RSS feed for comments on this post.

  1. Thanks for the report, I have one question:

    you say “The Jakobshavn is of particular importance as it has a bed below sea level for at least 80 km inland from the terminus. In this reach there are no significant pinning points, or abrupt changes in slope or width (Clarke and Echelmeyer, 1996) that would help stabilize the glacier during retreat.”
    yet your photo appears to show a widening at the 2006 position. Would you like to comment?

    Comment by Phil. Felton — 18 Apr 2008 @ 1:39 PM

  2. not a criticism per se … but this motion of glaciers is complex: at once moving in one direction, retreating in the other and thinning – it therefore seems like a constructive idea to make these motions clear at the outset, people can then form a type of cognitive map, and this will (of course) contribute to understanding, retention and so forth

    Comment by David Wilson — 18 Apr 2008 @ 2:12 PM

  3. Ice loss in Greenland seems to be constrained by outlet glacier calving rates. Is there a point where the mountains ringing Greenland open up such that as the calving fronts continue to retreat, they will begin to greatly widen? Essentially uncorking the plugs to the Greenland Ice Sheet? At that point, the Zwally Effect may become more important than the Jacobshavn Effect. I seem to recall that large parts of Greenland are a ring of mountains with a mushy center that is near or below sea level, like an atoll.

    Comment by Andrew — 18 Apr 2008 @ 2:18 PM

  4. Another relevant analysis is recent work presented at the AGU fall meeting which assessed how much Greenland’s outlet glaciers would need to speed up to get a sea-level contribution on the high side by 2100 or so. Basically all the outlet glaciers would have to get into a highly implausible state of utter overdrive and stay there with no slowdowns for the entire century to have Greenland contribute a meter or two (pretty sure I have that basic sketch right).

    As I wrote on Dot Earth in January:

    Some scientists assessing the recent acceleration of ice flows propose that the rates of increase can’t be sustained long enough to get a truly disastrous rise in seas by 2100 from a warming Greenland. Tad Pfeffer, of the University of Colorado, and Joel T. Harper of the University of Montana laid out their argument for caution at the American Geophysical Union meeting in San Francisco in December, and quite a few glaciologists seem to agree with them.

    I’d love to see Pfeffer or Harper weigh in here. Pfeffer has a nice pdf with his main findings.

    Comment by Andy Revkin — 18 Apr 2008 @ 3:12 PM

  5. 90 to 220 cubic kilometers/year translates into what in terms of increase in sea level?
    Is this a net figure for Greenland as a whole or a gross figure for the glaciers under consideration?

    [Response: That’s net Greenland I think, and translates to about 0.3 to 0.6 mm/yr (out of a total of ~3-4 mm/yr currently). – gavin]

    Comment by Bernie — 18 Apr 2008 @ 3:46 PM

  6. Does anyone know how long it took to empty Glacial Lake Agassiz into the Atlantic?

    Comment by pat n — 18 Apr 2008 @ 3:48 PM

  7. Glacial Lake “Agassiz”, link to article follows.

    Comment by pat n — 18 Apr 2008 @ 3:54 PM

  8. @Bernie. Earth surface = 511 million square km. Sea = 71% = 360 million square km. 90 cubic km = 1/4 of a millionth km = 0.25 mm. 220 cubic km = 0.6 mm. Both per year.

    Comment by Ark — 18 Apr 2008 @ 4:10 PM

  9. Our (me, Joel Harper, and Shad O’Neel) work on estimating the outlet glacier velocities required for Greenland to deliver a really big sea level rise (we looked at 2 m or more) has been submitted for review, so I don’t want to go into a lot of details now. However, the idea is simple, and I’ve talked about this much in many presentations this winter: Take the amount of ice you need to get rid of from Greenland to raise sea level 2 m in the next century, reduce it by your best estimate of the amount that would be removed by surface mass balance losses, and try to push the rest out of the aggregate cross-sectional area of Greenland’s marine-based outlet glaciers. You learn 2 things: 1) the outlet flow speed has to be *really* high, and 2) the volume required is so large that even generously inflated surface mass balance losses can’t touch it.

    This is a pretty simple calculation, conceptually anyway. The reason it’s valuable is that for all the recent discussion of fast dynamic contributions to sea level rise, no one has looked at the glacier dynamics that would be required to accomplish it. Rough analogies with past conditions (e.g. sea level rise was big 125,000 years ago, and fast 15,000 years ago, so therefore it makes sense that sea level rise should be big and fast now) don’t take into account any constraints imposed by glacier dynamics, and they neglect the fact that we live in a vastly different world glaciogical world than we did 15,000 years ago. In particular, we don’t have the Laurentide Ice Sheet hanging over us any more. The Jakobshavn Glacier is a monster on a human scale, but in terms of getting rid of the Greenland Ice Sheet fast it’s one of a small number of pretty tiny slots. There are no Hudson Straits in Greenland. There is a fairly large fraction of the interior that lies below sea level, but it’s very poorly connected to the ocean via marine based channels.

    The bottom line for me is that glacier dynamics is a very important and unresolved issue (and let’s not forget surface mass balance – it hasn’t gone way and it’s not small). I think there’s been a chicken-little effect at work, however, which tends to make us want to embrace potentially impressive processes rather uncritically. As several other posters have mentioned, we still have a lot of work to do. Let’s get the physics right (or as right as we can), and let’s consider all the sources.

    One other point: it’s *not* all Greenland and Antarctica. We neglect the other glaciers and ice caps at our peril. Glaciers and ice caps contributed 28% of total sea level rise compared to 14% combined for Greenland and Antarctica for 1993-2003, using IPCC 4th assessment numbers, or GIC 33%/Greenland and Antarctica 20% for 2006 using Meier et al (Science 2007) numbers. The total sea level rise potential from the glaciers and ice caps is tiny, of course, compared to the ice sheets (50, 60, 70 cm – the number is shifting around. Take your pick). But on the century time scale, 50-70 cm of sea level rise is not tiny. One comment I get a lot is that the glaciers and ice caps will run out of ice soon, so why bother to count them? True enough, they might run out of ice – but under what sea level rise conditions do we regard 50-70 cm as loose change, down in the noise level? The other comment I get is that the glaciers and ice caps aren’t subject to the same dynamic effects as Greenland and Antarctica. They’re not? What about the marine-grounded outlet glaciers draining the Russian and Canadian Arctic, Alaska, Patagonia, etc? How vulnerable are they? The fact is we don’t know – we don’t even have a decent inventory assembled of the marine-grounded glaciers in those areas. We need that information, and until we have it and can answer these questions I’m not ready to write off the glaciers and ice caps.

    Comment by Tad Pfeffer — 18 Apr 2008 @ 4:50 PM

  10. I interpret the picture as meaning that the Jakobshavn Glacier has become a liquid water river. Is this correct? If so, there would be no more solid ice Jakobshavn effect and the GIS is free to flow as liquid water down Jakobshavn Glacier’s riverbed. But the bulk GIS ice experiences Jakobshavn effect “shoulders” at the ends of the 2006 line so that the bulk ice can’t flow over a broader front? So the GIS is a “caged iceberg” unable to slide whole into the sea because of 50 or so kilometers of mountains holding it in place? I get the imaginary image that GIS is a dangerous monster trying to escape. Is the bottom of the GIS above or below sea level? If below, could a warm ocean current some day flow through Greenland, adding heat? Can you make for us a 3D picture of Greenland?

    Comment by Edward Greisch — 18 Apr 2008 @ 5:06 PM

  11. You have this net loss, and people tend to talk almost always about the loss side of it: melting, moulins, calving, lubrication, etc.

    How about Greenland’s precipitation? Is there any change in the rate of elevation increase in the interior? Also, as the perimeter of the GIS recedes, when, if ever, does rebound become a factor?

    [Response: Net loss includes the accumulation in the interior (that’s what makes it ‘net’). Rebound is already a little bit of an issue – but from deglaciation at the LGM, not today. – gavin]

    Comment by JCH — 18 Apr 2008 @ 5:17 PM

  12. Re #9 Tad Pfeffer you say:

    “The bottom line for me is that glacier dynamics is a very important and unresolved issue (and let’s not forget surface mass balance – it hasn’t gone way and it’s not small). I think there’s been a chicken-little effect at work, however, which tends to make us want to embrace potentially impressive processes rather uncritically.”

    As I see it there has been an inverse chicken-little effect at work. To avoid being called “Chicken Little” the scientists have refused to suggest stories if they are scary and they have emphasised any doubts they may have had.

    The IPCC AR4 left out ice melt from sea level rise completely, and scientists have been loath to accept that AGW is causing glaciers to melt. Obviously we are not going to have a rapid rise caused by a pro-glacial lake disgorging into the oceans. We have no pro-glacial lake to do that. However, we should consider the scenarios where a disastrous Greenland ice sheet collapse occurs in case that can happen.

    There are two positive feedbacks which have not been mentioned. First, as glaciers melt sea level rises and grounding lines retreat speeding the remaining glaciers’ advance into the sea. Second, as the glaciers melt their height reduces bringing a greater proportion of the glacier below the snow line. Sea level rise has a similar but lesser effect. Both of these cause more melting. Both could end with a runaway melt.

    But the biggest danger is the one which is obvious and urgent. When the Arctic sea ice goes the increase in water vapour in the Arctic region will accelerate the Greenland melt. Glacier ice melts by contact with water vapour not warm air. The latent heat in moist air is much greater than the sensible heat. Without the Arctic sea ice insulating the air from the water beneath, the increase in water vapour will be dramatic. This is an obvious and real danger. Why is it not being raised?

    Cheers, Alastair.

    Comment by Alastair McDonald — 18 Apr 2008 @ 6:30 PM

  13. Speaking of Ice Caps – this just in.

    “Only five large ice shelves remain in Arctic Canada, covering less than a tenth of the area than they did a century ago.”

    Trent U. report, 4/16/08

    Comment by d. beck — 18 Apr 2008 @ 6:54 PM

  14. I hope it is net since the meltwater amounts to an average rainfall for Greenland of about 2 to 3″ per year. I suspect, but don’t know, that this is a not excessive proportion of the annual precipitation.

    Comment by Bernie — 18 Apr 2008 @ 7:00 PM

  15. Mauri Pelto said:

    Jakobshavn Glacier, West Greenland, retreated 30 km from 1850-1964, followed by a stationary front for 35 years…. The glacier terminus region also had a consistent velocity of 19 meters/day (maximum of 26 m in glacier center), from season to season and year to year…. After 1997 it began to accelerate and thin rapidly, reaching an average velocity of 34 m/day in the terminus region. The glacier thinned at a rate of up to 15 m/year and retreated 5 km in six years. Jakobshavn has since slowed to near its pre-1997 speed, the terminus retreat is still occurring, but likewise is.

    What does the last sentence mean? Especially the part after “pre-1997 speed”?

    Comment by Meltwater — 18 Apr 2008 @ 7:35 PM

  16. Pfeffer’s “back-of-the-envelope” calculations are quite compelling (as presented at the Midwest Glaciological Meeting in Burlington, VT last month). Hopefully these calculations will be more widely available soon.

    His observation, “One other point: it’s *not* all Greenland and Antarctica. We neglect the other glaciers and ice caps at our peril.” is important. For example, the south dome of Barnes Ice Cap, Baffin Island (the last remnant of the vast Laurentide Ice Sheet) has experienced a rapid lowering in the last two decades. This is an ice cap, the size of Rhode Island, whose melt water flows into Baffin Bay & the Foxe Basin. (

    The small glaciers (less than 100 sq. km) of Svalbard do seem to be rapidly shrinking yet some of the larger ones are in near steady-state (unpublished thesis data). JCH asks, (#11) “How’s about Greenland precipitation?” Bamber,et al. (GRL V.31, 2004) found that the largest ice cap on Svalbard (Austfonna) was increasing in elevation and correlated the increase to the decline in sea ice which, in turn, provided more open water and, hence, more local precipitation. They do, however, caution that “large perturbations in the mass-balance of other Arctic ice masses may be expected.”

    It seems likely that the “small” ice masses of the Arctic archipelagos will, over the next 90 years, contribute more to sea level rise than Greenland. Yet, there is much (as Pfeffer points out) that we do not know about the state of the smaller, non-Greenland, Arctic ice masses.

    Comment by Bill S — 18 Apr 2008 @ 8:42 PM

  17. Rather than flow speeds (presumably at or near the terminus) I’d rather see the total flux of ice presented. That way corrections for the cross sectional area where the velocity has been measured (averaged?) are corrected for. Of course this isn’t directly observed.

    If I assume surface melting of 1M/year over the interior, say 500e3 KM**2 due to warmer climate & darker ice surface (old wet ice versus clean dry snow) that would contribute 1.4mm/year to sea levels. Significant, but several times smaller than the catastrophic rates in the popular (nonscientific) literature.

    Looking at your chart of the Jakobshavn, my concern would be that further retreat would lead to a much longer terminus as the confining topography is bypassed. Or is only a small part of the upstream terrain immediately inland of the current terminus is below sea level. A longer terminus could potentially lead to a large net calving rate. Is this a possibility, or does convergent flow mean that the present terminus should be very stable against further retreat?

    Comment by Thomas — 18 Apr 2008 @ 9:03 PM

  18. 10. Edward, here are some 3D pictures for you

    Comment by mg — 18 Apr 2008 @ 9:45 PM

  19. I wanted to ask a question very similar to one already asked by Alastair in #12: Is anyone looking at what the affect on GIS might be of a warm Arctic sea due to albedo effects once the sea ice is gone in the summer?

    Comment by Tenney Naumer — 18 Apr 2008 @ 10:23 PM

  20. Of the last great meltwater effects between 10,000 and 20,000 years ago there was massive sea level rises (5 meters per century) because there was massive ice sheets that stretched out to its continental shelf, before breaking off and scarring the edge of the shelf itself. Today the only place we have massive ice sheets is in the East Antarctic and to a lesser extent the West Antarctic, but now the world is twice as hot. The geography of Greenland is now preventing a rapid collapse of its glaciers ( something i call the Cascade Effect ) even though the Zwally Effect and the Jakobshavn Effect are at work now as they were 10 to 20 thousand years ago. My concerns are with the West and East Antarctic, especially the openings at the Ronnie and the Ross ice shelf’s in the West and the Amery ice shelf in the East Antarctic. Are my concerns founded.
    Cheers Simon.

    Comment by Simon Edmonds — 18 Apr 2008 @ 11:06 PM

  21. Tenney, that is THE question at hand, and I’ll bet no one will even touch it.

    Comment by d. beck — 19 Apr 2008 @ 1:07 AM

  22. AFP and AP and New Scienist news story here:

    Lakes on the surface of Greenland’s ice sheet are draining through the kilometre-thick ice and roaring to the bedrock with a flow rate exceeding that of Niagara Falls.

    Comment by Danny Bloom — 19 Apr 2008 @ 2:27 AM

  23. in reply to Thomas: “Rather than flow speeds (presumably at or near the terminus) I’d rather see the total flux of ice presented. That way corrections for the cross sectional area where the velocity has been measured (averaged?) are corrected for”

    You’ll have to wait for the paper to come out, but this is how we did the analysis I described above.

    Comment by Tad Pfeffer — 19 Apr 2008 @ 6:36 AM

  24. I wonder once the front has retreated from salt water and sea mist that ultracold deep ice could refreeze fresh water and slow the breakup. Notice if you tip dry looking frozen peas into a pot of water that they knit into a frozen block, albeit temporary.

    Comment by Johnno — 19 Apr 2008 @ 7:18 AM

  25. Re: #12

    Although increasing latent heat has been raised before *,
    the rapid opening of the Arctic Sea and latent heat effects on Greenland ice warrants immediate attention.

    * comment #135 at:

    Comment by pat n — 19 Apr 2008 @ 8:01 AM

  26. As so much of the GIS is grounded below sea-level, it seems to me that the key to any possible catastrophic mass loss is the Jacobshavn effect: the ungrounding of the marine front as the ice thins and becomes buoyant. It seems to me that this has potential for really dramatic sudden mass loss (“sudden” as in hours/days, not decades), because there’s no guarantee of monotonicity in the (hmm, search for a term) hypothetical net areal buoyancy of the grounded ice.

    That is, when a marine front becomes buoyant enough to unground, the section of ice behind the marine front becomes the new front. Until that moment, the buoyancy of that section (i.e. whether or not it has enough ice mass above sea level to remain grounded if exposed to the sea) has been irrelevant. There may be areas which already would float if they could, and which therefore will float very rapidly when the grounding line reaches them.

    I’m glad this effect has a name, by the way. I’ve been trying to describe it in blog comments for months now, rather ineffectively.

    So what we need is detailed topo maps of the bed and thickness of the GIS, and to work out a map of the “net buoyancy”, or some such (i.e. total ice area density subtracted from the area density of a hypothetical column of water resting on the bed and extending up to sea level). This will show the potential for catastrophic loss of this sort.

    Oh, and the same thing for WAIS. I gather there is NSIDC data which makes producing such maps a SMOP.

    I’m very much looking forward to seeing the paper to which Tad Pfeffer refers.

    Comment by Nick Barnes — 19 Apr 2008 @ 8:25 AM

  27. I would like to echo Mr. Edmonds inquiry as to the stability of the Pine Island and Thwaites glaciers which seem to connect directly to the Byrd Subpolar Basin, where the ice sheets are grounded far below sea level. I understand that acceleration of the Pine Island glacier has been observed, and there seem to be no natural pinning points. Coupled with the observations of surface melt from
    Antarctica, would someone care to comment on the mechanics of glaciers in these sectors ? I seem also to remember a comment by Bindschadler stating that while GIS glaciers could retreat inland from warming oceans, there could be no similar escape for Antatctica.

    Comment by sidd — 19 Apr 2008 @ 8:38 AM

  28. Unable to find a publisher for it, I put my saturation article on the web:


    Readers will see my debt to RealClimate articles which touched on this.

    Comment by Barton Paul Levenson — 19 Apr 2008 @ 8:38 AM

  29. I’m not a scientist, just a very interested citizen, but may I suggest another “chicken little scenario”:

    With the outlets currently limited in size and location effectively damning the flow, might there be a subsequent built-up of melt with accompanying positive feedbacks further inland – supporting a much greater rate of melt than what we see flowing out? Could this be a building gigantic reservoir? Could such a reservoir break-through at some point of “non-linearity” such as when the Mediterranean flowed back into the Black Sea, raising sea levels dramatically practically overnight?


    Comment by kenlevenson — 19 Apr 2008 @ 8:48 AM

  30. Re: #19 and #21

    Dear d.beck,

    Have you noticed that water vapour is pouring out of the cracks in the Arctic sea ice in stupendous quantities?

    Comment by Tenney Naumer — 19 Apr 2008 @ 9:46 AM

  31. Could the rapid melting of the ice also trigger mega-tsunamis? Ice can act as cement, stabilizing soil. After the ice is gone, you could perhaps have mountains that can break up and fall into the sea.

    Comment by Count Iblis — 19 Apr 2008 @ 9:53 AM

  32. Thank you Mauri Pelto and also Tad Pfeffer for your insightful explanations and comments. It is always good to get some good news even if it is slight.

