'Hidden plumbing' helps slow Greenland ice flow

Jan 26, 2011
'Hidden plumbing' helps slow Greenland ice flow
A moulin is a hole in a glacier that funnels meltwater from the surface to the bedrock beneath. This flow of water has importance consequences for the speed at which the glacier moves. Credits: J. Box

Hotter summers may not be as catastrophic for the Greenland ice sheet as previously feared and may actually slow down the flow of glaciers, according to new research.

A letter published in Nature on 27 January explains how increased melting in warmer years causes the internal drainage system of the ice sheet to 'adapt' and accommodate more melt-water, without speeding up the flow of ice toward the oceans. The findings have important implications for future assessments of rise.

The covers roughly 80% of the surface of the island and contains enough water to raise sea levels by 7 metres if it were to melt completely. Rising temperatures in the Arctic in recent years have caused the ice sheet to shrink, prompting fears that it may be close to a 'tipping point' of no return.

Some of the ice loss has been attributed to the speed-up of due to increased surface melting. Each summer, warmer temperatures cause ice at the surface of the sheet to melt. This water then runs down a series of channels to the base of the glacier where it acts as a lubricant, allowing the ice sheet to flow rapidly across the bedrock toward the sea.

Summertime acceleration of ice flow has proved difficult for scientists to model, leading to uncertainties in projections of future .

"It had been thought that more surface melting would cause the ice sheet to speed up and retreat faster, but our study suggests that the opposite could in fact be true," said Professor Andrew Shepherd from the University of Leeds School of Earth and Environment, who led the study.

An example of a two-dimensional ice-velocity map of the study area in southwest Greenland. The map is derived from Synthetic Aperture Radar images from ESA's veteran ERS-1 satellite separated by 35 days: 2 June to 7 July 1995. Credits: University of Leeds

"If that's the case, increases in surface melting expected over the 21st century may have no affect on the rate of ice loss through flow. However, this doesn't mean that the ice sheet is safe from , because the impact of ocean-driven melting remains uncertain."

The researchers used of six landlocked glaciers in south-west Greenland, acquired by the European Space Agency, to study how ice flow develops in years of markedly different melting.

Although the initial speed-up of ice was similar in all years, slowdown occurred sooner in the warmest ones. The authors suggest that in these years the abundance of melt-water triggers an early switch in the plumbing at the base of the ice, causing a pressure drop that leads to reduced ice speeds.

This behaviour is similar to that of mountain glaciers, where the summertime speed-up of ice reduces once melt-water can drain efficiently.

Explore further: Huge waves measured for first time in Arctic Ocean

More information: 'Melt-induced speed-up of Greenland ice-sheet offset by efficient subglacial drainage' by Aud Venke Sundal, et al. Nature, January 27, 2011.

Related Stories

Greenland and Antarctic ice sheet melting, rate unknown

Feb 16, 2009

The Greenland and Antarctica ice sheets are melting, but the amounts that will melt and the time it will take are still unknown, according to Richard Alley, Evan Pugh professor of geosciences, Penn State.

Recommended for you

Wave energy impact on harbour operations investigated

3 hours ago

Infragravity period oscillations—waves that occur between 25 and 300 seconds with a wavelength between 100m and 10km—can have an impact on berthing operations, depending on a harbour's geometry.

Huge waves measured for first time in Arctic Ocean

21 hours ago

As the climate warms and sea ice retreats, the North is changing. An ice-covered expanse now has a season of increasingly open water which is predicted to extend across the whole Arctic Ocean before the middle ...

User comments : 32

Adjust slider to filter visible comments by rank

Display comments: newest first

toyo
4 / 5 (4) Jan 26, 2011
Well well!
As we find out more, we learn to grow humble about our ability to forecast 'disasters' (I hope).
dinkster
4 / 5 (4) Jan 26, 2011
Well well!
As we find out more, we learn to grow humble about our ability to forecast 'disasters' (I hope).


A critically damped system doesn't oscillate, and the Earth (ironically?) is the best example in the world of a critically damped system. I tend to take predictions beyond 40 or 50 years with a grain of salt.
freethinking
2.2 / 5 (10) Jan 26, 2011
This is a lie. Al Gore said we all will drown then bought land close to the ocean. He Would never lie. Global Warming WILL make the ice sheets melt. AL GORE SAID SO. LONG LIVE GOBAL WARMING,errr LONG LIVE CLIMATE CHANGE!!!
ted208
2.5 / 5 (8) Jan 26, 2011
Another nail in the coffin for global warming alarmist and their casino models.
So much for "the science is settled" Al Gore
freethinking
2.1 / 5 (7) Jan 27, 2011
Ted208. The science is settled. Al Gore would not lie, my kids high school teacher said so. The world is warming, 2012 is coming.
GSwift7
2.6 / 5 (5) Jan 27, 2011
When they finally release the global sea level for 2010, there may be another big surprise in store. They are about 5 months behind in the data release, so we haven't seen the end of the year yet, but 2010 could become the biggest drop in sea level in the instrument history. The sea level rise is decelerating, which doesn't match up very well with IPCC predictions, or Al Gore's get rich quick fantasy.
thermodynamics
1 / 5 (1) Jan 28, 2011
GS7: Where are you looking at the sea level measurements? Can you send the URL?
GSwift7
2.3 / 5 (3) Jan 28, 2011
Yeah, sure.

Google this and take the first link:

university of colorado at boulder sea level change

Notice the decrease in the slope of the trend over the past 6 years, and the huge dip in the most recent data. There was an overall drop in sea level from 2006 to 2008 as well. Like I said, they don't have the last half of the year up yet, but i'm very interested to see it when they do.
MikeyK
3 / 5 (2) Jan 30, 2011
....and this drop in sea level has nothing to do with the sharp La Nina then.... Look at the graph again..ht[delete me]tp://sealevel.colorado.edu/current/sl_ib_ns_global.jpg
Skeptic_Heretic
3.7 / 5 (3) Jan 30, 2011
....and this drop in sea level has nothing to do with the sharp La Nina then.... Look at the graph again..ht[delete me]tp://sealevel.colorado.edu/current/sl_ib_ns_global.jpg

Correct.
The deal hwith the data graphs is that most people don't understand the multitude of systems that will affect the reading. ENSO variability provides a strong signal in sea level rise and fall due to thermal expansion. During cold phases the sea level is expected to drop from .5 to 2.5mm while it will rise .5 to 2.5mm above expected rise during a warm phase.

The same is seen in temperature readings. A warm cycle provides higher temperature readings, ie:98 and 05, while a cooling phase results in lower temps. Everything is cyclical, but whne removing the cyclical variation the trend is still positive over multidecadal timescales.
GSwift7
2.3 / 5 (3) Jan 31, 2011
but whne removing the cyclical variation the trend is still positive over multidecadal timescales


I challenge that. How many decades of "good" and "global" data do we have?

I challenge any sea level data prior to the past two or maybe three decades due to uncertainty of ground level measurement. You can't measure sea level when ground level changed in undocumented ways. A good example of inaccurate ground level measurements is the coast of Texas where it was found (after the last big hurricane) that ground height wasn't as high as they thought in some areas and the result was unexpected flooding of areas they thought were safe. And that's right here in the US, where we supposedly have good records, right? What about other places?

How long does the record need to be to show a trend? How long are the natural cycles you need to remove? How many of those do you need in order to eliminate anomolous years?
GSwift7
2.3 / 5 (3) Jan 31, 2011
What was the first year we started getting good satellite sea level data from TOPEX/Poseidon? 1992. How many multidecadal natural cycles have we seen in the past .9 decades? Hmmm.
Skeptic_Heretic
2.3 / 5 (3) Jan 31, 2011
How long does the record need to be to show a trend? How long are the natural cycles you need to remove? How many of those do you need in order to eliminate anomolous years?
In order:

Depends on the period of oscillation in the relevant defining factors.

Not all are known but the most impactful (with result swings of greater or lessor .01mm) are local temperature oscillation, (ENSO, AO, etc) Solar input, ground level, water surface and subsurface temp.

Anomalous years are removed via smoothing.
What was the first year we started getting good satellite sea level data from TOPEX/Poseidon?

Prior to satellites, tide gauges were the primary measurement device. The satellite data shows that tide gauge data was on the low side, not the high side, and as far tide gauges have given a statistically significant approximation when used over large areas to account for event variability, like the Texas incident you outline above.
GSwift7
2.3 / 5 (3) Jan 31, 2011
Anomalous years are removed via smoothing


If the only good records began in 1992, as I propose, then it is possible that you are smoothing numbers that are all too high or all too low, or which are part of an anomolous multi-decadal trend. You can't know if any of those cases are or are not true.

Tide height is notoriously effected by changes in a number of different outside factors. Those factors act on scales from hourly to geologic time, and they vary from one location to the next. You also can't really measure tides in mid-ocean without a satellite (which didn't happen till 1992).

Any claim that there is a trend in sea level without also stating the limitations of what we know about sea level is completely misleading.

Just because current tide guage readings are low it would be fallacious to assume previous ones are also low. You can't know that. It's an educated guess, but still possibly wrong.
GSwift7
2.3 / 5 (3) Jan 31, 2011
How can you prove that the trend of falling sea level from 2006 to 2008 was the anomoly and that the other years of record are the true trend? Of the 18 yeas of satellite record, that's about 16% of the record that shows a downward trend. That doesn't indicate that the slope of your trend is linear or that you can reliably predict future slope. I'll bet you didn't even know that sea level had fallen in any recent years. Most people don't. Just like most people think storms have become more severe and more intense. Misconception and deception.
MikeyK
3 / 5 (2) Jan 31, 2011
GSwift- FYI here is the latest UC seal level chart...who are you trying to kid? You do understand how ENSO can affect sea level do you?
htDELETEtp://sealevel.colorado.edu/current/sl_ib_ns_global.jpg
You are back to your old tricks of trying to discredit the science to 'muddy the waters' so to speak.
GSwift7
2.3 / 5 (3) Feb 01, 2011
Yes, MikeyK, I posted a reference to that data above. You will notice that from 2006 to 2008 the graph actually shows a decrease in sea level. That's the "official" numbers. This year is also shaping out to be a year of sea level decrease. That would make 3 of the past 5 years showing a decrease in sea level versus the previous year. While the overall trend has been increasing since we have begun making accurate records (18 years), I hardly think that's a long enough time to draw very many conclusions aside from the fact that sea level is chaotic and not very well understood.

When you say that I am trying to discredit the science, what are you talking about? I'm reading the same science that you are reading, and I'm saying they are right. The article on this site isn't science; it's journalism, and not very good journalism. As SH and I have pointed out many times, the articles are often VERY different from the actual study.
GSwift7
1.8 / 5 (5) Feb 01, 2011
The satellite data shows that tide gauge data was on the low side


Explain that please, and provide a reference. Do you suggest that the sea level rise shown by tide guage isn't accurate?

tide gauges have given a statistically significant approximation when used over large areas to account for event variability


Wait, let's go back to that first thing you said...

Are you really suggesting that if you take 1000 bad measurements and combine them into an average, that it eliminates the errors? That's not logical or correct.
MikeyK
3 / 5 (2) Feb 01, 2011
That would make 3 of the past 5 years showing a decrease in sea level versus the previous year.

Yes, that is because of the 2 La Nina's in that time frame!
Can't you see the trend or are you doing, and I will say this again, the usual of taking the high point of 2006 (the 2005-2006 El Nino was a strong and long lasting episode) and compare to the recent La Nina (one of the strongest in recent times) induced dip?
You are as bad as Loodt, SR and others that harp on about there being no warming since 1998 and using the peak as the proof of their myth....unfortunately you are doing exactly the same here.
Science isn't about cherrypicking...anti-science is.
GSwift7
2 / 5 (4) Feb 01, 2011
I'm looking at the 18 year record of sea level from satellites. The first half of that record has a relatively steep upward trend. The second half of the record is still increasing, but at a noticeably slower rate. The trend seems to be decelerating. I think the next 5 years will be of great interest, and another 10 will be even better. I want to see more data. I did a fun little exercise by taking the ENSO time series from 1992 to present (the time period of the satellite sea level record) and copied it into MS Paint. I then moved all the positive and negative parts together. It clearly shows an increase in La Nina and a decrease in El Nino over time, but El Nino dominates. The time series I'm looking at is the one from NOAA. Google Multivariate ENSO Index (MEI) to see it. That is ENSO index compared to a base period, which means we've had a lot more Nino compared to Nina than we usually have. If we are returning to a normal balance, sea level rise could surprise us and fall.
MikeyK
3 / 5 (2) Feb 01, 2011
Hang on, didn't you say that sea levels are falling? Now you say that the "record is still increasing, but at a noticeably slower rate".
I would agree with your latter statement, progress has been made!
Entirely speculatory but one reason for this slowdown could be the cooling effect caused by mass sea-ice melting. We are now at or very near the lowest global sea ice recorded, and ice depth measurements indicate that sea ice depth has decreased by a third in the last decade. I know that if you put an ice cube in a drink it cools it down....we have some pretty huge icecubes each end of the planet!
Just to add to your increase in La Nina point, this is one of the effects of -ive PDO, a phase we are in now; +ive PDO's tend to have more El Nino's.
GSwift7
1.8 / 5 (5) Feb 01, 2011
I shouldn't mention this, because I've already stated that I have little faith in the older records, but if you look at the whole ENSO time series, from 1950 to present, there's a trend.

I see about 30 years of almost all Nina and very little Nino (so that's a cool period). Then I see the next 30 years where there's almost all Nino and very little Nina (so that's a warm period). So, I ask this: Are we seeing a 30 year cycle here? Are we about to head back into the opposite direction and have dominant Nina again for the next 30 years? What would that do to temp and sea level trends?

Yes, I said that sea level has fallen in two of the last 5 years, and if this year ends low, then it will be 3 of 5. The trend is still upwards for the 10 year period around those years, but the record does not show a continuous increase in sea level. It is non-linear.
GSwift7
2 / 5 (4) Feb 01, 2011
I would agree with your latter statement, progress has been made!


Are you serious? Do you think that global CO2 output has decreased? I think you need to look at the Mauna Loa co2 record. We are not getting better, we are getting worse. China and India aren't really trying to slow down you know. Are you really saying that the ENSO cycle is driven by co2 now? wow.
MikeyK
3 / 5 (2) Feb 01, 2011
I would agree with your latter statement, progress has been made!


Are you serious? Do you think that global CO2 output has decreased? I think you need to look at the Mauna Loa co2 record. We are not getting better, we are getting worse. China and India aren't really trying to slow down you know. Are you really saying that the ENSO cycle is driven by co2 now? wow.

????? What are you talking about? We were discussing sea levels! Try not to use the denialist trick of putting wrong words in others mouths!
GSwift7
2 / 5 (4) Feb 01, 2011
Okay, so what kind of progress are you talking about and how is it related to sea level rise slowing down? What progress has been made?

Try not to use the alarmists trick of putting wrong words in others mouths!


I would never do that.

Skeptic_Heretic - Nov 3, 2010

I like to wear polka-dotted tights under my business clothes.
MikeyK
1 / 5 (2) Feb 01, 2011
The progress was you agreeing there was a rise in sea levels, you'r other statements were about a fall.
Where did you get the bulldust about me thinking that CO2 output was declining? What was the even greater bulldust about ENSO being driven by CO2?
I said that ENSO was primarily driven by the PDO, during a -ive PDO (which we are in at the moment) then La Nina's tend to be more prominent and frequent, during a +ive PDO the El Nino's become more prominent and frequent. Where did I mention CO2 in this?
What the H is the polka-dotted tights thing about?!!
GSwift7
1 / 5 (2) Feb 02, 2011
What the H is the polka-dotted tights thing about?!!


Humor: a comic, absurd, or incongruous quality causing amusement

I was demonstrating how I would never put words into someone else's mouth. You see, that's not a real quote.

Back to the 'serious stuff':

Now I understand your comment about progress. It sounded like you meant that sea level rise has slowed because we have made climate progress. That's the trouble with a conversation online. Easy to get confused with such a limit on what you can say in just a few minutes and just a few characters. I don't think your demonization of me is fair though. You make me sound like a dolt that can't read a graph. I believe I specifically stated that the graph showed an overall rise in sea level several times. It does also show an actual fall in sea level in the years I mentioned though. ENSO may explain the decreasing trend in sea level rise, but we need to see anther 30 years. My opinion about this hasn't changed.
GSwift7
1 / 5 (2) Feb 02, 2011
I said that ENSO was primarily driven by the PDO


If you're so sure about that, then you need to let all the people studying ENSO know about your great discovery. As far as I know, the causes of ENSO are still under investigation. I don't think we have enough evidence to say that PDO is driven by ENSO, or the other way around, or maybe they are both driven by some other thing like the sun. Correlation doesn't equal cause or effect. Same with the rate of sea level change and the ENSO cycle. The correlation isn't 100%, so it may not be as simple as you suggest. That's why they are still studying it, I think.
Skeptic_Heretic
3 / 5 (2) Feb 02, 2011

tide gauges have given a statistically significant approximation when used over large areas to account for event variability


Wait, let's go back to that first thing you said...

Are you really suggesting that if you take 1000 bad measurements and combine them into an average, that it eliminates the errors? That's not logical or correct.

If I have a ruler that is 14 inches long, I can use a better ruler (12 inches long) to deternmine the standard of deviation and determine the accurate measurement retroactively.

I thought you were an engineer, you should be very familiar with this practice and it's application.
I would never do that.
Skeptic_Heretic - Nov 3, 2010

I like to wear polka-dotted tights under my business clothes.
Who told you?!?!?!
GSwift7
1 / 5 (2) Feb 02, 2011
If I have a ruler that is 14 inches long, I can use a better ruler (12 inches long) to deternmine the standard of deviation and determine the accurate measurement retroactively


If you start out with 50 rulers and you have no idea how long they are. Then over time they are replaced with other rulers and you have no idea how long those are. Then you add another 500 rulers and you have a little bit of an idea about how long those are. Then you finally get a good ruler. You still don't know if the rulers you started with were too long or too short, then you also don't know if the rulers that replaced them are too long or too short, or if there was a trend in the size of the rulers increasing or decreasing over time. And the variations we are talking about here are on an order of magnitude so small that you wouldn't be able to see the difference in the size of any of those rulers with your eyes at arms length.
GSwift7
1 / 5 (2) Feb 02, 2011
continued:

to use the standard deviation method you have to make an assumption that the distribution of error over time has a flat slope. If there's an unknown slope in the distribution of error over time, then your results will be off. Right?
GSwift7
1 / 5 (1) Feb 03, 2011
Of course we are both ignoring the elephant standing between us: The relative size of the oceans versus the continents isn't constant. Any idea on whether the oceans are getting bigger or smaller? I've never read anything about it, but it can't be constant.