Thanks for this update on a rapidly evolving field of study. The linking of structural integrity (or its lack) and melt rates seems logical, though it is a connection that doesn’t always seem well understood by commentators here and elsewhere. It is also interesting how little fanfare the Wilkins collapse received.
Why is the floating ice getting thinner? It is being warmed from above and below if I read it right. It seems like it doesn’t take much to make a big change. Cracks grow and connect to each other, creating very long cracks rather than freeze back together as they did when the temperature was slightly cooler. This hints that we are at or near a tipping point. The ice must now be at a temperature that allows the ice to change shape much more easily than it did before. It seems like that indicates that the temperature is close to the melting point, at least for ice under the pressure found inside the glacier. Am I right?
Since melting point varies with pressure, can you make a theory that relates the thickness of the ice to the temperature? As the ice warms very slightly, it may be required to get a lot thinner. The end point is that at the melting point at zero pressure, the thickness of the ice has to go to zero.
Regarding 5 and 6 I am not making a case that the ice is weaker. I do not believe that the temperature change for the ice shelves has resulted in this. It is the thinning that weakens the ice shelf, leading to greater floatation and reduced contact with pinning points and consequent rift development and expansion. The thinning can come from bottom melting or surface melting or a combination for the ice shelves. The Greenland Ice Sheet ice properties are not changing, it is the balance of forces that is changing due to thinning. In this case we have no data to conclude that basal melting has changed and only small areas are actually afloat and exposed to the potential for additional melting via sea water.
One thing that is common to sea glaciers and ice shelves is that they are both affected by sea level rise. This will produce a vertical torque, especially at times of high tides (see my #3), resulting in increased calving.
As you are aware sea level rise also has the effect of moving the grounding line proximally, which reduces the friction on the ice shelf or valley glacier without reducing the gravitational force which is driving the ice from the slopes beyond the coastal plain. The acceleration is part of a positive feedback loop so its effect will increase with time. Is sea level rise only now emerging as a significant player?
However we have all seen movies of great chunks of ice suddenly breaking off to crash into sea to form iceburgs. Iceburg calving is an example of brittle failure of ice.
Brittle failure of ice apprears to have been studied as a result of increasing strain,with less attention paid to brittle failure of ice as a result of increasing temperature that change the strength of the ice.
Alistair (#7) mentions sea level rise. Has anyone put the pieces together yet and considered how these improvements in understanding ice shelf break-up translate to predicting sea level rise? Presumably the loss of ice shelves increases the discharge rate of land-based ice but can we learn anything from this as relates for example to the prospects of the West Antarctic Ice Sheet (WAIS), a substantial fraction of which is grounded below sea level?
The fact that thinning appears to be a major factor is a little reassuring as something as thick as WAIS (and hence with massive thermal inertia) is not going to be thinning any time soon …
Even temperate ice is quite strong and can sustain massive calving fronts. We have many temperate glaciers that rival the size of the terminus reach of many Greenland glaciers and the ice is somewhat but not greatly weaker. There are many in Palin’s home state for example. The Holland paper raises an interesting point. Is it warm ocean water impacting the Jakobshavn. The analysis in this Realclimate review points out it can be surface melting or basal melting that can induce thinning or both. It depends on the specific situation. The Holland paper hints for the first time that ocean temps may have a role. I would caution the reader that no temp observations right near the calving front are available to document this to be the case. The measurements are some distance from the calving front and may not have resulted in warming under the floating section. There is certainly a strong freshwater flow exiting from beneath the glacier to consider. Secondly as Howat et al., (2008) Sole et al., 2008 have noted the acceleration and thinning is occurring on over 40 outlet glaciers, many of which are not afloat at all and thus basal melting from warmer seawater cannot be the explanation. I am intrigued by the Holland paper and the concept deserves further examination, but it long way from being based on conclusive observational evidence that it is the cause of Jakobshavn Isbrae thinning.
Perhaps a bit off-topic– Getting away from the ice sheet surface (and below it), I’m curious as to what kind of effects the presence of atmospheric aerosols have in the polar winter. Presumably they have little to no shortwave albedo effect, but how significant is the downwelling infrared term (it’s very dry) at either ice sheet during the dark months?
Yes, heat released by condensation of water vapor weakens ice. Foggy periods turn lake and
river ice black in spring.
The significance of latent heat for snowmelt has been described by Dunne and Leopold (1978):
“If water from moist air condenses on a snowpack, 590 calories of heat are released by each gram of condensate. This is enough energy to melt approximately 7.5 gm of ice, which when added to the condensate yields a total of 8.5 gm of potential runoff”.
Dunne, T., Leopold, L.B. (1978) Water in Environmental Planning; pp. 477‑499
Horrendously complicated topic… thanks for tackling it. Here’s a bit more on the ocean component:
The subsurface has warmed, but why? Holland says it is due to a change in the atmospheric circulation resulting in a change in the North Atlantic gyre which then has allowed warmer water into the South Greenland region. That can increase the sea surface temperature and the air temperature, leading to more rapid melting and thinning at the glacier terminus. This effect would be countered to some degree by the cold freshwater melt coming off the glacier into the ocean, probably felt most strongly at the end of the melt season.
Thus, the mechanisms (terminus thinning and subsurface warming) don’t seem to rule each other out, and both might be adding to the overall observed result.
Where things get a little mysterious is in the attribution of the warming subsurface to a change in the NA gyre circulation, which is attributed to a switch in the NAO (North Atlantic Oscillation) from a positive to a negative phase. Here is a graph of the NAO index updated through 2007, from Tim Osborn: http://www.cru.uea.ac.uk/~timo/datapages/naoi.htm
We do know that ocean basins produce this oscillatory behavior – the El Nino / Southern Oscillation, the Arctic Oscillation, the Pacific Decadal Oscialltion, the NAO, the Madden-Julian Oscillation (Indian Ocean), but they seem to have a bit of a random component, their forcing mechanisms are poorly understood, their “phase changes” appear impossible to predict very far in advance, and they must also be sensitive to the overall climate warming.
What is most irksome is how these oscillations, which are real phenomena, have been abused by the tobacco science skeptic industry as an “explanation” of why the climate is warming. This leads to bias on the part of the observer, however… The real issue is probably the inverse – how is the warming climate going to alter / interact with these oscillations? What will happen to NAO, El Nino, etc., in a warmer world? That’s a big unknown.
Thus, while it may be possible to attribute the warming subsurface to a change in the NA gyre circulation to a shift in the NAO, to what do we attribute the shift in the NAO?
We might be seeing an increased equator-to-pole heat transport effect as well. Unfortunately, there is no detailed instrument record of subsurface changes in Gulf Stream heat transport into the region over the past decades, so it’s hard to say – and the atmospheric component?
Here is a paper in press which discuss the seismic effects to calving events in Greenland. The authors propose that glacial earthquakes are caused by icebergs overturning and scraping the fjord bottom during calving. What kind of implications could this have for seismic activity with further melting on the edges?
Pat you are correct on the impact of humidity on thin ice which has a good connection with the atmosphere. But not for an ice sheet where the vast bulk of the ice sheet is sealed from the atmosphere. Ike Excellent post. I agree whole heartedly with the Seager work that indicates a slowdown of the northern extension of the Gulf Stream is not the end all and be all for temperature change in Europe. I also have never been satisfied that the magnitude of temperature change identified in the abrupt cooling for the GIS during the Younger Dryas is not exaggerated. For example if Seager and Battisti are correct than a change in atmospheric ciruculation occurs, this would alter the moisture source and potentially the isotope signature of the snowpack falling on the GIS. Hence, the temperature calibration would be off. The role of the NAO and AO needs to be better quantified and I am sure it will be. The NAO is the dominant mode of winter climate variability in the North Atlantic region ranging from central North America to Europe and much into Northern Asia. The NAO is a large scale seesaw in atmospheric mass between the subtropical high and the polar low. The positive phase of the NAO that we have been in is supposed to yield warm and wet winters in Europe and in cold and dry winters in northern Canada and Greenland. But that is not what Greenland has generally seen. The AO is in a warm phase that appears to be the longest and strongest of the last 100 years. Which suggests that Ike you are right in asking how is global warming impacting the AO. http://jisao.washington.edu/ao/
Surely, this area of incredible research represents nothing more than reporting on regions of the planet where little analysis had been done previously. We now have the technology to view and document natural variability, big deal.
Since there was modest research and zero satellite images dating back to the beginning of this century (for obvious reasons), we lack sufficient time scale with glaciers in the Arctic and the Antarctic. Does it really matter what glaciers have done or not done over the past 10-years? The media constantly remind us how unprecedented these events are. But without any trustworthy historical data, it is meaningless.
It is curious that virtually no evidence on the thousand of glaciers that are advancing around the planet go unreported. It seems to me, observing a handful of cherry picked glaciers that are breaking up, as hard evidence to buttress global warning, is intellectually dishonest, despite the research acumen in a relatively new field.
[Response: Strong words for something that isn't true in the slightest. Records of glacier retreat go back much further than the 'last ten years' even in Greenland. Jakobshavn Isbrae has records going back to at least 1851 for instance. As for 'thousands of advancing glaciers', you should really try and find out for yourself what the truth is. Start at the World Glacier Monitoring Service or UNEP - gavin]
Re #21 Where Mauri wrote “Pat you are correct on the impact of humidity on thin ice which has a good connection with the atmosphere. But not for an ice sheet where the vast bulk of the ice sheet is sealed from the atmosphere.”
As Dr Hansen points out in that article, ice sheet melting is non-linear but the scientists want to draw straight lines and ignore any tipping points. Hansen calls this “reticence”!
But there seems little point me pointing out the foolishnes of the scientific world. Hansen has tried to avail. What chance have I to warn of the dangers ahead. The causes of this financial crisis were known long ago but no one could stop that. What chance is there of preventing a worse global catastrophe when abrupt climate change hits?
We must be the two worst students in class. I agree with you, that moist air can rapidly melt big ice.
I think that any ice subject to the formation of moulins cannot be considered, “isolated from the atmosphere.” Water falling down the moulins releases heat, and that heat softens the surrounding ice. Heat is being advected into the bulk of that ice, so the bulk of the ice is not sealed off. That is basic physics that I learned in grade school, and I cling to it.
True, the vast bulk of the GIS is still too high, too dry, and too cold to have surface water and form moulins. In that desert, the surface ice sublimates, leaving colder, harder, stronger ice behind. However, that high dry cold ice depends on the mechanical support of ice at lower altitudes.
Ice that has moulins in it is weak. The speed that moulins close at the end of the melt season shows how soft and plastic that ice becomes. Anything that depends on such ice for mechanical support is in trouble.
This is not our grandfather’s Greenland. Places where he had sea ice, we have have open water. He had an Arctic desert; now it rains at the North Pole. Models built for a cold desert, may not work in a new Greenland where the dew point may exceed 0C.
Richard that is a good point, but the energy is coming from the sun. By increasing the humidity, the solar energy is used more efficiently, or putting it another way there is less energy lost to space by the emission of longwave radiation because humid air has a strong greenhouse effect.
I obviously cannot use logic to convince that slightly warmer ice exposed to a bit more humidity will not weaken the ice notably. But lets review a few details. The number of lakes on the surface of the GIS is not significantly greater than in the 1950′s imagery or the 1980′s imagery. They must have a greater volume and flushing rate with higher melt of course, but they are not new. The moulins have always been there as well. Of course they also have a bit more volume now and maybe a bit higher distribution. But in general the area of lakes and moulins has been a zone of melting, thus we are not talking about a change in surface state. Also you need to look at, walk across glaciers in temperate climates with high melt rates, and far more moulins per unit area to realize that these features do not significantly weaken the ice, anymore than pulling a hair from your leg weakens your leg. That is how small moulins are compared to the entire ice sheet. They do carry lots of water and that is a different point. The theory is sound with holes and humidity, it just is not realistic as a big picture player. I have made plenty of observations of melt rate on the walls of moulins and it is similar to the surface rate, but is not very large.
Oh well, the idea of it all being caused by rising sea levels does not seem to grab you, and my (and Jim Hansen’s) wet surface seems to have an energy balance problem. So how about this? As CO2 concentration increase so will its greenhouse effect which will raise the height of the snow line globally. Since the snow line falls with latitude, raising the snow line will tend to melt the ice sheets from the lower latitudes first, for example the Antarctic Peninsula and southern Greenland.
Raising the snow line will also begin by removing the equitorial ice caps (Kilimanjaroo) and the sea ice, which is what we are seeing. It will also leave the tops of the high ice sheets, such as Greenland and East Antactica, unaltered as you are reporting.
Since the firn line line and the snowline are related, it should be possible to check whether the firn line has risen by the same amount for Jakobshavn Isbrae, Wilkins Ice Shelf and the Petermann Glacier. If so, then I would think that anthropogenic greenhouse gases may be the main culprit.
As sea level rise is on the order of a few mm/annum, and rapid glacial melting tends to be more like a meter/annum, SLR should be seen to be a much smaller effect.
There could be two impacts, from an increase in air temperature and/or atmospheric humidity. The most obvious one is heat transfer to the snow/ice surface. Less obviously and indirectly, the change in surface properties due to melting likely decreases the albedo. Could this second effect be important, and possibly underappreciated? I think Hansen referred to it as albedo flip?
However, the core of the glacier is an incompressible fluid (ice) with limited thermal conductivity. In the winter the skin of the glacier loses heat and becomes stronger. Now you have a fluid inside of a stronger skin which is a turgid structure. Turgid structures are stronger than would be estimated by the bulk properties of the internal fluid or the mechanical strength of the skin alone. Examples of turgid structures abound in biology and have been used in engineering such things as liquid fueled rocket boosters.
In the melt season, the skin of the glacier warms, softens, and the entire structure of the glacier loses strength. The entire structure loses mechanical strength despite the fact that only a tiny volume of ice on the surface changes temperature and thus its mechanical properties. This loss of strength allows the glacier to flow more rapidly or to surge despite relatively small heat fluxes over a brief period. Turgid structures are the ultimate in non-linear behavior.
If you consider a glacier in the winter as a fluid in a stronger skin, and consider the same glacier in the summer as a weaker surface around core that is now colder than the surface you will have a physical explanation for why glaciers surge in the summer.
Of course, basal lubrication has an effect, but mostly, surging is caused by gravity on a mechanical structure that rapidly loses strength as the skin softens.
You are going to say, “Oh that is impossible and silly.” How thick is the skin of a the guard cells that control the area of stomata? Glaciers change the strength of the skin rather than the contained volume, but the mechanical effect on the strength of the structure is the same How thick is the skin of a rocket booster held rigid by pressure of the fuel inside? How thick is the skin of a child’s balloon after the clown has tied it up into an animal shape?
Draw a picture of the problem and ask, “What would Feynman think?”
In Sole et al 2008 above it said that only 18% of the GIS is grounded below sea level, and that only 2-5% of it is subject to potential ocean contact. Is this data accurate and conclusive? Most everything I’ve seen on the net says that most of the GIS is marine and subject to ocean contact. A critical answer when it comes to predicting the fate of the GIS.
You mentioned reduced glacier inflow as a reason for ice shelf thinning. Is this because thinning results in reduction of cross sectional area of the inflow channels which offsets any gain due to increase in inflow velocity ?
I would like to make two points of order here. Moderator may wish to delete them as they are not germane to this article.
2)I had to reload the comment page three times to arrive at a captcha legible to my failing vision.
sidd (41) — The top of the three buttons just to the left of the reCAPTCHA logo gives you a new pair of words to try; repeatedly.
[Captcha modestly proclaims "slight powers".]
Comment by David B. Benson — 11 Oct 2008 @ 1:39 PM
#18. Phil, Downwelling IR is very interesting to have on a live basis as per link, what would be even more interesting is to have a similar sensor on the Polar ice cap during a known cooler period, say 1996-97 pre monster El-Nino period. IR Downwelling really seems to describe the heat of the entire lower atmosphere better than other traditional and none traditional methods. “The North Pole” graph, captures the missing data needed to explain some anomalies related to Polar ice cap and Polar glaciers melting. The peak IR of the melt season occurred mid-September as opposed to warmer surface temperature months, which is right, its the moment when the lower atmosphere is at its warmest. There are still significant warm spells in the High Arctic lately, last week over Cornwallis +4.8 C was measured along with freezing rain on the ground at -2 C, a plane flying over Lancaster Sound measured +10 C.. All the while ground temperatures were moderately above normal. Meaning, surface temperatures do not represent total heat of the entire atmosphere well, in this case the heat was really above, this drives surface temperature sensitivity quite wild over a longer time period. We are dealing with a vertically roving heat zone which appears and moves in mysterious ways, ultimately disabling our ability to discern any stable surface temperature trend in the short term.
