From the University of Colorado at Boulder Water flowing through ice sheets accelerates warming, could speed up ice flow
Caption: Standing melt water in Greenland crevasses can carry warmth to the ice sheet’s interior, accelerating the thermal response of the ice sheet to climate change.
Credit: Image courtesy Konrad Steffen, CIRES
Melt water flowing through ice sheets via crevasses, fractures and large drains called moulins can carry warmth into ice sheet interiors, greatly accelerating the thermal response of an ice sheet to climate change, according to a new study involving the University of Colorado at Boulder.
The new study showed ice sheets like the Greenland Ice Sheet can respond to such warming on the order of decades rather than the centuries projected by conventional thermal models. Ice flows more readily as it warms, so a warming climate can increase ice flows on ice sheets much faster than previously thought, said the study authors.
“We are finding that once such water flow is initiated through a new section of ice sheet, it can warm rather significantly and quickly, sometimes in just 10 years, ” said lead author Thomas Phillips, a research scientist with Cooperative Institute for Research in Environmental Sciences. CIRES is a joint institute between CU-Boulder and the National Oceanic and Atmospheric Administration.
Phillips, along with CU-Boulder civil, environmental and architectural engineering Professor Harihar Rajaram and CIRES Director Konrad Steffen described their results in a paper published online this week in Geophysical Research Letters.
Conventional thermal models of ice sheets do not factor in the presence of water within the ice sheet as a warming agent, but instead use models that primarily consider ice-sheet heating by warmer air on the ice sheet surface. In water’s absence, ice warms slowly in response to the increased surface temperatures from climate change, often requiring centuries to millennia to happen.
But the Greenland ice sheet is not one solid, smooth mass of ice. As the ice flows towards the coast, grating on bedrock, crevasses and new fractures form in the upper 100 feet of the ice sheet. Melt water flowing through these openings can create “ice caves” and networks of “pipes” that can carry water through the ice and spreading warmth, the authors concluded.
To quantify the influence of melt water, the scientists modeled what would happen to the ice sheet temperature if water flowed through it for eight weeks every summer — about the length of the active melt season. The result was a significantly faster-than-expected increase in ice sheet warming, which could take place on the order of years to decades depending on the spacing of crevasses and other “pipes” that bring warmer water into the ice sheet in summer.
“The key difference between our model and previous models is that we include heat exchange between water flowing through the ice sheet and the ice,” said Rajaram.
Several factors contributed to the warming and resulting acceleration of ice flow, including the fact that flowing water into the ice sheets can stay in liquid form even through the winter, slowing seasonal cooling. In addition, warmer ice sheets are more susceptible to increases of water flow, including the basal lubrication of ice that allows ice to flow more readily on bedrock.
A third factor is melt water cascading downward into the ice, which warms the surrounding ice. In this process the water can refreeze, creating additional cracks in the more vulnerable warm ice, according to the study.
Taken together, the interactions between water, temperature, and ice velocity spell even more rapid changes to ice sheets in a changing climate than currently anticipated, the authors concluded. After comparing observed temperature profiles from Greenland with the new model described in the paper, the authors concluded the observations were unexplainable unless they accounted for warming.
“The fact that the ice temperatures warm rather quickly is really the key piece that’s been overlooked in models currently being used to determine how Greenland responds to climate warming,” Steffen said. “However, this process is not the ‘death knell’ for the ice sheet. Even under such conditions, it would still take thousands of years for the Greenland ice sheet to disappear, Steffen said.
This study was funded by NASA’s Cryosphere Science Program.
Heat of Fusion still 333.55 kJ/Kg for water ice, is it?
Just checking.
@Mojo
> Heat of Fusion still 333.55 kJ/Kg for water ice, is it?
Thought it was a Night in Tunisia, Dizzy Gillespie, Decca 31193
Building stuff is hard. Real science is hard. It takes years and years of drudgery to build something well or do the basic science necessary to really contribute wealth to our world. And yet, even knowing this, I used to (develop software) love the fact that in almost no other endeavor but designing and building software was it so easy and fast to build something you could be proud of. As an engineer good at building things I was always easily tempted to send my creative side into a software ‘puzzle’ at the expense of more beneficial pursuits.
