At 85 degrees north latitude, specialist Nikolay Vokuev hung out of our helicopter’s open door, a strap tethered around his waist, and threw a burning flare onto the snow-covered ice floating on the Arctic Ocean a few meters below. The flare hit and started to smoke, darkening the pristine, crunchy white surface.

We were far north, hundreds of kilometers from the nearest landmass and roughly 50 kilometers from the Akademik Fedorov, the Russian icebreaker that was our home base. We were about to try a daring landing, setting the Russian Mi-8 helicopter down on a raft of ice that might be too thin to support the chopper’s weight. The maneuver was dangerous but unavoidable if we wanted to directly test the ice floe’s thickness.

The pilots swerved the helicopter into the wind (revealed by the flare’s smoke) and hovered just centimeters above the ice—touching down so softly that I barely felt it. They left the blades rotating to provide enough lift to lessen our weight and to stay poised, in case we needed to take off immediately.

The floe held firm, so Thomasz Petrovsky, a Russian sea-ice expert, jumped through the open door. He was light on the surface, working quickly to bore several holes next to the helicopter to find out how thick the ice was. Each time, his drill moved slowly but then jolted downward after merely a half-meter. He had hit the water underneath—and too soon. Petrovsky was looking for ice that was a meter and a half thick—solid enough to stabilize a massive ship—but every floe thus far had been thin. He returned to the helicopter, and we took off for the icebreaker as the fog began to close in.

The Search for Good Ice

Although it is easy to picture the Arctic ice cap as vast and impenetrable, it is actually a loosely knit collage of thin and small floes that constantly jostle about. From the air, it looks almost like a cracked eggshell at the top of our world. And those floes are shrinking.

On September 20 and 21, two icebreakers—the Akademik Fedorov from Russia and the Polarstern from Germany—sailed from Tromsø, Norway, on a mission taking them toward the central Arctic. The first task was to find an ice floe so strong and thick that it would ultimately freeze around the Polarstern, entrapping the vessel within the ice for a full year.

This unique platform will allow the mission (known as the Multidisciplinary Drifting Observatory for the Study of Arctic Climate, or MOSAiC) to study the Arctic in unprecedented detail—providing information that has been mostly nonexistent. Experts have never been able to study so many details of the climate system, from the ocean below to the clouds above, within the central Arctic across a full year. The findings will help scientists better understand how changes in the High North affect the rest of the planet.

Finding a suitable floe was far from easy. This summer very little ice survived the unusual Arctic heatwaves, following an ongoing downward trend in ice cover. I spent the initial two weeks of the mission onboard the Akademik Fedorov, first sailing through the open waters of the Barents Sea and then the Kara Sea—a region that has warmed by five to six degrees Celsius on average in recent decades, far above the global norm. “We’re in the hotspot of climate change,” says Michel Tsamados, a sea-ice expert at University College London, who is onboard the Russian ship.

As such, the ice edge has receded hundreds of kilometers farther north over the past few decades, forcing us to travel onward. Still, when we finally hit the edge on September 25, the ice was formidable. Every time we crashed through a particularly large chunk, the ship vibrated violently, sending drawers, personal items and even the bolted furniture into motion. But that was nothing compared with the racket. When ice sheared across our heavy steel hull, it sounded like a shovel scraped across hard pavement but hundreds of times louder. At its worst, the sound sent my heart rate skyrocketing. At its best, it kept me awake in the middle of the night.

Over the next few days, we snaked between the floes in order to travel through open water whenever possible. Although the Akademik Fedorov can break through ice 1.5 meters thick, the captain does so only if necessary. So we sailed slowly and methodically—allowing me time to take in the icy seascape. At its thinnest, the ice looked like a black-sand beach with sparkling white waves in the distance. But as we approached, the “waves” transformed into a sculpture garden of ice that was occasionally crossed by polar bears or dotted with a rare melt pond—a dazzling turquoise pool of water. It was a frozen wonderland that left me mesmerized.

The next few days passed quickly as we continued our eastward journey—crossing time zones that are very narrow at this latitude and losing an hour of sleep on a near-daily basis. But on September 28 the ship slowed to a stop at a point around 85 degrees north latitude and 125 degrees east longitude. It is from here that prevailing winds and waves, over an entire year, typically shepherd the overall mass of floes up toward the North Pole and then southward, in the direction of open water between Greenland and the Svalbard archipelago. And it was here that MOSAiC would begin its mission, as soon as project coordinators found a suitable floe.