> downwelling IR
Do the infrared astronomers keep track of this? I know some limited infrared astronomy is still done from Earth’s surface despite the infrared “sky glow” but I’d guess they must measure the infrared brightness of the night sky regularly, as a visible-light astronomer worries about sky glow in his or her observing frequencies.
Yeah, but it sounds like any infrared astro photograph made of an object with a known brightness will also be a record of the sky’s brightness at that time. I didn’t find mention with a quick search but anyone who knows anything about the subject will do better than I at answering whether there’s any information captured and carefully thrown away by people doing this:
“… all exposures with near-IR cameras have to be short ones, to avoid saturating the detector. …
… nearly all observations will be of objects much fainter than the sky. This means that accurate flat fielding and sky subtraction are quite vital. Indeed, unlike in the optical, you typically can’t see a thing in raw near-IR observations.”
I wonder if there are any “standard candle” infrared sources on the ground that are picked up by the infrared weather satellite imagery — they’d have to be something large to fill a pixel, maybe there’s a cooling pond for a reactor that has its temperature regularly documented? — if so, looking at the brightness of that known source viewed from above and compared might also be useful. Or, of course, there must be much higher resolution infrared imaging systems in orbit.
Perhaps we need a new ‘Gore Box’ initiative to open up military records for the atmosphere, the way then VP Gore did for the navy’s Arctic Ocean records — if someone can specify in a wishful-thinking way what would be relevant and useful, perhaps it’s there for the asking.
Pure speculation — but odds are there is data that can be pulled out and reused. Calling for a data mining expert in astronomy …..
I’ve two starking ocean current hypothesees to offer the climatific community.
1) It appears AGW that melts arctic freshwater also results in a negative feedback that acts against runaway warming; more warm Gulf Stream volume eventually results in less cool volume from Antarctic subpolar current to North equatorial at the base of Cuba. This is all from a googled 2007 news article. You’d expect more warm current to “churn” more cool current, but it actually churns more cold water to move around Antarctica in circles rather than head north. My insight is to suggest a mini-version of this historic effect (8200 yrs ago) might occur as arctic ice/glaciers melt, inducing a hiccup of lower temps in Eastern N.America and Western Europe. Were this to be predictable, farmers could maybe adapt by having cooler crop varieties handy.
2) Looking at a map of ocean currents in the Southern Hemisphere, the catastrophic worst case scenario in the warm direction (as bad as an Ice Age), one that would end our civilization, might occur if the Antarctic Subpolar current circling Antarctica breaks down and heads north. Right now parts of it branch off to form various cold currents: Peru, Benguela, W.Australia. More accurate to say if we AGW enough to cause a warm current; Brazil, Mozambique, E.Australia, to reach Antarctica rather than hitting the Antarctic Circumpolar and turning north…if the warm hits glaciers and the subpolar current turns north somewhere instead of circling Antarctica: this event if possible should be used by scientists as the worst-case baseline where our species becomes about as valuable as any other species when making normative economic calculations.
[Response: Neither of these things are remotely likely (and the second seems impossible under any circumstances). For more information on what might be more likely, look at the posts in the 'Oceans' category on the sidebar. - gavin]
Comment by Phillip Huggan — 13 Oct 2008 @ 11:45 PM
#50 This Chilean Lake drainage is unusual. It may become an annual source of GLOF’s now. It obviously is filling faster now. The question why, is either enhanced melting or reduced normal drainage. The cause of more frequent drainage could be retreat which weakens the ice dam. There is a Planet Action project on this lake, these projects are funded by SPOT Satellite Image Corp. http://www.planet-action.org/web/6-projects.php?projectID=626 . In terms of #51 not only is the 555 wrong but the 55% is just as incorrect. Take a look at 2005 in which WGMS notes that only 5% of the 275 glaciers and reported to them are advancing. This becomes worse if we expand to look at the fact that no glaciers were advancing in Switzerland of the more than 100 observed for the first time on record. In the North Cascades as well of the 107 we have observed not one is advancing and seven have now disappeared since 1984. http://www.nichols.edu/departments/glacier/Bill.htm
Anew paper on GIS mass balance due to surface losses and discharge via calving has been published by Rignot et al., in GRL http://www.agu.org/pubs/crossref/2008/2008GL035417.shtml
It updates this valuable record and again illustrates the magnitude of the sustained negative balances.