Log in | Jump |


News & Perspective from the Center for Environmental Journalism
This item was posted on October 23, 2010, and it was categorized as Arctic, Greenland, Tipping points, sea level rise.
You can follow comments through the RSS 2.0 feed. Both comments and trackbacks are closed.

Greenland’s Jakobshavn could be poised to speed up. So could even bigger glaciers farther north — with significant implications for global sea level

UPDATE 10/25/10: I’ve made a correction in this post. The original version said the fjord through which the Jakobshavn glacier flows is 1.6 meters below sea level. That was incorrect. It’s 1.6 kilometers.

The Jakobshavn Glacier in Greenland already flows so fast you can stand on a ridge above it and watch it move. In fact, it’s the fastest moving glacier on Earth, flowing from the land to the sea at more than 14 kilometers per year.

But we may not have seen anything yet. The map above shows why it could be primed to move even faster, and thereby dump even more ice into the sea.

And if the same process plays out with glaciers farther north — and there is growing evidence that it may do just that (see the map and discussion lower in this post) — there would be significant implications for global sea level.

The image above is a kind of topographic map showing the elevation of the bedrock below and around the Jakobshavn Glacier — as if Greenland’s overlying ice had been removed. Warm colors show areas of the bedrock that are above sea level. The green and blue colors show areas where it is below sea level. The fjord through which Jakobshavn flows out to sea on Greenland’s western coast is visible as the green and blue trough at the left side of the image. As is evident from the map, the fjord actually continues inland for another 50 miles, and to a depth of 1.6 kilometers below sea level. Why is this significant?

The calving front of the glacier marked on the map shows the current edge of the glacier. This is where ice bergs calve from Jakobshaven and float out to sea. Since 2001, the calving front has retreated about 6 miles (10 kilometers). During July 6 and 7 this past summer, it pulled back an entire mile. Now, the calving front is poised right at the western edge of the fjord’s continuation into the ice sheet.

“If the calving front retreats from where it is now, it can draw out a lot of ice,” says Konrad Steffen, director of the Cooperative Institute for Research in Environmental Sciences at the University of Colorado.

The effect will be like pulling the plug from the drain of a bathtub. It will mean that much more ice will be able to drain from the Greenland Ice Sheet through Jakobshavn into the sea.

“You can actually pull out a lot of ice with a fjord below sea level like this,” Steffen says. “And there are about four or five others like this.”

The image at left is from a presentation Steffen gave last Thursday at the Center for Environmental Journalism. Like the image at the top of this post, it is a topographic map of bedrock. This is the topography of Greenland with all the ice removed.

The blue areas on the map show regions where the enormous weight of the ice sheet — which is almost 2 miles thick at its maximum — pushes the bedrock below sea level. And the numbers mark the major fjords through which glaciers drain into the sea.

Because the ice must pass through such narrow outlets, these fjords are the bathtub plugs. But if the calving fronts of the glaciers were to retreat far enough into the interior, the plugs would be removed, and the interior of the ice sheet could then drain much more quickly.

Jakobshavn Glacier on the southwestern coast is marked number 4. And as the map shows, there are several more farther north. What’s happening to them?

The graphic below addresses this question. It is based on data collected by NASA’s Grace satellites. (The satellites, which fly in formation together and use gravity measurements to detect changes in mass below them, are seen in the illustration to the right.)

In the map on the left, pink and blue colors indicate where Greenland was losing ice mass between 2003 and 2007. Most of the loss was concentrated on the southeast coast. As the map to the right indicates, by this past March loss of ice had extended all the way up to the northwestern coast — where at least two of the big fjords of concern are located. One of these is the Petermann Glacier (marked number 2 on the bedrock topography map).

In early August, an ice island four times the size of Manhattan Island calved from the front of the Petermann glacier. With this, the glacier lost about one-quarter of its 43-mile long floating ice-shelf. “The freshwater stored in this ice island could keep the Delaware or Hudson rivers flowing for more than two years. It could also keep all U.S. public tap water flowing for 120 days,” says Andreas Muenchow of the University of Delaware’s College of Earth,Ocean and Environment (quoted in a press release).

Petermann and the other more northerly glaciers “go farther inland, are much broader, and go much lower below sea level than Jakobshavn,” Steffen says. (In fact, whereas Jakobshavn is about 7.5 miles across, Petermann is about 60 miles wide.) Right now, these larger rivers of ice are not moving as quickly as Jakobshavn. But if their ice fronts were to retreat more rapidly, and the glaciers were to speed up enough, they could “drain most of northern Greenland in decades or a century.”

And that could have a significant impact on global sea level. How significant? For insight into that question, please see my previous post, which was also based on Steffen’s presentation to our group.

This item was posted by .

You can follow comments through the RSS 2.0 feed. Both comments and trackbacks are closed.

This thing has 16 Comments

  1. Steve Bloom
    Posted October 24, 2010 at 12:33 pm | Permalink

    Ah, that depth would really be significant if it were 1.6 *kilo*meters! :)

    Did he mention Pfeffer et al. during his presentation, BTW? In the case of the Jakobshavn in particular, I think they rely on that tight spot to keep things from quickly backing up much farther that they have already.

    Anyway, really great post, but it still does leave that hanging question I mentioned in the previous post.

  2. Posted October 24, 2010 at 12:42 pm | Permalink

    Steve: I’m planning to speak with Pfeffer. But at the moment I’m sick as a dog, and behind on grading papers. So it’ll have to wait until mid-week maybe.

  3. Steve Bloom
    Posted October 24, 2010 at 2:46 pm | Permalink

    Thanks, I’ll look forward to it. Get better soon.

  4. Aaron Lewis
    Posted October 25, 2010 at 10:43 am | Permalink

    Ice is a fluid with structural and mechanical strength that is dependent on its heat content. The more heat, the weaker the ice. When ice warms to near its melting point, then the potential energy of the ice can provide the energy to drive a progressive structural failure with the ice sheet collapsing into a slurry.

    We now have 3 plumes of warm, moist air (seasonally) blowing across Greenland. Every gram of water vapor that condenses on the ice melts 7.5 gram of ice resulting in 8.5 gram of water absorbing sunlight. When that water falls through a moulin, its heat and potential energy is transferred to the interior of the ice sheet.

    Recently, we have not seen progressive structural failure of ice sheets because ice is a good insulator. Thus, glaciers melt from the outside in, with the interior retaining more mechanical strength as they flow form a colder region to a warmer (lower) region. In a glacier, only the lower tip, the calving face has a heat content of 0C plus some heat of fusion. However, in the current case of Greenland, we are warming very large volumes of ice all at once. And, this ice is currently supported by the (weak) ice around it.

    At some point, a moulin will disrupt this mutual support, and instead of the moulin healing after the surface water drains, the ice will start calving into the moulin, and the released potential energy will shock the surrounding ice, resulting in additional calving. Then, the ice around the moulin will undergo progressive structural collapse, rapidly forming a slurry of ice and water.

    Ice is not a good dam material. Ice will not dam water with a head of more than ~6 meters. Where we have large glacial lake outburst floods, part of the “ice dam” is actually terminal moraine – rock, not ice. And, such dams composed of terminal moraine and ice do not last very long and are less than 30 meters high rather than the many hundreds proposed for the great ice dams. Thus, the great melt water pulses were not ice dams breaking, but ice sheets undergoing progressive collapse. The wave benches from Lake Missoula were from shallow lakes sitting on of top hundreds of feet of ice. That canyon of ice broke up suddenly, and repeatedly, to form the Lake Missoula Floods. Similar, but much larger flow channels surround Greenland. Yes, over the next few years we can expect parts of the GIS to move much faster.

    The rock guys all say, “We have not seen progressive collapse of ice before,” to which I would reply, “Where are your notes from the Missoula Floods?

    The moulins on Greenland prove that ice dams cannot support large heads of liquid water.

    Feynman did the basic calcs in 1964.

  5. Steve Bloom
    Posted October 25, 2010 at 1:00 pm | Permalink

    Tom, just to clarify my comment above about the typo, it’s Steffen’s typo assuming that the graphic at the top of the post is one of his slides. Note that the correct 1.6 km depth figure appears at the lower right.

    Aaron, is that based on a paper? Also, could you provide a specific reference re Feynman?

  6. Posted October 25, 2010 at 2:11 pm | Permalink

    Steve: I’m checking with Konrad on a number of things. I’ve got some questions out to him, and will clarify things soon.

    First and foremost, I’m hoping to clarify his statement of up to 3 meters of sea level rise. I do not fully understand where this comes from. But 2 meters comes from Vermeer and Rahmstorf’s statistical modeling paper (PNAS 􏰀 December 22, 2009 􏰀 vol. 106 􏰀 no. 51 􏰀 21527–21532). What about the extra 1 meter? I suspect this is could be an educated guess based on the potential impact of dynamic processes (and also for beyond 2100). But I’m just not sure. I’ve asked Konrad. With a bit of luck, I’ll get an answer soon.

    As for meters versus kilometers, I’m clarifying that with Konrad as well. He did say 1.6 meters in his talk, but that does sound insignificant, especially considering the size of the calving icebergs from Jokobshavn. On the other hand, 1.6 kilometers sounds like quite an extreme depth. It may well be right, though. I hope to have an answer soon.

