the Air Vent

Because the world needs another opinion

Ice Sheet Loss Cut In Half

Posted by Jeff Id on August 27, 2010

Ice Sheet Loss Cut In Half

Submitted by Doug L. Hoffman on Thu, 08/26/2010 – 13:22

Much concern has been raised by climate scientists regarding ice loss from the world’s two remaining continental ice sheets. Rapid loss of ice-mass from the glaciers of Greenland and Antarctica are cited as proof positive of global warming’s onslaught. The latest measurements involve the use of satellite gravimetry, estimating the mass of terrain beneath by detecting slight changes in gravity as a satellite passes overhead. But gravity measurements of ice-mass loss are complicated by glacial isostatic adjustments—compensation for the rise or fall of the underlying crustal material. A new article in Nature Geoscience describes an innovative approach employed to derive ice-mass changes from GRACE data. The report suggests significantly smaller overall ice-mass losses than previous estimates.

The storage of water or ice on land—the presence of large bodies of water or glacial ice sheets—affect the Earth’s gravitational field. This effect is detected by the NASA Gravity Recovery and Climate Experiment (GRACE) satellites. Twin satellites were launched in March 2002, to make detailed measurements of Earth’s gravity field. Since then, GRACE has been used to study tectonic features, estimate ground water volumes and calculate the amount of ice contained in the Greenland and Antarctica ice sheets. However, other factors can contribute to the GRACE measurements than just the volume of ice in an ice sheet. These factors include the response of Earth’s crust (the lithosphere) to past changes in ice load.

Antarctica was found to be rising, but not as fast as previously thought.

As the weight of covering ice varies, the underlying surface rock can be pushed down or rise up, buoyed by the magma that the crust floats on. This would obviously impact efforts to measure the height of terrain, including glaciers. Compensating for the rise and fall of bedrock is termed glacial isostatic adjustment, and it can have a significant impact on estimated ice-mass losses. Changes in the spatial distribution of the atmospheric and oceanic masses can also enter into the picture. Correctly assessing these different factors is the key to accurately calculating ice-sheet mass balance. Xiaoping Wu and colleagues have proposed a new method for untangling these factors from GRACE measurements. In a News and Views commentary on the work by Wu et al., David H. Bromwich and Julien P. Nicolas sum up the problem:

The atmospheric and oceanic contributions are commonly derived from global reanalyses or other global climate models that assimilate observations. However, the contribution from glacial isostatic adjustment is more difficult to evaluate because the Earth’s mantle is viscoelastic and therefore responds to changes in surface loading with a long delay. Indeed, the variations of the mass and extent of the ice sheets since the Last Glacial Maximum, about 20,000 years ago, continue to affect present-day changes in bedrock elevation. Assessments of the glacial isostatic adjustment typically rely on deglaciation models—which simulate the evolution of the ice sheets since the Last Glacial Maximum—together with assumptions about the viscosity profile of the mantle. Much is still unknown regarding the history of the ice sheets, and even less is known about the behaviour of the mantle in response to loading and unloading.

The method used by Wu et al., in “Simultaneous estimation of global present-day water transport and glacial isostatic adjustment,” estimates ice-mass changes and glacial isostatic adjustment simultaneously, instead of estimating the latter separately from deglaciation models as had been done before. The problem is expressed in terms of a single matrix equation, with the observed surface-height changes decomposed into their different contributions. The equation is then solved for ice-mass changes using matrix inversion. While the glacial isostatic adjustment that results is not directly generated by deglaciation models, the inversion method still requires a first-guess estimate to begin the calculations.

Ice loss in Greenland has been significantly overestimated.

In describing their work, Wu et al. state: “Here we combine gravity measurements and geodetic data of surface movement with a data-assimilating model of ocean bottom pressure to simultaneously estimate present-day water transport and glacial isostatic adjustment. We determine their separate contributions to movements in the geocentre, which occur in response to changes in the Earth’s mass distribution, with uncertainties below 0.1 mm yr−1.” They further describe their methodology as follows:

Read the rest here

22 Responses to “Ice Sheet Loss Cut In Half”

  1. Jeff Id said

    If someone wouldn’t mind, I’d like a copy of the whole paper associated with this post. It seemed interesting because I’ve often wondered about the effects of floating rock on magma over a continent sized mass. It’s hard to believe that the relatively low density ice would create such a major revision by its weight. Actually, beyond the title I don’t even know if it’s a revision, I just am interested in the concept.

  2. Brian H said

    Loosing ice mass might refer to dropping ‘bergs into the sea — but the general process is losing ice mass. 😉

    Bottom line, I guess it’ll now take 6,000 yrs for the ice to melt, instead of just 3,000!

  3. Brian H said

    This page has some/most of the article duplicated. This page has some/most of the article duplicated. Most of the material, including the images, appears twice. Most of the material, including the images, appears twice.

    But the last half of the article is gone. But the last ha

  4. DeWitt Payne said

    Ice may be low density, but there’s a lot of it in the Antarctic ice cap, ~30 million cubic kilometers. A cubic meter weighs about 1 ton. The maximum thickness is reported to be about 4.8 km or nearly 3 miles. The pressure exerted by a column that thick is nearly 500 atmospheres. It’s little wonder that parts of Western Antarctica are now 2,500 feet below sea level.

  5. Eric Steig said

    If the ice sheets are losing mass less quickly then we thought, then steric sea level rise (due to thermal expansion) is greater than we thought. Hence the ocean is warming more than we thought.

    Just thought I’d point that out.

  6. DeWitt Payne said

    Re: Eric Steig (Aug 28 00:14),

    steric sea level rise (due to thermal expansion) is greater than we thought.

    Unless the GIA correction for sea level is wrong too.

  7. Gigih said

    visit also my blog….

  8. JamesG said

    Lucky the sea is cooling then..

  9. sod said

    Lucky the sea is cooling then..

    you are badly informed. are you taking yourinformations from this blog?

  10. Jeff Id said

    I don’t have any way to understand this paper without the rest of the text. The article caught my attention purely because of the data from the underlying rock. Conceptually if ice were to melt the rock would lift higher leaving the ice at a higher than actual altitude. The rock would have to sink more for the ice to loose LESS mass. So I wonder the mechanism.

  11. Geoff Sherrington said

    The GRACE system senses all of the Earth (ans places beyond). It senses the closest parts more strongly, that being the nature of inverse square physics. Has anyone yet found what footprint of the GRACE system is used for calculations? Clearly, it has diffuse boundaries and one has to apply a cut-off, making the measurements subjective. Given the heights of the satellites, it would not surprise me to find the working part of the gravity signal to be tens of thousands of sq km in area. If so, this would mean that maps like those shown are composites from multiple passes, with changes happening between passes that require more subjective assumptions, like where the Sun was and where the moon was.

    As we say here, it’s a bit like counting sheep by counting the legs you see and dividing by 4 or so.

  12. stan said

    Given: 1) all the degraded sat. sensors we keep learning about, 2) the horrific quality control we get periodic peeks at, and 3) the general lack of humility and overbearing hubris which seems to dominate so much of science, it would not surprise me at all to find that the system’s adherents have presented their findings with far, far more certainty than is warranted.

    I get this same sense from observing economists slice, dice, spindle and mutilate the latest numbers to the nth degree despite the ‘knowledge’ that the numbers they use are, at best, very rough guesstimates. What is it about ‘smart’ people that they get so deep into the minutiae of their theories and algorithms that they lose sight of that most basic reality — the numbers are only guesses?

  13. Steve Fitzpatrick said

    That was my first reaction as well. If the paper is correct about overstatements of mass loss, then ocean heat accumulation would have to be higher to balance the sea level budget. Either ARGO is all wet, this paper is wrong, “best estimates” of deep ocean heat accumulation are low, or maybe some combination of all three. My inclination is to be a bit skeptical of this paper’s result, since 1) it is reasonable to expect high latitude warming to accelerate glacial melting and so contribute more to sea level rise, and 2) there is an awful lot of AGRO data which indicates little heat accumulation in the upper 750 meters of ocean since 2004 or 2005. Not to mention that the ARGO group is probably more than a little paranoid about data quality due to the early data problems identified by Josh Willis.

    Accumulation of heat in the ocean is key to verifying Earth’s current radiative imbalance, and to estimate climate sensitivity, so this paper will without doubt be used by shrill voices (on both sides). It seems to me that a great deal of caution in interpreting these results is called for.

  14. DeWitt Payne said

    The calculation in the paper is another example of an ill-posed problem. There are an infinite number of solutions. That doesn’t mean that you can’t get a good solution, just that you have to be careful and you do need a best guess to start the process. Converting x-ray diffraction measurements into a crystal structure is another example of an ill-posed problem that is solved routinely.

    How GRACE works is interesting. There are two satellites in the same orbit at an altitude of 500 km. The distance between the satellites is continuously monitored by K band radar using phase difference rather than pulse transit time. Differences in local gravity will cause the distance to change. Because the satellites are close together, the resolution of the measurement along the line of the orbit is much finer than was obtainable from the orbit analysis of multiple individual satellites.

  15. Steve Fitzpatrick said


    Yup, ill posed problems turn up most everywhere. Too bad the solutions to these kinds or problems are too often presented as “the solution” instead of “a solution consistent with the data”, and with a suitable level of uncertainty noted.

  16. Thanks for highlighting this paper. I like it.

    Isn’t it selective reading when you conclude “Ice Sheet Loss Cut In Half”. It is true that their estimate much lower than Velicogna and Wahr’s, but there are other estimates. If you compare to the full set of recent grace estimates then their new estimate is on the low end, but hardly outside the full envelope of previous results. For Greenland see e.g. figure 2.13 (p.30) in the SWIPA/GRIS report. At present, I think we must conclude that there is considerable uncertainty in the GRACE based estimates. I am looking forward to the next GRACE papers to see how the new ideas change what other groups estimate.

    It is good news that the ice sheets might be loosing mass less rapidly. Yet, I think that Eric’s comment is very appropriate. Observed sea level rise has several contributors, and if the ice loss is contributing less then obviously something else must contribute with a large share in order to close the budget. The most obvious candidate would be deep ocean heating (very sparsely observed). On a side note, if ocean heating does play a greater role then it implies that the TOA radiative imbalance must be greater. This is because it takes much less energy to raise sea level by melting ice than by thermal expansion. (See Trenberth 2009).

  17. Steve Fitzpatrick said


    For sure, any reduction in estimated mass loss from Greenland implies greater heat accumulation in the ocean. But there is great uncertainty in both the deep ocean and glacial mass losses. What is clear is that temperatures around Greenland have risen significantly over the past 50 years, and especially over the last 20 years, so it seems reasonable for Greenland’s contribution to ocean level rise to be gradually increasing.

    There is no obvious trend in the rate of sea level rise over the last 17 years (, while rising global temperatures (especially at high latitudes) ought to be gradually increasing glacial melt contributions to sea level increase. Measuring the ocean heat content more exactly (specifically, more data below the depth of ARGO) seems to me the only reliable way to determine the true TOA radiative imbalance. Inferred estimates of ocean heat content from Grace and sea level measurements seem fraught with uncertainties.

  18. DeWitt Payne said

    Re: Steve Fitzpatrick (Aug 30 09:39),

    But if the heat is going into the deep ocean, then the time constant for equilibration gets very high. It won’t be 30 years but 300 or even higher. It’s unlikely the fossil fuel era will last that long. A well-mixed 2,000 m column of water heats at a rate of 0.0053 K/year at a radiative imbalance of 1 W/m2. Obviously it isn’t well mixed, but if it’s getting down to 2,000m it’s probably going even deeper. The coefficient of expansion of sea water drops with temperature, but even at 4 C, the approximate temperature of the deep ocean water, it’s still about 1/3 the level at 20 C.

    Another thing, if it is going deep into the ocean, then the ocean part of the Air-Ocean Coupled General Circulation Models is wrong, which would bias the short term temperature prediction high.

  19. Brian H said

    Doesn’t that “requirement” for heat accumulation assume the level of net energy influx is known? What if the wattage per sq. meter figures are biased or incorrect, as some papers by Spenser et al have suggested? Isn’t it “begging the question” to assume there is missing heat?

  20. Steve Fitzpatrick said


    Please don’t get me wrong… I am reasonably certain that there is only a minor quantity of heat going into the deep ocean. It is more than a bit of a stretch to suggest that vast quantities (>0.6 watt per sq meter, a la the endless Trenberth arm waves) are being deposited in the deep ocean. The more likely value is 0.1 watt or less.

    But the only way to terminate speculation is to measure the heat content of whole frickin’ ocean. Once that is done, if the heat is not found, then there there will be no alternative to low climate sensitivity.

  21. TomRude said

    #5 Eric Steig, very UNscientific deduction…

  22. mirundap said

    To post above get your info right

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: