Ocean warming threatens stability of Antarctic ice shelves by carving ‘upside-down rivers’ into their undersides

By Tom Yulsman | March 29, 2016 1:24 pm
Antarctic ice shelves

The Getz Ice Shelf extends several miles into the ocean along the western Antarctic coast. The vertical face of the ice shelf is almost 200 feet high and is estimated to extend another 1,000 feet below the ocean surface. This photo was taken from a NASA DC-8 by Ted Scambos, Lead Scientist at the National Snow and Ice Data Center.

Note: Thanks to a spring-break getaway, I’m just now catching up to this new research showing that warming ocean waters are threatening the stability of giant, floating shelves of ice fringing Antarctica. The post that follows offers a summary of the new findings, followed by a Q&A with the study’s main author. 

By carving giant channels into the undersides of Antarctica’s ice shelves, warming sea water is leaving some of them more vulnerable to disintegration — and raising new concerns about sea level rise.

“We found that warm ocean water is carving these ‘upside-down rivers,’ or basal channels, into the undersides of ice shelves all around the Antarctic continent,” says lead researcher, Karen Alley, a graduate research assistant at the National Snow and Ice Data Center, and a Ph.D. student at the University of Colorado in Boulder. (See note at the end of this post about my own connection to the University of Colorado.)

These shelves form over thousands of years as ice flows off the mighty ice sheets of Antarctica and into the ocean through outlet glaciers. They can extend over the water for many miles, with bergs calving from their faces. They are, in essence, floating extensions of the grounded ice on the continent.

Antarctic ice shelves

The cross-section above shows the transition from the grounded ice of an Antarctic ice sheet to a floating ice shelf. New research shows that warming Circumpolar Deep Water is carving channels into the undersides of Antarctic ice sheets, potentially weakening them and making them more vulnerable to disintegration. (Source: Bethan Davies, Antarcticglaciers.org. CC BY-NC-SA 3.0)

The ice shelves also act like dams that impound the glaciers behind them, slowing their movement to the sea. Previous research has shown that Antarctic ice shelves are thinning faster than previously thought. This is of concern because as they erode, glaciers can flow more quickly, releasing more ice into the sea and thereby raising sea level.

By weakening ice shelves from below, basal channels can only make the situation worse.

SEE ALSO: Antarctic ice shelves are thinning rapidly — and the losses are accelerating in West Antarctica

As ice flows off the continent and out over the water in fringing shelves, it slows down even more in places where it encounters obstacles such as islands and peninsulas. The new research, published in Nature Geoscience by Alley and her colleagues, suggests that as channels on the undersides of ice shelves grow, they are loosening the grip of these “pinning points.”

To understand the significance of this finding, imagine the proverbial Dutch boy who has stuck his finger in a dike to plug a little hole. If he removes his finger, the initial trickle of water through the hole would soon turn into a torrent. Similarly, if an ice shelf becomes unstuck from an island, peninsula or other pinning point, it will start flowing seaward faster — perhaps triggering an inevitable process of rapid disintegration.

As part of their research, Alley and her team used radar technology that can see into the ice,satellite imagery, and satellite laser altimetry, to map depressions indicative of channels in the undersides of ice shelves.

They discovered that the channels tend to form at the edges of pinning points. They also found that channels were carving upward into the rapidly thinning Getz Ice Shelf in West Antarctica at up to 33 feet a year. And in what might be the most worrisome finding, Alley and her colleagues discovered two places where ice shelves were actually fracturing along channels.

The following images offer graphic evidence of this process (please read the caption for details of what you’re looking at):

Antarctic ice shelves

The top image (a) shows a portion of the Getz Ice Shelf in West Antarctica, with the Scott Peninsula jutting out into it. The red line shows the path of a research flight over the area on Nov. 3, 2011. Directly below (b) is a cross section of the ice obtained with radar imaging during the flight. Two deep channels are visible in the base of the shelf on either side of the peninsula. Landsat satellite images of the area are next. The one on the left (b) was acquired in 2002; the one on the right (c) in 2014. Note the development of crevasses at the tip of the Scott Peninsula — where weakening from the channels has occurred. (Source: Karen Alley, NSIDC)

The situation in West Antarctica is particularly worrisome. Previous research has shown that the West Antarctic Ice Sheet has already passed a point of no return, beginning a long-term process of inevitable disintegration — scientists call it “collapse” — which could ultimately raise sea level by about 10 feet.

If sea level were to rise by that amount, the United States would lose 28,800 square miles of land on which some 12.3 million Americans live. Globally, literally hundreds of millions of people live in areas that would be inundated if sea level were to come up by that amount.

To explore these issues in greater depth, I emailed a series of questions to Karen Alley and her colleague Ted Scambos, lead scientist at the National Snow and Ice Data Center. Here is that Q&A:

Q: The West Antarctic Ice Sheet in particular has long been of concern. And in 2014 we had a pair of studies suggesting that the WAIS already had passed the point of no return and is now likely headed for inevitable “collapse.”  With that in mind, how do these new findings fit with the research conducted over the past few years that have raised concerns about the stability of the West Antarctic Ice Sheet?

Karen Alley: The recent research you mentioned does indeed strongly indicate that the West Antarctic Ice Sheet (WAIS) has entered a state of irreversible collapse. I think you’re also wise to put the word “collapse” in quotations, because to the general public that word tends to conjure up images of things falling apart very quickly, perhaps in a matter of minutes or seconds. The WAIS collapse will take much, much longer — on the order of many centuries. 

This means there are huge remaining questions to answer about exactly how soon and how fast the big changes will take place. These are currently very poorly constrained questions.

Ice on its own is remarkably complicated. In order to model ice flow in a truly accurate way, we need data on things like the temperature structure throughout the ice, the water content, how many impurities it has, what sort of bed it is sitting on, and even how the ice crystals are oriented. Models are, of course, complicated by the impacts that the atmosphere has on the ice, and whether air temperature is changing over time.

If you add an ocean to this equation (or, more accurately, this complex set of equations), the whole problem gets even more complicated. While we can understand the large picture of the West Antarctic Ice Sheet’s configuration, and we know that it is in a patently unstable position, predicting the details of exactly how that instability will play out is beyond our current data or computer capacity.

The basal channels in our study may be an important piece of the puzzle for predicting the big changes in WAIS and other parts of the Antarctic ice sheet. They are so far not included in most of the big ice sheet models.

We found the highest concentrations of large basal channels on the ice shelves bounding WAIS. We found that these channels are capable of growing very quickly. And we found that basal channels have a tendency to form along shear margins, which are usually already weak areas, raising stresses on the ice shelf to the point of fracturing the ice.

Could these basal channels be the weak points that break, allowing big pieces of WAIS to flow into the ocean more quickly and hastening the ice sheet’s inevitable collapse? Possibly. But we’ll need a lot more observation and modeling before we will know for sure.

Q: In the ‘upside down rivers’ identified in the study, how does water flow, in a sense, uphill? 

Karen Alley: On the earth’s surface, water obeys gravity. It flows into low spots, eroding those areas and making them lower as it continues on its way downhill. This directs even more water to these areas, and eventually you end up with a riverbed with lots of water flowing through it.

Beneath an ice shelf, water obeys buoyancy. More buoyant, or “lighter” water flows upward above “heavier” water. This light water finds high spots in the base of an ice shelf, melting those and making them higher, so more of the light water flows in. Because ice shelves are always melting as they flow, they get thinner towards the ice edge, so that direction is “uphill,” and it is the direction in which the water flows. 

In the Southern Ocean, the lightest waters tend to be relatively fresh, and the heavier waters are very salty. (Temperature can play a role, too, but some of the warmest waters around Antarctica — Circumpolar Deep Water, or CDW — actually tend to be quite dense and heavy because they are very salty).

When the deepest parts of an ice shelf melt a lot because of the presence of warm water, they create a lot of very fresh, light water. That water flows upwards, bringing some of the warm salty water with it, and melting a lot more ice as it goes. The continued melt keeps the plume fresh and buoyant, causing it to flow uphill towards the ice edge, and carving “upside-down rivers” as it goes.

Q: Where is the warmth that is responsible for carving these rivers coming from? How does it get to Antarctica, and what is its connection to human-caused climate change? 

