Icefin, a submersible robot, travels under Thwaites Glacier.Rob Robbins / United States Antarctic Program

The “grounding line” of Thwaites Glacier in West Antarctica is, without hyperbole, one of the most important places on one of the most important objects in the natural world. And scientists now have video of it for the first time.

Earlier this week, researchers published the first images of the place where this behemoth glacier, which is both one of the world’s largest and most vulnerable, sits on the deep seafloor and bleeds water into the ocean. In other words, they have made a film of sea-level rise in progress, showing ice from the land becoming water in the ocean.

But this may understate the importance of these visuals. Thwaites is one of a handful of glaciers in the world that could really matter in our lifetime: If it collapses, it could do so rapidly, cataclysmically spiking global sea levels during the next few decades. The question of whether it will collapse depends almost entirely on its grounding line and the processes shown in this video.

One of the scientists who drove Icefin, the submersible robot that actually took the video, told me that it felt like a “walking on the moon” moment for glaciology. But it struck me as more like the first photo of a black hole, published last year—the picture’s a little dusty and shadowy, sure, but it shows something ancient, mysterious, and capable of tremendous destruction. The difference is that this shadowy place is not safely far away in space, but only 10,000 miles or so from where you’re sitting right now.

Here’s the video:

The above footage debuted in a brief BBC report, which didn’t say much about what was actually happening in the video. So here’s a brief guide: What you’re looking at is warm ocean water working its way under Thwaites Glacier. The icy “ceiling” in the video is the bottom of the glacier’s ice shelf, the part of the glacier that floats in the water rather than sits on the seafloor. It’s more than 1,900 feet thick, and almost everything attached to the bottom of it was, only a few hours earlier, pressed into the Antarctic bedrock.

“The glacier is moving [several] meters a day, so that material we’re seeing at the grounding zone is brand new and is just exposed to the ocean,” Britney Schmidt, a glaciologist at the Georgia Institute of Technology, told me. She spoke by phone from McMurdo Station, the small American research base on the coast of Antarctica.

“We can definitely see it melting,” she said. “There are a few places where you can see streams of particles coming off the glaciers, textures and particles that tell us it’s melting pretty quickly and irregularly.” There’s also a moment, early in the video, that appears to show a barrier between the fresher water that’s coming off the ice shelf and the saltier water of the open ocean.

About 100 feet away from where the above video was shot, the glacier actually sits on the seafloor. The Icefin team hasn’t released video of that area yet, but they have released this still image:

This is the Icefin robot’s view of where Thwaites actually comes to rest on the seafloor, its “grounding zone.” It was taken in less than one meter (about 3.2 feet) of water. (Icefin / Britney Schmidt et al. / Michelle Babcock / Georgia Tech)

Thwaites Glacier is a scientific twofer. It is important, first, because it is huge. It contains enough fresh water to raise global sea levels by more than a foot and a half, and it braces the entire West Antarctic Ice Sheet, which could raise sea levels by almost 10 feet if it pooled away. But Thwaites is also important because it is physically mysterious. In its enormous size and ominous future rest the answers to some of the biggest unresolved questions in climate science. As Andrea Dutton, a geologist at the University of Wisconsin who was not connected to the new research, told me by email: “Understanding the dynamics where the Thwaites is pinned to the sea floor is crucial to predicting the future of the world’s coastlines.”

One of the biggest mysteries is more than 40 years old. In 1978, the British glaciologist John Mercer noticed that West Antarctica, which looks like a solid plain of ice from space, does not sit on a stable foundation. If you took all the ice away, he said, West Antarctica would look more like an archipelago of little rocky islands, and its seafloor would slope down toward the South Pole. That may sound like a geographic curiosity, but in a world like ours, it tees up a potential calamity.

If Thwaites and other West Antarctic glaciers were to start melting—perhaps due to a “CO2 greenhouse effect,” Mercer wrote—then they would begin to recede along that sloping seafloor. As the glaciers bled water, they would retreat, following the seabed as it got deeper and deeper. But as the seafloor deepened, so too would the glaciers, exposing more and more of their icy fronts to the warm ocean waters. These warm waters would cause the glaciers to melt even faster—leading them to retreat into even deeper waters, exposing even more of their ice. It would be an unstoppable feedback loop, Mercer warned. And once it started, it would result in the “rapid deglaciation of West Antarctica”—and become the “first disastrous consequence of rising CO2 levels.”

The cramped place shown in the film, with its dusty currents and dirty ceiling, is ground zero for that process. And before a few weeks ago, nobody had seen it before: “We’ve literally never been down to a grounding line with a camera,” Alexander Robel, a glaciologist at Georgia Tech who was not connected with the project, told me. Scientists have had to imagine it, using computer models and their knowledge of physics to constrain how it might work. Now there’s footage.

The grounding line was hard to access partly because Thwaites is hard to access. “It’s a thousand miles from the nearest field station, and in an area with notoriously bad weather,” Robel said. This year, the National Science Foundation and a number of other government-sponsored agencies tried to solve that problem by taking many teams there at once.

Schmidt and her colleagues spent four and a half weeks on the ice, from late December to January. But because of how long it took to set up the robot and drill through the ice, they conducted all of their five robotic dives—each of which takes nearly 12 hours—in just three days’ time. You can do the math on their sleep schedule. “We were pretty roasted,” she said.

But as they sat in a small tent in the blinding Antarctic midnight and drove the robot by remote, Schmidt and her colleagues were awed. “Oh my gosh, here’s the melting happening,” she remembers someone saying as they pointed at the screen. They had to invent names on the fly for shapes on the screen, noticing “cups,” “pebbles,” and dirty “scallops” in the ice.

Now will come the crucial work of taking the video and other measurements and figuring out what they mean for the world. “The grounding line is moving back—we know it’s happening but we don’t know how quickly,” Schmidt said. The difference could be enormous: Because small changes in sea level get amplified as storm surge, its retreat could mean the difference between seeing three and 15 feet of flooding during a hurricane, she said.

What’s already obvious from the images is that, when the glacier hits the ocean, it immediately begins to change. Part of the excitement of the video is that it forces scientists to ask some basic questions, starting with “What is that?” Robel, the Georgia Institute of Technology glaciologist, told me that he expects to see many of its images in conference presentations for years to come, as scientists debate what is even happening in them.

There’s already been at least one finding. Most glaciologists have called the area where a glacier meets the seafloor the “grounding line” (as I’ve done in this story). Others have argued that the area is more like the estuary of a river, with tides bringing in ocean water and carrying out meltwater, Robel said. The new footage seems to show something more like a “grounding zone,” a pockmarked area where the glacier makes contact and floats, before getting stuck on other boulders. But that is really only the beginning of what we may learn, said Schmidt, who will leave Antarctica for the first time this year on Friday, before flying home from New Zealand next week. “I haven’t had time to sit and think about this data at all,” she said. “But every time I watch these videos, I see something new.”

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.