Greenland holds enough water to raise global sea levels seven meters, and southern Greenland is already showing accelerated melting. But the rate of this melting and other ice dynamics are poorly understood, partly because Greenland’s surface is so inscrutably white and featureless in ordinary satellite images. Now, a new image-processing approach gives a clearer view of subtle inland features, providing sharper clues into glacial movements–and better insight into future sea-level increases.
The technology starts with as many as 94 red and infrared images of the same region, taken by two NASA satellites, called Terra and Aqua, that have polar orbits and cross Greenland several times a day. Each raw image–a measure of light from the surface–has a resolution of 250 meters per pixel. But by aligning and averaging values within areas of pixel overlap among multiple images of the same area, researchers at the National Snow and Ice Data Center at the University of Colorado at Boulder tightened the resolution to as little as 100 meters per pixel and roughly quadrupled contrast sensitivity.
As one example of a payoff, researchers are finally getting a clear picture of a 600-by-50-kilometer eyedropper-shaped ice formation informally known as NEGIS (for Northeast Greenland Ice Stream). This massive feature–which is sliding toward the sea at a few hundred meters per year–wasn’t even known to science until 1991. And it hasn’t been imaged in detail until recent months. “What we’ve done now is see how far upstream it goes, how close it comes to the summit of Greenland, and see some structures at the edges, to get an idea [of] how fast ice flows” and what directions it flows in, says Ted Scambos, lead scientist and glaciologist at the Boulder center, who codeveloped the image-processing approach.
Scambos says that such insights are everything when it comes to finding out how fast Greenland’s ice will pour into the ocean and begin inundating the world’s coastlines. The same technology is being applied to images of Antarctica, whose ice sheet contains enough water to raise sea levels 65 meters if all of it melts. The rate of such melting is one of the most poorly understood yet most high-impact effects of global warming.
“This gives us better resolution of subtle structures in the interior of the ice sheet,” says Scambos. “To the naked eye it looks like a smooth white plain. But there are hills, bumps, and ridges that show us how the ice is flowing, and how it will drain out from glaciers. Once we get away from the coast, the features that are important have to do with how the ice flows. They can be very subtle–hills and valleys that show you how the ice is moving off the continent. What we’ve got is a map that shows details much further inland, much further than before. Other images just show the interior of the ice sheet as a blank white surface without any features.”
The technology makes use of existing Earth-observing satellites. But these aren’t the only ones up there. Other satellites, notably NASA’s Landsat and ASTER sensors, are also well-known for making sharp images of the home planet. The main advantage of Terra and Aqua, though, is the greater sensitivity to subtle light contrasts–a big help when the photographic subject is a vast white surface. Plus, Terra and Aqua are available more often. Landsat doesn’t cross the same spot more than once every 16 days. Since many satellite images are unusable because of cloud cover, as a practical matter it would take many hundreds of Landsat images to make a similar map, Scambos says.
The new approach also allows rapid reevaluation of the entire sheet on Greenland to detect important short-term changes. In fact, the technology allows scientists to build a new high-res picture of the entire sheet every two months. And if scientists decide that they’d like another look at a small area, other satellites can potentially be brought to bear.
The subject is of more than academic interest, notes Mark Fahnestock, a geologist at the University of New Hampshire, in Durham, who collaborated with Scambos on the technology. “Basically, the Greenland ice sheet is putting out–in the last six, seven years–40 percent more ice than it was ten years earlier,” Fahnestock says. “We are trying to understand why, so we can have some idea of how to project it into the future.” Once this understanding becomes clearer, scientists will be able to tell the world how fast and how far sea levels might rise. This might even prod policymakers to reduce greenhouse-gas emissions and plan for receding coastlines and the inundation of populated areas.
One of the disturbing trends in Greenland is the growth of huge lakes of ice melt that form atop the ice sheet during summer months. These masses of water find cracks and drain deep into the ice sheet, to uncertain effect. The new imaging technology can see such cracks and how they are changing, Fahnestock says.
At a high level, the technology can show ice as a kind of slow-motion river. “In a river you can see standing waves and rapids,” Fahnestock says. “It’s the same sort of picture of the ice, even though it is moving much slower. You see this bumpiness because this ice is in motion.” The rate of ice-sheet melting is poorly understood, and “knowing where it’s bumpy lets us figure out why Greenland is changing as fast as it is today,” he says.
How AI is reinventing what computers are
Three key ways artificial intelligence is changing what it means to compute.
These weird virtual creatures evolve their bodies to solve problems
They show how intelligence and body plans are closely linked—and could unlock AI for robots.
We reviewed three at-home covid tests. The results were mixed.
Over-the-counter coronavirus tests are finally available in the US. Some are more accurate and easier to use than others.
A horrifying new AI app swaps women into porn videos with a click
Deepfake researchers have long feared the day this would arrive.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.