Ted Scambos: In essence, the warm water ultimately comes from tropical surface waters, which have warmed and increased in saltiness over the past few decades.

But in point of fact, Circumpolar Deep Water (CDW) was always there, and it was always warm. But its warmth is increasing slowly in the present climate-change era.

The real source of the change that is driving this new wave of erosion of the ice shelves (and we know it’s new because we can see new erosion of these ice shelves) is a change in mean wind patterns. These are driving the surface water generally away from the continent and leading to more frequent events where warmer Circumpolar Deep Water is pulled onto the continental shelves to replace surface water that has blown out northward. There is also a notion that once started, the creation of fresh water under the shelves leads to a kind of pump that pulls more warm deep water into the sub-shelf cavity.

Q: How concerned should we be?

Karen Alley: It’s pretty clear now that WAIS is in trouble, no matter what we do. We should absolutely be concerned about WAIS and the consequences of its collapse, especially in terms of sea level rise. We should also be concerned about other areas of Antarctica, such as the region flowing into the Totten Glacier in East Antarctica, that may also become vulnerable to destabilization in the future. 

Whether or not basal channels are present, a warming climate means that WAIS and other parts of Antarctica will eventually end up in the oceans. Studying features like basal channels is part of a much larger effort to answer questions like: Which parts of the ice sheet will fall apart first? How fast will we lose them? When will sea level rise become a major issue for humans?

Basal channels might give us some pieces of these answers, because they may help us identify some of the most vulnerable areas and how close they are to actually falling apart.  More research in the future could also show that other features and weak points are more important for answering these questions. But with so much more left to learn, developing our understanding of every piece of the puzzle is important.

Note: In addition to producing ImaGeo, I direct the Center for Environmental Journalism at the University of Colorado, where Karen Alley is a Ph.D. student, and where the NSIDC is located.


  • Mike Richardson

    Another interesting, and concerning, positive feedback mechanism. Even years with a growth of surface Antarctic ocean ice could conceal the fact that the ice is actually thinner, and extending a greater distance over ice that continues to melt on the undersides of the shelves. I’ve seen reports on melt channels inland on the Greenland ice cap that could be speeding the movement of glaciers to the ocean, but this shows that the ocean has an increasing role to play in accelerating the meltdown. On that note, a similar effect of ocean currents and wave action would result in “burning the candle at both ends” on the Greenland cap. Hopefully we’ll get more information from researchers there, as well.

    • Tom Yulsman

      Thank you for your comments Mike. I appreciate it.

  • OWilson

    Ah, another “threat”.

    Why does these studies never refer to the latest studies from NASA – Antarctic Ice Growing, or the major study by International Scientists that found the Antarctic Ice is much thicker than previously though, or even that Antarctic Ice Cover Is Setting Records – NASA.

    And of course, the qualifier to at least relate wishful thinking on the part of a global warmer, to some semblance of reality”

    “”While NO ice shelves have completely disintegrated due to carving by basal channels, the study points to the NEED FOR MORE OBSERVATION AND STUDY OF THESE FEATURES, said co-author Helen Amanda Fricker of Scripps Institution of Oceanography at UC San Diego. “It’s FEASIBLE that as ocean temperatures around Antarctica continue to rise, melting in basal channels COULD contribute to increased erosion of ice shelves from below.” (caps mine)

    Now you know the rest of the story.

    But NASA also says Antarctic



ImaGeo is a visual blog focusing on the intersection of imagery, imagination and Earth. It focuses on spectacular visuals related to the science of our planet, with an emphasis (although not an exclusive one) on the unfolding Anthropocene Epoch.

About Tom Yulsman

Tom Yulsman is Director of the Center for Environmental Journalism and a Professor of Journalism at the University of Colorado, Boulder. He also continues to work as a science and environmental journalist with more than 30 years of experience producing content for major publications. His work has appeared in the New York Times, Washington Post, Audubon, Climate Central, Columbia Journalism Review, Discover, Nieman Reports, and many other publications. He has held a variety of editorial positions over the years, including a stint as editor-in-chief of Earth magazine. Yulsman has written one book: Origins: the Quest for Our Cosmic Roots, published by the Institute of Physics in 2003.


See More


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar