Cloud albedo: what does it respond to?

Posted: September 17, 2011 by Rog Tallbloke in atmosphere, climate, Energy, Ocean dynamics

There has been much discussion recently about clouds and feedback. The Spencer and Dessler debate, and the blog hosted science being done around the issue has captured a lot of attention.

Let’s take a look at the primary cause of the change in temperature over the last few decades. This graph is put together from two graphs which appeared on Skeptical Science, with a correction to the Y axis, which John Cook has two orders of magnitude too small. It’s been that way for over a year, despite two separate efforts on my part to get him to correct it. The data comes from two sources; the ISCCP international cloud project which uses weather satellites and integrates the data to produce a global series, plus data from the Earthshine project, which measures the amount of light being reflected from the Earth and bouncing back off the new Moon.

According to the Palle et al graph of Earthshine from 1999.5 to 2007.5, a percentage point change in albedo is equivalent to around 3.5W/m^2 entering the atmosphere from the Sun. As you can see, from the early ’80′s to around 1997.5 this means a big change of around 10W/m^2 at the top of the atmosphere. Allowing for the Sun only illuminating one side of Earth and for obliquity towards the limbs and poles by dividing by 4, this means an average additional insolation forcing of around 1.25W/m^2 on the atmosphere, land and most importantly, the oceans over the period 1984.5-1997.5. Clearly, it was a very significant contributor to the surface warming seen over the period, which amounted to approximately 0.2C globally. My calculations on sea level change indicate this forcing would account for most if not all the increase in the steric sea level measured by TOPEX/JASON. Something had to be responsible, because as we know from previous lengthy debates here and on WUWT, back radiationdoesn’t heat the ocean, and the time to equilibrium from a lowered air/ocean temperature differential would be too long to account for the accelerated sea level rise.

Because it is such a strong factor in the surface warming, we naturally want to know what caused the changed in cloud albedo. This is a contentious area, the IPCC barely discusses the change, let alone the cause of the change. The Svensmark effect is still a long way from confirmation although it looks promising. Another possible explanation is the drop in specific humidity at the 300 millibar level up near the tropopause, noted in a previous post on this blog. The specific humidity level corresponds closely to the variation in solar activity, according to the NCEP re-analysis data.

If as the IPCC claims, increased temperature results in more evaporation and so more cloud, this is clearly a second order effect, since cloud amount reduced as temperature rose over the 1980′s and 1990′s. But temperature is only a measured symptom of physical causes, not a real physical force in itself. If as the IPCC claims, the cloud feedback is positive, then logically, this is due to a reduction in cloud amount as empirically measured. This means the cause of this ‘positive cloud feedback’ is the extra energy from the Sun entering the oceans, not ‘temperature’. The increase in temperature is therefore largely a consequence of the reduction in cloud amount related to changes in solar activity, not the cause.

  1. Stephen Wilde says:

    Now we are really getting into my area of interest.

    To summarise stuff that I have set out in detail elsewhere:

    I) An increase in solar activity (probably causing chemical changes involving ozone) changes the vertical temperature profile of the atmosphere so as to COOL the mesosphere and stratosphere rather than warming them as per standard climatology.

    ii) That sign reversal for the solar effect is required to achieve the actual observed consequences from a more active sun. The cooler stratosphere raises the height of the tropopause especially at the poles and invigorates the upward flow through the polar vortices.

    iii) That makes those vorticess more intense (positive) but the enhanced upward flow draws the surface pressure distribution towards it as the surface extent of the vortex actually reduces. The surface air pressure pattern is drawn poleward (especially the mid latitude jets).

    iv) That allows the tropical air masses to widen especially if there are warm ocean urfaces at the same time (El Nino) and that enhances convective activity along the ITCZ. The extra uplift along the ITCZ increases descending air in the widened high pressure cells either side of the equator and cloudiness over the tropical oceans decreases.

    v) At the same time the more poleward mid latitude jets become more zonal, waving about less and the length of the air mass lines of mixing decreases for a reduction of cloudiness in the higher latitudes too.

    vi) As a result of the reduced cloudiness and albedo more energy enters the oceans which then skews ENSO in favour of El Nino events for a positive warming feedback.

    vii) However that poleward shift in the surface air pressure distribution represents a speeding up of the water cycle which transfers energy to space faster by non radiative processes.

    That poleward shift therefore provides the stabilising negative response to the effect of the more active sun on the vertical temperature profile of the atmosphere.

    Exactly the opposite occurs when the sun is less active when in that case the surface air pressure distribution then moves equatorward to provide a negtive system response to less energy entering the oceans. That is, when less energy is entering the oceans the water cycle slows down to offset the cooling effect of less energy entering the system.

    That is the only mix of positive and negative effects and feedbacks that could account for actual observations.

  2. tallbloke says:

    Hi Stephen,
    looking pretty good, apart from vii) which needs etending a bit. All energy ultimately is radiated to space. There is no other way.

    So if you feel you’ve got there with the phenomenological description, it’s time to do an Erl Happ and start combing the data for supporting evidence. Sounds like a project I could enjoy collaborating with you on.

  3. Stephen Wilde says:

    Yes as regards point vii).

    I should have said that energy gets shifted upwards by the faster water cycle, by passing the greenhouse effect because it is in latent form and so does not affect surface temperature, so that it is more readily radiated to space from the higher location.

    Thus the effect of the non radiative processes is to ADD to the energy transmission capabilities of surface to space radiation.

    Hence the ability of the water cycle to speed up or slow down to adjust net transfer of energy in response to changes in GHG quantities and any other forcing agent.

    Help with linking that description to real world data would be appreciated. There should be lots of bits of data that confirm it over and above the ones I have already linked to in my articles on the subject.

  4. adolfogiurfa says:

    Tallbloke Talkshop´s regular Michele Casati has a new article about this issue, at NIA:
    (translate it with Google)

    google translate here

    [added translation, tim]

  5. P.G. Sharrow says:

    Yes Stephen, an excellent summation. Now it is time to “snow them with what you know”. Dazzling with BS is for the Experts. *-) pg

  6. Stephen Wilde says:

    Tricky to follow the translation accurately but Michele is clearly going for cosmic rays as the cause of more clouds as per Svensmark.

    I find that unlikely because we also see changes in the vertical temperature profile of the atmosphere which is unlikely to be a result of cosmic rays.

    So, I think the temperature changes force the latitudinal air circulation shifts and the movement of the cloud bands resulting in a redistribution of surface pressure which is what changes cloud quantities and albedo.

    Cosmic rays may help a bit but I think they are just along for the ride.

    As regards “snowing them with what I know” that is an appealing prospect but first I think I need to get incontrovertible evidence that the stratosphere does indeed tend to warm when the sun is less active. I am hoping that will not be long. It has certainly stopped cooling.

  7. tallbloke says:

    Thinking about this a bit further, I think I made an error doing the divide by four on the reflectivity effect. Clouds near the lims and poles cast big shadows, and the albedo only applies to the side of Earth facing the sun anyway. If the real effect is nearer, say, 8W/m^2, then the ocean is far more dynamic in the way it maintains atmospheric temperature than we previously thought.

  8. P.G. Sharrow says:

    TB: I think you are right. As a “mad dog Englishman” that works in the noonday sun, as a gold diver that has worked many hours in clear cold water, and as a engineer of solar energy collection panels , I can attest to the fact that the 4 hours around high noon, by far, are the most effective for solar energy transfer. A “bullseye target of concentric rings of percentage of transfer would be the best visual. pg

  9. Orson Olson says:

    Perhaps I’ve missed it (because I’m very tired at the moment of writing), but doesn’t this period of lower albedo nicely fit with the super-El Nino of 1998? As well as the step change seen in the MSU-satellite temp records at around 2002?

    It appears that at least some portion of warming over the past decade or so is due to a measurable change in the earth’s albedo, related to diminished TOA cloud cover.

    I would like to see the top chart ‘Earth’s Albedo Variation” overlaid with the satellite Temp chart, and smoothed. The time lag issue — repeatedly sussed out in various fora, recently — bears interest here.

  10. tallbloke says:

    Hi Orson, sure, I’ll produce that graph. The lag would be around 4-5 years I would guess. Heat comes back out of the ocean when the Sun goes quiet, or cloud increases. This keeps temps up (or even increases them) for a while. How long, we don’t know. It seems there was quite a lag between the Sun going quiet and the onset of the really cold period in the Maunder Minimum.

  11. tallbloke says:

    I’ve inverted the cloud change and used southern hemisphere sst for the graph, because that’s where most of the important action is I think:

    It looks like there isn’t much lag between cloud changes and the SST response after all. Once the El nino ‘bounce’ settles down, the trend is clear.

  12. Stephen Wilde says:

    “Heat comes back out of the ocean when the Sun goes quiet, or cloud increases.”

    Not so sure about that unless the quiet sun coincides with a positive Pacific Mutidecadal Oscillation (PMO) of about 30 years with dominant El Ninos.

    When the sun was active in the late 20th there was also a positive PMO (not PDO as Bob Tisdale keeps telling us).

    However a negative PMO did occur around the time of the mid century cooling and again there was a positive PMO during the early 20th warming.

    Nonetheless I think the oceans have a degree of independence in their cycling so one can get EITHER heat being held back by the oceans OR heat being released by the oceans at a time of quiet sun.

    Now you might be right about the NET position when the sun is quiet and/or cloudiness increases because at such times the solar input to the oceans could well drop below replacement whether the oceans are releasing or witholding energy.

    I think the reality is a bit more subtle. The oceanic PMO continues as it pleases but the level of solar activity skews the NET balance between El Nino and La Nina events to produce a stepping up or stepping down of temperature between successive PMO phases of the same sign.

    Thus from LIA to date upward steps of which two were observed during the 20th century and from MWP to LIA a succession of downward steps which are sadly beyond our recording periods.

  13. Roger Andrews says:

    “Heat comes back out of the ocean when the Sun goes quiet, or cloud increases.”

    I’m not sure exactly how much heat comes back out of the ocean into the air, but according to Bob Tisdale a disproportionate amount of the stored heat in the oceans came to the surface in the North Atlantic between 1970 and 2005 (

    Bearing in mind that the North Atlantic is an end-point in the thermohaline circulation, is there any chance that this could be a delayed response to the increase in solar activity between 1910 and 1950?

  14. kim says:

    So what happens if sunspots get ‘large, sparse, and primarily southern hemispheric’?

  15. Stephen Wilde says:

    “Bearing in mind that the North Atlantic is an end-point in the thermohaline circulation, is there any chance that this could be a delayed response to the increase in solar activity between 1910 and 1950?”

    Not in view of the THC taking about 1000 to 1500 years depending on who is guessing.

    The surface or near surface currents are much quicker and I would suggest 10 years or so from ENSO regions of the Pacific to the Arctic Ocean.

    Hence the very low sea ice figures of 2007 about ten years after the El Nino peak of 1998.

    The effect of the long run of strong El Ninos only fades away slowly though because the route from Atlantic to Arctic Ocean is a bottleneck which only allows the warmth through slowly.

    Hence the slowness of the ice recovery since 2007.

  16. tallbloke says:

    We’re talking about a lot of different timescales and sizes of effect here, and it’s easy to start talking at cross purposes if we’re not careful. I commented on WUWT earlier on relevant threads and also quoted Harald Yndestad from the second ever article on the talkshop:

    Yndestad says:

    “ In this analysis we may understand the forced gravitation oscillation between the earth, sun and the moon as a forced coupled oscillation system to the earth. The tide and the earth rotation responds as a non-linear coupled oscillation to the forced gravity periods from the moon and the sun. This is a complex oscillation in periods between hours and thousands of years. The forced gravitation introduces a tidal mixing in the Atlantic Ocean. This tidal mixing introduces temperature and salinity fluctuations that influences climate and the eco system.”

  17. Tenuc says:

    tallbloke says:
    September 19, 2011 at 4:50 pm
    “We’re talking about a lot of different timescales and sizes of effect here, and it’s easy to start talking at cross purposes if we’re not careful…”

    I agree Rog. We don’t yet have a full understanding of how the multitude of different energy transfer systems operate together to produce the various observed effects. Ocean currents are poorly understood and I suspect there are many ‘unknown unknowns’ in how our planets non-linear heat engine works. We need to take an holistic approach to how the system operates. I think that trying to decompose this complex and changeable system into a number of small discrete elements is unlikely to provide insights as to what is happening. I also think Stephen Wilde is on the right track with his macro approach to understanding the system.

    The big question remains as to whether our observation systems can produce enough good quality data at a sufficient temporal and spacial granularity to confirm how things hang together. As ever, we will just have to blunder along with the information available, but I don’t expect to see any definitive answers. Unfortunately the whole area is ripe for cherry picking information to support ones own conjectures.

  18. Orson Olson says:

    “Fascinating…” he said in his best Leonard Nimoy voice.

    This cloud/albedo data I have never come across before. I’m sure the topic has been broached many times at CA, but the data? NEVER.

  19. Tenuc says:

    Following my last comment, I’ve just come across this…

    Uncertain Climate Models Impair Long-Term Climate Strategies: New Calibration Satellite Required to Make Accurate Predictions, Say Scientists

    “…The solution put forward by the paper is the TRUTHS (Traceable Radiometry Underpinning Terrestrial- and Helio- Studies) mission, a concept conceived and designed at NPL. This which would see a satellite launched into orbit with the ability to not only make very high accuracy measurements itself (a factor ten improvement) but also to calibrate and upgrade the performance of other Earth Observation (EO) satellites in space…”

    Should this proposed mission get the go-ahead, this would go some way to improving the quality/granularity of weather/climate data. This would be a good first step to understanding.

  20. tallbloke says:

    CA concentrates on claims made by IPCC scientists, who for obvious reasons don’t talk about the cloud data.

    Tenuc, I wonder if they envisage this satellite enabling re-analyses of older satellite data?

  21. Orson Olson says:

    A brief observation. Looking at the albedo(clouds proxy)/sst chart, is appears that SST leads. Of course, this is only an artifact of the data. Albedo measures are too crude to separate lead-lags. And smoothing of SST blends the noise too much.

    This sets us back to wonder how much refinement in available measures are needed in order to resolve valuable priority, which might well establish causal direction.

  22. Stephen Wilde says:

    Just to clarify a point on the cause/effect issue.

    My view is that latitudinal air mass shifting will occur from either or both of:

    i) A top down change in solar activity altering the vertical temperature profile of the upper atmosphere above the tropopause.

    ii) A bottom up change in the rate at which oceans release energy to the air altering the temperature differential between surface and tropopause.

    In practice both effects are competing with one another all the time with the latitudinal shift constantly adjusting to maintain the overall energy budget by changing the speed of the water cycle. That shifting is always a negative system response.

    In consequence it will be very hard to separate cause and effect as regards cloudiness and albedo changes.

    Solar influences will disguise oceanic effects and vice versa.

    However I would suggest that overall the cloudiness changes are a result of latitudinal shifts and those shifts arise from the combined net effect on the energy budget from the solar and oceanic forces.

    Poleward shifts cause decreasing cloudiness overall by widening the equatorial high pressure cells either side of the ITCZ and thereby increasing the speed of the water cycle globally.

    Equatorward shifts cause increasing cloudiness overall so that the equatorial high pressure cells shrink thereby decreasing the speed of the water cycle globally.

    That is clearly the opposite of initial expectations because one would normally anticipate that a faster water cycle would involve more clouds but that appears not to be the case due to the latitudinal shifting.

    I think the best way to explain it is to point out that when the jets are more poleward they become more zonal with a greater temperature contrast between polar and equatorial air masses and more vigour in the zonal flow. That increased vigour increases the speed of the water cycle by more than the reduced cloudiness decreases it. In effect the faster water cycle is concentrated into a smaller area.

    In contrast when the jets are more equatorward they become more meridional with a reduced temperature contrast between polar and equatorial air masses for less vigour in the zonal flow. Thus one then sees a slower water cycle but more clouds overall for the opposite effect.

    That results in a faster water cycle offsetting warming from whatever cause and a slower water cycle trying to reduce energy loss offsetting cooling from any cause (but only temporarily because energy reflected never gets into the system in the first place).

  23. tallbloke says:

    SST leads atmospheric temp by a few months as SST falls. Warming of the atmosphere when the ocean releases heat is pretty much instantaneous, reflecting the vastly higher heat capacity of the ocean.

