On the Debunking of Spencer’s Feedback Ideas: An Appeal to Physical Scientists Everywhere

September 1st, 2010 by Roy W. Spencer, Ph. D.

I am seeing increasing chatter about one or more papers that will (or already have) debunked my ideas on feedbacks in the climate system.

Yet, I cannot remember a climate issue of which I have ever been so certain.

I understand that most people interested in the climate debate will simply believe what their favorite science pundits at RealClimate tell them to believe, which is fine, and I can’t do anything about that.

But for those who want to investigate for themselves, I recommend reading only our latest and most comprehensive paper in Journal of Geophysical Research. It takes you from the very basics of feedback estimation — which I found I had to include because even the experts in the field apparently did not understand them — and for the first time explains why satellite observations of the climate system behave the way they do.

No one has ever done this before to anywhere near the level of detail we do.

[Unfortunately, our 2008 paper in Journal of Climate, I now realize, had insufficient evidence to make the case we were trying to make in 2008. I believe our claims were correct, but the evidence we presented could not unequivocally support those claims. Only after finishing our most recent 2010 paper did I realize the insufficiency of that previous work on the subject.]

Then, once you think you understand the main points we make in the new JGR paper, read any other critiques or criticisms that catch your fancy.

As a teaser, one of the clear conclusions the new paper supports is this: The only times that there is clear evidence of feedback in global satellite data, that feedback is strongly negative.

All I ask is that you evaluate whether anyone can come up with a better explanation than what we have given for the structures we see in the satellite observations of natural climate variations. Do not settle for others’ vague arm-waving dismissals based upon preconceived notions or what others have told them.

You engineers and scientists from other fields are capable of understanding this, and I am appealing to you to bring fresh eyes to a field where the research establishment has become hopelessly inbred and too beholden to special interests to see that which is staring them in the face.

This is the main reason why I wrote The Great Global Warming Blunder…the evidence is simple enough for the science-savvy public to understand. But the experts do not see the evidence because they refuse to open their eyes.

WeatherShop.com Gifts, gadgets, weather stations, software and more...click here!

42 Responses to “On the Debunking of Spencer’s Feedback Ideas: An Appeal to Physical Scientists Everywhere”

Toggle Trackbacks

  1. BCC says:

    The gauntlet is thrown!

  2. Joe Born says:

    Although you are to be commended on the paper, which I find compelling and, for a technical paper, relatively clear, I had hoped to find a supplemental-information link to the data and programs behind your graphs and tables so that interested parties would be able to crunch the numbers for themselves.

    Such supplemental information would be helpful not only in dispelling evident ambiguities–which are almost inevitable in even the clearest papers–but also in revealing latent ambiguities.

    As an example of latent ambiguity, I’ll observe that it was only by actually studying your simple-model spreadsheet (which antedates the paper) that I recognized your having used (in my opinion, harmlessly) a less-than-desirable smoothing operation. (Because you used a 30-year average for the smoothing, the pseudo-random radiative-anomaly signal you ended up with had a significant negative autocorrelation at a 30-year lag.)

    In short, your audience would have greater confidence in, and a deeper understanding of, your paper’s results if you provided data and programs that enabled your audience to crunch the numbers for itself.

  3. Alexander Davidson says:

    I’ve followed your work with interest. There is a way to allow your analysis to co-exist with others and it comes from my initial discipline, phase transitions and kinetics.

    The way to understand it is an English anecdote “Three hot days then a thunderstorm”. In the tropics it’s a thunderstorm every day! And therein lies the clue.

    Positive feedback is the process of greenhouse gas warming, evaporation increasing absolute water vapour concentration in turn triggering more IR trapping. Eventually the system enters an unstable equilibrium whereby even a small perturbation triggers negative feedback – weather. That’s a phase transition with its own kinetics. The water precipitates in various ways and the latent heat release triggers convection right the way to the stratosphere. It’s much faster in the tropics.

    But there is another issue: how can the likes of a collapsed thermosphere [e.g. the 2009 fall in extreme UV output of the sun] and the arrival of solar flares from an active sun trigger significant changes of weather at the ground, and they do.

    The clue here is the concept of the ‘heat valve’, an analogy with the grid in a vacuum tube which can modulate electronic charge transfer. Modify the radiative thermal impedance from the upper atmosphere and you change the heat transfer all the way to the ground. There may well be delays, but that modulation is the cause – a bit like tickling the scalp causing the toes to move…..

    PS the IPCC’s predicted AGW forcing is based on an assumption about cloud albedo change by pollution which cannot be substantiated experimentally. The idea appears to have been based on extrapolating the ‘Twomey Effect’. Look at his theory, Mie scattering, and it breaks down for thick clouds. Also, it predicts a maximum albedo of 0.5 when measured values are greater. What is amusing, and possibly very serious, is that NASA literature claims ‘more smaller water droplets have higher surface area so reflect up to 90% of sunlight’. That completely false physics appears to be widely believed in climate science. The real explanation is that there are two optical process in clouds. The lidar people know it.