    Those interested in some more information about this topic might be interested in the article in this week’s Nature (available at least at this time without a subscription):

    Comment by Arch Stanton — 19 Apr 2008 @ 10:03 AM

  33. Count Iblis: There may be links between tsunamis and deglaciation (Storegga slide is touted) and there is a well-know paraglacial response of rock slopes to deglaciation. But any likely tsunami will probably be rather localised.

    Comment by san quintin — 19 Apr 2008 @ 10:34 AM

  34. Tad Pfeffer (9) — I doubt that there are any (significant) marine grounded glaciers in Alaska or as part of the Southern Patagonian ice cap. Even Hubbard Glacier

    appears to terminate in quite shallow water.

    Comment by David B. Benson — 19 Apr 2008 @ 1:08 PM

  35. Barton Paul Levenson (#28) wrote:

    Unable to find a publisher for it, I put my saturation article on the web:


    Readers will see my debt to RealClimate articles which touched on this.

    Thank you!

    It’s a good essay — and I have bookmarked it for future reference…

    Comment by Timothy Chase — 19 Apr 2008 @ 2:30 PM

  36. Tad Pfeffer:#9: “…Greenland to deliver a really big sea level rise (we looked at 2 m or more)…”

    Why choose this particular (unquestionably catastrophic) situation? IMHO this would relate to an overall sea level rise of about 5 meters within 100 years, West Antarctic and higher latitude glaciers included. I never see such a potential rise discussed seriously.

    This approach is of course interesting as it might provide an upper boundary for the risk – something that is badly missing. So why not re-compute the situation based on a 0.7 meter contribution from the GIS? That is bad enough for most practical purposes.

    Comment by Pekka Kostamo — 19 Apr 2008 @ 3:17 PM

  37. Re No. 34, There are a number of tidewater-calving glaciers on the west coast of the South Patagonian Icefield. The only similar glacier on the Northern Icefield, is the San Rafael (which used to have a high calving flux and velocity until it started receding into shallow water).

    Comment by san quintin — 19 Apr 2008 @ 4:55 PM

  38. san quintin (37) — Thank you for the information. The issue, as I take it, is whether or not the glacier has dug a fairly deep submarine canyon, one which the terminus still occupies.

    Comment by David B. Benson — 19 Apr 2008 @ 6:28 PM

  39. Thanks for a great post. From my perspective the interesting item you raise with both the Zwally and Jakobshavn effects is the persistent increase in the volume of water moving under, over and through these glaciers.

    Obviously in the short term the ZnJ effects will give rise to variations in the rate of movement of the glacier, but it is the longer term effects relating to energy transfer that worry me.

    Alastair notes that increased water vapour will carry more energy to the surface of the glaciers, likewise these increased water flows over, through and under the glaciers is also transferring vast amounts of energy into the ice. And turbulent flow is WORK being done on the water which generates more heat which is likewise dissipated into the ice.

    Do you know of any investigations of current deep ice temperatures of the glaciers and ice caps? I understand that global surface temperatures are not responding as rapidly as they should be when the atmospheric models are considered. Is this ice the sink that is absorbing the currently ‘missing’ global temperature rise?

    Among the ice sheet dynamics to fret about I see this change in the temperature of the ice from say -30C to ice-at-0C and the subsequent uptake of the heat to go from ice-at-0C to water-at-0C as the ‘dark matter’ of the cryosphere.

    For me this raises the potential of items like the Jakobshaven changing from a ‘cork in the bottle’ of the GIS to being a warm run of open water sloshing about 80km into the GIS, and rather than the cross section available for movement of ice/water being confined to the cross section of the glacier it will increase to the side face area – which is much greater and exposes that new face to the same ZnJ effects..

    So if as Arc says the ocean is 360 million sq km, then 2m rise is 720,000 cubic kilometres in a hundred years.
    Roughly Greenland has a perimeter of 6600km
    Say 10% of this is open to the present-day ocean, gives a ‘gap’ 660km wide for GIS to empty though.
    720,000 cukm per hundred years = 19.7 cukm per day = 228,300 cubic metres per second.
    228,300 cubic metres per second thru a 660km wide gap = about 350 litres per second per metre of gap.
    (Check the numbers, please!)

    A sheet flow 350mm deep moving at a leisurely one metre per second. Hydraulically that is VERY possible.

    Interestingly the Amazon River – which gets up to 45km wide during the rainy season – discharges at an average rate of 218,000 cubic metres per second. That flow (if not balanced by the water cycle) gives (wait for it!) 1.9m sea rise over 100 years.

    So 2m in a century from GIS plus other glaciers plus Antarctic is trivial – hydraulically at least.

    And of course as soon as the melt starts the ‘gap’ runs up the Jakobshavn and its cousins and for every km back it runs the gap increases by 2km.

    And of course the existing wet surface area of the Greenland ice is in fact evidenced by the wetted perimeter others have spoken about – the water on top of, beside, inside, through beneath and abutting the glaciers and ice sheet. It is huge – far more than a mere 660sq km. And its warmer than the ice.

    And theres no need to confine this consideration to Greenland:-

    Ditto all the above with Antarctica, of course, with bells on…

    …the deep ocean around the Antarctica is changing, in particular becoming less salty and less dense.
    **The cause of the freshening is the next challenge for us to work out, the leading hypothesis at the moment is that the water is becoming fresher because the ice around the edges of Antarctica is melting more rapidly,** Dr Rintoul said.
    Oh dear..!

    Comment by Nigel Williams — 19 Apr 2008 @ 7:05 PM

  40. Replying to David Benson #34: in Alaska, Columbia Glacier has retreated ~16 km up a marine-grounded fjord since ca. 1982, and has ~14 km left to retreat before its bed rises out of sea level. The glacier’s calving flux has *averaged* 4 cubic kilometers/year since the onset of retreat with a maximum of 7 kilometers/year in the early 2000s. Hubbard does indeed terminate in shallow water at present (as is presently advancing), but the terminus is on a moraine; upstream from the terminus there is a long marine-grounded reach, although details are lacking… there has been some airborne radio-echo sounding done there recently with more planned. We should know more about Hubbard before long.

    Comment by Tad Pfeffer — 19 Apr 2008 @ 7:19 PM

  41. Replying to Pekka Kostamo #36: Your questions are good ones, and we address these in the paper. I need to wait for the paper to pass through the review process, though, and not leak out its contents in little bits!

    Comment by Tad Pfeffer — 19 Apr 2008 @ 7:22 PM

  42. Re 18. Thanks mg for the link to the 3D pictures.

    It looks like the Jacobshavn will be getting much wider soon. Is that correct? How much could that affect the need for extreme glacial acceleration to raise sea levels by a meter?

    Is there a good breakdown of how much ice is where on the globe and then projections on how much melting might be expected for a given area of ice?

    In complete jest:

    Why don’t we create a giant damn across the outlet of the Jakobshavn Glacier. It would only be 1 1/2 times bigger than the Three Gorges Damn. This will not only hold back some of the water but could also be used to generate electricity. The electricity could be used to create hydrogen that would be shipped to the US to power all those hydrogen cars that we don’t have the hydrogen for.

    Comment by Eric Hacker — 19 Apr 2008 @ 8:25 PM

  43. No worries mate. Pat Michaels has the floor.

    Disko Bay, Greenland: Temperatures on the island are no warmer than they were in the mid-20th century.

    As long as this is what makes it into the press, the truth won’t matter. Write a response. Anyone?

    Comment by Mark A. York — 19 Apr 2008 @ 10:53 PM

  44. Eric (#42). I’d surely prefer a giant dam over a giant damn.

    Comment by Ark — 20 Apr 2008 @ 1:07 AM

  45. To what extent might long term response to changes hundreds of years ago during the Little Ice Age (LIA) result in glacier surging on the Greenland Ice Sheet and the West Antarctic Ice sheet streams surging into the Ross Sea? More specifically, can one define the response time of a particular ice sheet? Are we seeing the results of AGW in the last 100 years or a change over the last 200 or 500 years?

    In addition, this same question could be asked about the valley glaciers in Alaska and elsewhere. How much early 20th century surging and flow could be attributed to LIA addition of mass? And is the current large scale ablation seen on these glaciers due to these glaciers coming to some equilibrium with a warmer world due to coming out of the LIA and response times associated with the large masses involved?

    Re 12: The reduced ice mass over the Arctic Sea and increased water vapor as an agent for melting makes sense. But also keep in mind that Lambeck et al (2002) have suggested that loss of permanent ice over the Arctic Sea at the end of the last interglacial allowed for increased accumulation of snow and ice at high latitudes because of the increased amount of moisture available. This seems like a lucid argument for another possible complex feedback in the climate system. See: Lambeck et al (2002). “Links between climate and sea levels for the past three million years.” Nature 419: 199-206

    Comment by Drew Latta — 20 Apr 2008 @ 1:19 AM

  46. As I have reacted to the speed increase of the Ilulisat/Jakobshavn glacier in the past, herewith some more comments:

    – There is a discrepancy between the above map showing the retreat of the ice breakup points and the NASA map over 150 years (7 MB .tiff file, smaller .jpg file here).
    According to the NASA map, the retreat of the calving front in the period 1929-1953 (24 years) was as large as the retreat in the period 1953-2003 (50 years). According to the map at the top of this article, the recent retreat was much faster.
    I suppose that may have to do with in-between advancing of the glacier’s calving front?

    – The previous speedup in the period 1930-1950 might have been caused by (summer) temperatures which were as high or even higher than in the current period. See the around Greenland (coastal) temperatures up to 2005 here

    – Greenland (summer) temperatures seems to be more influenced by AMO/NAO than by the general global warming trend.

    – Interesting that the retreat speed in the past few years is decreasing. This is quite unexpected, as the resistance of the downstream ice is reducing, thus one should see an increase in retreat speed…

    Comment by Ferdinand Engelbeen — 20 Apr 2008 @ 3:53 AM

  47. 42 Eric, data on the volume of ice and potential sea level rise (pslr) associated with the various components are given in a table on this page:

    East Antarctic ice sheet vol=26,039,200 km3 pslr=64.80m
    West Antarctic ice sheet vol=3,262,000 km3 pslr=8.06m
    Antarctic Peninsula vol=227,100 km3 pslr=0.46m
    Greenland vol=2,620,000 km3 pslr=6.55m
    All other ice caps, ice fields and valley glaciers vol=180,000 pslr=0.45

    Total vol=32,328,300 km3 pslr=80.32m

    Comment by mg — 20 Apr 2008 @ 3:55 AM

  48. Re #42,


    You can forget to build a dam at the end of the Ilulisat/Jacobshavn glacier. The water there is about 7 km wide and 700 m deep (therefore the icebergs have a height of about 100 m above the water line!). The fjord ends with an underwater morene, where the (about 800 m thick) icebergs are pushed over by tides, wind and the presuure of the icebergs behind them.

    As the glacier/icefjord icebergs advance with a speed of over 30 m/day, any dam (how heavy it may be made) will be pushed away by the forces behind the ice front… Not to be forgotten the harsh winter freezing there…

    Comment by Ferdinand Engelbeen — 20 Apr 2008 @ 4:05 AM

  49. Figure 4 on this page

    gives some very interesting sea ice visualisations from the 1982 to 2007. Around 2002 the sea ice dynamics shows some interesting changes, with what appear to be orthogonal shock waves across the ice pumping cycles. It is interesting to note that it was around this time the Larsen ice shelf broke off. Does the oceanic system with its circulatory reconfigurations strongly couple ice dynamics at the two poles? Do these involve other interesting features (eg formation of large oceanic scale mills) around the antarctic circulatory system (can’t find the link I had, but will post when found)? Thus, could ice discharge at Greenland destabilise ice in Antarctica? [Apologies for the amateur questions].

    Comment by mg — 20 Apr 2008 @ 4:27 AM

  50. OFF TOPIC:

    This article was from another article posted by the Daily Telegraph today about the NIPCC (Non IPCC). The names make familiar reading and so does the rhetoric really.

    Maybe someone would like to reply to them but I guess scientifically the debate ended some time ago but getting political and economic action on climate change still requires a response from those who continue to befuddle the public.

    Comment by pete best — 20 Apr 2008 @ 4:49 AM

  51. Why this need to guess about sub-ice water pressures? Technology to measure pore water pressures at kilometre depths is highly developed – oil & gas drilling, mining, even civil geotechnics. Grout in some fibre-optic piezos and measure it.

    Comment by GlenFergus — 20 Apr 2008 @ 5:21 AM

  52. An example link for my earlier post concerning large-scale oceanic mills is

    Comment by mg — 20 Apr 2008 @ 5:29 AM

  53. Thank you, Tim.


    Comment by Barton Paul Levenson — 20 Apr 2008 @ 6:08 AM

  54. Looking at Google Maps in conjunction with the map in this post, it appears that the Jacobshavn ice front is now inland of the mountains that ring the GIS. Wild.

    Comment by Nick Barnes — 20 Apr 2008 @ 6:33 AM

  55. replying to GlenFergus #51: Drilled access to the bed and direct measurements of subglacier water pressure have been made many times over the years (for a recent example see Harper et al, Geophysical Research Letters, Vol 34, L12503, 2007). Logistically and technically, however, making these measurements on fast-moving, highly crevassed glaciers is infeasible. (It has been done a few times; for an example see Meier et al, Journal of Geophysical Research, 99(B8), 15,219-15,229, 1994).

    Doing such measurements at a place like Jakobshavn would be a tough task to say the least. Having an Oil & Gas industry budget might help, but even then I would have my doubts about success…

    Comment by Tad Pfeffer — 20 Apr 2008 @ 9:28 AM

  56. I have a map of Greenland’s basal topography. Interesting, the centre is below sea level – almost certainly depressed due to the weight of ice. The east coast/south coast have relatively high mountains which act as a curtain/retaining wall – hence glacial flow via narrow channels. By contrast parts of the west cost seem to have no curtain wall of mountains. A possible scenario is that this area, not “narrow valleys” would see substantial ice loss. Anyway, just a scenario – has anybody else seen the basal maps – worth looking at.

    Comment by Mike Parr — 20 Apr 2008 @ 9:39 AM

  57. 54..Nick Barnes, dont forget that google earth or maps are definately not up to date and can be 3-5years old so further retreat will have happened since then. On the subject of calving glaciers and the subsequent added freshwater being released into the sea, already there is clear evidence and measurable decrease in the salinity of the deep waters of the north atlantic current just east of canada. That coupled to a 0.7degC increase in the deep water current in just 1 year. This implies that a slowdown of the current is imminent as occured 20k yrs ago. Only this time the global level of CO2 is considerably higher than it was 20ky ago most likely resulting in a greatly prolonged ice-age, much longer than in the past. Does any contributer to RC know what effect an ice-age in the northern hemi will have on the southern hemisphere ? As the NA current moves futher south and/or stops. I also read that the jet stream winds at around 30-40k ft have changed direction and they are now much nearer the poles than previously- this could well cause much bigger and more destructive tornadoes and hurricanes as they wont get their heads chopped off by the high velocity jet streams and can grow to any size the want..within natural limits of course.

    Comment by Lawrence Coleman — 20 Apr 2008 @ 10:12 AM

  58. So given a worst case scenario — 6C increase by 2100 or 2150 — how long would it take for all ice to melt in Greenland? 200 years, 300 years, longer? Under a worst case.

    Comment by Lynn Vincentnathan — 20 Apr 2008 @ 10:34 AM

  59. To Tenney, someone did actually respond to your very valid question about “what the affect on GIS might be of a warm Arctic sea due to albedo effects once the sea ice is gone in the summer?.”

    (#25 Pat N. said:

    Although increasing latent heat has been raised before *,
    the rapid opening of the Arctic Sea and latent heat effects on Greenland ice warrants immediate attention.

    * comment #135 at:

    What Pat points to, I believe, is this statement from the article (2006):

    “Looking forward, the current (small) imbalance (whether positive or negative) of the Greenland ice sheet is not terribly important. What matters is if the melting were to increase significantly.”

    The Arctic Ice Sheet gone in the summer absolutely means an increase in GIS melting. I doubt that even Singer would dispute that (without a large accompanying check).

    Pat is one of the United States’ few heroes. He was fired from the National Weather Service for telling a bit too much of the truth.

    As to ‘water vapor from the cracks in the Arctic’……uh,…..well….no….uh… .Haven’t been there in a while…….alright, actually never been.

    What are you talking about, anyway?

    Comment by d. beck — 20 Apr 2008 @ 10:37 AM

  60. #43 Mark, I equally can write such dribble, although
    I would make it less opaque and easier to understand. I can take any location on Earth, show a surprising cooling effect, and point out that Moulins are for Don Quixote and Sancho Panza….. I could write this way,
    but I wont. Because there is more to life than just being paid to be a Don Quixote… There are actually real events, far more interesting… The surface temperature record is vulnerable to attack, with cut and run pot shots, any record is vulnerable to ridicule, as long as no one else is there to respond to the criticism. But what WSJ Quixote’s consistently forget to write about is the disappearance of the perennial ancient Glaciers found all over the world, and especially multi year ice of the Arctic Ocean, true metrics against their arrogant assault on reality……

    Comment by wayne Davidson — 20 Apr 2008 @ 11:35 AM

  61. Tad Pfeffer (40) — Thank you for the information.

    Comment by David B. Benson — 20 Apr 2008 @ 1:25 PM

  62. Global sea level rise is currently at 2.4-3.8 mm/yr (since 1993, IPCC 2007). Assuming that rate stays steady (highly unlikely), you’d get a lower bound of 24-38 cm of sea level rise by 2100.

    Now, the 2007 IPCC estimate:

    Sea level is projected to rise between the present (1980–1999) and the end of this century (2090–2099) under the SRES B1 scenario by 0.18 to 0.38 m, B2 by 0.20 to 0.43 m, A1B by 0.21 to 0.48 m, A1T by 0.20 to 0.45 m, A2 by 0.23 to 0.51 m, and A1FI by 0.26 to 0.59 m. These are 5 to 95% ranges based on the spread of AOGCM results, not including uncertainty in carbon cycle feedbacks.

    It seems their lower bounds are kind of implausible… and we already know that the AOGCMs missed some important features of the Arctic climate system, the sea ice response at least (IPCC 2007):

    There is a projected reduction of sea ice in the 21st century in both the Arctic and Antarctic with a rather large range of model responses. The projected reduction is accelerated in the Arctic, where some models project summer sea ice cover to disappear entirely in the high-emission A2 scenario in the latter part of the 21st century.

    The take home point for reporters is that the IPCC has underestimated these trends and their projections are also “very likely” to be underestimates.

    Another issue has to do with the rate that the rest of the cryosphere, the mountain glaciers and icecaps, will melt at. The causes of sea level rise are the expansion of ocean water caused by warmer ocean temperatures, melting of the Greenland Ice Sheet and the Antarctic Ice Sheet, and the glaciers.

    If we look at Science 16 March 2007, Recent Sea-Level Contributions of the Antarctic and Greenland Ice Sheets, Shepherd & Wingham, we find that:

    Although the balance between these opposing processes has varied considerably on a regional scale, data show that Antarctica and Greenland are each losing mass overall. Our best estimate of their combined imbalance is about 125 gigatons per year of ice, enough to raise sea level by 0.35 millimeters per year. This is only a modest contribution to the present rate of sea-level rise of 3.0 millimeters per year.

    For the fraction due to thermal expansion of ocean water, see Thermosteric sea level rise, 1955–2003, J. I. Antonov, S. Levitus, and T. P. Boyer, GRL 2005

    For the 0–3000 m layer of the entire World Ocean, the linear trend of thermosteric sea level is 0.40 mm/year for 1955–1959 through 1994–1998.

    If so, then the total sea level rise, 3.0 mm/yr, equals 0.40 mm/yr + 0.35 mm/yr + X

    Where X is the amount of water entering the oceans from melting glaciers and small icecaps. This number is probably of greater immediate practical importance to human populations. Is it safe to say that the current contribution is then ~2.3 mm/yr?

    To quote Tad Pfeffer, #9:

    Glaciers and ice caps contributed 28% of total sea level rise compared to 14% combined for Greenland and Antarctica for 1993-2003, using IPCC 4th assessment numbers, or GIC 33%/Greenland and Antarctica 20% for 2006 using Meier et al (Science 2007) numbers. The total sea level rise potential from the glaciers and ice caps is tiny, of course, compared to the ice sheets (50, 60, 70 cm – the number is shifting around. Take your pick). But on the century time scale, 50-70 cm of sea level rise is not tiny.

    For a nice photo of this, see: River of melting water springing from Nigardsbreen, Norway

    Thus, by the time contributions from Greenland and Antarctic sea ice melt become dominant, will all the world’s mountain glaciers and icecaps be almost gone? If so, then there will be a huge water crisis, because many highly populated regions rely entirely on glacial melt for some period of the year for their water supplies. Where will they go? To the coasts, which will eventually be inundated, whether in 100 years or 200 years?

    Comment by Ike Solem — 20 Apr 2008 @ 2:38 PM

  63. RE: #60

    And they won’t Wayne, that’s the problem. They moderate responses too. Only friendly facts allowed. That’s the definition of propaganda.

    Comment by Mark A. York — 20 Apr 2008 @ 4:13 PM

  64. The International Earth Rotation and Reference Systems Service (IERS) suggest an ongoing transfer of mass to the pole(s) since 1998. Does this present a problem for AGW?

    [Response: That’s funny. It’s rare that such a specific conclusion is drawn from the flimsiest of evidence despite direct observations contradicting it (google GRACE). Unfortunately, the adding of leap seconds is not as objective as you might think – (see here), and the complete lack of alternate hypotheses underlines the agenda of the writer. Any water shifts – whether they are towards higher precipitation in higher latitudes, or sea level changes due to circulation or thermal expansion (which is not uniform) etc would seem to me to be candidates for the explanation (even if one were needed). Additionally, it’s not obvious that solar or planetary tidal effects couldn’t be playing a role. And of course, the amount of mass concerned would need to be quantified. Meanwhile there is plenty of evidence indicating that the length of day responds to climate changes (especially ENSO) etc….. – gavin]

    Comment by fieldnorth — 20 Apr 2008 @ 5:25 PM

  65. Mark & Wayne (#43 & 60), I figured it out — no warming trend bec they were sticking their thermometers in the ice, so it has been remaining fairly steady, around 32F.

    My bigger problem with the article was:

    Pending legislation in the Senate from Joe Lieberman and John Warner would cut emissions even further – by 66% by 2050. No one has a clue how to do this [emphasis mine].

    I could tell my CF lightbulb joke again….Nah.

    Comment by Lynn Vincentnathan — 20 Apr 2008 @ 7:14 PM

  66. I’m a bit confused by the post. I understand the “Zwally effect” would be enhanced by warming. But it also seems the “Jakobshavn effect” is at least partly enhanced by warming — warming that thins the glacier and makes it more bouyant. And perhaps by way of increased ocean pertubations at the calving front?? Is there an increase in that type of perturbation due to global warming? Then sea rise would also increase that perturbation and bouyancy (but sea rise is very slow).

    Also I read somewhere that reduction of the ice load on Greenland may be causing small local earthquakes. If so, would this also be considered “perturbations” that could enhance the “Jakobshavn effect”?

    Aside from this I just read about the storehouse of methane possibly opening from Siberian methane hydrates (see:,1518,547976,00.html ). This could cause a lot more warming, and if it does, would it be expected that the Zwally effect would overtake the Jakobshavn effect?

    Finally, my meager understanding is that a lot of energy goes into melting ice (is this akin to kinetic energy?), rather than heating the surroundings (another reason why in some places in Greenland there might not be warming), but once the ice is melted the energy then goes into the atmosphere and ocean more quickly. That might be wrong, but as long as there is one small ice cube in a glass, the water stays cool, and as soon as that melts away, the water quickly warms to room temp.

    Comment by Lynn Vincentnathan — 20 Apr 2008 @ 7:39 PM

  67. To d. beck, comment #59

    When the sea ice cracks open, the temperature of the water below it is so much higher than the air above it, that water vapor is created. At least, that is what seemed to happen in the beginning of April when the cracks really began to open up, and it seems to be very much an ongoing process, judging by the IR satellite photos that get updated every couple of hours — see:

    Also, see the link a guy from Holland posted on my blog showing the current ice situation compared to last year,
    from the National Snow and Ice Data Center, Boulder, Colorado:

    Let’s face it, you expose above freezing temp water to dry air that is at 10 degrees Fahrenheit — what is gonna happen?

    There are so many cracks now that it is difficult to see what is going on, but it was very clear a few weeks ago.

    Also, if you happen to get a chance to look at a good google map that shows the topography around the coast of Greenland, you will get a good shock — the place looks like it is just made for ice to slide out of it.

    Comment by Tenney Naumer — 20 Apr 2008 @ 8:12 PM

  68. Re: comment #59

    To: d. beck,

    sorry, that was the wrong link — this one is to the Arctic:

    Comment by Tenney Naumer — 20 Apr 2008 @ 8:27 PM

  69. Gavin, how many computer models actually incorporate the predicted increase of CO2/CO and CH4 and fine particulate smog from china and india and the other developing countries over the next 50 yrs. even if the US went cold turkey tomorrow re: emmissions, that would be cold comfort if china and india continued along their merry way.

    Comment by Lawrence Coleman — 20 Apr 2008 @ 9:55 PM

  70. I wonder if it is necessary to invoke the Zwally effect every time one sees a sudden advance of a glacier. Recrystallization of ice due to local strain accumulation or presence of water can shift the primary deformation mechanism from dislocation creep to some other mechanism, such as diffusional creep or superplastic creep. David Goldsby, now at Brown University, has done some wonderful research where he identified a grain boundary sliding (GBS) mechanism in ice, accompanied by dislocation motion.

    Comment by Jeff — 20 Apr 2008 @ 10:44 PM

  71. Re # 66 Lynn V: “…my meager understanding is that a lot of energy goes into melting ice (is this akin to kinetic energy?)…”

    It is called latent heat:

    Comment by Chuck Booth — 20 Apr 2008 @ 10:45 PM

  72. One of the reasons for authoring this post, was a misconception among a number of some glacier scientists even that the Zwally effect was driving the outlet glacier acceleration of the 2001-2005 period. That this is not the case is key to document. It should however, be clear that though we can eliminate that as the key dynamic for acceleration the specific glacier dyanmic underpinnings have not been identified. It is evident that an imbalance at the calving front is one of the key ingredients, but not the only one. Sustaining the observed acceleration for an extended period is difficult, having an outlet glacier better instrumented at the onset of the next acceleration will be a key to fully understanding the process. The widening of the fjord of the Jakobshavn Isbrae above the 2006 terminus can lead to additional calving and further thinning. To what extent I will not hazard an estimate at this moment. It will be worthy of its own post. It will be interesting to examine Pfeffer at all in that light. Total volume flux is a critical parameter to observe over a long time period. In the 1990 paper referenced for this post we did calculate volume flux. It will be nice to compare to Pfeffer’s new work. This post was focussed specifically on Greenland outlet glaciers, not even Greenland as a whole. Meltwater inputs can and will be more important in others regions of the ice sheet. My main focus in on alpine glaciers not ice sheets normally, and I agree with Pfeffer that for now they remain critical.

    Comment by mauri pelto — 21 Apr 2008 @ 7:11 AM

  73. Re #59 True: “What matters is if the melting were to increase significantly”.

    Also true: I was removed from federal civil service by NWS (July, 2005). The removal was later reversed
    and a retirement certificate by NWS for “30 years of Loyal Service rendered to the government of the United States” was awarded (Feb. 2006).

    and, “Evaporation is greatest during the winter because of the greater difference
    between air and water temperatures” Cynthia Sellinger, deputy director of NOAA’s Great Lakes Laboratory.

    and that cold air and open surface waters in the Upper Midwest
    results in water vapor released from lakes and rivers (photos available).

    Comment by Pat N — 21 Apr 2008 @ 11:00 AM

  74. I’m afraid I am just not understanding the point here. It is not surprising that thinning would be important in setting at what place an outlet would float, nor that thinning might be increasing at an altitude where snow does not persist year-round. But, it is fairly clear that glaciers do not cease to move once their termination is above sea-level. And it is that aspect of their motion that seems to hold the key for future sea level rise. Is there a process that transports ice to an altitude where it must melt rapidly or will it tend to stick at an altitude where it must melt slowly? Whether that lower rapid-melt altitude is sea level or not does not seem to be the key issue: something made Jakobshavn thin so that it floats. Once it floats, who cares? It’s gone already. That Jakobshavn thinned is important because it indicates that at whatever rate it is being fed, it is melting faster than it is piling up. If it happens to be fed at a faster rate from higer altitudes with insignificant thinning owing to melt, but rapid mass transfer owing to melt lubrication then this seems to me to be the more important effect.

    Maybe I’ve missed something? Tad Pfeffer (#9) seems to insist that only marine grounded outlets are important, but then points to all the other non-Greenland or Antarctic ice as being more important at present. Many of these reach the sea through rivers, not glaciers.

    Comment by Chris Dudley — 21 Apr 2008 @ 12:48 PM

  75. Have I gone blind?

    What I am not seeing here is a discussion of the recent reports of supraglacial lakes’ hydrofracture of kilometer thick ice resulting in the advection of heat to basement ice. This means the basement ice suddenly warms, and warmer ice is weaker. Weaker ice under the stress of supporting kilometer(s) of ice is subject progressive collapse energetically driven by the potential energy of the supported ice structure.

    The result is a slurry of ice and water, moving at a high speed. The high speed is generated by the release of kinetic energy as the face of the (warmed and now weak) basement ice progressively fails under the weight of the ice above. I call this the “Missoula effect.”

    The Missoula Effect could move a lot of ice, water, (and rock) through a narrow outlet rather rapidly. If the material still has any significant speed when it gets the shore, it can cause a tsunami.

    OK, we have not seen this example of physics in a while. On the other hand, we have not had big chunks of ice suddenly warming in a while either. Now, we are back to a situation where big chunks of ice are suddenly warming.

    Comment by Aaron Lewis — 21 Apr 2008 @ 1:02 PM

  76. Chris, the large Greenland outlet glaciers are not thinning significantly via melting, it is by moving faster and calving more rapidly. It is much faster to lose mass via calving than melting. Note the breakup of Larsen B or the disintegration of the terminus reach of the Jakobshavn from 2001-2006. What matters for calving rate is only if the front of the glacier is afloat. The key for future sea level rise rate of the large ice sheets is not in place melting, but calving.

    Comment by mauri pelto — 21 Apr 2008 @ 1:31 PM

  77. Chris Dudley:

    it is fairly clear that glaciers do not cease to move once their termination is above sea-level. And it is that aspect of their motion that seems to hold the key for future sea level rise

    Most of the potential for catastrophic sea-level rise in this century lies in the GIS and WAIS, which are both mostly grounded below sea-level. The centre of the WAIS is grounded kilometres below sea-level.

    Comment by Nick Barnes — 21 Apr 2008 @ 2:37 PM

  78. What Aaron Lewis was mentioning is this:

    News Release : Lakes of Meltwater Can Crack Greenland’s Ice and Contribute to Faster Ice Sheet Flow
    Researchers Make First Observations of Surface Meltwater Cutting through the Ice Sheet to Lubricate the Bottom

    Mauri Pelto says:
    The key for future sea level rise rate of the large ice sheets is not in place melting, but calving.

    First, for the glacial and small icecap melt, the 50-70 cm of sea level rise is all mostly inplace melting, not calving, right? So, while you qualify with “large ice sheets”, if the issue is how fast sea level will rise then inplace melting is playing the key role right now.

    Second, isn’t there a tendency for glaciologists to neglect the effects of changes in the global planetary circulation (i.e. oceanic and atmospheric heat transport to poleward regions)? This is an area that glaciologists traditionally don’t study – but it seems possible that a warming ocean and atmosphere could indeed lead to high rates of in-place melting.

    Another thing is that glaciologists seem to be relying fairly heavily on past records of change – but according to the ice core CO2 measurements, the increase in CO2 is now 30X greater, rate-wise, than anything seen in the record. Thus, we might see things happening on a very fast timeframe relative to what has happened in the past.

    P.S. I’ve been trying to figure out what the relative contributions to sea level rise are between large ice sheets, mountain glaciers, and thermal expansion, and I get 0.35 mm/yr, 2.3 mm/year and 0.40 mm/year, based on the papers linked to. Tad Pfeffer says it is 24% glaciers and 14% for ice sheets, leaving 62% for thermal expansion – the numbers I came up with were more like 75% glaciers, 13% thermal expansion, and 12% ice sheets… any comments that might clear this up?

    Comment by Ike Solem — 21 Apr 2008 @ 3:05 PM

  79. My understanding has always been that ice masses, whether ice sheets, ice caps or linear valley glaciers, are divided into zones of accumulation – where ice formation from compressed annual snowfall is greater than melting and evaporation – and ablation zones – where melting and evaporation exceed snowfall and ice formation. Between the two zones is the equilibrium line which in detail varies in location season to season and year to year, but in the longer term either moves higher up the valley glacier or further towards the centre of the ice sheet/cap. Because there is a pressure gradient between the accumulation and ablation zones, the ice moves from te first to the second – from effectively the centre to the edge of the sheet. Movement almost always occurs – even if the ice mass is totally frozen to the bedrock, the upper layers are able to slide by plastic deformation et al over the lower layers; however the movement will be more rapid if the lowest layers melt due to the pressure of the overlying ice, to geothermal heat leakage, and to meltwater from the surface reaching the ice base.

    What I am not aware of from the scientific literature is whether there is any evidence of the shrinkage of the accumulation zones on the Greenland or even Antarctica ice masses – whether the equilibrium line is contracting towards the centre of the ice sheets. Ascertaining this would involve a calculation of the mass gain and loss of ice over large areas. Can this be done? Has it been done? – by remote sensing perhaps, or detailed long term measurements at random locations scattered across the ice sheets. It is a matter of speculation whether air masses, warmer now in consequence of global warming processes, carry moister air into the centres of ice masses (warmer air has a greater potential capacity for water vapour than cold). This might result in an actual increase in ice formation in the accumulation zones; at the same time, the warmer air might cause more rapid melting and evaporation and iceberg calving at the margins of the ice sheets. The result would be a steeper lateral pressure gradient within the ice between the accumulation and ablation zones, which in turn would bring about greater ice velocities as recorded at so many valley glaciers and ice sheets/caps around the globe.

    Comment by Mike — 21 Apr 2008 @ 3:07 PM

  80. A loss of 200 km3 per year from the total 2,650,000 km3 to 3,000,000 km3 of total ice in Greenland is only 0.007% per year.

    At that rate, it will take 13,000 years for all the ice to melt.

    Increasing sea levels by 0.6 mm per year, it will take 1,600 years to raise sea level by 1 metre.

    Basic math indicates that everyone is making this issue out to be way more serious than it is.

    [Response: Hmmm…. and so you expect the ice loss rate to stay constant even as temperature rises? And you aren’t concerned about the Eemian sea level rise (4 to 6 meters) with only a few more degrees warmth around Greenland? – gavin]

    Comment by John Lang — 21 Apr 2008 @ 3:44 PM

  81. Re #80 John Lang – that it will take 1,600 years to raise sea level by 1 metre and 13,000 years for all the ice to melt – basic maths calculation. True – if you assume the ice just stays where it is – which it doesn’t. It slides out from the centre of the ice sheets towards the warmer margins – and it will probably slide out faster the warmer the climate gets – see #79. In other words, glacial melting may be an accelerating process. And this acceleration is likely to be accentuated by the replacement of white snowy/ice surfaces by darker bare rock – and also the rise in sea level prises up or floats the snouts of glaciers and the ice sheets initially anchored to the sea bed, exposing more of the ice bottom surface to melting effects and increasing the downslope velocity of the ice from the interior of the land mass. More and more feedback effects of this nature may be responsible for ever more rapid melting, quite probably at rates far greater than the 200 cubic metres a year quoted in #80.

    Comment by Mike — 21 Apr 2008 @ 4:34 PM

  82. Re #60 — change in w (omega or rotation of earth). From moment of inertia: dw/w = – dr/r assuming conservation of rotational energy and mass. One possible explanation of dw being negative (slow down) is that dr is positive, namely increase in r due to expansion of ocean (not redistribution of mass) due to increase global T (as already measured). There are probably many other, perhaps more valid, hypotheses.


    Comment by jgradie — 21 Apr 2008 @ 5:13 PM

  83. Oopps! My comment should have referenced #64 not #60

    Comment by jgradie — 21 Apr 2008 @ 5:22 PM

  84. At that rate, it will take 13,000 years for all the ice to melt.

    The Hansen (et al) paper last summer completely discounted a period of thousands of years for a calamitous sea level rise. Of the 3 broad spans – decade, hundreds, and thousands – they found hundreds to be most likely and couldn’t rule out a period of decades once broad scale melting was under way. The two of you might be arguing two related things. Your statement was for “all” Greenland ice melt, and the Hansen paper just examined the scale of time for sea levels to rise calamitously. As for the difference, if Florida winds up looking like a wasp stinger, I won’t begrudge Greenland it’s remaining ice.

    Comment by Jeffrey Davis — 21 Apr 2008 @ 5:30 PM

  85. Good question Mike. In the case of Antartica there is not a substantial traditional ablation zone. Mass loss is dominantly via calving and sublimation. Ice shelves in particular are not compatible with ablation zones. On the Greenland Ice Sheet the extent of the area experiencing ablation is determined each year using passive satellite microwave data which distinguishes wet snow from dry snow. It is evident that in the short record available the mean extent has expanded. You are correct that if accumulation increases and ablation increases, the higher balance gradient would necessitate that the glaciers accelerate. It is not evident that the balance gradient has steepened significantly. Even if it does, a large ice sheet outlet glaciers response time is not rapid for such a change to propogate downglacier in the time interval for which we have data. A steeper balance gradient cannot be responsible for the recent observed accelerations. An acceleration from a steeper balance gradient would not be at its maximum at the terminus either. It would be at its maximum near the equilibrium line, where no notable accelerations have been observed.

    Comment by Mauri Pelto — 21 Apr 2008 @ 7:02 PM

  86. Mauri and Nick,

    Thanks for your responses. I can understand that a uniformally accelerating glacier would thin regardless of melting. I had been thinking that in the case of 25 meters of sea level rise over a few centuries (3 oC of warming), perhaps most of the melting in Greenland would occur on land at low altitude rather than in the sea. But, perhaps the transition would not occur in this century as Nick suggests.

    I guess what I am still not understanding is: if the surface melting and resulting lubrication is the origin of the acceleration at the outlet, is not the thinning there better thought of as a result of the increased pressure rather than the “plug” being somehow diminished? What is going on upstream seems more important.

    Comment by Chris Dudley — 21 Apr 2008 @ 7:18 PM

  87. Gavin, You continue to say that current sea level rise is 3 to 4 mm per year. However NASA says since 1992 average is just at 3.05 mm and over the past year there is really no rise or possibly a fall Other types of measurements have run about 1mm to 2mm per year. You are also surely aware that NASA’s Argos system found no rise in sea temperature in the past five years and the U.N. admits there has been no atmospheric increase in temperature since 1998

    Comment by Gary — 21 Apr 2008 @ 7:43 PM

  88. Dear Pat (#73) and d. beck (#59),

    I think that the point that I am trying to make is this:

    That very thin new Arctic sea ice has pretty much been going all to smitherines over the last 3 weeks, and tons of water vapor has been created, which has to go somewhere, and the part of it that ends up hanging over Greenland can hasten the melting, and it can even cause rain. (I think I read somewhere last year that it actually rained on Greenland at altitudes previously thought unlikely.)

    Of course, this cracking up of the sea ice goes on each summer, but now it is occurring earlier in the year and faster over a much greater extent.

    OK, and this is weird: I just looked at the sea ice extent graph, and the updated graph from today shows completely different data compared to yesterday’s graph. That’s a really big change. Fortunately, I saved a copy of yesterday’s graph on my HD, which I will copy both to my blog — I think we all deserve an explanation from the National Snow and Ice Data Center.

    Comment by Tenney Naumer — 21 Apr 2008 @ 9:04 PM

  89. 87:
    The ocean temperature that matters for sea level rise would be the (coefficient of expansion weighted) average sea temperature. Only a tiny fraction of this water is close enough to the surface to be measurable by sea surface temperature observations. If indeed sea surface temperatures have in fact stabilized (this is likely a temporary effect) the conduction/advection of heat into the deep oceans would still continue for many hundreds of years. It may be possible to indirectly measure ocean basin average temperatures by sound travel times. In principal gravity measurements would do the trick, but I suspect the noise due to ordinary geological processes might overwhelm the warming signal.

    Comment by Thomas — 21 Apr 2008 @ 9:06 PM

  90. Re #88

    “OK, and this is weird: I just looked at the sea ice extent graph, and the updated graph from today shows completely different data compared to yesterday’s graph. That’s a really big change. Fortunately, I saved a copy of yesterday’s graph on my HD, which I will copy both to my blog — I think we all deserve an explanation from the National Snow and Ice Data Center.”

    I noticed the sudden downturn on the graph yesterday and also that it coincided with images that were missing signal in a couple of regions. I anticipated that the graph would shortly be adjusted and as you have shown they were.

    Comment by Phil. Felton — 21 Apr 2008 @ 10:02 PM

  91. Aaron Lewis (#75)

    What I am not seeing here is a discussion of the recent reports of supraglacial lakes’ hydrofracture of kilometer thick ice resulting in the advection of heat to basement ice. This means the basement ice suddenly warms, and warmer ice is weaker. Weaker ice under the stress of supporting kilometer(s) of ice is subject progressive collapse energetically driven by the potential energy of the supported ice structure.

    Storm tracks will be moving progressively north over the course of the century, more rainfall at higher latitudes — not to mention the polar amplification and extended “growing” (“melting”) season. Rain and ice makes slush. Of course what I am wondering about is on the other side of the earth, where ocean currents are circulating more heat content to greater depths, where rivers run under glaciers, the base of which are below sea level — particularly along the West Antarctic Peninsula.

    Additionally, the deeper the heat content goes, the more likely it would seem that it will destabilize methane cathrate deposits sooner rather than later. Not that I have heard anything about such deposits there, but higher oxygen content in the water implies more organic material, which means more material from which to generate methane. And we have found plenty of such deposits in the Arctic region.

    Comment by Timothy Chase — 22 Apr 2008 @ 12:55 AM

  92. Thanks Gavin, I thought you were on vacation or something.

    Comment by d. beck — 22 Apr 2008 @ 1:24 AM

  93. Tenney@88:
    Yes, I noticed the NSIDC graph change too. I assume that the earlier graph, which was extreme, was a data processing error. The current graph is still plenty alarming.

    These things happen at the cutting edge. We had similarly alarming sea ice area graphs from Cryosphere Today for a short while earlier in the year, before they found and fixed a bug.

    Comment by Nick Barnes — 22 Apr 2008 @ 3:34 AM

  94. Aaron 75, your description takes my concerns a useful step further, as you have identified a collapse mechanism for deep warming ice. As you say if there is water flowing beneath ice then the roof of ice over that flow is warming, and that heat will move upwards progressively weakening and softening the ice mass – even though the ice near the surface looks ok.

    Your Missoula Effect sounds like a classic non-linear response. I think the risk posed by these effects utterly overwhelms concerns over calving vs melting.

    The recent discoveries of great rivers flowing beneath EAIS and WAIS – including vertical surface movement of plus/minus 3 metres in a matter of a few weeks on ice sheet surfaces which were coupled with changes in basement water flows demonstrates that the security of grounded ice is a myth. Where water flows there energy flows.

    The grounded ice can still be full of voids above sea level that allow melt water to circulate bringing its energy contribution, and below sea level where melt water runs out as basement flows under upstream pressure and sea water flows in and out under tidal differential heads. Any crack in the ice joining with the ocean will be influenced by tidal effects, and these cracks will not self-heal. They will widen laterally as the sides and base melts away, and as the ceiling takes heat out of the water vapour. Below sea level water circulation within the grounded ice mass can also be driven by salinity and temperature gradients.

    It is not hard to envisage an ice sheet like a hair-brush sitting with its bristles in the water, and the back of the brush being the above high tide solid ‘lid’ on a very unstable structure.

    Comment by Nigel Williams — 22 Apr 2008 @ 6:33 AM

  95. Re #88, even without the strange downward line on the olf chart the nww chart seems bad enough and that downward turn could perhaps be significant becaue the ice cannot reform until next winter once gone I presume?

    Comment by pete best — 22 Apr 2008 @ 7:05 AM

  96. Chris the point you miss, is that it is not the surface melting that has driven the acceleration. End of story. It is not what is happening upglacier that has been the key. If either was the case then flow would be higher in the summer and acceleration would not have begun at the calving front. Get rid of your false assumption that the meltwater is driving the acceleration. Aaron and Nigel: If you look at the 1962 maps of Greenland or any satellite image from the 1970-1990’s period, the glacier lakes are present. They can deliver some heat, but it is not a new phenomenon. Their extent and volume you would think would be increasing, but this has not been documented. Thus, we do not have a mechanism for additional significant basal warming. The water at 0 C also cannot warm the basal ice which is at the pressure melting point much.

    Comment by mauri pelto — 22 Apr 2008 @ 9:40 AM

  97. There’s a lot written about heat from friction at the glacier/rock interface; don’t assume a glacier is cold at the bottom, it may be warmer.
    Some tidbits:

    “Warm based glaciers are at the pressure melting point at their bed. Heat from the Earth and from basal friction provides energy to melt ice at the bed, …”

    And I’ve been getting in William’s hair* about this over at Stoat

    *If I have seen further than others, it’s by getting in their hair

    Comment by Hank Roberts — 22 Apr 2008 @ 9:44 AM

  98. I have a question about this part of the post:

    “The second mechanism is a ‘Jakobshavn effect,’ coined by Terry Hughes (1986), where a small imbalance of forces caused by some perturbation can cause a substantial non-linear response. In this case an imbalance of forces at the calving front propagates up-glacier.”

    But, we know that Dr. Robert Corell witnessed a “substantial non-linear response” to something when 5 km of the Ilulissat glacier roared past in only 90 minutes.

    And, while presently it appears that events at the front of the glaciers are the significant areas, won’t it actually be the effects of things such as glacier lake collapses upstream that become ever more frequent and thus more significant, as time goes on?

    Comment by Tenney Naumer — 22 Apr 2008 @ 9:58 AM

  99. Re: #95

    Pete, the remnants of the older sea ice have all already broken up and appear ready to weaken substantially.

    What I would like to know is if long-term, time-series data are available concerning the temperatures of the Northern Ocean, at various depths (surely the U.S. Navy has something on this), and/or if there are in place instruments to measure these temperatures this year and in the future.

    Comment by Tenney Naumer — 22 Apr 2008 @ 10:06 AM

  100. Re: #96

    Sorry Dr. Pelto, I missed your remark about the glacier lakes. But the number and size of the moulins are increasing somewhat exponentially, are they not?

    Comment by Tenney Naumer — 22 Apr 2008 @ 10:09 AM

  101. Another contributing factor here appears to be the Arctic Spring Haze of airborne pollutants –

    Industry, transportation, and biomass burning in North America, Europe, and Asia are emitting trace gases and tiny airborne particles that are polluting the polar region, forming an “Arctic Haze” every winter and spring. Scientists suspect these pollutants are speeding up the polar melt. . .

    “This is our first airborne deployment of a powerful new suite of instruments in the Arctic,” said ARCPAC lead scientist Dan Murphy, also of NOAA’s Earth System Research Laboratory. “When we analyze all the data, we’ll be able to piece together the equivalent of a ‘high-def’ movie of the atmosphere as springtime sunlight warms the region and sparks a chain of chemical reactions.”

    Warming air melts the ice faster, thin ice is blown apart by the winds and currents in the Arctic Ocean, leading to large areas where the ocean can freely exchange heat with the atmosphere (the rate being dependent on wind and other factors – for more see ).

    The actual effect of all that is not intuitive, but has to be analyzed using computers for data collection, data analysis and modeling. A lot of “skeptics” don’t seem to understand the basic process of model design and refinement – make a model, see what it misses, refine the model, see what it misses, and keep iterating. It is not a “curve-fitting process” – real data is used to initialize the model, but then no comparison is made to the real data until afterwards. So, the question is, has anyone made a model, which, when initialized with say, 1980 conditions, predicts the rapid loss of sea ice in the Arctic? Such a model might have a difficult time predicting the amount of Arctic spring haze pollution each year – so that might be an external variable, set by the researcher.

    It’s also worth looking at some glaciers that are not going with the overall melting trend:
    Howat et. al 2006: A precipitation-dominated, mid-latitude glacier system: Mount Shasta, California

    Temperature is often seen as the dominant control on inter-decadal glacier volume changes. However, despite regional warming over the past half-century, the glaciers of Mount Shasta have continued to expand following a contraction during a prolonged drought in the early twentieth century, indicating a greater sensitivity to precipitation than temperature. We use the 110 year record of fluctuations in Mount Shasta’s glaciers and climate to calibrate numerical glacier models of the two largest glaciers. The reconstructed balance and volume histories show a much greater correlation to precipitation than temperature and significant correlation to oscillatory modes of Pacific Ocean climate.

    An approximately 20% increase in precipitation is needed for every 1°C increase in temperature to maintain stability. Under continued historical trends, oscillations in climate modes and random variability will dominate inter-decadal variability in ice volume. Under the strong warming trend predicted by a regional climate model, the temperature trend will be the dominant forcing resulting in near total loss of Mount Shasta’s glaciers by the end of the twenty-first century.

    So, that’s a very useful number – 1C rise in temp must be countered by a 20% increase in precipitation. Is it widely applicable, or limited to mountain glaciers, or to Shasta in particular?

    Comment by Ike Solem — 22 Apr 2008 @ 10:52 AM

  102. Re 91
    Greenland has the potential to make its own weather. It rises to produce orogenic precipitation, whchever way the wind blows. Its ice provides a temperature diferential to steer winds from either the North Atlantic or the newly ice free Arctic Ocean.

    These days, I think of Greenland less as a pile of ice accumulating snow, and more as a water shed with a summer (March to November) rainy season – with no plants to retard runoff. Where does that water go? How much energy does water flowing through a moulin release?

    Comment by Aaron Lewis — 22 Apr 2008 @ 11:08 AM

  103. Re #101

    Regarding the comments about arctic haze, our ‘Johnny on the spot’, Wayne Davidson has been observing very clear skies for the last couple of years and associates that with the warming and melting.

    Regarding the 1º/20% increase number I would expect that to be site dependent although there must be a link everywhere.

    Comment by Phil. Felton — 22 Apr 2008 @ 12:08 PM

  104. Mauri Pelto,
    Thanks for the article and your active involvement in this thread.

    #87 Gary,

    Didn’t you think to check around for the uncertainty involved in the 3.05mm/yr you quote? Or do you really think it’s 3.05 exactly with no uncertainty?

    From IPCC AR4 Chapter 5
    “Numerous papers on the altimetry results (see Cazenave and Nerem, 2004, for a review) show a current rate of sea level rise of 3.1 ± 0.7 mm yr–1 over 1993 to 2003 (Cazenave and Nerem, 2004; Leuliette et al., 2004)”
    So that’s 2.4 to 2.8mm per year. If you’re worried about supposed IPCC bias you may want to try reading the references from which those AR4 statements are drawn.

    The second item you link to merely shows that a lot of work is being done by the experts, so it is pointless to draw conclusions at this stage with respect to Argo, deployed in 2003. However it should hhave been apparent to you from your first link that the altimeters show a sea level rise from 2003 to 2007 at the same rate as previously. And the previous rise was in part due to thermal expansion. So if the oceans have really cooled…

    Well it’s discussed in the article, which is an interesting lay person’s read through to the final paragraph(hint).

    Sorry but the Investor’s Business Daily page at the end of your final link is a complete joke (It’s your use of it that suggests to me you’re in denial. Getting suckered by such carp is almost on the same level as believing a schoolkid can undermine a whole branch of science.)

    You state “the U.N. admits there has been no atmospheric increase in temperature since 1998” The Investor’s Business Daily(IBD) article you link to implicitly supports this contention. Yet it is only the article itself that makes this incorrect assertion, not the UN!

    Have a look at the CRU global average temperature: Have a look at that graph, it’s telling the true sceptic something crucial. (scientists should be sceptical, denialists cling to their precious beliefs in the face of evidence)

    By any of the 3 main datasets it is clear that 1998 was an outlier, this is accepted, it is only those in denial (like the author that article) who make a straw man of 1998. i.e.

    From the IBD article:
    “If that’s the case, then why can’t the Pacific’s El Nino current, which played a large part in the warm reading for 1998, simply been seen as a “variability” and not part of a greater warming trend? Because it doesn’t fit the agenda?”
    To which I answer:
    “El Nino IS seem as natural variability, you dolt.”

    It would be correct to say “global average temperature remains below the 1998 peak in the CRU dataset”. But it is clearly wrong to say “There has been no increase in global average temperature since 1998.” Both of those statements are factual and as Mark Twain once said (paraphrased); a person is entitled to their own opinion, but they are not entitled to their own facts.

    Click on this link
    Select “Hadley/Reyn_v2” as ocean data source. Then select year/month or year and annual/seasonal averages. You will find that the cooling is predominantly in the southern hemisphere’s oceans, mainly the Pacific – that’s the La Nina.

    This supported by considering the GISS 3 latitude banded plot:
    And is the detail I suggested is also available to a sceptic looking at the CRU plot. The northern hemisphere continues to warm on a multi-year timescale. Tamino also covers this here: NB the last graph and text around it.

    In short there is NO evidence that Global Warming has stopped.

    To underline the errors you’ll get (either way – alarmist or denialist) in fussing over blips, in the IBD article they also state:
    “Of course the IPCC spins the news.

    You should look at trends over a pretty long period,” said WMO Secretary-General Michel Jarraud, “and the trend of temperature globally is still very much indicative of warming.”

    Jarraud is of course not spinning, in both the reference to La Nina and the need to look at the relevant timescales is a warming to view data in it’s relevant context. The weakness of focussing on year to year fluctuation is demonstrated by playing “time machine” with one of the temperature datasets. Pretend we’re in 1991 on the CRU dataset, after that you have 3 cooler years cooling, then the temperature rises again. 2007 was only the 4th year below the previous peak. Without a La Nina it may have been higher, as for 2008, we’ll see. You could also try “Time Machine” with your altimeter SLR graphs.

    In the IBD article they refer vaguely to a ‘bombshell’ paper that shows “”the weight of the current evidence . . . supports the conclusion” there is no agreement between the models and the observation temperatures.”
    I can only presume this is a reference to Douglass et al which was done by RC last year The headline results of Douglass are (IMHO) not worth the paper they’re written on.

    Anyway, at least the last sentence gave me a giggle:
    “The global warming debate is not over. Indeed, the debate is beginning to favor the skeptics.”

    Yes that’s it old bean, if you keep on repeating it, it’ll come true.

    Soon the denialists will be clicking the heels of their shoes together as they wish with closed eyes. Unfortunately for them a physical system like the Earth’s climate is not amenable to sleight of hand and “Wormtongue” council.

    Comment by CobblyWorlds — 22 Apr 2008 @ 1:54 PM

  105. Re#99
    “Pete, the remnants of the older sea ice have all already broken up and appear ready to weaken substantially.”

    Just look here for evidence of that (particularly the Beaufort sea):


    Last year

    The white sea areas are perennial ice, the grey ice formed this winter

    Comment by Phil. Felton — 22 Apr 2008 @ 2:18 PM

  106. Phil, I don’t find any discussion of aerosols or spring haze in the Wayne Davidson reports – mostly he is talking about the effects of warm Arctic air. For a better discussion of that, try this news report:

    So, he hasn’t been observing “very clear skies” – just a warm air mass. In fact, there’s no way he could get comprehensive data on aerosols from his weather station.

    From the NOAA people:

    Analyses of observations and recent climate simulations suggest that, in addition to greenhouse gas-induced warming and feedbacks, Arctic warming may also be caused by shorter-lived climate forcing agents. In particular, four processes have been postulated to contribute significantly to observed atmospheric warming in the Arctic and reductions in sea ice there. These processes include:

    1) direct warming of the lower troposphere by the absorption of solar radiation and IR emission by aerosol particles from anthropogenic and biomass burning sources (e.g., Treffeisen, 2005; Ritter et al., 2005),

    2) changes in snow melt due to deposition of soot (light-absorbing carbon) to thesurface in springtime (Hansen and Nazarenko, 2004; Flanner et al., 2007),

    3) increases in IR emissivity of wintertime and springtime clouds in the Arctic due to the effects of anthropogenic aerosol particles on cloud properties (Lubin and Vogelmann, 2006; Garrett and Zhao, 2006), and

    4) direct radiative effects of tropospheric ozone in the Arctic (Mickley et al., 1999; Hansen et al., 2005).

    During the International Polar Year of 2008, NOAA will engage in an airborne field measurement campaign targeted at improving understanding of these four climate-relevant processes. This effort will be focused on direct measurements of properties and processes that can be used to reduce uncertainty in radiation and climate models. . .

    NOAA’s “State of the Arctic” reports of the past few years make for interesting reading in light of recent events – and now they may be wondering why they got it so wrong. (Quote: “There are indications that some components of the physical system may be recovering and returning to the recent climatological norms observed from 1950 to 1980.” – Oct 2006, NOAA State of the Arctic Report)

    Comment by Ike Solem — 22 Apr 2008 @ 3:19 PM

  107. Re 96

    If water falls a kilometer down a moulin, where does its kinetic energy go? Not to the basal ice? If the Melt water is 0C and the basal ice is -10C, why can’t the falling water warm the colder ice below? Suppose the basal ice is near the maximum temperature at which it has the mechanical strength to support the ice above it. (This would be some temperature below its pressure melting point.) Then, additional heat is advected to the basal ice. The additional heat weakens the ice. It is not necessary to melt the ice in-situ, only warm to it the point where it loses the required structural strength to support its overburden. The situation at the base of the GIS where ice is under significant pressure is somewhat different from a glacier. (The ice in a glacier is more likely to relive stress by creep. Except see

    What is the mechanical process of ice failing? Is that process different when ice is near its pressure melting point? What are the molecular physics of ice in structural failure under high pressure? By and large, the engineering community has only looked at the smaller loads found in the near surface environment. (My calculation is that under very high pressures, ice failure releases heat. This is not inconsistent with and

    When I look at ( ), I see increasing areas of melt and increasing days of melt across the GIS, melt season by melt season. Since melt drains off ice, or down through moulins in the ice very rapidly, area and depth of supraglacial lakes is not proportional to total surface melt. Moreover, colder ice tends to hold water on its surface better than warmer, weaker ice. Thus, as a percentage of total melt, I would expect less water on the surface of the Greenland Ice Sheet now, than in the past. On the other hand, now and in the near future, I would expect a larger volume of melt water advecting heat to the base of the ice.

    Comment by Aaron Lewis — 22 Apr 2008 @ 3:26 PM

  108. Re: #105

    Phil, thanks for the images. Yikes!

    OK, I am going to go out on a limb here and say that, in my opinion, due to lack of funds, personnel, equipment, rapidity of the changes, etc., research so far on Greenland has largely been limited to the tail of the elephant. We need more indepth surveys of the ice streams and what their dynamics are.

    Comment by Tenney Naumer — 22 Apr 2008 @ 4:54 PM

  109. The amount of meltwater generated has increased, the amount draining through moulins has increased. The amount of increase is not known, but it has not increased by anything close to 50%. In the rapidly moving outlet glaciers the base of the glacier is at the pressure melting temperature not -10 C. The cold water cannot provide that much heat to the ice via advection. The amount of water is also not that much greater. Focus on what is happening at the glacial front, that is where the critical changes have been occurring. The point of this point was to clear up the misconception that increased meltwater delivered to the bottom of the GIS is driving the acceleration of the large outlet glacier. It is not, for all of the reasons noted.

    Comment by Mauri Pelto — 22 Apr 2008 @ 6:05 PM

  110. Despite all our speculation, this seems conclusive (I hadn’t seen this when I posted immediately after it):

    mauri pelto Says:
    22 April 2008 at 9:40 AM

    “… it is not the surface melting that has driven the acceleration. End of story. It is not what is happening upglacier that has been the key. …..”

    and goes into detail about why.

    Comment by Hank Roberts — 22 Apr 2008 @ 7:38 PM

  111. “Phil, I don’t find any discussion of aerosols or spring haze in the Wayne Davidson reports – mostly he is talking about the effects of warm Arctic air. For a better discussion of that, try this news report:

    So, he hasn’t been observing “very clear skies” – just a warm air mass. In fact, there’s no way he could get comprehensive data on aerosols from his weather station.”

    So I was dreaming when I read this:

    “Now in very warm Resolute April, again with clear skies,”
    “especially clear Arctic skies”
    “its hard for me to understand what LaNina has to do with Arctic clear air.”
    “Extraordinary persistence of the cold temperature North Pole since last March 07 has its roots in clear air over Ellesmere North Baffin Islands area,”
    et cetera?
    It’s clear from reading his reports that Wayne thinks the Arctic air is currently unusually clear and that this is related to the melting. No what instrumentation he has I’m sure that he would notice ‘haze’.

    Comment by Phil. Felton — 22 Apr 2008 @ 8:31 PM

  112. Perhaps I’ve missed this, but someone upthread asked how we know this melting isn’t a return to equilibrium from an earlier cool period.

    I looked at the terrain near the glacier’s outflow area and saw a significant number of marks from where a much larger glacier had been there, presumably millenia ago.

    Obviously, the glacier is melting, my question is — how much and for how long? Because I’m starting to get concerned by the rhetoric about “We have to do even more, because it might be even worse!” and the prospects in that another century or two there’s a “Little Ice Age” and entire world has shifted into “It’s getting warmer! It’s getting warmer!” The worst possible scenario is often not the one that isn’t planned for — it’s the one that was intentionally ruled out. Knowing how much and for how long would be a good thing in terms of understanding how much this might be a response to emerging from the early 1800’s.

    Comment by FurryCatherder — 22 Apr 2008 @ 9:11 PM

  113. #111 Phil and Ike. There is definitely warming Up Here, along with remarkably clear air, an astronomer friend calls it: “molecular air” . First there seems to be a small misunderstanding of the Arctic . There are no highways, no factories, nothing but Antarctic like wilderness without all the ice. A single station observing an increase in air quality has a range much greater than a simple dot on a map. The atmosphere moves, not the station, so a continuous stream of observations of clear air is very significant. There are two basic ways of seeing clear air, by colour and also by brightness during the long night. Over the last few years, both colour, so crisp “molecular colours” were noted along with brighter twilights ( really startling High Arctic dwellers) at the horizon (thickest air possible), indicating a reduction in air turbidity amongst other refraction effects.

    using a photometer one can observe various brightness depending on Aerosols, Ice crystals ,clouds and refraction effect as well. Clear cloudless air has been a dominant player here
    since March 07. Not only at one station, but as reported during the great melt of 07.

    I would also like to thank Mr Pelto for his helpful presentation and feedback, I still wonder if there are there are deep inside the core temperature measurements within Ice sheets…

    Comment by Wayne Davidson — 22 Apr 2008 @ 9:52 PM

  114. Mauri,

    (#96) I think this puts a finger on where I am not understanding the argument. I don’t see why a seasonal signal 35 km upstream would be preserved as seasonal and not just show up as a general acceleration.

    The thinning at the outlet is not due to melting, correct? Thus, I am picturing it as owing to acceleration through mass conservation. Is this a problem?

    After this portion, I think I follow: the thinner ice floats closer to the shore and thus the calving front retreats. This then can provide a feedback for more acceleration, more thinning and further retreat of the front.

    But, what was the origin of the original acceleration and the initial thinning? If it is merely chaotic, then the vissitudes at the outlet have no importance. The calving front will advance and retreat with no particular effect on average flow. If it is driven by an upstream requirement for greater outlet capacity then, again, it would seem that the upstream effects, where more material is being supplied by the effects of AGW is the thing to watch. If, contrary to the idea that enhanced local melting has little to do with the behavior of the outlet, warming is the cause of the thinning, then we see two potentially independent behaviors that could both contribute to sea level rise: A clearing of the plug and a larger supply of material to bubble on out.

    Further thoughts would be appreciated.


    Comment by Chris Dudley — 22 Apr 2008 @ 10:05 PM

  115. Regarding the ongoing changes in the Arctic:
    Hoye et. al. Rapid advancement of spring in the High Arctic (2007)
    U.S. scientists to study Arctic smog

    Globally, the net impact of aerosols is to cool climate, partly offsetting the warming from a build up of human-generated greenhouse-gases, explains Ravi Ravishankara, who heads the chemical sciences division of the National Oceanic and Atmospheric Administration’s Earth System Research Laboratory in Boulder, Colo.

    “But,” he adds, “the Arctic is a strange place.” Aerosols that form Arctic haze appear to warm the region, he says.
    Although the project is roughly halfway through its first three weeks in the field, researchers are already noting the region’s role as a caldron for emissions flowing up from lower latitudes.

    “We’ve seen European pollution, North American pollution, Russian pollution. We’ve seen Siberian forest-fire plumes already, in April. We’ve seen plumes coming all the way up from Indochina,” where locals use fire to clear farmland, Dr. Jacob says. This stew is aging in the Arctic, combining to form Arctic haze.

    So, has spring sprung across the northern Arctic? Probably around now – see the satellite view at

    One aspect of this is the poleward atmospheric transport of organic pollutants. It’s the reason that so many organochlorine compounds are found in Inuit peoples and mammals in the region. They might actually have a minimal aerosol effect themselves – but they do condense out in the cold. See for example:
    Kidd et al “High Concentrations of Toxaphene in Fishes from a Subarctic Lake(1995), Science

    Concentrations of toxaphene and other organochlorine compounds are high in fishes from subarctic Lake Laberge, Yukon Territory, Canada. . . . A combination of low inputs of toxaphene from the atmosphere and transfer through an exceptionally long food chain has resulted in concentrations of toxaphene in fishes that are considered hazardous to human health.

    As a result of this phenomenon, polar bears have among the highest levels of persistent organochlorine pollutants of any species:

    The mechanism is also understood:
    Blais et al 1998 “Accumulation of persistent organochlorine compounds in mountains of western Canada”, Nature

    Higher than expected occurrences of these compounds in remote regions are the result of long-range transport in the atmosphere, precipitation and ‘cold condensation’ — the progressive volatilization in relatively warm locations and subsequent condensation in cooler environments3,4 which leads to enhanced concentrations at high latitudes.

    Organochlorines are only one part of what gets transported north:

    Garrett and Verzella say the first report of Arctic haze pollution usually is credited to a U.S. Air Force meteorologist J. Murray Mitchell, who in 1957 described “the high incidence of haze at flight altitudes” during weather reconnaissance missions from Alaska over the Arctic Ocean during the late 1940s and 1950s. Mitchell was credited in the 1970s by Glenn Shaw from the University of Alaska, Fairbanks, and his collaborators Kenneth Rahn and Randolf Borys, from the University of Rhode Island, who were the first to discover the haze contained high levels of heavy metals, including vanadium, suggestive of heavy oil combustion. In a later study, Rahn and Shaw said: “Arctic haze is the end product of massive transport of air pollution from various mid-latitude sources to the northern polar regions, on a scale that could never have been imagined, even by the most pessimistic observer.”

    So, it’s easy to see how it could have a contributing influence. Indeed, it seems to be having a large influence in Alaska right now:

    April 22nd, 2008 – The cloudy, off-white haze crept into Anchorage over the weekend, obscuring the once-crisp view of the Chugach Mountains with a smog-like quality more suited to a view of the Los Angeles skyline. Murky skies have spanned the state, showing up from Fairbanks to Kodiak and Valdez to the Aleutian Islands.

    Comment by Ike Solem — 23 Apr 2008 @ 1:39 AM

  116. #105, 108:

    Sea ice images shrunk and animated here.

    Comment by GlenFergus — 23 Apr 2008 @ 2:00 AM

  117. There is of course no lack of energy in water at 0C. Because that’s actually water at 273Absolute running past ice at say only 243A. There’s no shortage of dT there, and the transfer will happen. Interesting, when you think of it in those terms; liquid water – the bete noir of the cryosphere – is perforce always at or above 273A while precious ice is always colder. Where water flows the energy goes, absolutely.

    Comment by Nigel Williams — 23 Apr 2008 @ 3:39 AM

  118. There is mention above and elsewhere that parts of interior Greenland (and Antarctica) beneath the glacial ice sheets are below sea level. Are there topographic maps which show this? And are the below-sea-level interior basins potential seas like the Aral or Caspian or are they potentially joined to the ocean like Hudson Bay or the Mediterranean? If the latter is there a below-present sea level connection (blocked by glacial ice) or is some amount of sea level rise required?

    Comment by George — 23 Apr 2008 @ 7:04 AM

  119. George, have a look at:

    Check out the centre of Australia at around +70m, and google-fly over Ukraine through the Caspian then north to the Artic ocean looking for land levels above/below 80m Florida and Amazonia long gone. Nice places to sail, but you wouldnt want to live there.

    Comment by Nigel Williams — 23 Apr 2008 @ 8:49 AM

  120. It has been understood for sometime based on Investigations by the University of Alaska and ETH Zurich that the majority of the fast flow of Jakobshavns Isbrae is from deformation of the basal ice not sliding (Luthi et al, 2002) Borehole measurements revealed the existence of a substantial temperate layer, at the pressure melting point near 0oC at the base of the ice stream. The thickness of the temperate basal layer is between 40 and 50 m at the drill site, with a very high temperature gradient (0.11 K/m) in the ice above it.

    Continued thinning of the terminus reach of outlet glaciers has been observed by Pfeffer (2007) to explain rapid retreat and acceleration of tidewater glaciers such as Columbia Glacier, Alaska.
    The comparatively short duration of speed-up on Jakobshavns, Kangerdlugssuaq and
    Helheim Glacier along with the speed up beginning at the calving front and working up glacier suggest this as the cause (Thomas , 2004; Howat and others, 2007). Howat and others (2005) proposed that the observed speed-up resulted from two effects, increased effective stress over the glacier’s main trunk following retreat of the calving front, with subsequent thinning propagating upglacier resulting in steeper surface slopes and velocity increase. Retreat of the calving front terminated where the bed slope reversed, allowing the glacier to re-equilibrate (Howat and others,2007). The reversed bed slope would inhibit steeper surface slopes from propagating upglacier.

    Csatho, (2007) note that the periodic thinning near the terminus of Jakobshavn has led to accelerated retreat in the past as well. Thinning can and does occur via increased ablation, which can then lead to greater flotation and acceleration.

    Csatho, B., A.F. Schenk, C.J. van der Veen, and W. Krabill (2007), Intermittent
    thinning of Jakobshavn Isbræ, west Greenland, since the Little Ice Age. Journal
    of Glaciology

    Howat, I.M., I. Joughin, S. Tulaczyk, and S. Gogineni (2005), Rapid retreat and
    acceleration of Helheim Glacier, east Greenland. Geophysical Research Letters
    32, L22502, doi:10.1029/2005GL024737.

    Howat, I.M., I. Joughin, and T.A. Scambos (2007), Rapid changes in ice discharge
    from Greenland outlet glaciers. Science 315, 1559-1561.

    Joughin, I., W. Abdalati, and M. Fahnestock (2004a), Large fluctuations in speed on
    Greenland’s Jakobshavn Isbræ glacier. Nature 432, 608-610.
    LÜTHI, M. et al, 2002; Mechanisms of fast flow in Jakobshavn Isbræ, West Greenland: Part III. Measurements of ice deformation, temperature and cross-borehole conductivity in boreholes to the bedrock Journal of Glaciology, 48,162, 369-385

    Pfeffer, W.T. (2007), A simple mechanism for irreversible tidewater glacier retreat.
    Journal of Geophysical Research 112, F03S25, doi: 10.1029/2006JF000590.
    Thomas, R.H. (2004), Force-perturbation analysis of recent thinning and
    acceleration of Jakobshavn Isbræ, Greenland. Journal of Glaciology 50, 57-66.

    Comment by mauri pelto — 23 Apr 2008 @ 9:20 AM

  121. Dr. Pelto, do you think there is anything that can be said about changes in behavior of meltwater on the surface of the ice (Greenland and Antarctic)? Is there a change in the amount or is it just being noticed and reported? If there’s a change on the surface, is there more meltwater going down into the ice? Is water in a crack able to make cracks larger? and if so, is anything known about what’s going on deeper?

    I get the impression that “no data” is the answer about this, and wonder if I’m missing something — is it being talked about, even if nothing’s yet being published?

    Comment by Hank Roberts — 23 Apr 2008 @ 10:46 AM

  122. Re: #90 Phil. Fenton, #93 Nick Barnes, #105 Phil, #116 GlenFergus

    The National Snow and Ice Data Center continue to update their graph in strange and miraculous ways. The graph of April 23 is now nearly identical to the graph of April 20, and now, once again, shows a very precipitous drop in Arctic sea ice relative to 2007.

    At this point, it seems illogical to assume that processing errors and missing signals are at the root of these modifications.

    All three graphs can be viewed here:

    Thank you to Leon in Holland for giving me the heads up, again.

    Comment by Tenney Naumer — 23 Apr 2008 @ 11:55 AM

  123. A few months ago there was a discussion about Greenland, and somebody post a link to graphics they had made of Greenland’s bedrock topography and a rendition of what Greenland would look like if ice free. The center of the island was open ocean, and there was a ring of mountainous islands around it. I’ve tried to find that link for george, but have had no luck.

    Comment by JCH — 23 Apr 2008 @ 12:31 PM

  124. Re: #123 JCH

    See the link in comment #18

    Comment by Tenney Naumer — 23 Apr 2008 @ 1:43 PM

  125. Using a value of 3m SLR for BAU-induced structural and dynamical suppression of the world economy (estimated by mapping deep into characteristic supply chains through specific plants and linking these up through the building blocks of the current global economy, and assessing capex, operational integrity, and safety margins etc as a function of SLR) and taking a value for world GDP as US$ 54,000,000,000,000, and equating 3 m SLR to 1.2 million km3 of ice (from the data in post 47), one can estimate the economic value of each km of ice. The value is about 44 million US $ per km3 of ice. That’s a lot of money being melted away.

    Comment by mg — 23 Apr 2008 @ 3:16 PM

  126. mg (125) — A current estimate of the world’s GDP is $76 trillion per year, not 54 any more…

    Comment by David B. Benson — 23 Apr 2008 @ 3:34 PM

  127. >National Snow and Ice Data Center charts

    page says: “temporary spurious results may occur.”

    “The satellite data sources for these products, while generally providing complete coverage, are subject to gaps (shown in dark grey) in coverage because of satellite operations. In the daily extent timeseries, gaps are replaced with values interpolated from surrounding days, but temporary spurious results may occur. The current satellite source is aging and showing more frequent data gaps. NSIDC is investigating a reliable replacement data source. —Credit: National Snow and Ice Data Center”

    Comment by Hank Roberts — 23 Apr 2008 @ 5:04 PM

  128. Ike, #115, I find no contradictions with all your info, Arctic Haze is simply not as strong as it use to be, at least here, there are still toxins falling mixed with ice crystals, the recent air clarity may be attributed in part to a dominant wind change from Northwesterlies to Easterlies. A warmer atmosphere everywhere and in Russia especially may have reduced heating pollution a great deal, also abating haze. Over all, its been better “smog” wise for a few years now.

    Comment by Wayne Davidson — 23 Apr 2008 @ 9:28 PM

  129. #125 mg,

    Sorry, but in view of the uncertainties involved, I (and many others) find all such attempts to “value” the “fixtures and fittings” of this planet utterly unconvincing.

    In light of our continued inadvertent “terraforming” of the Earth, it’s worth remembering that when one says of a person that they “know the cost of everything and the value of nothing.” It is not meant as a compliment.

    Comment by CobblyWorlds — 24 Apr 2008 @ 12:52 AM

  130. Thanks, Tenney, for the link at #18 to the ice-free Greenland simulation. I was hoping for something more like a topo map, but even from the link it appears that the interior below sea level basin(s) of Greenland are connected to the ocean by below sea level channels. If this is the case, is it only the weight and strength of the ice which currently fills the channels that prevents the ocean from reaching into the interior basins? And are there conditions of glacial deterioration under which the North Atlantic will be able to force its way below the ice?

    Comment by George — 24 Apr 2008 @ 6:44 AM

  131. 126. it doesn’t really matter which value for world gdp value one takes although care should be taken not to use estimates too close to the point of global economy phase transition. the point of post 125 is to highlight that economic/financial connection can be made between phase transformation of ice sheets and BAU suppression of circuit configurations underpinning the current form of the world economy. more complex analysis of this looks at autocatalysis mechanisms during the time-lagged damage cascades and potential opportunities for recovery & reconstruction from BAU-derived economic residue. Better to mitigate.

    Comment by mg — 24 Apr 2008 @ 7:04 AM

  132. George, try this:

    Comment by Phil. Felton — 24 Apr 2008 @ 7:54 AM

  133. George, this is another one for you:

    Comment by JCH — 24 Apr 2008 @ 8:27 AM

  134. 129. there is no attempt whatsover to attempt to suggest that anyone knows the cost of everything or even the cost of a lot of things or even the cost of a few things. the issue to be addressed is to find ways of creating dialogue between those who are observing the disastrous events unfolding in nature and those whose world view is driven by financial and economic matters, irrespective of the planetary implications. if the communications disconnect cannot be addressed urgently and if people driving the massive engines of GHG emissions cannot be woken up then that is very sad, for many will suffer. at the end of the day, each of us in our own way trys to do something to give others a way of looking at this situation and using tools & techniques to create common forms of understanding of this perilous situation. It is through common forms of understanding that hopefully the greatest and fastest and most effective forms of mitigatory actions can be achieved. the planet is in great peril and everything living on the planet is in great peril; no-one in their right mind would want any compliments. these are the days when all of us should be feeling shame for what is happening and for what is being done to the earth.

    Comment by mg — 24 Apr 2008 @ 8:35 AM

  135. Dear George,

    First there is the sheer enormity of Greenland and the amount and thickness of the ice sheet on top of it. Its weight depresses the land, but it is estimated that if the ice were to disappear, it would take 10,000 to 20,000 years for the land to spring back, which is not on a scale that is of relevance to our current problem.

    Next there is the fact that we are only just beginning to get enough data to even talk about what the glaciers, ice streams, ice shelves, and ice sheets are going to do. It is all still pretty vague. Some new satellites are supposed to help with this lack of data.

    And we can also see that in the last 15 years, things have changed much more rapidly than expected and in ways that were not predicted. So, pretty much no one knows with any certainty what is going on (in my opinion).

    But in answer to your question, I will just make an educated guess as I am not a scientist. I would imagine that the ice sheet will remain so large and heavy that there is little chance of the surrounding ocean penetrating and lifting up the ice sheet, at least not on a time scale that would be of interest to us now.

    Comment by Tenney Naumer — 24 Apr 2008 @ 12:16 PM

  136. Phil. Felton, JCH, Tenney Naumer ,mg and Edward Greisch, thank you for the info on the bedrock topography of Greenland as I had also wondered about the frequent claim that it is x% underwater.

    There would seem to be quite a discrepancy between the 2 sources linked too (18, 132) however (even when you leave out the rebound)…

    Comment by Arch Stanton — 24 Apr 2008 @ 2:51 PM

  137. > discrepancy

    Maybe; I think most of that comes from deep ice penetrating radar work, of which there’s not a lot available. What are you looking at?

    Seems to me the pictures are illustrations, and you’d want to see numbers for topography in the original data set (and know the accuracy) — looks like much vertical exaggeration in the former’s color graphic (#18); in the latter, Schneider (#132), “at the horizontal resolution of the atmospheric grid, the coastal mountain ranges are represented with a reduced altitude, as illustrated by Fig. 2.”

    Both show a good bit of the area’s under ice.

    I wonder if the ring of land around an almost landlocked body of water accelerates freezing — isn’t that true for the Arctic generally, that it’s protected to some extent so easier for it to cool off? Same for the area we call Greenland?

    Comment by Hank Roberts — 24 Apr 2008 @ 3:47 PM

  138. Mauri (#120),

    Thanks again. Ablation (melting at this altitude) may have been the spur for the initial thinning that then propogates itself through floatation further and further in the inland direction.

    I think though that one does want to watch smaller accelerations in the larger masses of ice which are available at higher altitude. The depth of ice is more than a km, and apparently all of it is accelerating.


    Comment by Chris Dudley — 24 Apr 2008 @ 6:34 PM

  139. Those who are interested in Greenland bedrock, look at the maps in this paper:

    Comment by JCH — 24 Apr 2008 @ 10:20 PM

  140. JCH@139: thanks. So the submarine grounding is only in the interior: although the ice streams have worn channels in the bedrock, and of course the outlet glaciers have submarine grounding, this erosion is not sufficient to link the sea to the inland basin.

    Comment by Nick Barnes — 25 Apr 2008 @ 6:36 AM

  141. Nick, I’m completely unqualified to comment. I lack the background to understand scientific papers, so I just look at the pictures.

    There are some more pictures in this one; in fact, Googling Bamber in scholar yields a lot of visual goodies of Greenland:

    I do wonder if they’re incorporated SLR into those maps. Is it?

    Comment by JCH — 25 Apr 2008 @ 9:08 AM

  142. 131. An example which illustrates very clearly the connection between plant location and meltwaters from the ice sheets is illustrated by the map on this page

    By accumulating maps of critical nodes across different sectors (refineries, power plants, etc) one can build up a variety of indices (eg economic disengagement probabilities) for physical nodes as a function of sea level rise. Since SLR is a global attribute, engulfment of nodes on the industrial staircase is global. CBRN release profiles are global. Damage cascades through what are currently considered as “value adding chains” are near-enough spontaneous with physical node take-outs. There isn’t enough money in the global economy to shift ensembles of coastal industrial plant simultaneously to higher grounds.

    The observations being made of ice sheet dynamics and meltwater rates are critical for a proper assessment of the risks involved and the necessity to engage industrial leaders in full dialogue about the potential impacts.

    Comment by mg — 25 Apr 2008 @ 4:25 PM

  143. JCH@141: that’s a great paper, thank you, and I find it very reassuring about GIS collapse.

    There is indeed a submarine basin in central Greenland, but (a) it’s not very deep (a few hundred metres at most) and (b) it is not connected to the sea by any channel below sea level.

    The outlet glaciers are mostly grounded above sea-level, even hundreds of metres above sea-level, within a few dozen kilometres of the sea. The main exception is the NEGIS, which flows down its own basin, much of which is submarine, to the sea. The paper asserts that the basin causes the ice-stream rather than vice versa.

    The paper includes a look at the “excess thickness” (above flotation thickness) of submarine-grounded outlet glaciers (Humboldt, Petermann, NEGIS), and finds it above 500m within 25km of the sea. This is all good news, because it bounds the possible magnitude of Jakobshavn-effect mass loss (the Jakobshavn effect can’t operate if an outlet glacier’s terminus isn’t floating or nearly floating). And if the current paper is right and the Jakobshavn effect is much greater than the Zwally effect, then this is all very reassuring for GIS.

    Another paper, this one specific to the NEGIS:

    To answer your question, SLR would be almost invisible on these maps. The contour lines are 200m apart, and even total ice loss at both poles adds up to less than 100m.

    Comment by Nick Barnes — 25 Apr 2008 @ 5:05 PM

  144. My mistake: the paper I linked to in my last comment is not NEGIS-specific. I do highly recommend it, though.

    Comment by Nick Barnes — 25 Apr 2008 @ 5:12 PM

  145. Thanks JCH (#141)!

    Comment by Tenney Naumer — 25 Apr 2008 @ 9:48 PM

  146. Has anyone mentioned the very recent Joughin, Das, et al. stuff that shows meltwater acceleration on one glacier taken over the year had a much smaller effect on the glacier outlet? This claims that the meltwater speeds up flow by silly amounts (50-100%)inland but only ~15% near the outflow

    From an article in the 4/18 issue of Science by Richard Kerr Richard Alley of Penn State is quoted:

    “Could things go two times faster than we thought 10 years ago? Yes. They can go faster but not ridiculously faster”

    Comment by Eli Rabett — 26 Apr 2008 @ 8:11 AM

  147. Mauri,

    Great post. But looking at these 3D images of Greenland:

    I technically conclude (I’m not a scientist): Greenland’s a great big bathtub with great big holes in its sides!

    That signals to me intuitively, that calving, while dominant so far, is destined to play a very minor role. I’d describe the calving as like splashing a bit of water over the tub’s edge.

    Seems to me that ice melt, internal reservoirs and internal collapse, while minor thus far, will soon dominate. We could very much face a situation of non-linear dumping into the oceans and resulting catastrophic sea rise.

    Comment by kenlevenson — 26 Apr 2008 @ 11:58 AM

  148. Dear Eli,

    That depends on how you define “ridiculously.”

    Did you see the NewScientist article on the possibility of the North Pole being ice free this year?

    Comment by Tenney Naumer — 26 Apr 2008 @ 12:49 PM

  149. > internal reservoirs and internal collapse

    How would we tell?

    Is there any GRACE information showing loss of mass, but without a corresponding decrease in elevation of the surface?

    Anyone doing ice-penetrating radar that would detect voids in the ice below the surface?

    Tracing those moulin lake waterfalls to find out where that amount of water could go without increasing the outfall or lifting the ice surface?

    Other ideas?

    Anyone who has whacked a solid-looking piece of wood and found it had been riddled by termites — or walked out on a sheet of ice and found that it had a big void under it — has had this kind of surprise.

    But how can we look for the problem if it’s there?

    Comment by Hank Roberts — 26 Apr 2008 @ 1:19 PM

  150. Send your contributions to:

    The Society for Connecting all of Greenland’s Mountains with Dikes
    General Delivery
    New Orleans, LA

    We’re going to make that sucker into a great big bowl of ice water.

    Comment by JCH — 26 Apr 2008 @ 2:06 PM

  151. kenlevenson@147: I was very struck by the 3D images at, but note that the vertical scale and choice of colour make it impossible to resolve heights within 100m of sea-level, where it really matters. For a much clearer view, see the paper by Bamber et al at NSIDC:
    As you can see from that, the central basin is fairly shallow and nowhere connects with the sea below sea-level.

    Comment by Nick Barnes — 26 Apr 2008 @ 4:19 PM

  152. Re #146: That’s fair enough based on the present melt rate, but extrapolating it to much higher rates seems a little shaky. Recall that RA was probably the main target of Hansen’s “reticence” comment a while back and was more responsible than anyone for the conservative SLR estimate in the AR4.

    Comment by Steve Bloom — 26 Apr 2008 @ 6:59 PM

  153. Re: #149

    Dear Hank,

    I assume you already know this stuff, but this link might provide some googling ideas, vis-à-vis GRACE measurements:

    Comment by Tenney Naumer — 26 Apr 2008 @ 7:40 PM

  154. d18O is thought to be a proxy for ice volume.
    Using the measurements in

    from 30 yap (1980 CE) back to about 10300 ypb, we
    can obtain some idea regarding possible futures for
    ice volume. First I scanned the orginal data for
    the Younger Dryas (which is before the Holocene)
    and the 8.2 ypb event (which is in the Holocene).
    In both intervals indeed the d18O level was lower
    in comparison to dates both earlier and later.

    I averaged the data in 100 year intervals and placed
    each interval into one of 7 bins, bin00 (lowest d18O,
    most ice) to bin06 (highest d18O, least ice). The
    last 100 year interval, corresponding to 1881 CE to
    1980 CE, is in bin03, the median bin for the Holocene.
    As for potential futures, consider the up-transitions
    from bin to bin in the Holocene, in 100-year jumps from

    up 1 bin:
    from ~1670 to ~1570
    from ~3770 to ~3670
    from ~6470 to ~6370
    from ~7170 to ~7070
    from ~9970 to ~9870

    up 2 bins:
    from ~2570 to ~2470
    from ~3370 to ~3270
    from ~4270 to ~4170
    from ~5270 to ~5170
    from ~9770 to ~9670

    up 3 bins:
    from ~9370 to ~9270

    where the ‘~’ reminds us these are units of years
    before present (ybp). Now up 3 bins takes us to
    a climatic local maximum for d18O (in Central
    Greenland at least), although around ~6870 may
    well be the climatic optimum for high d18O there.
    The point is that sea stands rose to about 3 meters
    higher than now during the mid-Holocene. Of course
    nothing here says that all of this rise occurred
    in a mere 100 years.

    One the other hand, such a large transition has
    only occured the once in the past, out of the 23
    different centuries that d18O was in bin03 in the
    103 centuries studied; that’s 4% of the transitions.

    Disclaimer: I am an amateur with regard to the
    subject at hand, especially with respect to
    determining ice volume, and so sea stand, from d18O.

    Comment by David B. Benson — 26 Apr 2008 @ 8:21 PM

  155. 149. the Bamber paper linked in 151 has an interesting balance velocity plot which can be compared with the location and frequency of glacial earthquakes on Greenland (eg Ekstrom et al 2006). presumably the icequakes may play an important role in development of the percolation threshold as moulin networks interconnect prior to sheet collapse, or at least through the transitional architectures. not sure what the order parameters might be given the degree of nonergodicity and nonmarkovian nature of the system. presumably techniques such as Hurst and rescaled range analyses may provide some insight but the whole system is so far far-from-equilibrium. heard that IBM are setting up a supercomputer facility here in UK to model these things – an interesting development!

    Comment by mg — 27 Apr 2008 @ 5:12 AM

  156. 149. Hank Roberts – good questions, can anyone answer them?

    150. JHC – Since it doesn’t look like we’ll be pumping sulfur into the stratosphere perhaps this is the geo-engineering feat needed! ;)

    151. Nick Barnes – I don’t think the bottom needs to connect to the ocean, and this idea makes it scarier to me. It looks like an elevated tub where sides are made of mountain ranges and big “holes” where ice must be currently damming the outflows. What happens when these ice dams are penetrated, crack, and open? Images of sudden catastrophic earthen dam collapses come to mind. Based on the Bamber paper the northern half of the western coast seems, to my amateur eyes, where this sort of “break out” would most likely occur. Pure “chicken little” conjecture, maybe – but is it possible?

    Comment by kenlevenson — 27 Apr 2008 @ 7:54 AM

  157. There would not be substantial voids in the ice. There is some. Thus, GRACE tells the story. The Joughin and Das papers was cited in this article, and was one reason behind the timing of this post, as it indicates the limits on the Zwally Effect. There are sufficient outlets to the ocean for substantial drawdown via calving. This post does not try to address how far before outlet glacier calving would be limited. But calving is a more rapid means to reduce melt and thus, the prospect of more rapid sea level rise is heightened by the Jakobshavn effect in the near term. Eventually maybe melting will be the key, but in the near term it is calving that is the dynamic response mechanism that we have observed is the key to rapid retreat of tidewater glaciers around the world. Melting in place of adjacent non-calving glaciers has been much slower. this is evident on the glaciers adjacent to the Jakobshavn that are not calving.

    Comment by Mauri Pelto — 27 Apr 2008 @ 10:30 AM

  158. kenlevenson@156: my point is that this bedrock data shows that the ocean cannot undermine the GIS — the Jakobshavn effect (which this post is about) — except at the coasts and a little way upstream on some of the outlet glaciers. This particular mechanism was giving me particular concern about Greenland, as it was one way in which a huge amount of heat might be delivered to the basement ice under the GIS, leading to a catastrophic collapse. It seems that in fact this is not possible. We shouldn’t be complacent about the GIS, but if considering the risk of catastrophic collapse, we can restrict ourselves to other mechanisms which might deliver that heat to the basement. Principally, drainage of surface meltwater and rain (the Zwally effect).

    Comment by Nick Barnes — 27 Apr 2008 @ 12:28 PM

  159. Mauri, I’m curious if we could tell a void from a water-filled space in the ice, do you know? Do you think it’s possible to rule out an outburst flood of meltwater from underneath the Greenland ice, by knowing how much liquid water may be accumulating as the ice cracks and lakes drain?

    I’m thinking of how decades ago the big meltwater outbursts were pictured as lakes of water sitting on top of ice and behind barriers, but the more recent descriptions I’ve seen do talk about the meltwater accumulating underneath rather than above the ice.

    I know about some of the fluid movement mapped under the Antarctic ice but haven’t seen much about that from Greenland, except this most recent article in Science about seeing the lake water make the ice move as it drained down.

    Comment by Hank Roberts — 27 Apr 2008 @ 12:45 PM

  160. In contrast, much of the WAIS is grounded deep below sea-level, which is certainly connected to the Southern Ocean not just by shallow water but by very deep areas.

    Comment by Nick Barnes — 27 Apr 2008 @ 1:41 PM

  161. Re #154: “Over longer time periods, this ratio [d18O] indicates the average temperature of the regions between the evaporation site and the coring site.”


    So it appears to be of no immediate predictive value. Apologies for wasting your time.

    Comment by David B. Benson — 27 Apr 2008 @ 1:50 PM

  162. artic sea ice cover has been shown to be in a state of self-organised citicality

    might it not be true also for the ice sheet and its features?

    Comment by mg — 27 Apr 2008 @ 2:41 PM

  163. Does this sentence from the article need some attention:

    If the Zwally effect is the key than since meltwater is a seasonal input, velocity would have a seasonal signal.

    On the 3D rendition, if you tilt your screen, and look carefully at the colors, I think you’ll agree that it’s a more accurate representation of an ice-free Greenland than it might have seemed at first blush. My description of it showing a ring of islands is incorrect.

    Comment by JCH — 27 Apr 2008 @ 4:04 PM

  164. Mauri and Nick,
    Thank you for your clarifying responses and patience. It’s starting to make more sense to me. Although my questions still go in the direction of Hank’s #159. Unfounded fear or no?

    Comment by kenlevenson — 27 Apr 2008 @ 4:38 PM

  165. kenlevenson (164) — In the mid-Holocene, temperatures were much warmer than even now in Greenland. The sea stand was about 3 meters higher than now. So the Greenland ice sheet was still mostly there.

    Similarly for the Eem (Eemian interglacial) with sea stands estimated to be about 4–6 meters higher than now.

    Comment by David B. Benson — 27 Apr 2008 @ 5:21 PM

  166. Guys, isn’t the question really whether or not at some point in the future the GIS could become so riddled all the way to the bottom that it would collapse here and there under its own weight, and how to determine if GIS is becoming more and more prone to collapse? Or, at some point, will the glacier fronts just keep melting away and pull back until they no longer calve?

    Comment by Tenney Naumer — 27 Apr 2008 @ 5:22 PM

  167. “Moreover, work on Greenland ice cores suggests
    an inverse relationship between temperature and
    accumulation rates during the last 7 ka., and the ice sheet is
    therefore especially prone to volume reduction during
    periods of warm climate, such as the middle Holocene
    (CUFFEYand CLOW, 1997). FUNDER’s (1989) summary
    of Greenland glacial records supports this view, noting that
    middle Holocene ice margins were considerably inland
    from present-day, and re-advanced after ca. 4 ka.”

    Journal of Coastal Research SI 36 65-80 (ICS 2002 Proceedings) Northern Ireland ISSN 0749-0208
    Middle Holocene Sea-Level and Evolution of The Gulf of Mexico Coast (USA)
    Michael D. Blum, Amy E. Carter, Tracy Zayac, and Ron Goble

    Comment by David B. Benson — 27 Apr 2008 @ 7:38 PM

  168. This paper looks to be somewhat on topic:

    Is the mid-Holocene so simply invoked?

    Comment by JCH — 27 Apr 2008 @ 10:49 PM

  169. 168 JCH thanks for that link. If the west antarctica ice sheet were to partially disintegrate first (since many commentators seem to think that is the most vulnerable) giving a rise, say, of a couple of metres, what might the impact be on the oceanic / moulin-riddled-GIS interfacial dynamics? Might the sequence of events be the signature of an MWP (giving a WAIS contribution of say 2 metres and a GIS contribution of – as per the link – about 5 metres? The linked paper suggests that GIS was much steeper – does that tie in with the glacial earthquakes now being observed compromising the ice sheet edges?

    Comment by mg — 28 Apr 2008 @ 6:51 AM

  170. JCH, interesting Nature link, that looks helpful:

    Comment by Hank Roberts — 28 Apr 2008 @ 9:45 AM

  171. Lakes form at the bottom of a glacier or on the surface. Because ice crystals deform under pressure, and pressure is substantial within a glacier or ice sheet it is not possible to have substantial void volumes. Ice under pressure would deform and flow into this void. This happens to much of the seasonal hydrology system each winter. Without water flow to keep tunnels open, they close, then in spring maximum water pressures often occur befor the conduit system redevelops. Once opened the flowing meltwater can maintain these narrow conduits. However, the meltwater does not have enough heat to melt much. At the base of the glaciers even in the summer next to these streams, you will see new ice coating the bedrock in places. The moulin ice riddling is science fiction. No ice sheet or glacier collapses due to riddling by moulins. I still see a persistent misconception about the ability of meltwater to melt glacier ice and riddle the glacier with holes. I work on glaciers with lots of melt and they are not weakened by all the meltwater drainage. The meltwater is not a very capable melter of ice. Ice is unlike rock which does not deform under the pressure and temperatures observed on glaciers.

    Comment by Mauri Pelto — 28 Apr 2008 @ 2:52 PM

  172. That’s helpful, Dr. Pelto. I think this is one of those trivial questions. I asked a system operator once what he meant when he said I’d asked a trivial question. He thought for a moment and then said “it’s not worth my time to explain, and you’d never figure it out for yourself.”

    I very much appreciate your taking the time to address it clearly. I think we amateurs don’t have much of a sense of what’s going on deep under thick ice, and so imagine by analogy to the very different experience we have of shallow thin ice melting.

    Seeing all the recent reports of water moving under the icecaps, it’s hard to have a sense of what’s going on without hearing from someone like you who’s actually looked into it.

    Comment by Hank Roberts — 28 Apr 2008 @ 4:48 PM

  173. Re: #171

    Dear Dr. Pelto,

    Thank you for your explanation. I think I have finally been disabused of my misapprehension of the ability of moulins to increase to the point that they endanger the structural integrity of the GIS.

    Comment by Tenney Naumer — 28 Apr 2008 @ 5:44 PM

  174. Ice flows under pressure. There are no voids in thick ice. However, hydro-fracture can be a relatively rapid process. Alternatively, volumes of water can sink through ice in a process very much like the classic physics demonstration where a wire is passed through a block of ice using weights and gravity. This happens when water on top of ice is more than just over 6 meters deep.

    The heat advected into the ice by the stream of water in a moulin may be small in comparison to the heat required for substantial melting. However, advection results in conditions very different from those when conduction is assumed to be the primary mechanism for heat transfer to the base of the ice and our thinking must be adjusted accordingly.

    With conduction as the primary heat transfer mechanism, the colder (and stronger) ice will always be toward the bottom of the or the ice sheet. In this case, there is no possibility of the ice sheet undergoing rapid collapse. This is the conventional wisdom.

    However, if advection by melt water is the primary heat transfer mechanism, then the basal ice may be as warm as the upper layers of ice, and be too warm (and too weak) to support the weight of the ice above it (without extensive ice buttresses around it.) In this case, there is a weak foundation that is subject to progressive collapse, if the surrounding buttress weakens. Remember, ice flows under pressure. As the foundation ice warms, it will put increasing stress on the buttress. As the buttress warms, it will become weaker and the system will collapse. Thus, the real question is “How much heat is being advected?”, and “How much heat is being conducted into the GIS?” My intuition is that advection will increase much faster than conduction. I would be much happier if I did not trust my physics. Unfortunately the very first lesson that I learned at the university was to trust the “the physics” rather than any particular textbook or author.

    Finally, as water or ice falls under the force of gravity, potential energy is converted to kinetic energy. When the falling object strikes another object, the kinetic energy is converted to heat. If that heat is transferred to ice, that ice becomes warmer and weaker. The heat may not be enough to melt the ice, but the ice will be weakened. (On the other hand, measuring such melt would be difficult.) That water at the base of a glacier is at 0C says only that it has transferred the heat that it gained in falling to the base of the glacier, to other, colder ice. The ice, that gained that heat, is now weaker. In fact, that weaker ice may have become so weak that it flowed together to fill the moulin that acted as the transfer conduit. Thereby, it may be hard to identify just where the heat from the falling water now resides. Nevertheless, somewhere in the glacier, there is a volume of weaker ice. Heat does not just disappear – even in a glacier.

    Comment by Aaron Lewis — 28 Apr 2008 @ 8:46 PM

  175. Aaron Lewis @ 174: on potential energy. Worst case scenario: a meltwater pond on a very thick ice sheet which falls 3km through a moulin to the basement. Each kg of water converts 30kJ of potential energy to heat in the basement (in fact the heat will be distributed through the depth of the ice sheet, as friction slows the falling water). That same kg of water also carries 334 kJ of latent heat. So even in this worst-case scenario the potential energy is negligible.

    Comment by Nick Barnes — 29 Apr 2008 @ 8:42 AM

  176. Re: #171

    Dr. Pelto,

    I greatly appreciate your time and patience in helping us laypeople better understand ice sheet dynamics. But I confess I’m confused by your statements that meltwater and moulins don’t significantly weaken glaciers or ice sheets.

    After the Larsen B ice shelf broke up a number of popular press articles pointed to the meltwater lakes and moulins as having been a major factor in precipitating or magnifying the disintegration of the ice. Because the meltwater was only present for several months each year it would have had to act quickly and powerfully to weaken the ice shelf. Did the writers simply misrepresent the science, or do meltwater and moulins act differently on a floating ice shelf than on a grounded ice sheet?

    Comment by Phillip Shaw — 29 Apr 2008 @ 10:49 AM

  177. Phillip Shaw (176) wrote “… do meltwater and moulins act differently on a floating ice shelf than on a grounded ice sheet?” The meltwater at the bottom of a floating ice shelf just freshens the sea water at the outlet. The meltwater at the bottom of a glacier digs a tunnel downhill.

    Disclaimer: I’m an amateur.

    Comment by David B. Benson — 29 Apr 2008 @ 12:51 PM

  178. Would there be enough pressure in a floating ice shelf to plasticize the ice to fill the voids?

    Comment by JCH — 29 Apr 2008 @ 1:47 PM

  179. Re 175
    The energy from one little melt pond is negligible. The melt from weeks of warm, moist onshore winds from the North Atlantic or the Arctic Ocean every melt season for season after season is not negligable.

    Advection can move more heat into the ice during the melt season than conduction can remove during the winter. Thus, the heat accumulation can be cummulative year after year and decade after decade.

    Look at the weather patterns on Greenland’s East Coast last May and June. Then, promise me that they will not occur again.

    Comment by Aaron Lewis — 29 Apr 2008 @ 2:06 PM

  180. this is what Jim Hansen wrote in a lengthy paper here
    which at paragraph 49 has:

    “49. The primary question, however, is whether the rate of melt has the potential to accelerate in a rapid, nonlinear fashion. Some hint of that possibility is contained in recent earthquake data for Greenland. Seismometers around the world have detected an increasing number of earthquakes on Greenland between 1993 and 2005, as shown by the green bargraph in Figure 22. The location of these earthquakes is near the outlets of major ice streams on Greenland, as shown by the red circles in Figure 22. The earthquakes, which have magnitudes between 4.6 and 5.1, are an indication that large pieces of the ice sheet lurch forward and then grind to a halt from friction with the ground.”

    the Figure 22 (in the linked pdf) referred to can be compared with the figure linked in 170.

    what is the precursor / trigger for the “rapid, nonlinear fashion” ? is it to do with collective dynamics across the ice sheet / oceanic combine ?

    Comment by mg — 29 Apr 2008 @ 2:24 PM

  181. Good point P. Shaw and JCH. Yes, surface water on an ice shelf behaves much differently, and I have seen no mention of moulins for Larsen B or Wlikins Ice Shelf. The surface water filling crevasses helps weaken the ice shelf. But yes the ice shelves in this case are fairly thin and ice deformation is slower, but it still was not the internal drainage system that seemed to breakup the ice sheet. If you look at the disintegration it produces good sized tabular icebergs, which indicates crevasse control. Again this happened from the calving front, and represents a rapid calving retreat not an internal collapse. Thus, it the surface water played a role, but it was an inbalance in forces at the calving front that allowed the retreat to propogate. Nick, it has never been demonstrated that advection occurs in ice sheets. It may but the signal is too weak to be identified yet. And even warm ice is not weak. Note many glaciers are temperate, and the bottom of all rapidly moving glaciers are at the pressure melting point, warm, they do not collapse at that point. Nick, trust your physics but that means you must understand the properties of ice too.

    Comment by Mauri Pelto — 29 Apr 2008 @ 2:51 PM

  182. While puzzling over this earlier I left the bits I found in this thread over at William’s blog Stoat:

    As I puzzled I found mention of what seemed to me to be three different situations, at least (each with varying amounts of overburden ice pressing down)

    Continental icecap sitting more or less where it is, bottom below sea level, pushing down on rock or sediment, where water at the base can run uphill and get squirted out at the edges due to pressure at the middle, and the pressure/temperature at the bottom allows liquid water.

    Look at some of the radar maps of Antarctica (I’m looking for links); I recall the topography under the ice is very crisp looking at the center, you can see dendritic drainage patterns extending from the central ‘sea’ out to the edges. And out at the edge all that clearly discernible rock is no longer visible — there are layers of sediment that look like they were pushed out from under the central ice, around the perimeter.

    I don’t know if Greenland has anything similar, though I wonder.

    A glacier above sea level — where once water entering from above declines air space is left as water drains out below and the pressure of the ice can force the openings to close, and air presumably can circulate as well cooling off the void by evaporation — “ice caves” for a while, closing up under time and pressure.

    Floating sea ice. I know there are observations of cracks forming in ice sheets due to tidal change and to episodes of waves from distant storms. I’d wondered if there were enough rise in sea level to crack them at the point where they ‘hinge’ and become floating. There are observations that the cracks once opened don’t weld themselves shut with equal strength; they fill up with a loose mass of snow and ice and air debris that blows in.

    Then there’s more — circulating seawater opening channels under the ice, perhaps up into the ice flows (the place an unmanned submersible was lost in Antarctica might be one such, I don’t know).

    This is a quick memory dump, cites would be in the very tolerant William Connolley’s thread over at Stoat.

    Comment by Hank Roberts — 29 Apr 2008 @ 2:56 PM

  183. So — can we set the other stuff aside, til our hosts open a thread for those again? The International Polar Year probably has most of the likely people out in the field or frantically writing papers on it now (grin).

    But — here and now:

    We have Mauri Pelto here to talk about his area of expertise — “Moulins, Calving Fronts and Greenland Outlet Glacier Acceleration” — and it’s a big area.

    Dr. Pelto, let me try pointing back to the topic, what question _should_ one of us amateurs have thought to ask you that no one has? Anything obvious that hasn’t been paid enough attention in the news, that you find remarkable or puzzling or instructive?

    Comment by Hank Roberts — 29 Apr 2008 @ 3:12 PM

  184. Check out these photos showing the break up of multiyear ice, notably in the Beaufort sea, it was a solid piece last year at this time:
    A month ago:

    Comment by Phil. Felton — 29 Apr 2008 @ 3:28 PM

  185. You have this situation where the GIS is accumulating snow/ice at elevations above 1500 meters, and losing elevation below. I assume this was a big part of the reason why the GIS was much steeper during the mid-Holocene as the combination would have that result.

    What understand Mauri to be saying is the GIS sort of recreates its drainage system each melt season, and that the drainage systems tubing size is somewhat self-limiting because of the water’s low temperature. In my experience with water, it sort of settles for the easiest way out, so it would not really be looking for additional glacier to vandalize. In the winter, what I take him to be saying, is the incredible pressure deforms the ice and it sort of heals itself. Is it possible warm ice is even better at doing that?

    I sort of doubt the ice sheet is healing itself much under around 1000 meters, but I don’t understand how that pressure extends out into the boundary ice. Also, it seems unlikely this deformation healing would be as efficient above some elevation – like maybe 2000 meters.

    So back to this steeper GIS situation, which it sounds like will develop as the coming decades pass by. The only things I can imagine causing non-linear melting, is a situation where significant amounts of ice that is above 1500 meters suddenly falls below 1500 meters, or if the 1500 meter “tree line” goes up quite a bit.

    Is the that 1500 meter line going to move up? I don’t understand all of this stuff, but I don’t see how it could not move up as it gets warmer. If it moves up, would further additional precipitation offset that situation? So far, that seems to be the case. Will it continue to be the case – more warming, more precipitation on Greenland? It’s pretty interesting how the GIS can defend its heights.

    Assuming it gets steeper, is a steeper ice sheet as strong? Was the missing boundary ice buttressing the domes? It seems like it would have to have been. Would the domes of a steeper GIS eventually be susceptible to a form of calving? I realize ice hitting the ground at lower elevations does not equal calving into the ocean, but it’s unlikely to accumulate ice again, and the area it represented above 1500 meters will not accumulate again, and it’s going to melt quickly enough. So it would be a sort of double-edged melting event.

    This paper answered most of my precipitation questions:

    Comment by JCH — 29 Apr 2008 @ 4:03 PM

  186. I appreciate the notes of thanks for responses. But I appreciate the questions, it makes it evident that scientists in communicating with each other leave gaps in our explanations of scientific processes, that allow ample room for alternative visulatizations. I will now be able to make a more comprehensive explanation of this topic than I did one week ago. Hank in #182 your observation #1 I did not really follow. Observation #2 is true as to the closing off. the cooling via evaporation, while in the bulk of the ice sheet the ice is cold already, at the base it is not. Observation three, calving is affected by tides, Jakobshavns has a tidal flexure point you can identify. It is an area of weakness, which is why pinning points are important. Pinning points provide a support that a free floating ice tongue would lack, and where tidal flexure would be more dominant.

    Comment by Mauri Pelto — 29 Apr 2008 @ 4:42 PM

  187. Mauri, my #1 was thinking almost entirely of Antarctica (as far as I know).

    I’ll try to clarify it briefly but, seriously, if this isn’t possibly relevant to Greenland, let’s not go off to Antarctica til the Polar Year news is in!

    So to answer re my #1, I put earlier speculation here:

    I then found I’d been scooped (grin) by New Scientist
    and typed some quotes into the following post at Stoat. Below I’ll put a shorter excerpt. The NS link here only goes to a teaser, I was copying from the paper magazine from our library.

    ‘You can have lakes sloping down the sides of mountains, you can have uphill waterfalls, it’s wacky’ [Don Blankenship, geophysicist at U. Texas] …. Blankenship fears that warming since the end of the last ice age has melted the base of the ice, and this may already be priming some parts of the ice sheet to slip. East Antarctica could be ready to open its floodgates.

    “David Marchant from Boston University believes this may have happened before…..
    —–end excerpt—-

    While there’s a link, it’s only to a teaser online:

    (PS, I pointed back to this thread and your summary, in a recent addition to that Stoat thread)

    —-So, again, if this is off topic, file for later?

    I’m still hunting for those ice/radar pictures I recall seeing that showed the under-ice topography of Antarctica and the surrounding ocean floor.

    Comment by Hank Roberts — 29 Apr 2008 @ 5:22 PM

  188. Ok, this looks like part of the image I remember

    — you can see the fairly sharp topography inside the line that marks the edge of the ice cap, and the very soft contours outside that. I imagine the accumulating ice pressing down on the middle of the continent and slowly extruding all the sediment out the underlying drainage channels to the surrounding seafloor.

    Articles with this image:

    And a bit more:

    Comment by Hank Roberts — 29 Apr 2008 @ 5:36 PM

  189. Aaron Lewis @179:
    You entirely miss my point @175. The potential energy of surface water, turning into heat as it falls into the ice, will always be negligible compared to the latent heat of that same water. There’s a factor of at least 10. The latent heat is the biggie, and can warm a lot of the interior and base of the ice sheet as the surface water re-freezes.
    It might not be able to melt much of the ice (because of course to melt a kg of ice requires the latent heat of a kg of water), but it can raise its temperature significantly. I guess the limiting case here is when the ice sheet is all at melting point.
    Of course, if the water leaves the ice without re-freezing (i.e. if it drains into the sea or into lakes under the ice) then it retains its latent heat.

    Comment by Nick Barnes — 29 Apr 2008 @ 5:40 PM

  190. Mauri@181: Advection of what? Of heat, by the draining of surface water into the ice sheet? It might not have been demonstrated, but the physics makes it obvious that this happens, and lets us put limits on the amount of heat which can be transported in this way. I’m entirely ready to believe that it hasn’t been a dominant process in the past, and even that it is still not so today, despite the rather alarming pictures of huge moulins and talk of arctic Niagaras. However, now that the bedrock topography has been clarified for me, some sort of advection of heat to the interior and the basement is the only mechanism I can see for catastrophic collapse of the GIS (about which I am getting more sceptical by the day, as can be seen in comments). Conduction surely isn’t going to do it, as the colossal latent heat of the ice sheet will keep surface temperatures at or below melting point until there’s no ice sheet left.
    There are various ways in which this heat advection through the ice sheet can become much larger. One is a great deal of rain. Another is a great increase in surface melting.

    Comment by Nick Barnes — 29 Apr 2008 @ 5:51 PM

  191. Phil Felton@184: That’s a great pair of images; the Beaufort Gyre as plain as day. But the main contrast with a year ago is surely the area around the pole, and in the eastern Arctic basin. I recommending loading this pair of images into adjacent browser tabs and flicking back and forth:

    Comment by Nick Barnes — 29 Apr 2008 @ 6:04 PM

  192. On the Larsen B collapse:
    Oceanographic Conditions at the Larsen B Ice Shelf Front Before and After the 2002 Breakout,
    Huber, B.; Domack, E.; Padman, L.

