Through the Ice, Darkly

Guest Post by Willis Eschenbach

As always, I get distracted by the daily news. The weather news today is a lovely rainy morning here in drought-plagued California, we got just under an inch (2cm) in last night’s storm, and the outer world is green and happy. Regarding the climate news, Anthony highlighted a claimed recent darkening of the Greenland ice cap.  This is said to be reducing the ice cap’s albedo, which is the percentage of sunshine reflected back to space, and thus leading to more solar absorption and more melting.

Being an inherently suspectful type oif fellow, I thought I’d take a look at the albedo and other anatomical features of Greenland. First, the big view. Let me start with a map showing the global “all-sky” albedo from the CERES satellite data. It shows the average of all satellite observations, both when the sky is clear and when it is cloudy.

greenland all sky average albedoFigure 1. All-sky average albedo, CERES top-of-atmosphere data Mar 2000 – Feb 2015. 

Overall, the combination of the clouds and the surface reflect just under a third of all the sunlight that hits the planet. In general the albedo is smallest in the tropics and increases towards both poles. In Figure 1, you can see the inter-tropical convergence zone just above the equator. You can also see Greenland, bright red up near the north pole, with an average albedo of about 65%

The CERES data provides us with another view of the albedo, which is just the measurements taken when the sky is clear. Figure 2 shows that clear-sky albedo, the solar reflection from the surface when there are no clouds..

greenland clear sky average albedoFigure 2. Clear-sky average albedo, CERES top-of-atmosphere data Mar 2000 – Feb 2015. 

As you can see, without the clouds there is much less sunlight reflected from the surface. For example, the ocean reflects less than 10% of the incident sunlight … but even without clouds, Greenland still has an albedo of about 65% because like Antarctica, it has a permanent ice cap. It is the darkening of this Greenland ice cap that I set out to investigate.

Now, there’s a problem with measuring albedo near the poles. Albedo is a ratio. It is a fraction with reflected solar energy on the top and the incoming sunshine on the bottom. Most of Greenland is above the Arctic Circle. So when the sun gets to very near zero in the winter, the albedo gets very uncertain and averages get distorted. As a result, I look instead at the total amount of sunshine that is reflected from Greenland. The incoming sun is constant on an annual basis, so any change in the albedo will be reflected as a change in the total amount of sunshine reflected.

Figure 3 below shows the month-by-month changes in the all-sky reflections from Greenland. I masked out the ocean, so Figure 3 represents solar reflections of just the area of the island itself.

greenland toa reflected solar all skyFigure 3. All-sky average reflected solar energy, CERES top-of-atmosphere data. Units are watts per metre squared (W/m2). Mean value is 119.8 W/m2. Top panel shows raw data. Middle panel shows the average seasonal variation in the data. Bottom panel shows the residuals, which are the raw data minus the seasonal component. Standard deviation of the residuals is indicated by the horizontal gold dashed lines.

The average amount of energy reflected by the clouds plus the surface is about 120 W/m2. There is no trend visible over the period, and the standard deviation of the residuals (bottom panel) is only about ± 2.5 W/m2.

“Ah”, I hear you thinking, “but that includes the clouds”. Indeed it does, it is not the surface albedo from the ice cap. I like to look at what is happening overall before I look at the specifics. Having seen that there is no overall albedo trend in Greenland, Figure 4 shows the Greenaland surface reflections when the sky is clear.

greenland toa reflected solar clear skyFigure 4. Clear-sky average reflected solar energy, CERES top-of-atmosphere data. Units are watts per metre squared (W/m2). Mean value is 115.9 W/m2. Top panel shows raw data. Middle panel shows the average seasonal variation in the data. Bottom panel shows the residuals, which are the raw data minus the seasonal component. Standard deviation of the residuals is indicated by the horizontal gold dashed lines.

I note first that the surface average reflection is about 116 W/m2, only slightly smaller than the 120 W/m2 we saw in the all-sky data in Figure 3. This shows that the albedo of the surface and the albedo of the clouds are quite similar, with the clouds reflecting slightly more than the ice cap

And just like with the all-sky data, there is no trend in the surface data either. There is no indication at all of the claimed darkening of the surface.

Finally, I was interested in what to me was the most curious feature of Figure 4. This is the large dip in surface reflection in the summer of 2012 that reaches a minimum in July. I seemed to remember some oddity that year, and a bit of searching found this from the National Snow and Ice Data Center:

An intense Greenland melt season: 2012 in review

February 5, 2013

Greenland’s surface melting in 2012 was intense, far in excess of any earlier year in the satellite record since 1979. In July 2012, a very unusual weather event occurred. For a few days, 97% of the entire ice sheet indicated surface melting. 

Now, we know from Figure 2 that water has a much lower albedo than ice. So we can see that meltwater on the icecap reduced the reflection of sunlight, and led to the 2012 summer drop in reflected solar energy shown in the CERES data.

The appearance of this July 2012 event in the CERES data supports the validity of the data, and also shows that the data should be more than precise enough to show any trend in the solar reflection over the fifteen-year period of the record … and despite that, there is no such trend visible.

Go figure … I don’t know why the original researchers are claiming a darkening of Greenland, but I’m unable to find it in the CERES data.

w.

My Usual Request: If you disagree with me or anyone, please quote the exact words you disagree with. I can defend my own words. I cannot defend someone else’s interpretation of some unidentified words of mine.

My Other Request: If you think that e.g. I’m using the wrong method on the wrong dataset, please educate me and others by demonstrating the proper use of the right method on the right dataset. Simply claiming I’m wrong doesn’t advance the discussion.

128 thoughts on “Through the Ice, Darkly

    • Why is this here? Did you mean to post it elsewhere, say for example on the thread discussing the subject of your link?

      Or is this just drive-by nonsense? The world wonders …

      w.

      • “The world wonders …”
        Halsey was set off in the wrong direction by such nonsense… 8-)

    • So, oldnwise – any comments on your link? Since you posted it twice, you must agree with the source. And this relates to albedo ….. how?

    • OH my, that link is so frightening: “Global warming is going into overdrive.” Isn’t it amazing how much more powerful mother nature is with her El Nino that can do that? I hope you noticed that, 1oldnwise4me@reagan.com.

  1. Willis,
    “Go figure … I don’t know why the original researchers are claiming a darkening of Greenland, but I’m unable to find it in the CERES data.”

    They looked at the period 1981-2012. It looks like a downtrend in your plot up to 2012, and there’s almost twice as long you haven’t looked at.

    “So we can see that meltwater on the icecap reduced the reflection of sunlight, and led to the 2012 summer drop in reflected solar energy shown in the CERES data.”
    Yes. Wet ice is part of their feedback mechanism.

    • Yes, there is a clear drop from 2000-2003 in Willis’ graphs. About 4 W/m^2 by eye.

      Pretty flat since but may well have also changed leading up to y2k. However, not much evidence of Greenland getting “locked into a feedback loop” as suggested by the paper.

      That is the old “tipping point” argument again. Recovery after the sharp dip in 2012 is not in any way consistent with that claim.

      Good work Willis.

      • Actually this is just standard Climate Science® practice – start at an outlier year in one direction and end at an outlier year in the other direction and you can get a nice trend of your choice.

        They were lucky they had 2012 available though. Large melt events only occur about once a century (the previous one was in 1888). During the previous (warmer) interglacial they came about once per decade.

    • Willis,

      “Go figure … I don’t know why the original researchers are claiming a darkening of Greenland, but I’m unable to find it in the CERES data.”

      They looked at the period 1981-2012. It looks like a downtrend in your plot up to 2012, and there’s almost twice as long you haven’t looked at.

      Sorry, Nick, but their data claims there is a decrease in the albedo from 1996 onwards which accelerates after 2006. See their Figure 1 here.

      Instead of a steady decrease followed by an accelerating drop after 2006, the CERES data shows little change since 2000. Other than a couple of upticks at the start and a downtick at the end, it’s basically flat.

      So no, it’s NOT because I looked at a shorter period. Your unending desire to discredit me has overcome your caution, you are just making stuff up.

      w.

      • I’d give Nick the benefit of the doubt, Willis. He’s knowledgeable & unflappable. That said, he does appear to have an agenda at times…

      • Willis,
        “you are just making stuff up”
        Not made up. The time intervals are different. But below I’ve aligned the albedo plot from their Fig 1 with your Fig 4, scaled. Their grey is the MAR result, black is GLASS. The main features are similar, especially for MAR. Theirs has less detail, being annual. But there is the steep dip in 2012, peak about 2006 and high values around 2000. As for trend, you probably should add an actual 2000-2012 trend line.

      • Thanks Nick, I think you’ve put your finger on it. I was curious about their data stopping in 2012 in a 2016 paper. See how easy it is to fool the eye of the reader.

        They generate the impression of runaway melting : a tipping point by stopping the data at the bottom of the 2012 spike. Their paper fails to see the rebound the following year.

        It’s like plotting Arctic sea ice area and stopping in the OMG 2012 minimum and not reporting the massive increase in ice volume that followed.

        Anyway, don’t forget : “More than half of the Greenland ice sheet melted last summer”
        http://www.theguardian.com/environment/2016/mar/03/greenland-ice-sheet-melting-global-warming-feedback-loop

      • Their paper was “Received: 6 September 2015“. I find it very suspicious indeed that the paper’s study period ended in mid-2012, at exactly the moment that albedo hit a minimum before rapidly and completely rebounding. Given the date of the paper, how is this not a blatant and fraudulent cherry-pick?

      • Nick Stokes March 3, 2016 at 5:17 pm

        Willis,

        “you are just making stuff up”

        Not made up. The time intervals are different. But below I’ve aligned the albedo plot from their Fig 1 with your Fig 4, scaled.

        Nick, this is why I rarely engage with you, because it is so frustrating. Your previous claim was that I was wrong because:

        They looked at the period 1981-2012. It looks like a downtrend in your plot up to 2012, and there’s almost twice as long you haven’t looked at.

        I pointed out that that wasn’t the problem at all. The problem is that my data shows no decrease post 2001, and their data claims a decreasing trend up until about 2006, a decrease which accelerates from 2006 up to 2012.

        You come back, and you don’t say a word about all that previous BS about the problem being that I hadn’t looked at a long enough record. That was not the problem, you realize that now after I pointed it out … and despite that, you refuse to acknowledge that you were wrong. Instead, you’ve blithely and adroitly moved on to another steaming pile of misdirection to claim that I’m still wrong, so nobody will notice your previous nonsense.

        So I’m going to leave the field to you, Nick. I have no time for a man who is always right even when he’s wrong. You are no fun to discuss science with, because you never, ever admit your errors. I’m not in this for the money or the fame. It’s a tragedy, because you’re a smart guy, but you are unbearable. I’m in it for the joy of intellectual discussion … and instead of joy, you provide pain. I’ve tried, because as I’ve said you are a smart guy. And I’ve been disappointed, and I’ve said the same thing to you that I say in this comment, and I’ve tried again … nothing.

        You are aware, I know, that your nickname on the intarwebs is “Racehorse” Stokes, in honor of “Racehorse” Haynes, who was famous for never admitting he was wrong, and who famously said:

        Say you sue me because you say my dog bit you. Well, now this is my defense:

        My dog doesn’t bite.
        And second, in the alternative, my dog was tied up that night.
        And third, I don’t believe you really got bit.
        And fourth, I don’t have a dog.

        That’s your reputation, Nick, that’s the bed you’ve made for yourself. And while it’s funny when you read about Haynes doing it … when you pull that stuff on me in a scientific discussion, it is a massive PITA.

        So I’m giving it up. Fool me now, my fault.

        Pass …

        w.

      • Willis,
        “the problem being that I hadn’t looked at a long enough record”
        I simply pointed out that the record they analysed was quite different to yours. That is true, and was a relevant observation. You didn’t mention anywhere in your post what period they were analysing. It matters.

      • Exactly my point, Nick. You are clearly 100% correct that you were totally right in your claim that your logic was perfectly flawless when you said that all of your previous comments were without the slightest flaw or blemish … and besides, you don’t have a dog.

        Pass …

        w.

      • Willis/Nick,

        Willis appears to be plotting the full year data but the plot being discussed is only J(une), J(uly), A(ugust).

        Isn’t this apples vs oranges?

      • John,
        “Isn’t this apples vs oranges”
        Willis is plotting total solar reflection. As his plots show, that mostly happens in summertime, so summer albedo is about right.

        OTOH, it shows a reason why his plot looks different to the paper, and with flattish trend. Willis is plottig monthly, and the winter months are bound to be at zero. So you see a succession of spikes, with these flat bits in between. The plot from the paper is annual.

      • I would be interested to get an idea of the error bars on the anomoly data. It seems a small difference compared to the absolute magnitude.

      • “Abstract
        The surface energy balance and meltwater production of the Greenland ice sheet
        (GrIS) are modulated by snow and ice albedo through the amount of absorbed solar
        radiation. Here we show, using spaceborne multispectral data collected during the
        three decades from 1981 to 2012, that summertime surface albedo over the GrIS
        decreased at a statistically significant (99 %) rate of 0.02 decade−1
        between 1996 and 2012. The negative trend is confined to the regions of the GrIS that undergo
        melting in summer with the dry-snow zone showing no trend. The period 1981–1996
        showed no statistically significant trend. The analysis of the outputs of a regional
        climate model indicates that the drivers of the observed albedo decrease is imputable
        to a combination of increased near-surface temperatures, which enhanced melt and
        promoted growth in snow grain size and the expansion of bare ice areas, as well
        as by trends in light-absorbing impurities on the snow and ice surfaces. Neither
        aerosol models nor in situ observations indicate increasing trends in impurities in the atmosphere over Greenland, suggesting that their apparent increase in snow and ice
        might be related to the exposure of a “dark band” of dirty ice and to the consolidation
        of impurities at the surface with melt. Albedo projections through the end of the
        century under different warming scenarios consistently point to continued darkening,
        with albedo anomalies in 2100 averaged over the whole ice sheet lower than in 2000
        by 0.08, driven solely by a warming climate. Future darkening is likely underestimated
        because of known underestimates in projected melting and because the model albedo
        scheme does not currently include light-absorbing impurities and the effect of biological
        activity, which themselves have a positive feedback, leading to increased melting, grain
        growth and darkening.”
        Let me repeat:
        “Albedo projections through the end of the
        century under different warming scenarios consistently point to continued darkening,
        with albedo anomalies in 2100 averaged over the whole ice sheet lower than in 2000
        by 0.08, driven solely by a warming climate. ”
        This is clearly cherrypicking, making trends and projections based on —what?
        So: M. Tedesco, S. Doherty, X. Fettweis, P. Alexander, J. Jeyaratnam, E. Noble and J. Stroeve:
        Is this science or is it mission impossible?

      • Mike Jonas March 3, 2016 at 6:36 pm
        Their paper was “Received: 6 September 2015“. I find it very suspicious indeed that the paper’s study period ended in mid-2012, at exactly the moment that albedo hit a minimum before rapidly and completely rebounding. Given the date of the paper, how is this not a blatant and fraudulent cherry-pick?

        As far as I can tell the GLASS data ends in 2012 so it would not be possible to continue further.

      • nick stokes, being one of the most respectful and knowledgeable people posting on climate blogs today i cannot believe what you just posted in reply to willis. my gob is smacked.

    • I am under the impression that ” albedo ” is the solar energy reflectance of planet earth, that is the percentage of total incident solar energy reflected back out into space still as solar spectrum energy.

      So I don’t see how Antarctica can be reflecting 72% of 342 W.m^2 back into space. I don’t think it even receives that much energy from the sun.

      If they are confusing albedo with reflection coefficient, they should say so.

      There is a very basic reason why there is all of that ice there in Antarctica with its high reflection coefficient (which might be 72% for fresh snow).

      That reason is that very little solar energy is incident there. It would be even colder there in Antarctica and also in the Arctic, if it wasn’t for the astronomical amounts of heat energy that is convected from the tropics to the polar regions, by warm ocean currents, and atmospheric currents.

      Their phony color map of painting the polar regions red, gives the impression that lots of energy is being reflected from the poles. It isn’t.

      G

      Albedo should be (and is in my view) a SINGLE number for the entire planet; it isn’t regional. Reflection coefficient can be regional.

      • george e. smith March 3, 2016 at 4:33 pm

        Their phony color map of painting the polar regions red, gives the impression that lots of energy is being reflected from the poles. It isn’t.

        “Their phony color”??? That’s my graph. In addition, I gave you a graph of exactly how much energy is being reflected by the red area of Greenland. It is about 120 W/m2, which is indeed a significant amount of energy.

        Albedo should be (and is in my view) a SINGLE number for the entire planet; it isn’t regional. Reflection coefficient can be regional.

        Perhaps it should be, but it isn’t a single number. Here’s a typical list:

        Surface, Albedo
        beets, 0.18
        boreal forest with snow, 0.12 – 0.30
        coniferous forest, 0.05 – 0.15
        deciduous forest, 0.10 – 0.20
        dry dark soil, 0.13
        dry sand, 0.35
        fresh snow, 0.75 – 0.95
        grass, 0.17 – 0.28
        maize, 0.18 – 0.22
        old snow, 0.40 – 0.70
        potato, 0.19
        rain forest, 0.12
        savanna, 0.16 – 0.21
        steppe, 0.2
        sub-arctic, 0.09 – 0.20
        wet dark soil, 0.08
        wheat, 0.16 – 0.26

        The use of “albedo” is by no means restricted to the entire planet. It is widely used in the earth sciences to mean reflected/incident sunlight on particular types of surfaces, as in the example above.

        w.

      • “Perhaps it should be, but it isn’t a single number. Here’s a typical list:”

        1. Water seems to be missing from the list. But between oceans and lakes isn’t water by far the dominant surface material on Earth?

        2. What causes the dramatic peaks in the residuals in the Summer(?) of 2000, 2001, etc? It’s probably obvious if one is smart. But I’m not.

        3. Surely sun elevation angle is a major variable, especially with water, but probably with ice, snow and even other materials. You do address the case of angles near zero (sort of), but it’s unclear to me exactly how sun elevation gets handled.

        4. I’m a bit hazy on the geometry of this thing. Are you/we/they plotting the location of the satellite or the location of the reflection point? I think the latter is directly under the satellites only near the equator? And it’s location varies with the elevation of the reflection? Is the difference between the positions significant? It would seem to me that when the sun is low in the sky, it might be quite a ways, but I could be visualizing entirely the wrong thing.

        I see that Clyde Spencer has addressed some of my concerns below. Regretably, he seems to have even more questions than I do.

        And as always, thanks for taking all this on.

      • Willis
        I kinda hate the term ‘albedo’. Reason is that it doesn’t actually say much, or perhaps it is so general that to me it becomes meaningless or in fact aggravating. I am currently doing some private research involving integrating spheres.
        My research to date has shown that there is a lot of confusion regarding terms ie. albedo, reflectance, reflectivity. All of these are actually different in the specifics. Reflectivity is like a mirror. Albedo takes a look at the entire spectrum (it should) and determines the light bouncing off. Reflectance is about the wavelengths that are absorbed/reflected.
        What am I really saying?
        Be careful about what you are looking at.
        Never forget the cosine rule.
        Take atmospheric thickness into account.
        The above three things are involved in the ‘product’ you use for calculations. Algorithms all the way down.
        Accuracy of algorithms is up in the air as far as I’m concerned.
        They have a certain usefulness.
        As NASA says useful for remote sensing but not accurate for qualitative work.

        Keep up your good work

      • “Water seems to be missing from the list. But between oceans and lakes isn’t water by far the dominant surface material on Earth?”

        Yes, but it is difficult to put a single number to water albedo. The amount of suspended particles (largely plankton, but also sediment) affects the albedo, and at low sun-angels specular reflection is also a significant factor. However the albedo of oceans is low – on the order of 0.1 – so the vast majority of heat is absorbed by the oceans.

      • tty: “and at low sun-angels specular reflection is also a significant factor. However the albedo of oceans is low – on the order of 0.1 – so the vast majority of heat is absorbed by the oceans.”

        Near the equator, that’s probably true. But we’re dealing with far Northern latitudes here where the sun elevation angle is always pretty low assuming that the sun is even visible. The lower the elevation angle, the higher the reflection. You can test that by walking toward any nearby body of water whose surface isn’t rippled. A rainwater pond will do just fine. From a distance what you see is a reflection of whatever is beyond the water. You have to get pretty close before the bottom under the water becomes visible.

    • You mean a downtrend between 2000 and 2002? Or are you looking at the minimum in July 2012? Was CERES around in 1981?

    • Nick Stokes: They looked at the period 1981-2012. It looks like a downtrend in your plot up to 2012, and there’s almost twice as long you haven’t looked at.

      The record before 2000 is pretty flat, and the record since 2012 shows a rebound to what might be called nearly “average”. Why submit a paper in Sept 2015 with data truncated at an observed minimum in 2012? We must look forward to their update to learn why, perhaps.

    • An albedo figure for all of Greenland is not going to show a reducing trend. As the paper demonstrates, the albedo of the highlands is static or increasing, and the reductions are only on the margins and in the south. See their figure 2.

      If you want to see the albedo reductions, you will need a dataset for southern Greenland.

      R

  2. The British would say “Spot-on!!!”

    Well done & well researched!!! The 2012 melt has other implications but I lack the time to address them here!

    Nice job Willis!!!

  3. george e. smith March 3, 2016 at 4:33 pm Edit

    I am under the impression that ” albedo ” is the solar energy reflectance of planet earth, that is the percentage of total incident solar energy reflected back out into space still as solar spectrum energy.

    So I don’t see how Antarctica can be reflecting 72% of 342 W.m^2 back into space. I don’t think it even receives that much energy from the sun.

    Thanks, George. Albedo is reflected sunlight divided by incident sunlight … and incident sunlight in Antarctica is far below 342 W/m2. Hang on … OK, mean incident sunlight on Antarctica is only 182 W/m2.

    Remember, individual objects have an albedo … so it cannot be just a planetary number.

    w

  4. Thanks Willis, I have to wonder too. I took that study’s conclusions at face value because I have been seeing so many photographs of surface darkening in Greenland, and so many moulins with leftover soot and ash at the bottom after they drain meltwater, so this seems to only make sense that there would be an Albedo change going on.

    The 2012 Instamelt that created the spike was triggered by forest fire soot along with a weather event, see: https://wattsupwiththat.com/2014/06/06/study-greenlands-july-2012-insta-melt-was-triggered-by-a-combination-of-warm-weather-and-carbon-soot/

    The last time an event like that happened was 1889. According to this researcher:

    https://wattsupwiththat.com/2012/07/24/greenland-ice-melt-every-150-years-is-right-on-time/

    So, we know this can happen and it makes sense…but

    From the study I posted today:

    The study used satellite data to compare summertime changes in Greenland’s albedo from 1981 to 2012. The first decade showed little change, but starting around 1996, the data show that due to darkening, the ice began absorbing about 2 percent more solar radiation per decade.

    And from the paper:

    …surface albedo retrieved under the Global LAnd Surface Satellite (GLASS) project

    I have to wonder, could we be missing something in the CERES data? Is the CERES resolution good enough to detect the 1-2% change they note from the GLASS data?

    http://www.mdpi.com/2072-4292/5/5/2436/pdf

    • Also note that they are only looking at JJA summer data. The change in albedo could be because of surface melt water not black carbon.

      BC would presumably affect the annual data W was looking at, melt water not. So that 1-2% would be 1/2 to 1/4 of that in the annual data.

      2% of 182 W/m^2 would be detectable like the downward slope from 2000-2003 in Willis’ graph; 0.25 – 0.5% probably not.

      To question their result it would be better to do apples to apples comparisons. ie extract JJA from CERES.

      • Good point about the JJA and melt water Greg. Maybe Willis will take your challenge and look at JJA for an apples to apples comparison. The melt water could certainly be a confounding factor.

    • Anthony,
      There are further confounding factors. The things that Willis listed with representative albedos all have diffuse reflectance, albeit often with a strong forward scattering. The proper and complete characterization of them is with a bidirectional reflectance distribution function (BRDF), which I doubt that CERES is capable of measuring. In the laboratory it is measured with a specimen at the center of a sphere or at least a hemisphere. Calm water and smooth ice have specular reflectance where the reflectivity varies with the angle of incidence of the sunlight (as well as the index of refraction, which in ice and water are similar). One often gets the impression that the total reflectance of water is much lower than what it is unless they are opposite the illumination and viewing the surface at the same angle at which it is being illuminated. I doubt that CERES is taking its readings by looking across Greenland into the sun. It would probably blind the sensors because all of the sunlight is concentrated into a narrow cone. All other things being equal, if one is observing a diffuse reflector, and sprinkle something else on it that has a lower diffuse reflectance, the composite will look darker. I suspect that what CERES is providing is relative diffuse reflectance, or a ratio to some material such as snow. It isn’t telling us everything about how much light the object is actually absorbing.

      • Algorithms. You only need a small tweek to make major changes. I don’t have a problem with algorithms, just the people designing them. You never get to see this stuff. Just take it on faith. BS. If I don’t see the algorithm then I dismiss it. Throw it on the table so everyone can see it. There are various ways to do things and some are better than others. I hate a produced ‘product’. Willis accepts it as gospel. Nothing against Willis.

      • Clyde
        Hemispherical reflectance is the main standard. That is qualitative. BRDF is ok for computer simulations of games

      • Clyde. What you are saying makes sense to me. This is far too complex for me to absorb in a few minutes. I’ve never worked in this area. But, I suspect that what CERES is measuring is indirect, diffuse reflection from under the satellite and that probably that is not easily related to albedo because a large, indeterminable(?) amount of the energy being reflected is going elsewhere. If so, CERES might be of great use for a lot of things. But likely not for determining albedo.

      • Alex,
        Those involved in computer graphics use BRDF modeling because it is the only way to get realistic-appearing results. The corollary of that is if you aren’t using BRDF in the real world, you aren’t getting accurate results. What CERES and most satellites are measuring is an apparent reflectance — only what is seen by the satellite overhead. What the satellite doesn’t see is a part of the energy balance also, however. The question is whether it is a significant portion. Most diffuse reflectance, such as with snow, has a strong forward reflectance. If one isn’t measuring that forward lobe, then a lot of the energy is missed. For materials at the poles of Earth, the sun is never overhead! Therefore, there is going to be a substantial amount of reflected light that is not observed. For background on BRDF, see the following link:

        https://en.wikipedia.org/wiki/Bidirectional_reflectance_distribution_function

      • Don,
        See https://en.wikipedia.org/wiki/Albedo . In particular, see the section on Terrestrial Albedo. They confirm that the satellites cannot see the total albedo and a model is used to estimate the total albedo based on BRDF of the assumed material. I have no personal experience with the model used and how effective it is or whether or not they take into account the sun elevation and topography.

      • Clyde:

        Agreed. Just to reinforce your point: BRDF is the standard for the modeling of light scatter from surfaces in optical systems. Albedo often assumes a Lambertian scatterer (reflected radiance is uniform in angle). Snow is a Lambertian reflector for light incident from directly overhead. Whenever one approaches steep angles of incidence, even a rough surface will develop a strong specular (forward) component. This could have a significant influence on the interpretation of the data. We know this in everyday experience, just by looking at a very quiet pond in the early morning. The smoother the surface, the darker it will appear from above, even if a significant amount of light is reflected. (By the way, the reflected light is also polarized, but that may be beside the point of this discussion.)

        Regards,
        Tom Brown

    • Good points Anthony. Having skimmed the above paper, I see much description of the computing systems used to resolve the data from the satallites, but I see no details of the capabilities of the sensor, even basic stuff like response vs wavelength and noise figures. Presumably these are available in the references somewhere.

      It is important, for example the work I am involved with at present involved using remote satelite sensing to validate agricultural useage, such as areas and crop type. But there are some crops/land types that cannot be distinguished. So the utility of satellite data gathered, as with all experimental data, must be carefully looked for each use case.

  5. Willis

    What is significant here is that the albedo is lowest in the area covered by the tropical ocean, just where the impact of energy from the sun is most important. The absorption of solar irradiance into the tropical ocean is is powering/charging the heat pump of the planet, and one reason why the K&T energy budget cartoon is incapable of explaining planet Earth.

    It is also noteworthy that albedo is highest where solar irradiance is at is weakest.

  6. It is clear from the photographs of moulins that BC gets wasted down and out by melting, so melting has a self correcting effect on albedo.

  7. Willis, thank you. Presenting these beautiful albedo graphics, figures 1 and 2. They are new to my eyes, and are worth, as you have done, deep pondering. For me the NH role for Greenland and SH Antarctica ice is massive as one can see from the red (72% Albedo). The recent posting on soot on/in ice is very interesting.

  8. Nick has posted the paper’s fig 1 against Willis’ graph upthread. They are not too dissimilar. It’s just that they cut off their graph at the bottom of the spike.

    2012 was the end of what could have been read as “runaway” melting of Arctic sea ice. Any current analysis in 2016 that stops in 2012 will be very misleading.

    • Its interesting when you flip the NH sea ice graph (not that one , but a similar one) and put it against the AMO.

    • Isn’t it a pity that you cannot easily at the data pre 1980.

      The warmists have the data, but do not want people to see it since it shows the extent of natural variability. Increasing NH ice extent in the 1970s does not fit in well with the manmade global warming mantra and is difficult to explain given the increasing CO2 during that time.

      For example see:

      See how the NH ice extent rose rapidly from 1974. Hence the promotion of the fear of a coming ice age.

      This data is no longer included in recent IPCC Reports. Do you wonder why?

      • Has anyone tried to get the 70s data? FOIA? One for the USA congressional probe? Also, I have never seen this graph extended to today.

  9. Here’s a quote from the previous story:

    The study also raises questions about whether Greenland’s high plateau is darkening as previous reports have suggested. The scientists found no long-term trend of darkening at the top, and they suspect that the Terra MODIS satellite sensor that has detected darkening in the past may actually be degrading, as previous studies have suggested. At lower elevations, the signal is much stronger.

    It seems clear, to me at least, that the albedo change applies only to part of Greenland. In fact, this image makes it look like the high plateau covers most of Greenland.

    In light of the above it seems unlikely that the albedo of Greenland, as a whole, changes much (which is pretty much what Willis found).

    • No such thing as a “high plateau” on Greenland. Large icecaps are flattish domes, not plateaus. Ice can only flow downhill, so a completely flat glacier is dynamically impossible. The Greenland cap consists of two coalescing domes, one southern and one northern. I presume by “high plateau” they mean the higher northern dome.

      • Perhaps they are referring to the rock underlying the glaciers rather than the glaciers themselves.

        Other plateaus are … and the Canadian Shield or Laurentian Plateau, a U-shaped region of ancient rock, the nucleus of North America, stretching north from the Great Lakes to the Arctic Ocean. Covering more than half of Canada, it also includes most of Greenland … link

        On the other hand, here’s a link to a map with contour lines. It shows Greenland as a big pile of ice with a kind of rounded top that falls off sharply on all sides.

        My (not so well) calibrated eyeball tells me that about a third of Greenland would qualify as high, a third would qualify as low, and the rest would be the sloped bit that connects the two.

        Your guess is as good as mine. :-)

      • The bedrock in central Greenland is a Precambrian shield, so it’s fairly flat. However it is definitely not a high plateau, much of it is actually below sea-level.

    • I’m pretty sure Willis’ results are correct. The problem seems to be that more than one viewpoint is represented in the story.

      Yet the darkening of Greenland around its periphery remains a source of concern because it contributes to making the ice sheet melt away faster.

      Willis’ results show that there is no measurable overall change in Greenland’s albedo. At most any albedo change would be local, which is implied by the above quote.

  10. “The average amount of energy reflected by the clouds plus the surface is about 120 W/m2. There is no trend visible over the period, and the standard deviation of the residuals (bottom panel) is only about ± 2.5 W/m2.”

    CO2 collective RF 1750 to 2011 is 2 W/m^2. That’s less than one half the uncertainty in albedo.

  11. Anthony, have you seen this :

    “More than half of the Greenland ice sheet melted last summer”
    http://www.theguardian.com/environment/2016/mar/03/greenland-ice-sheet-melting-global-warming-feedback-loop

    That is so stupid and ignorant that it deserves a fairly loud and public ridiculing.

    I don’t recall reading about the resulting 3 metre high tsunami inundating most of the world last year.

    I notified the editor about it but seem quite happy to carry on misleading their readers with this nonsense.

    • Greg march 3 @ 5;55 pm, thanks. This is the first time I actually read a complete article by the “Guardian” and read some of the comments, your evaluation of “stupid and ignorant” is very polite. I have a few other words that AW probably would not print. The comment are priceless, though I truly didn’t think people were THAT ignorant. Why are most of these people not fleeing to higher ground? I thought the right was always the “conspiracy”, “Trailer park” group, I have changed my mind on that one.( but maybe these people live in one room apartments and share the bathroom at the end of the hall on alternating days).

      • The Guardian is a very leftist rag. Leftists are stupid and/or ignorant, if they weren’t, they wouldn’t be leftists.

        Remember what Clemenceau said “If You aren’t a socialist at twenty you have no heart, if you are still is a socialist at thirty you have no brain”

  12. It is just incredible to me that the original “darkening” paper cut off their analysis at 2012. That is lying by omission.

  13. Terrific post. Dunno how accurate and precise (two different things) CERES is. Dunno how accurate the ‘black snow’ data is. Do know this is an impecably delightful post. Thanks, Willis.

    • Willis: Thank you for the careful, thorough work as usual. Keep it up.
      The point of the original was that “we” are darkening Greenland and the world is subject to “runaway” warming. A slightly different “we” from 1888-1889. Still, soot is soot so the Chinese (:-)) are at it again. Pretty soon them thar commies will be blamed for EVERYTHING. Err…well… modern free market economics will get the blame.

  14. I think you may have indirectly answered why ice age glacials end so abruptly.

    Once insolation is high enough, the first melt causes a dramatic water driven albedo shift to latch up the melting process. Not in Greenland, as we are at too low an insolation level right now, but 10,000 years ago, insolation over New York in summer was enough to have full surface melt… add some spring rains too and the melt would be dramatic. Shifting rapudly to more melt than winter snow.

    That 2012 on steroids.

    Oh, and per the Greenland graph: Their ending in 2012 is scientific malpractice. You caught them at it. Way to go!

    • E.M.Smith March 3, 2016 at 7:23 pm

      I think you may have indirectly answered why ice age glacials end so abruptly.

      Thanks, E. M. Funny, I thought the same thing when I was considering the size of the impact that the meltwater was having on the absorption in 2012. I had an image of the Laurentide Ice Sheet with millions of meltwater ponds on the surface collecting sunshine, and making water that ends up either running off or punching straight down through the sheet …

      w.

      • But there was always a vast amount of melting on the margins of the big icecaps in summer, even at Glacial Maximum. In Europe you have a huge belt of outwash sandplains all along the edge of the old icecap. In the US the Nebraska Sandhills are much the same, though the high winds off the Rockies have blown the sand into dunes.
        A really big volcanic ashfall and a warm summer might do the trick though.

      • Yep. That huge latent heat of liquefaction/solidification is one heck of a powerful tipper truck. I notice the radiative myopia merchants (who seem to forget water rapidly washes dust away or fresh snow rapidly covers it up) are back with their fag packets in hand. Diffusion through ice still has to happen; solid ice is far from a perfect insulator as its fluid behaviour shows. And my guess is it’s a lot more complex than we think, especially in thick ice sheets.

    • Never mind the water driven albedo shift (whatever that is). The greatest change is the concentration of many previous dust layers further down through the ice sheet. This concentration of dust decreases the albedo almost exponentionally, and increases insolation absorption likewise.

    • I wonder occasionally exactly how the continental ice sheets could ever have formed and have extended as far South as NYC (40N). I always come to the same answers. Either a dimmer sun. Or constant cloud cover accompanied by lots of snow. A few weeks ago, I watched the Winter sun at 45N take out 5 cm of fresh snow on a day when the temperature never got above -15C. If the Summer sun at 40N gets down to the surface any significant amount of the time, that glacial ice would seem to be doomed.

      • Yeah but remember that last year the last snow clearings in Boston didn’t go away until July! So obviously survival and accumulation to the next year doesn’t take a huge change.

  15. Great work Willis. Very interesting. Seems to me that changing albedo and complicated ocean effects coupled with small variations in incoming solar radiation over time are the major drivers in climate. These effects apparently have driven the numerous glacial cycles in our current ice age that began about 3 million years ago. Large explosive volcanic events, large meteor impacts, and maybe even galactic dust are wild cards to add into the long-term mix. But CO2 seems to be a minor player and possibly more from how it effects plant growth than how it might possibly directly effect the radiation balance.

    Unfortunately, it may take decades or even centuries to develop a real consensus and understanding of the complicated workings of climate and that may only happen when Earth inevitably begins trending into the next glacial period. I see nothing to indicate that the glacial cycles are going to be broken, although it would be greatly helpful for humans to break that cycle at the high temperature end as we are now. The next glacial period will be devastating for humanity, though certainly not insurmountable, especially considering our ancestors survived them … but with tiny populations compared to today. All this fuss about a small amount of likely net beneficial warming and CO2 increase seems such a waste of resources that could be much better spent. Perhaps we should be thinking more about ways to prevent the next glacial cycle. I don’t think more CO2 will help in that regard unfortunately. That would be too easy.

  16. Some people wanted to examine the Jun-Jul-Aug (JJA) averages. Let me say I always resist doing that because as soon as you do that, you need to adjust your significance level. This is because if you look at every month of the year separately, some will go up more and some up less, and some might go down more or less. This makes it much more likely that we’ll find spurious “trends” which appear significant but are not.

    To adjust for this you use the “Bonferroni correction”, where you divide your initial significance level (typically a p-value of 0.05) by the number of subsets into which you’ve divided the data. Since you’ve gone to monthly subsets, you need to find something significant at the 0.05 / 12 ≈ 0.004 level … and that is hard to find in the climate world.

    In any case, the p-value trend of the JJA averages, far from being below p = 0.004, is 0.06, not significant in the slightest.

    Having said that, here are the JJA averages.

    Like I said … not statistically significant.

    w.

    • I fully agree Willis, I would much rather work with 12 months per year and analyse it correctly. Your method was far better. I was just curious as to whether the JJA approach accounted for the difference. It seems not.

      The main problem with the study is that it stops at the bottom of the spike and gives a very misleading impression.

    • But oblique Sun angles do not show the true albedo, as the soot and dust particles bury themselves in the ice and are not visible at low Sun angles. And so the effective low albedo is only visible at midday in midsummer, and melting will only occur then (slope angle of the ice sheet notwithstanding). I will try to dig out the paper, when I get back.

      R

    • ralfellis March 4, 2016 at 2:17 am

      But oblique Sun angles do not show the true albedo, as the soot and dust particles bury themselves in the ice and are not visible at low Sun angles.

      Thanks, Ralph, as always good to hear from you.I don’t understand this objection. If the measured albedo is not the “true” albedo, then what is being measured? The “false” albedo?

      w.

      • >>False albedo.

        Yeah, I see what you mean. Poorly explained. It highlights a couple of things.

        If the satellite is measuring from above, and the Sun is lower on the horizon, then the albedo measured by the satellite will not be the albedo ‘observed’ by the Sun. The Sun will see a much higher albedo. And vice versa.

        More importantly, the ‘albedo masking’ effect suggests that only a the three months around the summer solstice are important, when the Sun is in the zenith. At any other time of year, the effective albedo will be high no matter how much dust and soot there is on the surface, and no matter what the observed or calculated albedo may say.

        R

      • Regards, Ralph. The measurement of upwelling surface radiation (for the albedo calculations) in the CERES dataset is quite detailed and complicated, specifically to address the issues you raise. It is described here. From that document:

        Monthly regional CERES clear-sky SW TOA fluxes in the CMIP5 archive are from the CERES Energy Balanced and Filled (EBAF) Ed2.6r data product. The approach used to determine clear- sky SW TOA flux is described in detail in Loeb et al. (2009). We determine gridbox mean clear- sky fluxes using an area-weighted average of: (i) CERES/Terra broadband fluxes from completely cloud-free CERES footprints (20-km equivalent diameter at nadir), and (ii) MODIS/Terra-derived ‘‘broadband’’ clear-sky fluxes estimated from the cloud-free portions of partly and mostly cloudy CERES footprints.

        In both cases, clear regions are identified using the CERES cloud algorithm applied to MODIS pixel data (Minnis et al. 2011). Clear-sky fluxes in partly and mostly cloudy CERES footprints are derived using MODIS–CERES narrow-to- broadband regressions to convert MODIS narrowband radiances averaged over the clear portions of footprints to broadband SW radiances. The narrow-to-broadband regressions applied to MODIS are developed independently for each month in order to ensure that the final product’s calibration is tied to CERES. The ‘‘broadband’’ MODIS radiances are then converted to TOA radiative fluxes using CERES clear-sky ADMs (Loeb et al. 2005).

        Monthly mean clear-sky TOA fluxes are determined from instantaneous values using the same approach as clear-sky fluxes in the CERES SSF1deg product. In that product, SW radiative fluxes between CERES observation times are determined from the observed fluxes by using scene-dependent diurnal albedo models, which describe how TOA albedo (and therefore flux) changes with solar zenith angle for each local time, assuming the scene properties remain invariant throughout the day. The sun angle– dependent diurnal albedo models are based upon the CERES ADMs developed for the Tropical Rainfall Measuring Mission (TRMM) satellite (Loeb et al. 2003).

        Best to you,

        w.

      • using scene-dependent diurnal albedo models, which describe how TOA albedo (and therefore flux) changes with solar zenith angle for each local time, assuming the scene properties remain invariant throughout the day.
        _____________________________

        Interesting. So they do try.

        However, are these adjustments simply for solar zenith angle changes on standard snow? Or do they actually allow for embedded and recessed soot deposits? I might suspect the former. No mention of soot or dust in that advice sheet.

        R

    • I always liked the Sidak correction more, as it is is universally more powerful than Bonferroni (hence Bonferroni correction got modified into Holm-Bonferroni which is also universally more powerful than naive Bonferroni; though Sidak is still slightly more powerful). Not that it has any relevance here, since the trend can’t even pass a 0.05 alpha, but I definitely encourage Sidak over Bonferroni whenever possible. Unless you need a per-family Type I error control, where one penalizes multiple simultaneous Type I errors as greater than just each Type I error on its own, which is where Bonferroni does actually excel over Holm-Bonferroni and Sidak.

      Anyways, sorry for the random statistics ramblings!

      • Thanks, Ged. I never took a statistics class in my life, so when I started studying the climate I had to figure out the multiple-trial correction on my own. Afer some thought, I used the standard formula from rolling one of a pair of dice, where if you roll twice your odds of finding e.g. a six gets better. This relies on the use of the odds of it not happening. So let me go through it for rolling a six-sided die, using 6 as my target.

        Chance of a 6 in one roll = 1 / 6

        Chance of not getting a 6 = ( 1 – 1 / 6 ) = 5 / 6

        Chance of getting a 6 in two rolls = 1 – ( 5 / 6 )2

        Chance of getting a 6 in three rolls = 1 – ( 5 / 6 )3

        Similarly with p-values. Suppose your initial significance level is p = 0.05, so we use 1 – 0.05 = 0.95 as our value. Then for two trials you need a corresponding significance level of 1 – (0.95)1/2, for three trials you need a p-value of 1 – (0.95)1/3, and so forth.

        Anyhow, that’s what I figured out for myself. Then, some years ago when I was posting here at WUWT someone said I should just use the Bonferroni correction.

        When I looked it up I found that the Bonferroni correction is just a much easier method for determining the required p-value—you simply divide your original significance level by the number of trials. Here’s a comparison of the two methods, the actual calculation of the odds versus Bonferroni

        Trials, p-value, Bonferroni
         1, 0.0500, 0.0500
         2, 0.0253, 0.0250
         3, 0.0170, 0.0167
         4, 0.0127, 0.0125
         5, 0.0102, 0.0100
         6, 0.0085, 0.0083
         7, 0.0073, 0.0071
         8, 0.0064, 0.0063
         9, 0.0057, 0.0056
        10, 0.0051, 0.0050
        11, 0.0047, 0.0045
        12, 0.0043, 0.0042 

        As you can see, there is only a trivial difference between the two methods so I use Bonferroni, I can do the math in my head.

        If I had to guess I’d say that the Bonferroni correction was the first term of a series expansion of (1 – x1/n) … but my math-fu isn’t strong enough to determine if that is true, and it’s just idle speculation anyhow.

        As to the Sidak method, I am totally clueless. I took one year each of college physics and chemistry and one semester of calculus, and that is the full and complete record of my formal scientific education. However, google assuredly knows, and a nudge is as good as a wink to a blind horse, so thanks for the tip, when I get some time I’ll look it up.

        Always more to learn,

        w.

      • Well, that’s funny. When I researched the Sidak Correction, I found that it was the method that I had derived for myself and that I describe just above … I love it when I find out that something I derived myself is already known. It lets me know that my instincts are good and my method is valid.

        w.

  17. http://earth.columbia.edu/articles/view/3275
    http://www.the-cryosphere.net/10/477/2016/

    The change in albedo began ca. 1996 and may have been a stepwise change before your data began. This occurred exactly when the AMO flipped positive. They mention the NAO as a possible cause of this sudden change, but then continue with “Later records show those conditions shifted in 2013-2014 to favor less melting, but the damage was already done – the ice sheet had become more sensitive.”

    Certainly a climate scientist would know that Greenland summer melt isn’t explained by the NAO, is strongly correlated to the AMO, and is very strongly correlated to the Greenland Blocking Index, right? They know that the AMO is still positive, right?

    • “They know that the AMO is still positive, right?”

      But just started to turn.

      Arctic sea ice is pretty much exactly where it should be for the phase of the AMO.

      Just outside 1sd from the mean.

      PioMass shows increase in the last few years. We are pretty much out of the trough.
      (unlike the climate alarmista, who’s nature place is in the trough)

      This year will be a bit tough because of the temperature anomaly in northern Russia, but once the El Nino effect subsides, Arctic sea ice will be on the climb again..

      And won’t it be fun to watch the AGW alarmist PANIC then !! :-)

      • Andy: Mention of the significance of the phase of the AMO reminded me of a comment in the (somewhat gushing) wiki biography of ME Mann:

        “A paper published in April 2014 by Mann and co-authors set out a new method of defining the Atlantic Multidecadal Oscillation (AMO) in place of a problematic method based on detrending the climate signal. They found that in recent decades the AMO had been in a cooling phase, rather than a warming phase as researchers had thought. This cooling had contributed towards the recent Global warming hiatus in surface temperatures, and would change to enhanced surface warming in the next phase of the oscillation.[28] ”

        It seems to me that the period 2000- present (the Pause) is actually in the positive trending part of AMO according to the chart above, contrary to Mann, so am I reading the AMO charts the wrong way round?

      • The thing Mann was trying play on was that a perfect cycle on top of a upward slope will have it’s max a little later than the max of cycle itself.

        It’s very little different , certainly not “decades”. But when you are desperate any little helps.

      • I’m glad in a way that my life expectancy does not take me much past 2040. If I can hit my late 80s then out past mid century. I suspect by the 2060s the world will be fighting many unexpected issues having nothing to do with AGW. Some may have to do with cooling.

  18. Willis
    Low ice cover in Arctic is said to heat up the ocean and accelerate the melting. I always say the opposite.
    Low Ice cover is typical in fall, late September.
    Is it possible to figure out this using DMIs chart combined with Your figures?

    • Oppti, I started out to measure that, but as I said I got sidetractored … still working my way back upstream.

      w.

    • Since the rebound from 2012, it seems clear that the idea of open sea adding warming and leading to a postive feedback is demonstrably wrong.

      This is just a handwaving alarmist claim that does not match observations.

      while open water does have a much lower albedo than ice it is not the only process. What also happens is that water is almost “black” in the IR spectrum. This means much more outgoing IR and that happens 24/7/365 .

      The rebound from the 2012 low in sea ice seem unequivocal in terms of which effect dominates.

      If Willis can pull something out of CERES that will be interesting confirmation.

      • Plus evaporative cooling of course and the latent heat of evaporation is huge.

        It takes almost as much energy to change the state of a gram of water as it requires to get it from 20C to boiling.

  19. This is a great post and I would not dare question things I no little about but would like to ask a question.
    I’m just a layman so I’m sure many here might find the following a silly question but I find albedo confusing.

    As I understand it, albedo is frequency dependant. For example, fresh snow/ice is a white body for visible or shortwave radiation but it is a blackbody for infrared wavelengths. Snow and ice melt due to infrared rather than sunlight because it reflects almost all SWR. Shouldn’t we be looking for sources of LWR to explain snow and ice melt (Such as warm ocean/water or rocks or air currents etc).

    Does albedo say more about reflection and thus grey body calculations and should or do energy balance equations take frequency dependance into account when considering albedo?

    • Thanks, Scott. On my planet, the only foolish question is the one you don’t ask.

      The term “albedo” is usually used only for visible light. When you are discussing thermal (longwave infrared) radiation, it’s usually called “reflectance”. However, for most substances typical IR (thermal or infrared) radiation reflectance is quite low, a few percent, so it is not discussed much. Instead another measure, called “emissivity” is used.

      As to whether snow melts more from IR or visible light, I’d suspect that both contribute about equally, but that’s just a SWAG … but yes, the energy balance equations do take frequency dependance into account.

      w.

      • Willis,
        Emissivity is different from reflectance. Emissivity is the radiation from a heated body, e.g. the classic Black Body. Infrared can be reflected from an object, notably opaque bodies, as with an IR mirror.

      • Willis,
        I should have also remarked about the classic relationship between emissivity and reflectivity: E = 1 – R

      • Thanks for responding Willis, I was feeling a little foolish!

        I was mixing up ideas about the adaptions some animals use to cope with heat exchange.

        White seems like a bad choice for an animal at the poles and growing up I used to discuss this (However unscientifically!) with a mate, who also enjoyed musings about such things! ;-)

        If you want gain warmth in a cold place from insolation, black would be a better colour but it is a bad choice for heat loss! Polar bears have black skin and the fur is actually transparent. Their white appearance is caused by partial reflection and absorption. The fur is able to block all outgoing radiation such that they are invisible against background temperatures (Ice and snow) and impossible to detect with night-vision goggles (Except for a small patch visible when they exhale).

        This made me think about the trade-off of darkening ice. Clean Ice absorbs visible light to such a depth that there is insufficient energy to melt it. Apparently, only thermal energy will do the trick. But dirty snow and ice will melt quickly from the thermal radiation provided by direct insolation alone. The trade off I guess, is that dirty ice and snow will also lose heat quicker but is the effect big enough to make a measurable difference to the slowing of a melt, enough to provide a feedback?

        I’m not asking again, just musing out loud.

        If I had any “swag’ I’d guess that no is the answer to my question!! ;-)

        cheers,

        Scott

  20. “Being an inherently suspectful type oif fellow”

    I suspect your detractors may agree, but I believe you meant “I am a naturally suspicious type of person”
    A suspect label is usually applied to the person in question as a negative term, not their thought processes, though it can be used in specific situations, as in “I think my wife suspects me”.

    I know, I know, I need to get out more, but the weather here in Wales has been so bad for so ruddy long.
    On the other hand, I’ll go and get my coat :

    • Gareth Phillips March 4, 2016 at 1:17 am

      “Being an inherently suspectful type oif fellow”

      I suspect your detractors may agree, but I believe you meant “I am a naturally suspicious type of person”

      Thanks, Gareth. I vacillated between describing myself as “inherently suspectful” and “chronically suspectish”, and I finally settled on the former, figuring I didn’t want to make a suspectacle of myself …

      w.

  21. Looking at the data for albedo I think the trend may not just tell you what is happening.
    I think there is a “Flush” and “Cover” effect hiden also in the data which may make looking at the trend slightly less usefull.

    What I am think is oncde the snow melts at the surface the black carbon hidden in the snow gets exposed and increase the energy absorbed. Eventually there will be water flowing on the surface and wash the black carbon away and it accumulates in certain spots or gets washed out in to the ocean.
    After the meltflush is over there is less black carbon left and albedo increases and the trend gets pushed up again.
    Also new snowfall will cover the black carbon and if the next years melt season does not melt all of last years snow then the underlying darker snow will cause less meltning.

    So by just looking at the data I think that it is hard to say if the amount of black carbon has increased or decreased due to the “Flush” and “Cover” effect. You relly need to take snow samples every season before the melt season to see how much black carbon it contains to see if it changes.

    It looks like there was a darkening 2010-2012 due to more exposed dirty snow but once it was flushed away and covered with new snow the did not melt entirely next season the albedo resumed to previous levels.

    • There has been problem with instrumental drift in albedo detection I recall. Darkening lenses.

      • I would imagine that huge dust storms rose up during each Ice Age and made the ice dirty. This wouldn’t explain the sudden warming/sudden cooling of every Interglacial cycle.

  22. Some observations on the paper cited here.

    They are using the GLASS dataset, which uses information from the MODIS TERRA and AQUA sensors. This is the dataset. http://glcf.umd.edu/data/abd

    And the data indicates that the albedo reductions are only occurring on the ice-sheet margins, and the southern portion of the ice-sheet. This was not made clear in the summary. However, this is to be expected, as significant albedo reductions will only occur where there is melting or ablation, and the dust and soot can concentrate on the surface. In the highlands, the dust is covered with fresh snow each year and is not visible (and thus not effective). So these albedo reductions only effect a small proportion of Greenland.

    I am surprised that they are using ‘degrading’ satellite data, with precious little real-world confirmation. Greenland is a region that could easily be overflown with an aircraft bourne sensor. But I suppose that no grants are available for albedo research, rather than co2 research. However, they do report that ice core data suggests that soot deposites were worse in the early 20th century than today. Presumably, this was USA emissions, from when the USA had a proper manufacturing industry.

    This lack of funding confirmed by the Dark Snow project, which could not get any grants and was forced to crowd-fund their research. And I am surprised that this study (which did get a grant) makes no real mention of Jason Box and the Dark Snow project, despite Box highlighting this topic since 2005. Is professional jealousy that strong in academia?

    http://www.darksnow.org

    One other thing. I am not sure if this effects any of these results, but one study I saw said that you can only measure albedo from directly above. The soot grains sink into the snow and ice, and are not visible obliquely. So oblique albedo measurements will always be higher than zenith measurements. This also means that albedo melting will only really occur at midday in mid summer.

    Ralph

    • Thanks,

      “The version 1.0 of released GLASS albedo product has a temporal resolution of 8 days, and is available from 1981 to 2010. ”

      So how did they run it out to 2012 ?

    • … maybe by using data that the GLASS team regarded as unreliable. Maybe by graphing in another satellite.

    • Ralfellis,
      For a second opinion on what is happening in Greenland, check this out: http://www.digitaltrends.com/cool-tech/high-tech-satellites-show-thick-clouds-melting-greenlands-ice-sheet-faster/

      Actually, whether there is soot/dust present or not, texturing of the surface from differential melting will create shadows if the sun is not directly overhead (which it never is). Thus, the apparent reflectance will be decreased in proportion to the shadow component, which is a function of the sun altitude and micro topography. It is all very complex!

      • But clouds reflect insolation, and I rather think the redused insolation effect is greater than the increased insulation effect.

        And regards albedo measuremnts. What this means is that measurements should be taken at the equal and opposite declination and azimuth as the Sun, otherwise the observed albedo will be incorrect. But that is not what has been done. Which rather suggests that aircraft should be used for observations, rather than satellites.

        R

  23. The residuals demonstrate sizeable peaks. The solar component is pretty constant. The differencies in the peaks represent more or less meltwater. The process of melting ice can be described as intial melting by changes in the atmospheric condition and accelerated by the subsequent change in albedo.
    The number of peaks are limited so it seems that the local conditions are controlling the entire process

  24. Tks Willis,
    The take that I find important is the surface melt attributed to July 2012.
    In decades to come, that will appear in ice cores as a paraconformity or disconformity or unconformity, a happening very well known to stratigraphers in geology. (Wiki defines terms. The period in which normal deposition was not happening was named a “hiatus” in anticipation of future fun: not).
    Of interest is not so much a single season with melting, but a succession of seasons with melting, when non-melting is showing above and below in the core. The hiatus would be expected in almost any case where drilling was done where ice cover is transient. Less expected would be a hiatus in a well frozen area like inland Antarctica.
    When we do not know the length of the hiatus in ice core, we have a timing problem because in ice core, as opposed to rock core, one relies much on layer counting to determine ages and time intervals. Hiatus can mean missing layers, missing years, a mistimed sequence.
    It would be so easy to miss an unconformity in ice core. How can a core logger tell easily between an unconformity and a plastic flow layer of reduced thickness?
    There is even the possibility of a hiatus lasting thousands of years. Therefore emphasis is placed on regional or global markers, where for example dust deposits from volcanos as matched in ice cores in time and space.
    Personally, I think that there is too little mention of the possibility, even likely occurrence, of a hiatus or two in the ice core literature. Is this because it is overlooked, because it is seldom found in the chosen locations, or why? A recognised hiatus in Antarctic ice has important implications.
    From readers, I’d be grateful for references in ice core literature to unconformities etc and hiatus periods.
    Geoff.

    • Actually a major melt unconformity would be very conspicuous since it would collect all insolubles (dust, volcanic ash, micrometeorites etc) from the missing period in a single layer. Unconformities do occur and are a major problem in the deeper part of the cores, but they are caused by deformation of the ice as it moves over uneven bedrock.

      • Thanks, tty
        The dust etc. ‘should’ accumulate and aid detection, but one can envisage scenarios where accumulation ‘might not’ happen. Any references in mind?

  25. Glacier Girl (a Lockheed P-38F) was lost in 1942, 10 miles from the South-Eastern shore of Greenland on the ice sheet. Search began in 1981, it was found in 1988 using ground penetrating radar, two miles away under 264 feet of ice. It was dug out and flown again eventually.

    Therefore average annual ice accumulation at that site was almost 6 feet for half a century, meaning much more snow, because it gets compacted into ice once buried.

    I can hardly believe a single season’s surface melt may have lasting effect on albedo under such circumstances.

  26. Planes were found at a depth of 80 m.

    Average density of firn down to that depth is much higher than that of fresh snow, because as pressure builds up, it gets compressed ever more.

    To reach a depth of 80 m in half a century at least ten feet of annual snowfall is needed. It is never going to melt back sufficiently during the next summer to expose darker layers beneath, no way.

Comments are closed.