Antarctica’s glaciers are under threat, but not in the way you’re thinking: The problem isn’t so much that the sun’s beating down on them, but that the warming sea is uppercutting them. The bit of a glacier that’s resting on land is known as an ice sheet, and the bit floating on the ocean is the ice shelf. The exact divider between them, where the ice lifts off, is called the grounding line. As the world rapidly warms, that line is falling back. And as a result, Antarctica’s glaciers may be degrading far faster than scientists anticipated.
Think of an ice shelf as a cork that’s keeping the rest of the glacier, that ice sheet, from sliding into the ocean. The Florida-sized Thwaites Glacier, for instance, is known as the “Doomsday Glacier” for good reason: It’s attached to a seamount off the coast and is holding back ice that would raise global sea levels by two feet if it all melted. Last month, scientists reported that Thwaites’ ice shelf could crumble in three to five years.
But current glacier melt models don’t account for a phenomenon called tidal pumping. Whenever the tide rises, it heaves Thwaites’ ice sheet upward, allowing relatively warm seawater to rush farther upstream underneath the glacier. That drives melting along its belly, making the ice sheet more prone to fracture. “It means that warm water that is at the bottom of the glacier can infiltrate up to several kilometers upstream,” says University of Houston physicist Pietro Milillo, who is studying Antarctic glaciers. “And all of a sudden you start realizing, ‘Wait a minute! The models that actually predict the future state of the glaciers do not have these kinds of phenomena. They basically have a grounding line that is fixed.’”
Last month, Milillo and other scientists reported that tidal pumping is forcing the rapid retreat of the grounding lines of other West Antarctica glaciers—Pope, Smith, and Kohler. Using a satellite that fired radar waves at the ice, the scientists could detect minute changes in elevation along each grounding line. “When the tide gets higher, the entire ice shelf lifts up,” says Milillo, lead author of a paper describing the work in the journal Nature Geoscience. “So by measuring how much it moves at the top because of the tides, we’re able to actually see where the grounding line is at the bottom of the glacier.”
The measurements are dire. In 2017, Pope’s grounding line fell back over two miles in just three and a half months. Between 2016 and 2018, Smith logged a mile and a quarter retreat a year, while Kohler pulled back three quarters of a mile. And when that grounding line starts retreating, it initiates a cascade of catastrophes: The more of the glacier’s underside that’s exposed to seawater, the more melting. “Once you trigger a subtle retreat, they're going to just keep retreating and retreating, which means that they're going to keep speeding up,” says Milillo. “Speeding up the glacier acts like a chewing gum: The glacier thins, and by thinning also you have a speed-up, because while not in contact with the bed, there is less resistance to the flow. Which means the glacier [movement] will accelerate and in turn will inject more ice into the ocean.”
The grounding lines of these neighboring glaciers might even withdraw to the point where they actually merge. “That will take a long time, probably. But if that were to happen—I'm not saying it's going to—that's when you get this mega problem,” says Peter Washam, an oceanographer and climate scientist at Cornell University, who studies Thwaites but wasn’t involved in this new research. “The fear with Thwaites is that as you move upstream, it pulls such a large area of ice that, once you begin to pull that quickly, you can sort of envelop the glaciers around it.”
Think of it like a watershed, in which several creeks drain into a larger river, but instead of liquid water it's (slowly) flowing ice. “If you unplug Thwaites, you're pulling the cork out of the drain,” says Lizzy Clyne, a geophysicist and glaciologist at Lewis and Clark College, who researches the glacier but wasn't involved in this new work. “Then you allow the ice that was previously flowing in different directions to be like, 'Well, the wall behind me went away, so now I'm going to fall back into Thwaites.' And you can therefore in theory tap on a lot more ice.” If Thwaites and its surrounding glaciers are destroyed, collectively they could add 10 feet to sea levels.
Last week, another paper from researchers at the Georgia Institute of Technology, CalTech, and Dartmouth College modeled how warm seawater is likely even squeezing past the grounding line, accelerating melting even further. Scientists previously thought that the grounding line acts as a kind of barrier to keep seawater from slipping underneath the ice sheet resting on the ground. But this new mathematical modeling suggests that if the ground is flat or “retrograde,” meaning it slopes deeper into the interior of the ice sheet—and both apply to these glaciers in West Antarctica—saltwater can indeed intrude past the grounding line. Like, way past.
In these conditions, and if the freshwater flow from melting ice is not too fast, seawater should be able to invade at least hundreds of feet past the grounding line, and probably miles, says Alexander Robel, head of the Ice and Climate Group at Georgia Tech and lead author of the new paper, published in the journal The Cryosphere. Yet, like tidal pumping, this phenomenon also isn’t represented in current models of glacial melt in Antarctica. “This is based on the prior assumption that basically there's a hydraulic barrier at the grounding line, and seawater never gets upstream,” says Robel.
There is one modeling exception, but it happened by chance. A 2019 paper from an international team of scientists compared a bunch of different models and noted that one accidentally produced the same kind of melt as intrusion, Robel says. (The reason some models diverge over these factors has to do with technical quirks regarding how to represent a glacier as a grid.) This paper showed intrusion could double the amount of glacier melt. “If seawater intrusion is causing melt upstream of the grounding line, the rates of sea level rise that you would project from places like Antarctica would be up to twice as much,” says Robel.
Specifically, without factoring in this kind of melting, the model projected that Antarctica’s glaciers might contribute between 3.5 and 6.7 inches to sea level rise by the year 2100. But with intrusion-like melting, that doubles to 8.3 and 11 inches. If his team’s new paper is correct in showing that seawater is indeed pushing past the grounding line and causing melt upstream, Robel says, “then it's not crazy that these models could be producing much higher rates of sea level rise.” (It’s worth noting that even small changes in sea level are catastrophic, particularly in low-lying areas where a fraction of an inch goes a long way.)
The model that accounts for extra melting also happened to better explain extreme sea level rise in the past. Some 3 million years ago, for instance, the world was a 3 degrees Celsius warmer (the Paris Agreement calls for keeping temperatures below 1.5 degrees C above preindustrial levels) and the seas 100 feet higher. “That's been a puzzle, to explain exactly why sea levels were so much higher,” says Robel. Previously, he says, if you tried using ice sheet models that didn’t account for seawater intrusion and the associated glacial melt, “when you subjected them to these warmer temperatures, they would not melt enough to explain this much higher sea level during past warm periods.” (As an alternative, some other models can achieve the same result by greatly increasing the rate at which icebergs fracture at the edge of ice sheets.)
While Robel’s group was making a mathematical prediction, other scientists have also been accidentally finding hints of evidence of seawater intrusion from their fieldwork in Antarctica. Using ground-penetrating radar, they send pings through glaciers and analyze what bounces back, or they set off explosions in the ice and analyze the seismic data. Both are good ways to measure where the grounding line is: The signal is different if it bounces off underlying rock versus saltwater from the sea. If the grounding line is indeed acting as a barrier to keep out saltwater, you’d expect the signal to change as you cross the line.
But that signal doesn’t tend to change at the grounding line, says Robel. Instead, the change often becomes noticeable miles upstream. “I think there is this now diversity of evidence, particularly in West Antarctica, using different observational methods, different instrumental methods, that indicate that there are definitely places where it looks like there's seawater and melt upstream of the grounding line,” he says.
Since these studies by other groups weren’t actually looking for this signal, now the next step is to get teams out on these glaciers to do experiments specifically designed to hunt for seawater intrusion. “It's more of the beginning of a scientific story than it is the end of a scientific story—of, Aha, we solved the problem!” says Robel. “We think that there is something interesting here. Now we need to really go figure out whether this is something that's happening in the real world.”
- 📩 The latest on tech, science, and more: Get our newsletters!
- Ada Palmer and the weird hand of progress
- You (might) need a patent for that woolly mammoth
- Sony's AI drives a race car like a champ
- How to sell your old smartwatch or fitness tracker
- Crypto is funding Ukraine's defense and hacktivists
- 👁️ Explore AI like never before with our new database
- 🏃🏽♀️ Want the best tools to get healthy? Check out our Gear team’s picks for the best fitness trackers, running gear (including shoes and socks), and best headphones