It’s a temptress, software is. And I wonder how many smart people have been tempted by it to the tipping point where they’ve abandoned the cold, hard, real world for the ‘soft’ world where you can do most of your work in your PJs in your mom’s basement, sipping on a Kool Aid.
If you graphed the number of scientists building models over the last 30 years would get a hockey stick? If you did, would that be a proxy the scientific consensus?
Maybe some of these guys ought to actually climb down into a glacier and see what is really going on there.
So what’s new? Just go to the terminal face of any glacier and observe the amount of water flowing away, and more importantly, where it is coming from. The flow isn’t just drip water from the glacier surface, but it is flowing from pipes, crevasses and the glacier base.
As tty already commented, not including this in melt projections is just incompetence.
tty,
Thanks for your explanation of why the crevasse water appears so blue.
How can water that must be 0 degC when it enters the ice possibly warm the inside an ice cap? This is only possible where the ice inside the glacier is well below 0 degC. In the process of warming the inside of the ice cap, however, some of the water must be turned into ice. Instead of having some water at 0 degC and some ice at say -10 degC, we get less water and more ice at say -9 degC. To a first approximation, changing how thermal energy is distributed within an ice cap will have no impact on how long it takes for the ice cap to melt. The interesting question is how much faster will the ice flow because it is warmer. Instead of the press release, try the real paper:
Thomas Phillips, Harihar Rajaram, Konrad Steffen
GEOPHYSICAL RESEARCH LETTERS, VOL. 37, L20503, 5 PP., 2010
doi:10.1029/2010GL044397
Cryo-hydrologic warming: A potential mechanism for rapid thermal response of ice sheets
Cryo-Hydrologic (CH) warming is proposed as a potential mechanism for rapid thermal response of glaciers and ice sheets to climate warming. We present a simple parameterization to incorporate CH warming in thermal models of ice sheets using a dual-continuum concept, which treats ice and the cryo-hydrologic system (CHS) as overlapping continua with heat exchange between them. The presence of liquid water in the CHS due to surface melt leads to warming of the ice. The magnitude and time-scale of CH warming is controlled by the average spacing between elements of the CHS, which is often of the order of just 10’s of meters. The corresponding time-scale of thermal response is of the order of years-decades, in contrast to conventional estimates of thermal response time-scales based on vertical conduction through ice (∼102–3 m thick), which are of the order of centuries to millennia. We show that CH warming is already occurring along the west coast of Greenland. Increased temperatures resulting from CH warming will reduce ice viscosity and thus contribute to faster ice flow.
Some people see ice in a different way:
http://www.auburn.edu/academic/science_math/cosam/research/antarctica/2006/journal/6/8/index.php
Group from Auburn, at Palmer station, Antarctica:
“…Several gifts and mementos later, we were up in the station’s Lounge for a congenial dinner, followed by a summary (so far) talk about the cruise by Ken Halanych and then some social time. One very special thing was the “bar ice” used to make cold drinks colder. This was a chunk of 20,000 year old ice from the glacier that had been craned up to the upper deck. Periodically, someone would go out to chip off some chunks into a rubber basin, and that was the ice bucket. Imagine using, the clearest, coldest oldest ice you’ll ever see ahhhhh, refreshing!!…”
With other scientists, a 20,000 year old piece of ice could have given clues about the history of the earth.
Added:
Sounds like they may have done some post-grad time working as a bartender:
“…An ice luge is a large block of ice with a narrow channel carved through it used to cool a beverage for drinking. A determined quantity of liquid, typically liquor, is poured into a channel at the top of the luge. A few seconds later the drink is dispensed at the bottom of the channel at an ice cold temperature, either directly into an open mouth of a participant or a waiting glass. An ice luge can be carved from a block of ice or cast from a mold, which are sold commercially…”
So if the alcohol is chilled by the time it hits the glass, wouldn’t the water also be super-chilled by the time it hit the bottom?
So to sum up…
We already knew that:
1. ice melts and turns to water.
2. a glacier has all sorts of stress fractures from its constant movement
3. water seeks the lowest point
4. water has far more capacity to carry and transfer heat than air.
5. ice can be an insulator
6. water flowing through ice can warm and melt the surrounding ice
What we discover from scientists playing with a new model
1. previous glacier models did not know about 1-6 above.
On the one hand I commend them for making a better model. On the other I would be ashamed to admit that the previous models were so simple that they did not bother to process what happened to the ice once it melted.
Richard111,
You ask:
“So what value then is the Greenland Ice Core Data?”
The GRIP and GISP2 ice cores were taken at very cold sites where there is no meltwater. GISP2 is at 72N, 38W at an altitude of 3,205 meters (10,500 feet) so the average annual temperature is -29 degrees Celsius.
IMHO the Greenland ice core data is the best we have for the northern hemisphere covering the last 50,000 years.
John Day,
As you say, the temperature in Greenland peaked in 1940. However, the melting of ice in Greenland has been going on for over 10,000 years. The rate of melting goes up and down with temperature and other factors.
Please note that while the modern warm period has increased the rate of melting it has also increased the amount of precipitation in central Greenland by almost a factor of 5 compared to the last glacial:
http://www.ncdc.noaa.gov/paleo/pubs/alley2000/alley2000.gif
I thought pure liquid water was blue – but you need an appreciable path length to observe that. It is the suspended solids in water and the properties of the illuminating ambient light that are usually responsible for the significant variation in colour that water can have. ttys comment makes perfect sense though – is old ice blue because it includes particles of rock, or because of a phase change in the molecular structure of the ice?
Paul Bahlin you make an excellent point about software. Sadly I’ve witnessed a computer model used in lieu of real experiments in undergraduate Chemistry practicals – the model’s results are so much more reliable than the real experiments!
kadaka (KD Knoebel) I liked your comment too. Somewhere Climate Science lost its way and instead of getting back to basics and reviewing the body of evidence critically in order to decide where more work was needed and where the ‘established’ science was not as sound as it may appear (ie when it could have matured) it jumped to a politically motivated conclusion and abandoned any pretext at scientific method. Or perhaps Geography just got delusions of grandeur?! I think if most scientists working in fields where there is money at stake were frank, they ought to recognise parallels in their own disciplines with what went wrong in Climate Science and I just hope that now Science as a whole can take a long hard look at itself. As other commentators on this blog have pointed out, perhaps the blogosphere really will become part of that review.
Anyhow, models can be fantastically useful in prioritising real experiments, but when they replace the real experiments Science has lost its way.
Rational Debate says
———–
But we’re supposed to trust the accuracy of predictions and conclusions based on these various models? The science is incontrovertible and undebatable, the scientists all in utter consensus.
Call me skeptical (pun intended).
————–
I think a bit more skepticism is need by checking the original paper. Nothing is as obvious as it seems, contrary to the views of the local navel gazers.
In this case the issue is not whether there us melt. The issue is how important is it? Navel gazing does not answer this question.
The process in science is simple:
1. Come up with the simplest model possible.
2. Compare it’s predictions with observations.
3. If they don’t match find out why and incorporate the new understanding into the model.
4. Repeat until satisfied.
This is what happened here.
And why do I have to explain this? A couple of centuries of science and people still don’t get it.
From LazyTeenager on November 6, 2010 at 3:57 pm:
That is what should have happened at the start, but there were massive failures at the very first step. Some small amount of field experience or even studying the existing literature would have revealed the true nature of the glacial ice. Heck, a study of pictures from National Geographic should have done it. Instead, more than once and by multiple people, too-simple models were used that yielded inaccurate results.
Then there came the screw-ups at the second step, as somehow no discrepancy appears to have been noticed for quite a long time, thus the third step was never engaged. Indeed, it looks like they went right from the first step straight to the fourth, convinced their work was inherently correct.
For those who missed it, this is more of the same story covered quite frequently on this site: What happened decades ago, billed as the nascent climate science, buttressed by computer models that yielded their own consensus, wasn’t science. Now, finally, are the real questions asked, the real world closely studied with the models made to match it, and real science emerging.
Several omissions, errors even? Certainly, Greenland’s geography is mis-represented in the writeup presented.
Contrary to most grade-school era maps used across the world, Greenland is merely a large island. Its land are is comparable in magnitude to Peru, Sudan, the Congo, Mexico. In the US, it is as far north to south (tip to tip) as north Idaho through Utah and Nevada to Arizona, and – like these Great Basin states – is bordered by mountains on all sides.
The writers ignore – or don’t know (doubtful!) that only the narrow 15 – 30 mile strip strip of coastal glaciers could move by the so-called basal flow of water suspension theory.
The rest of the island? Its interior is surrounded by mountain ranges. Any movement of ice in the interior “down and westward” (from the eastern side of the island’s interior) is resisted completely and permanently by a “down and eastward” movement from the western mountains. Any attempted movement north is resisted by ice masses moving south.
The much talked about “basal flow” theory is likewise flawed: each paper (and each text book) discussing the theorectical glacier ice flow used for this computerized model uses simplified assumptions of equal water pressure across the bottom of the glacier. Each flow calculation uses approximations of resistance across the base of the rock under the glacier – and then extrapolates this assumption all the way up to the near and far sides with resistance at the ice-to-rock interface assumed equal. Each approximations assumes equal pressure across the ice, and equal resistance across the glacier to flow – as is from an ice mass that is constant and of equal structure, density, stress resistance and tensile strength.
Real ice is a com[plex broken field of “small” individual ice block that may – or may not – extend from top of the ice to the bedrock below. Each block moves down by gravity independent of the block above, but pushed by the block above once resisted from below. Is is NOT a constant flow – nor anything like a constant flow of water, gas, or even a solid block.
The rough bedrock below the ice itself contributes its own character that is ignored by a simplified “basal flow” assumption. The rough rock (full of 1 meter depressions and dips, 2 and 3 meter “notches” between and inside of the larger ten and twenty meter valleys allows any water that does get to the bottom of the ice (and some always will of course) to flow unimpeded to lower areas – without ever being able to “build up” into lakes or ponds of sufficient mass to “flow” the tons of sliding ice above. Any water trapped momentarily runs around the obstacles on the bedrock through these diversions and spillways and canals – it is not trapped and dammed behind these long enough (deep enough) to float the ice.
Also, when ice builds up, it (obviously) builds up mass that forces the ice directly down against the bedrock that supports the glacier. Thus, unlike a high-pressure oil being forced into a “wedge” between a micron-tolerance ultra-smooth bearing surface and a rapidly rotating shaft, the water has nothing but a few feet (at most!) of gravity head (gravity pressure) behind it. It cannot “build up” to any depth under the glacier ice (which is required to “float” the ice under the basal flow theory) because there is no way to pressurize this water to force it under the ice into the gap between the ice and the irregular bedrock. It (the water on the bottom of the glacier) will always move to the lowest point (the “valleys” and irregular dips and scratches in the bedrock) and then flow downhill. If trapped, the still water loses heat immediately to the ice (at -15 to -30 degrees) and simply re-freezes in place.
Do glacier ever flow rapidly – as if suspended by water? Yes. It is wrong to claim they do not. But that infrequent, less-than-1% of the time) flow is not the default or normal condition used by the CAGW community to create a “catastrophic sudden melting – of Greenland’s small mass of coastal glaciers. Rather, it is like a very long threaded screw several miles in length. That threaded screw is broken and scarred by tens of thousands of random cuts and nicks and scars on the threads. Randomly, can a section be free of scars and the nut be turned easily? yes. Certainly. But then immediately thereafter, the scars return and the nut jams again. It doesn’t “turn off” the threaded rod, but must be torques and twisted cutting new metal the rest of the way. The glacier moves slowly this way, cutting new “threads” in the steel rod every inch. Never spinning off in massive long bursts of theoretical, well-lubricated speed.
Did the writers eventually admit their “new” theory is limited? In a way, yes. They said only that it might take centuries to melt the glaciers.
Well, in centuries – with an massive ice age already overdue, and with the modern warming era coming to a close in 2000-2010 or 2060-2070 – their fears are not well-founded.