    Take away the unproven cloud albedo cooling, known by 2004, and the IPCC’s predicted AGW forcing has to be halved.

  4. Harold Pierce Jr says:

    RE: The Late President Eisenhower’s Second Warning

    This passage from “The Farwell Address to American People”
    follows the text where he mentions the familair “military-industrial complex.”

    IIRC, this passage in part was based on his confrontations with the ruthless egomaniac Edward Teller who convined the Congress and American people to pursue the development of the H-bomb, which he recognized asa true weapon mass of destruction, and believed was unnecessary and would result in an extremely expensive weapons program and greatly increase then severity of the cold war with the Russians.

    “Akin to, and largely responsible for the sweeping
    changes in our industrial-military posture, has been
    the technological revolution during recent decades.
    In this revolution, research has become central; it
    also becomes more formalized, complex, and costly. A steadily increasing share is conducted for, by, or at
    the direction of, the Federal government.

    Today, the solitary inventor, tinkering in his shop
    has been overshadowed by task forces of scientists in laboratories and testing fields. In the same fashion,
    the free university, historically the fountainhead of
    free ideas and scientific discovery, has experienced a revolution in the conduct of research. Partly because
    of the huge costs involved, a government contract becomes virtually a substitute for intellectual curiosity. For every old blackboard there are now hundreds of new electronic computers. The prospect of domination of
    the nation’s scholars by Federal employment, project allocations, and the power of money is ever present
    and is gravely to be regarded.

    Yet, in holding scientific research and discovery in respect, as we should, we must also be alert to the
    equal and opposite danger that public policy could
    itself become the captive of a scientific-technological elite.”

  5. Sean Houlihane says:

    I think you maybe over-estimate the average scientist’s understanding of this complex field. Whilst your paper seems consistent and well presented, I think a degree of familiarity with the data is necessary to have much chance of poking holes in it. Certainly, I will be interested to see if anyone attempts more than a hand-waving exercise to dismiss it (and I find it easier to evaluate the counter argument by how specific it is).
    As a follow-up, some bounds on testability would be interesting. What is the shortest data-set which could be analysed in isolation and be shown to be consistent or not with your simple model? I have a feeling that the next step is to investigate the detailed alignment of the grown-up models with actual observations (and that seems very non-trivial). Without that step, your simple model remains just a possible approximation, and difficult to prove (seeing as it is hard to decide if surface temperatures are consistent with the current models).

  6. marco says:

    Dr Roy

    I’m probably not your ideal reader in that my science education aint the best and so I won’t be offended if you are not waiting with anxious anticipation for my perspective.

    I simply want to point out that you really don’t need to over egg the significance of your argument. If it stands it will stand if it doesn’t …well I suppose that’s just the way science works.

    You want an honest opinion of your paper but then you immediately say that you’d rather people didn’t swallow any real climate arm waving (please excuse the mixed metaphor).

    This could lead to the uncomfortable conclusion that a negative critque of your paper (which is far from a foregone conclusion) will be dismissed as partisan posturing. Was that your response to the criticisms of your original paper which you now concede was flawed? Do you see what I’m getting at?

    I enjoy your blog as you explain complex principles clearly and patiently. I also admire, for the most part, your good humored moderation of the debates that ensue.

    The passions around the AGW debate are both understanable but limiting. From my reading once you concede that the increase in CO2 is predominantly anthropogenic, once you concede that CO2 as a greenhouse gas ‘warms’ the climate then the big issue to be resolved is the question of climate sensitivity? Am I wielding Ockhams razor carelessly?

    If this characterisation is correct then all the attention your paper garners may not be because the establishment wants to bring you down but perhaps simply because it is the biggest question which needs to be addressed and as such not only requires intense scruitiny but demands it.

    But it would be lovely if we could all concede to the other good faith but at the same time when that good faith is lacking remain a little indulgent for in these matters regardless of our passions science has the habit of rendering them innocuous.

    Please forgive any presumption if it appears a puppy is trying to teach granny to suck eggs.

    • All I am saying is ask for the evidence if someone claims to have refuted our ideas…don’t settle for just arm waving.

      And, to answer your question, when we saw a written critique of our first (2008 J. Climate) paper, we looked into it and verified that the objection was valid…that we had NOT adequately demonstrated that feedback diagnoses lead to a bias in the direction of high climate sensitivity. When I am wrong, I admit it and move on. That’s science.

    • Anonymous says:

      Plants can’t grow without carbon.
      When the climate warms, more plants grow (feeding more animals and people).
      When the climate cools, fewer plants grow (starving animals and people and causing starvation and epidemics of plague and such-like).