  24. suyts says:

    TB, nice little write up! Cook isn’t going to change his graph. I think he and Trenberth are smoking the same stuff. It seems to me they’re both becoming more and more unreasonable and close minded. I think they actually believed it when they were told the science was settled, and now with these new studies contradicting the claim, the false reality they’ve built is unraveling their minds.

  25. Orson Olson says:

    So, to recap: cloud changes are reflected in SST changes (within months), followed by LT-temps.
    And cloud cover is hypothesized to be regulated by GCRs (modulated by solar-magnetic field forces as per Svensmark), as well as biogenic aerosols (oceanic algae blooms, sandstorms, volcanic, plus human black carbon, sulfur dioxide, at least regionally).

    Is this correct? We can also add other elements, known and unknown.

    This is a long causal chain! But, observationally, more plausible than the “single biggest control knob” (ie, CO2) of Richard Alley (Penn State University, Geosciences Prof).

    The latter is getting long-term seal of approval in the US from the Public Broadcasting System (amounting to maybe one-tenth of the BBC in funding/audience), in a three-part, two year program (and book) called “Earth: The Operators Manual” or (ETOM).

    You know that Alley follows Mann closely (if not so pessimistically), when you read in the book that the Hockey Stick’s “Hide The Decline” was just a kerfuffle over the exclusion of “data known to be false.” Nothing serious, move along, folks.

    Furthermore, Alley also believes that CO2 changes can almost account for the ice ages.

  26. tallbloke says:

    Orson, good summary.

    The main point of my post is that any slight change in the cloud feedback due to increased co2 is a second order effect, because increasing temps and the alleged water vapour feedback didn’t lead to increased cloud, which was alleged to be a positive feedback to temperature.

    The warmies are all washed up on cloud feedback.

  27. Orson Olson says:

    Thanks Rog-

    “The main point of my post is that any slight change in the cloud feedback due to increased co2 is a second order effect, because increasing temps and the alleged water vapour feedback didn’t lead to increased cloud, which was alleged to be a positive feedback to temperature.”

    You mean that the hypothesized Enhanced Greenhouse Effect has failed to be confirmed, again.

  28. tallbloke says:

    Yes, Enhanced Greenhouse Gas theory isn’t looking good. The ‘experts’ have EGG all over their faces.

  29. Stephen Wilde says:

    I agree with Spencer that it would be impossible, currently, to separate the top down effect on cloudiness from the solar variations from the bottom up effect on cloudiness from oceanic variations.

    However the net outturn would clearly be a consequence of the interaction between both top down and bottom up processes.

    If we could calculate the total cloudiness at any given time and monitor changes as they occur then that should give us an indication aa to when changes in global temperature trend occur and the speed of such changes and in due course maybe a figure for global cloud quantity at which the system would be approximately in thermal balance

  30. SW The inter actions you seek for the connection from the top and bottom of your analysis is due to the churning of the atmosphere by the actions of the Lunar declinational tides in all layers of the atmosphere and the oceans, coupled with the annual solar seasonal declinational tidal influences.

    All of this is further modulated by the geomagnetic responses to the shifts in solar wind speed density most of which is due to the interactions with the outer planets. These tidal and inductive inputs to the global circulation drives the trades, and zonal flows in the mid-latitudes and the 18.6 year pattern of the variation in the declinational culmination angle of the Lunar effects shifts the patterns of the blocking highs.

    I think most of the pole to pole oscillation of the atmospheric pressure results from the positions of the outer planets relative to the ecliptic plane, as well as the synod conjunctions between them.

    With the common focus upon the near surface 100Km atmospheric movements the view point that results is much the same as trying to shoot pool without taking your chin off of the felt.

  31. Stephen Wilde says:

    Well it’s possible, RH but I like to start with the simplest reasonable diagnosis which to my mind is solar induced chemical processes involving ozone.

    If one has a scenario as complex and multifaceted as the one you suggest I doubt that one would see such a good correlation to solar varaibility over centennial timescales.

    Still, it doesn’t really matter to me how the observed variations are actually caused. My description of the consequences in climate mechanism terms looks pretty good to me.

    Unless we see a cooling stratosphere while the sun stays inactive that is. That would be a problem for me.

  32. tchannon says:

    Discussion is taking place around the web, mostly by people majoring in Statistical math.

    Steve Mcintyre references Troyca as interesting.