    The Larsen B ice shelf in the Northwest Weddell Sea experienced a catastrophic breakout in early 2002, resulting in an open-water embayment in the region formerly covered by shelf ice, with remnant ice fronts and exposed tidewater glacier regimes. Oceanographic station data occupied at the Larsen B ice front in December 2001 and at the remnant fronts in March 2005 are examined for changes in water mass properties which might be related to the ice shelf breakout of 2002.

    The region is characterized in both years by a dominance of near-freezing point remnant Winter Water capped by a warmer seasonal mixed layer. At several locations along the ice fronts there is evidence of sub-freezing Ice Shelf Water (ISW) emanating from beneath the ice shelf. In the 2005 observations near the remnant tidewater glacial fronts, some plumes of ISW are marked by significant reductions in optical transmittance, implying that the plumes may be carrying glacial debris.

    Modified Warm Deep Water (MWDW) is present in only small quantities at some stations, consistent with published observations of conditions at the Larsen C ice shelf front to the south.

    See also:

    “Accelerated ice discharge from the Antarctic Peninsula following the collapse of Larsen B ice shelf, E. Rignot, G. Casassa, P. Gogineni, W. Krabill, A. Rivera, and R. Thomas GRL 2004

    Warmer air temperatures increase surface melt water production, which may reach the bed and increase basal lubrication near grounding lines [e.g., Zwally et al., 2002]. Similarly, ice shelf thinning and accelerated flow and calving reduce ice shelf buttressing, which allows faster flow. It is not clear a priori which process exerts the greatest control on ice flow.

    We note however that most glacier equilibrium lines in the eastern Peninsula are close to grounding lines [Morris and Vaughan, 2003], so surface melt should not be a major factor in the glacier acceleration because it is mostly confined to the ice shelves.

    Furthermore, there has been no marked change in air temperature in the last two years compared to the prior decade that could explain a sudden and widespread glacier acceleration from enhanced surface melting. Indeed, Flask/Leppard glaciers have remained more or less steady although they experience the same climate conditions as Crane, Evans, Green and Hektoria. In contrast, the widespread glacier acceleration coincides with the ice shelf removal, and glaciers still buttressed by an ice shelf did not accelerate.

    These results are in broad agreement with the theory of ice shelf buttressing. The most likely explanation for the glacier acceleration is a major reduction in ice-shelf buttressing. . .

    Further south, glaciers drain larger reservoirs of ice, and the thinning of Larsen C [Shepherd et al., 2003] may trigger an even larger contribution to sea level. These observations are also particularly relevant to the evolution of ice streams and glaciers draining West Antarctica. Although the climate conditions of the Antarctic continent are colder and drier than in the Peninsula, ice shelf thinning could be caused by a warmer ocean instead of warmer air temperatures.

    A news release on more recent work is here:

    The theme seems to be that rising ocean temperatures are playing at least as large a role as air temperature increases in the breakup of Antarctic ice sheets, but that the subsequent acceleration of glacial outflow is not a temperature-driven phenomenon.

    Comment by Ike Solem — 29 Apr 2008 @ 7:55 PM

  193. > ome plumes of ISW are marked by significant reductions
    > in optical transmittance, implying that the plumes may
    > be carrying glacial debris.

    Intriguing. One of the articles I noticed recently suggested that in past cycles glaciers were grinding away a layer of soft sedimentary rock, and eventually that gets removed and a later glaciation will be working on a harder, older, deeper layer of igneous or metamorphic rock on which its behavior will change.

    Comment by Hank Roberts — 29 Apr 2008 @ 8:45 PM

  194. Re #191

    Nick, agreed the loss from North of Greenland via the Fram Strait this winter has been dramatic, it’s hard to think that the sea ice can survive another year or two of this. My guesstimate is that the perennial sea ice has dropped by ~half this winter. This movie is pretty impressive and it still falls short of the present situation!

    Comment by Phil. Felton — 29 Apr 2008 @ 9:08 PM

  195. In 171 ice flows to fill in moulins and in 181,it is noted “that even warm ice is not weak.” If this seems a bit confusing, try
    Ice is the most interesting stuff in the world.

    Comment by Aaron Lewis — 29 Apr 2008 @ 9:12 PM

  196. re 189
    Given a moulin, if the water in it freezes, it delivers 10X heat, but it plugs the moulin. If the water does not freeze, then maybe 11X units of water can flow, and deliver their units of heat in the course of a melt season. Thus, ultimately more heat is advected if the water does not freeze. Since I tend to take worse case, I assumed that a lot of water flowed through without freezing. It helps that I first thought about this back when I worked just downstream of the Priest Rapids hydroelectric dam

    We need a field trip with lots and lots of expensive toys. I think this would be very tough thing to measure in the near future.

    Comment by Aaron Lewis — 29 Apr 2008 @ 10:04 PM

  197. Mauri@186 and elsewhere, you say that the base of the ice is at melting point (unlike the middle depths of the ice, which are much colder). Why is this? Is there a source of heat (geothermal?) which keeps the ice so warm?

    Comment by Nick Barnes — 30 Apr 2008 @ 7:09 AM

  198. Now all that’s quite interesting, because from,_Antarctica

    It looks like ice temperature near the surface of the Antarctic ice sheet around Vostok Station is typically around -55 C. However Vostok Station altitude about 3490 m. The surface of Lake Vostok (or the ceiling of the ice above or confining the lake) is at about 3750 below Vostok Station and so 260 m below sea level. The usable core only extends down to 3310 m, which is to a position 180 m ABOVE sea level. Below that the cored ice consisted of refrozen waters from Lake Vostok. Lake Vostock ranges in depth between 400 m to 800 m with an average depth of about 340 m.

    Presumably the water in Lake Vostok is around 0 C, and the ice above it (at least for the first 440 m) is ranging from ice at 0 C above the lake surface towards the -55 C found near the top of the ice sheet. Since the lake is probably connected via other water ways to a discharge/recharge channel to the ocean, then pore pressure will be forcing water up through the ceiling towards sea level, and beyond by capillary action.

    So here at the Pole of Cold we find that the ice sheet is at around 0 C at its base, that the bottom 10% of the ice sheet is comprised of refrozen lake water – and the temperature gradient runs from 0 C at the base to -55 C at the surface. This refrozen water tops at 180 m above sea level, so the ice sheet is poised above sea level on a 400 m thick wad of refrozen lake ice which in turn is sitting above a vast lake of water at 0 C. The proximity of the interface between old ice and refrozen ice to existing sea level is also disconcerting since Vostok is 1400 km from the coast. That’s hardly the sort of structural base I would like to think is supporting 50 metres worth of sea level rise.

    Comment by Nigel Williams — 2 May 2008 @ 3:47 AM

  199. I’m an environmental studies major at UMKC so I have a bit of general background; but upon reading others’ comments to this blog, I’m sensing that many people reading this know a great deal more about glaciers than I do, as I had to consult for “moulins” and went to Wikipedia for a better understanding of “calving fronts”. I found this very informative, especially the information about the Zwally and Jakobshavn effects. This is a bit specialized for my level of expertise however.

    I wonder if there’s a single repository of stockpiled glacier data: a “Glacier Watch” forum where data is compiled and contrasted, including analysis of historical patterns and also future projections. I’ve found many individual glacier resources, but I didn’t find one website to link them all… I think that would be great if it doesn’t already exist.

    I’d like to re-ask Mauri’s question posed in comment #15: I don’t understand the sentence “Jakobshavn has since slowed to near its pre-1997 speed, the terminus retreat is still occurring, but likewise is.” I thought maybe the sentence would be continued somewhere.

    I also was hoping to find more elaboration on Alestair’s comment #12:
    “There are two positive feedbacks which have not been mentioned. First, as glaciers melt sea level rises and grounding lines retreat speeding the remaining glaciers’ advance into the sea. Second, as the glaciers melt their height reduces bringing a greater proportion of the glacier below the snow line. Sea level rise has a similar but lesser effect. Both of these cause more melting. Both could end with a runaway melt.” Sounds like this chain reaction could accelerate or at least complicate calculations.

    I’ve read a lot of these discussions including calculations and attempts at drawing conclusions. I believe a consensus exists that regardless of the specific anticipated rate of glacial melting, the consequences aren’t good. What about some solutions? I think this is a fascinating and ingenious approach to the problem:
    While it may be slapping a bandage on a compound fracture, at least I can derive a sense of hope from this.

    I’m just gonna put this crazy idea out there: what if there was a reality TV show about glaciers? Public interest in nature is growing, as evidenced by the box office success of March of the Penguins and Disney’s recent addition of a nature documentary department. What better way to relate to the “common folk” than with a popular “reality” format? Or: imagine a mock commercial with a “receding glacier” as the symptom and some a pr3scr!p+!0n for greenhouse gas reduction techniques?

    My simple theory is this: if we want to “save nature” we have to really relate to “human nature” as it has become. At least in my experience, many Americans are generally reactionary and don’t go out of their way to acknowledge problems of this magnitude, especially if there’s any room to hide in ambiguity or debate. But if we pump the entertainment industry full of images they can relate to, say: instead of keeping track of c3lebr!+!3s’ life… we might just find more voices, more signatures on petitions and subsequent laws or policies, more trend altering behaviors…

    [some words include symbols in place of letters due to over-active spam filter]

    Comment by Andy Siebert — 2 May 2008 @ 1:24 PM

  200. The base of the ice sheet is at the pressure melting point because of geothermal heat. The heat travels up the thermal gradient to the base of ice sheet. Sorry, about the discontinued sentence, I think I just failed to remove the “but likewise the”. From “12 it is true that sea level rise shifts the grounding line, but this is insignificant over the time span of a decade or two for the Jakobshavns. More important for ice shelves that rely on pinning points, and lack good lateral margin pinning points. It is true that surface lowering at the equilibrium line would cause it to expand inland. The thinning has not been large enough to cause any significant shift to date. But during many rapid calving retreats this does occur.

    Comment by Mauri Pelto — 3 May 2008 @ 7:12 AM

  201. > a reality TV show about glaciers?
    That’s a scary thought. I’m old enough to remember when “watching glaciers melt” was a simile for something very slow and very boring, like “watching the grass grow.” Now, you’re right, people would tune in to watch. Good grief.

    Here’s one for the paleo-Arctic:
    Megatides in the Arctic Ocean under glacial conditions

    Comment by Hank Roberts — 3 May 2008 @ 1:18 PM

  202. Several named effects have been discussed in this thread, although there is a further named effect concerning the ice sheets which may be worthy of note. The following is an extract from an as yet unwritten draft of the sea level rise section of the IPCC’s 10th report on climate change, expected to be published some time around 2060. There was a bit of a delay because the authors couldn’t agree on the wording. An extract from the technical section on sea level rise reads:

    “Each of the ice caps sat on a pole

    All the world’s governments and all the world’s industrialists couldn’t put the icecaps together again.”

    After a lot of deliberation they did agree on a name for the effect, but for the moment it is being kept secret, although one suspects it might get leaked.

    As the icesheets slip off their perch and crash into lots of little pieces (just like an egg that falls – or is pushed? – onto the ground), it is about time some the world’s top industrialists had the courage to convene through a globally broadcast internet conference and tell the world’s public what they are going to do specifically about the icesheets.

    Comment by mg — 4 May 2008 @ 3:23 AM

  203. #194 “My guesstimate is that the perennial sea ice has dropped by ~half this winter.”
    Josefino C. Comiso
    Cryospheric Sciences Branch, NASA Goddard Space Flight Center
    Journal of Geophysical Research
    22 February, 2008.
    “Arguably, the most remarkable manifestation of change in the polar regions is the rapid decline in the Arctic perennial ice cover. Changes in the global sea ice cover, however, have been more modest, being only slightly negative in the Northern Hemisphere and even slightly positive in the Southern Hemisphere”

    This paper confirms that there is at least some thickening ice in the Antarctic :
    H. Jay Zwally
    Cryospheric Sciences Branch, Goddard Space Flight Center
    Journal of Geophysical Research
    19 February, 2008.

    Comment by Mike — 7 May 2008 @ 2:25 AM

  204. If you want to so something “chilling,” watch this Japanese video of polar satellite images, showing the disintegration of multi-year ice during last winter.

    From the video:

    The image is a low-resolution reproduction of a sequence of satellite images of Arctic ice this past fall and winter. The sequence runs in a continuous loop from October 01, 2007, to March 15, 2008. A link to the high-resolution video file is also provided.

    Note the stream of multi-year ice flowing out of the Arctic basin down the east coast of Greenland at one o’clock in the image. As of the middle of March, most of the basin, including the pole itself, appears to be covered only by seasonal ice.

    Comment by Ron Taylor — 7 May 2008 @ 8:01 AM

  205. > This paper confirms that there is … some thickening … Zwally

    “May–June (fall) and October–November (late winter) of 2004 and 2005” and “0.28 m in 2004 and 0.29 m in 2005” — but are you claiming a trend based on these numbers? The paper’s not about that.

    Comment by Hank Roberts — 7 May 2008 @ 9:12 AM

  206. Re #205

    Nice video similar to the Quikscat one I showed in #194 (interesting that the Japanese look at it upsidedown ;) ) The shots of the ice breaking up and squeezing through the Nares strait are really good.

    Comment by Phil. Felton — 7 May 2008 @ 12:47 PM

  207. The paper referred to sea level contribution, and post #194 speculated on the loss of sea ice. I posted a link to a paper that indicated that net loss of global sea ice was slight. That paper made the, perhaps surprising to some, that Antarctic sea ice had increased, so I looked for and found a confirming paper.

    No I’m not claiming a trend.

    It is interesting, though, that the Arctic and Antarctic often seem to go in different directions.

    Comment by Mike — 7 May 2008 @ 5:05 PM

  208. On Antarctica:

    Comment by JCH — 8 May 2008 @ 8:28 AM

  209. links for sea surface temperature anomaly plots to go with the satellite images linked in post 204 can be found here

    Comment by mg — 8 May 2008 @ 8:39 AM

  210. Re: #207

    Dear Mike,

    Keep looking into things and you will rapidly find information that will tell you that it is not at all surprising that the Arctic and the Antarctic would tend to go in different directions with regard to sea ice extent (at the present time, but perhaps not so much in the future when the ozone hole over the Antarctic is repaired).

    Comment by Tenney Naumer — 9 May 2008 @ 12:10 AM

  211. some interesting data on bursts of seismic waves in an ice stream (60 miles wide and one-half mile thick)

    ” … combined seismological and global positioning system (GPS) analyses to reveal two bursts of seismic waves from an ice stream in Antarctica every day, each one equivalent to a magnitude seven earthquake”

    Comment by mg — 5 Jun 2008 @ 4:24 PM


    An interesting title for the article “Something’s shaking in Antarctica”

    I recall reading somewhere that ice quakes don’t occur in Antarctica. Well that’s another tick in the nonlinearity box.

    Comment by mg — 9 Jun 2008 @ 4:15 AM

  213. >”…Well that’s another tick in the nonlinearity box.”

    Doubtful, the article implies that the quakes were previously undetected, not that they are new.

    Comment by Arch Stanton — 9 Jun 2008 @ 10:12 AM

  214. I propose another mechanism which needs to be considered regarding ice flows. Glacier ice is under extreme pressure, and it is also just slightly below freezing. It should not be considered solid like a diamond, but rather solid like a candle. Waxy. When an individual molecule of water happens to have enough energy to jump to a neighboring location it may do so. Remember, the ice is not a rigid crystaline structure, but an amorphous solid. It is possible for a molecule to ‘wiggle around’ in this solid. When the headend pressure is relieved, the pressure differential (head to tail) is increased, this will accelerate the flow rate and increase the thinning. One must consider the pressure, temperature, viscosity, and perhaps chemical composition of the particular ice to properly model this.

    Comment by Steve — 21 Jun 2008 @ 4:19 PM

  215. Steve, the physical properties of ice are known over a broad range of properties that encompasses all conditions occurring terrestrially. Were the ice homogeneous (and yes, amorphous solids can be homogeneous) and the terrain it advances over even moderately smooth, modeling would be a tractable problem. As it is, you have a variety of processes going on, and which are most important is still moot.

    Comment by Ray Ladbury — 21 Jun 2008 @ 5:49 PM


    Excerpt follows, click link for full article from New Scientist.
    Cite to Science magazine is at end of excerpt.

    Van de Waal believes that the channels that carry the meltwater out to sea freeze up during the winter months. In summer, pulses of water rushing down the moulins to the bedrock overwhelm the narrowed channels, and the increased pressure lifts the ice sheet off the rock, enabling it to move faster.

    However, after a few days the channels are forced open by the water, and it drains away from the glacier. As a result, the ice grinds back down against the bedrock and the lubricant effect is lost.
    No lubrication

    Van De Waal says this indicates that, overall, meltwater has a negligible effect on the rate at which the ice sheet moves.

    Not all scientists agree. Jay Zwally of NASA’s Goddard Space Flight Center in Greenbelt, Maryland, US, says that averaging data over the last 17 years does not make sense because the most rapid melting at the edges of the ice sheet did not start until recently.

    “It’s only in the last five years or so that the warming signal has really been visible,” he says.

    Zwally told New Scientist that unpublished data from the eastern edge of the ice sheet suggests between 3% and 5% more ice is being lost because of lubrication than would otherwise happen. That is less than the 25% that was previously calculated, but still significant, he says.

    Journal reference: Science, vol 321 p 111

    [Response: Mauri Pelto discussed this exact issue here in one of his guest posts a couple of months back. – gavin]

    Comment by Hank Roberts — 3 Jul 2008 @ 6:44 PM

  217. Does anyone know what is going on at Jakobshavn? Between the 28th and 29th of July, the satellite images appear to show that something massive occurred (large white mass representing ice) in the waters at the outlet point.


    Sometimes, you have to start at this link to get to the composite satellite images:

    On the 28th the waters were clear, on the 29th the little bay is full. Anyone know anything about this? Thanks

    Click on the farthest left image, which will take you to the pages of the latest satellite photos.

    Comment by Tenney Naumer — 31 Jul 2008 @ 9:09 AM

  218. Oh, and the images from the 28th and the 29th are no longer available, but if anyone wants them, I saved them and can e-mail them. In any case, today’s image shows that the bay is full.

    Comment by Tenney Naumer — 31 Jul 2008 @ 9:12 AM

  219. Well, obviously my knowledge of the geography of Greenland is pitiful. In fact, the area I am talking about is to the east of Qeqetarsuag. Sorry.

    Comment by Tenney Naumer — 31 Jul 2008 @ 9:23 AM

  220. OK, I finally figured it out — it is the outlet for the Ilulissat glacier. Anybody know what is going on there?

    Comment by Tenney Naumer — 31 Jul 2008 @ 9:30 AM

  221. OK, I know I can be a complete dummy — in any case, I cropped the images from the 28th and the 30th and posted them on my blog so anybody can see what I am going on about:

    Comment by Tenney Naumer — 31 Jul 2008 @ 9:46 AM

Sorry, the comment form is closed at this time.

Close this window.

0.340 Powered by WordPress