For the past several months, Petrovsky and others had scoured satellite images for potential ice floes. But the only way to assess a floe’s true character is to see it firsthand. So on the morning of October 1 we flew the helicopter north of the Akademik Fedorov to a region where we could drill into five different floes to measure their thickness. But every one was less than half a meter on average—too flimsy to keep a ship stable.

We returned to the icebreaker, where the atmosphere onboard—already dispirited from previous searches—turned sour. “This is also one of the consequences of climate change,” said Jari Haapala, a sea-ice scientist at the Finnish Meteorological Institute, who is also part of MOSAiC. “The ice is not healthy.”

From Ice-Bound to Ice-Covered

Little did we know that the Polarstern had caught a lucky break. While we on the Akademik Fedorov had been scanning one local area, scientists onboard the nearby German ship had been checking a strange floe southeast of us. On radar images, it appeared as dark as the other potential sites (a sign that it was so thin that it was probably flooded), but it also hosted an unusual bright region. Scientists made landfall—hiking across the floe one day and taking snowmobiles the next—and found that ice in the region was four to five meters thick.

Because communication was mostly limited to short e-mails, those of us onboard the Akademik Fedorov did not know about the Polarstern’s discovery until October 2, when the two icebreakers met side by side, and the crews could discuss their options. Together they had assessed 16 floes, but only the Polarstern’s find, with its thick, white upper region—now dubbed the “Fortress”—could stabilize the German icebreaker. The news sent a wave of relief across everyone present. “It’s like a hidden treasure,” said Thomas Krumpen, our cruise leader on the Akademik Fedorov, during a meeting. “We can be quite lucky that something like this was discovered.”

With the floe selected, the mission could finally begin in earnest. The ships remained tied together for two days, swapping gas cylinders, snowcats (truck-sized vehicles on treads) and even passengers via a large crane. I moved to the Polarstern, where I would spend the next two weeks, via a “mummy chair”—a large birdcage that lifted me high into the air, moved over the slit of open water between the ships and placed me carefully on the neighboring helicopter deck. As I settled into my new cabin, the two ships went their separate ways—the Polarstern toward the chosen floe, and the Akademik Fedorov toward another one on the horizon. The latter would stop there and at several other distant points so researchers could set up a network of drifting instrument stations that circle the Polarstern as the central observatory.

We reached the Fortress early in the evening of October 4, crashing into the 1.5-meter-thick ice, which caused the ship to rattle as that ice crumbled beneath the hull and was thrust to the sides. After 300 meters of this action, the captain slowed us to a stop and killed the engines. Their hum faded as the ship fell back ever so slightly, settling into place. We were finally entombed within the ice.

For the first time on our voyage, the ship was still—almost as though we were back in port, except our surroundings were truly alien. The “skyline” in the distance around us was white and relatively smooth, with rolling hills, but the ice right next to the ship was craggy and raised—it had been overturned when we plowed through it. On the starboard side, there was a stunning chunk of ice with a turquoise stripe embedded between layers of white. The ship, too, quickly transformed into a frozen castle, with frost flowers dangling from the railings and ropes. But no sight was as foreign as the sun’s oblate disk, which hugged the horizon, traveling not up or down but sideways. Later it  disappeared, and polar night would quickly settle over the floe.

With that transformation began the scramble to set up the central observatory while scientists could still work in twilight. They began building roads with snowmobiles, shuffling equipment to a number of sites where different teams would operate and running power lines from the ship to those sites. The work was vast, and on most days we felt as though we were building a city in the Arctic. Soon there would be a remotely operated vehicle that swims in the ocean, instruments for monitoring the sea ice and a tower that measures all aspects of the atmosphere. Over the next year, hundreds of scientists will come and go as they conduct research. The effort is the largest and most ambitious climate-change research expedition the Arctic has ever seen, providing data that will be fed into climate models for years to come.

But all of that hinges on the stability of the ice. On October 5 the crew drilled holes into the floe, dropped anchors through them and cinched the ropes—causing the Polarstern to shudder one last time as it became as stable as possible. The little open water behind the ship froze, further locking us in place. But the Arctic is unpredictable. Three days later a storm with winds of roughly 35 kilometers per hour began to pummel the floe—opening a crack within the ice that ran toward our bow, along the port side of the Polarstern and behind its stern. In one hour, the crack widened by five meters, and the icebreaker dipped several meters to the port side, pulling the starboard ropes taut. The storm left a ship-shaped hole in the floe that was full of open water and fallen ice—including that beautiful turquoise piece that had once been pressed tightly against us.

Since then the winds have ceased and the temperature has dropped even further, causing the crack of ice to freeze again and even close, pushing up a ridge of small mountains. But scientists cannot say how long we might remain stable. We are now in the clutches of the Arctic.