  7. Steve Bloom
    Posted October 25, 2010 at 4:28 pm | Permalink

    Thanks, Tom.

    It sounds as if the 2-3 meters was a reference to a north Greenland contribution only, which would make for a possible global SLR of at least ~5 meters given everything else (which note, oddly, is the same figure Hansen pulled out of his nether region as a blue-sky worst-case scenario for this century and got huge flack for). But anyway, I assume we’ll be getting a definitive statement that will clarify all.

    I just saw a one-liner from Jason Box in an MSNBC story that SLR estimates would have to go up, but there were no specifics.

    If these guys are speaking out based on a radical departure of observed Greenland behavior in just this season, then yikes.

    BTW, a minor point I forgot to mention before: Your first post implied that the ice was all that’s keeping central Greenland below sea level. My understanding (no reference, sorry, but obviously Steffen would know) is that post-melt Greenland would be an inland sea surrounded by a mountainous archipelago. Possibly the inner part would come above current sea level, but the final configuration would include a 70-meter increase in sea level (and whatever degree of PGR is involved). Possibly there would be more exposure at intermediate stages of melt, though.

  8. Steve Bloom
    Posted October 25, 2010 at 4:39 pm | Permalink

    But looking again I see that the 2-3 meters was in reference to the statistical modeling, which in turn has to be a reference to V&R’s global figure.

    Even so a loss of the north Greenland ice has to be on the order of three meters, and the idea that such a thing is within the realm of possibility within decades is alarming in the extreme.

    I look forward to all being clarified.

  9. Posted October 25, 2010 at 6:42 pm | Permalink

    I finally got some clarification from Steffen. 1.6 kilometers is correct, so thank you Steve for pointing that out. I’ve made the correction. Also, I clarified the sea level rise issue in the previous post. I won’t try to explain it here. Just head over there and see the corrected version.

  10. spyder
    Posted October 26, 2010 at 11:25 am | Permalink

    Given that a one meter rise in sea level will induce staggering costs to lives around the Earth, the shock that a two meter rise is possible in the next few decades (assuming that Antarctica is having similar glacial melt issues), is unthinkably bad. Keep up the good and dedicated work Tom.

  11. Susan Anderson
    Posted October 27, 2010 at 8:33 am | Permalink

    This is terrific! Thanks.

    And thanks also to Aaron Lewis, along with the correct inference from Feynman, a darling of deniers, which always gets my nanny because he would spin in his grave.

  12. Icecap
    Posted October 27, 2010 at 10:44 am | Permalink

    Good article and interesting comments (Aron)!

    Has the moulins been proven to end up in the sea?

    If, not there might be some truble in the comming years?

  13. Icecap
    Posted October 27, 2010 at 11:01 am | Permalink

    One more point…

    I would like to see an article “Every Thing You Always Wanted to Know About Moulins, but where afraid to ask”.

  14. Posted October 28, 2010 at 3:34 am | Permalink

    The above Aaron comment just does not make sense. We have glacier in all sorts of climate settings including relatively temperate settings with glaciers calving into the ocean and in tropical settings with very intense solar radiation. In none of these settings is the structure of the ice weakened beyond the immediate by melting in a substantial way that would lead to its losing its solid strength and becoming somewhat of a slurry as suggested. I have spent plenty of days on glacier when the temperature was in the 70′s and melwater is running everywhere and the ice we were on was plenty strong. Drop an ice cube in a glass and the ice does not disinegrate into shards. The glacier on Eyjafjallajökull did not just turn into a slurry despite this massive heat induced meltwater running over and through the Gigjokull Glacier

  15. Susan Anderson
    Posted November 1, 2010 at 11:43 am | Permalink

    re Lewis-Pelto comments above. I thought about this and finally got up the courage to ask a physicist friend some dumb questions, like does an ice cube ever melt from the center. Answer: no

    Nonetheless, I think it’s worth keeping in mind that there are multiple influences, including the different sources of warmer air/currents, the mix of other materials, and the honeycomb effect of some of the moulins.

    Thanks for the education, and please forgive me for my sloppy – almost nonexistent – “science”.

  16. Susan Anderson
    Posted November 1, 2010 at 11:46 am | Permalink

    Forgot to mention, another issue is timeframe. When a scientists says rapid, he may mean thousands (or more) of years, or at least hundred or tens of years (a thunderclap of time in geological time) where we tend to think of rapid means immediate in a more human sense. This tends to contribute to the confusion and difficulty of misinterpretation.

Comments are currently closed