shoes and put on felt slippers. Inside, the design is pure Scandinavian minimalism.

Sandgreen met me in the museum's window-lined café. We chatted for a bit about Ilulissat's history prior to colonization. Just beyond the museum lie the remains of an ancient settlement that was serially occupied by three different cultures: the Saqqaq, the Dorset, and the Thule. The first two groups died out—the Saqqaq around three thousand years ago, the Dorset around a thousand years ago—for reasons that are unknown. Contemporary Greenlanders are descendants of the Thule, who arrived on the island from what's now the Canadian Arctic around the year 1200. (The Thule are sometimes called the proto-Inuit.) "We are from the Thule culture," Sandgreen told me.

Sandgreen, who is forty-five, was born and raised in Ilulissat, which was—and still is—an important fishery. His father fished for prawns in Disko Bay, and he would have done the same had his parents not encouraged him to get more education, in Denmark. "I'm very happy I listened to them," he said.

We went to take a look at the Centre's exhibits. There were some Thule artifacts and an animation showing how the calving front of the Jakobshavn Glacier—the source of Ilulissat's icebergs—has moved over time. Since the mid-nineteen-nineties, the front has retreated some fifteen miles up the fjord.

At the Centre's center, a set of glass cases displayed sections of an ice core from a site known as EGRIP. (The core, which was drilled by a Danish team about two hundred miles northeast of Summit, was completed just last year.) Though the cases were refrigerated, the ice cylinders were dripping. "We are very concerned about how they're melting so fast," Sandgreen said. "We're going to have to get some new ones."

Sandgreen told me that Ilulissat's climate had changed dramatically since he was a boy. It used to be that Disko Bay froze over every winter and people rode dogsleds over the ice to hunt. Now the bay no longer freezes, and it doesn't pay to keep dogs, so the town's canine population, which used to number almost ten thousand—twice as high as the number of humans—has dropped to around seventeen hundred.

"Also, the temperature," Sandgreen said. "When I was a teen-ager, I remember minus forty degrees Celsius was just normal. But after this climate change the air has become moister. So now, when it's, like, minus ten degrees, it feels colder than minus forty." (Minus forty degrees Celsius is also minus forty degrees Fahrenheit; minus ten degrees C is fourteen degrees F.)

I asked Sandgreen what message he hoped people would take from the Icefjord Centre. He told me that a lot of politicians had visited Ilulissat, including the former U.S. Secretary of State John Kerry. "They're coming here to see the climate change, to tell the rest of the people in the world what is happening," he said. But he didn't believe that it would make much difference: "I think we are too few people in Greenland to tell the people in the rest of the world to do something."

The next day, I went back to the Icefjord Centre, and then kept on walking, along a boardwalk that crosses a stretch of spongy tundra. The stretch, which borders the fjord, was the site of the ancient settlement, though you wouldn't know that by looking at it. In an apparent violation of the rules, two women in red vests were trudging along on the tundra itself. They turned out to be American scientists, there to study erosion. As the permafrost in the area degrades, the remains of the old settlement are, they told me, collapsing into the sea.

I kept walking. It was a spectacular day, and there was a wonderful scent in the air—a bit like thyme and a bit like lemon. I thought of Tedesco's museum of Arctic smells. The boardwalk curled east and then ascended a rocky ridge. From the ridgetop, there was a view directly onto the ice jam: a floating mountain range with slopes of pure white. The reflections of the icebergs quavered in the water, which was blue to the edge of purple. The smaller bergs were the size of a house; the bigger ones, I figured, were the size of Grand Central Terminal. A couple arrived right behind me. "Oh, my God," the woman exclaimed in American-accented English. "This is unbelievable!"

The icebergs' source, the Jakobshavn Glacier, is sometimes called an outlet glacier and sometimes an ice stream. Ice is always flowing, albeit slowly, from the center of the ice sheet toward the edges. In an ice stream, it flows particularly quickly. As Jakobshavn's calving front retreated, it also thinned, and the glacier's speed—already brisk for a block of ice—increased. In 2012, its flow rate exceeded a hundred and fifty feet a day, which is believed to be a glacial world record. Though it has since slowed down again, the glacier has still been losing a lot of ice—during the past few decades, some ninety billion tons of it. (All on its own, Jakobshavn is believed to be responsible for more than one per cent of global sea-level rise since 2000.)

More recently, one of the researchers involved in that proposal, John C. Moore, of the University of Lapland, in Finland, suggested an alternative fix: a three-mile-long curtain across the fjord. In an published earlier this year, in the journal Climatic Change, Moore and several colleagues examined dozens of "emergency measures" that have been proposed to help conserve the Arctic, which is warming four times faster than the global average. In addition to the underwater curtain, the schemes include pumping water onto the Arctic sea ice to thicken it and "brightening" the region's clouds so that they reflect more sunlight. The group argued that studying such "emergency measures" was important not because all—or even any—of them would work but because options are dwindling.

Interest in exploring such interventions is "most definitely growing," the group wrote, owing to the "increasingly dire findings of the effects of warming in the North, and the obvious global impacts of climate-related disasters."

I spent a few hours at the ice jam, basically just admiring the view. A cruise ship had docked in Ilulissat, and several people wearing jackets with the tour company's insignia joined me on the ridge. I asked a few of them why they had come to Greenland. An American woman told me she had already been to so many other exotic places, including Antarctica, that this was one of the few destinations left. A British woman said that climate change had influenced her choice: "We definitely made a point to see it now, before we lose it." A Norwegian man who had travelled on his own told me that he had come to visit his brother, who was working at a destination restaurant just south of Ilulissat. (When I later looked up the restaurant, I found that the tasting menu, wine included, ran to seven hundred and forty-five dollars and might feature whale skin, musk ox, and reindeer.)

Until the Second World War, Greenland was more or less inaccessible to outsiders. It could be reached only by boat, and the Danes, who at that point considered it a colony, made it hard for foreigners to visit, ostensibly to protect the island's population from the destructive trends of modernity. Many Greenlanders still lived, if not exactly as their ancestors had, without electricity or running water.

Following the war, Greenland modernized rapidly—so rapidly that one scholar described it as having transitioned "from the Middle Ages to the twentieth century in less than twenty-five years." Today, Greenland is eagerly courting foreigners. A new international airport has been built in Nuuk, and another is under construction in Ilulissat. From the ridge overlooking the ice jam, I could see a cloud of dust rising from the construction site.

The new airports will bring more visitors to Greenland to see the melting ice, and the increase in air travel will melt more ice—another potential feedback loop. No one desires this outcome—not the travellers and certainly not the Greenlanders, whose attachment to the ice is profound. But everyone pushes ahead anyway.

Climate change is not like other problems, and that is part of the problem. What it lacks in vividness and immediacy it makes up for in reach. Once the world's remaining mountain glaciers disappear, they won't be coming back. Nor will the coral reefs or the Amazon rain forest. If we cross the tipping point for the Greenland ice sheet, we may not even notice. And yet the world as we know it will be gone.