      Pray for global warming and enough carbon in the atmosphere to make plants grow well and lushly.
      If reduction of atmospheric carbon is possible, we don’t want it because we want to continue to eat. Less carbon, less food. Eliminate politicians who continue to prattle about this. They are dangerous because they hold pursestrings that can cause very bad things to happen.

  7. John says:

    Dr. Spencer –

    Does the chatter have any discussion of a basis for the debunking your ideas, or are they just general statements that the paper will be debunked?

    • I have yet to see any basis…except for my aforementioned agreement that out 2008 J. Climate paper did not adequately demonstrate what we were claiming. Beyond that, I have seen nothing that would refute anything in our new (2010) JGR paper.

  8. Jim Cripwell says:

    Let me emphasise what Sean Houlihane wrote “I have a feeling that the next step is to investigate the detailed alignment of the grown-up models with actual observations (and that seems very non-trivial). Without that step, your simple model remains just a possible approximation, and difficult to prove (seeing as it is hard to decide if surface temperatures are consistent with the current models).”

    No model has much meaning until it has been validated. The way you validate models is to accurately and quantitatively predict the future. Surely the next step is to predict what some future experimental data will be. And then let us see whether it turns out to be accurate.

    • Jim, what you suggest has yet to be done in the climate arena. And it is not clear what kind of forecast would even validate a climate prediction. Warming could continue for 20 years, as the IPCC predicts, but that doesn’t mean they have the correct reason why it warmed.

      I can flip a coin to predict whether the stock market will be higher or lower tomorrow, and I will on average be right 50% of the time. Doesn’t that make my coin-based model pretty good? 🙂

  9. Christopher Game says:

    Christopher Game commenting on Dr Spencer’s article On the Debunking of Spencer’s Feedback Ideas: An Appeal to Physical Scientists Everywhere, posted 2010 Sep 1.

    I would like to take this opportunity to say something about feedbacks, especially since I think that perhaps Dr Spencer thinks I am opposed to the use of the idea of feedback in the study of climate dynamics.

    No fear! I am keen on the idea of feedback in the study of climate dynamics. There is no other way. I think Drs Spencer and Braswell are right to work on feedbacks in climate dynamics. I applaud them for their work in this.

    I have attacked the IPCC’s meretricious and seductive “forcing and feedbacks” formalism, not because it works on “feedbacks”, but because it is structured about the confounding notion of “forcings” and because it works its “feedbacks” in unphysical ways, and because it does not use explicit system-independent driving functions. The IPCC formalism’s “forcings” are really feedbacks dressed up as if they were driving functions. This makes them practically impossible to interpret in terms of cause and effect. I interpret Drs Spencer and Braswell’s 2010 JGR paper as evidence in support of my attack on the IPCC formalism. I have also other reasons for finding the IPCC formalism hopelessly flawed, but I will not detail those reasons here.

    The purpose of my attack on the IPCC formalism is to alert people to its flaws, so as to stimulate people to try to invent, and to use instead, simple tailor-made physically based theories for the study of the physics of the effects of CO2 on the energy transport process of the earth’s atmosphere.

  10. vboring says:


    If climate scientists agree that anthropogenic climate change isn’t dangerous, then the public will be convinced that it is and CO2 policy will be passed.

  11. Doug Proctor says:

    Dr. Spencer,

    It is not a wonder that the journalists don’t get your paper or those of others when they are like this one: you need two, not one. The first is dry-as-a-bone, non-contentious, so that it passes the editors. The second that says, “We established X in our paper. The IPCC claim Y, and others say Z. Because X trumps Y and Z, the following fails and the subsequent is going to happen.”

    But that would be having an opinion. Stats are inviolable, but conclusions can be challenged. The odd thing is that “science” is all about conclusions. I’m a geologist faced with a lot of data coming from others. Data is only useful when it combined with context, when it becomes “information”. Information then leads to conclusions. The non-science based observer doesn’t understand this process. He thinks data is hard and leads to unique solutions. The simpler, the better (misunderstanding Occam’s Razor along the way).

    I hope your work gets translated into something that a RealClimate reader can understand, otherwise other number-crunchers can ignore it.

    Recently, using the brain that my mother told us God gave us for the purpose of figuring things out for ourselves, I looked into the simple depiction of how the TOA insolation was broken down into its warming and non-warming parts. I was shocked to see the annual variation of 6.8%, or +/- 47 W/m2. A post of yours mentioned that the global temperatures, however, were backwards to the aphelion and perihelion positions by about 2.3C*. The IPCC claims that a doubling of pCO2 from 384 ppmv generates 5.6W/m2 of radiative forcing, yet the annual variation in TOA insolation is almost 8X that, and still a 2.3C* reverse differnce exists. The difference is, by IPCC pCO2 extimates, about 2.8 W/m2 of forcing. Amazing. 47 extra and the world is 2.8 COOLER.

    The buffering of the global temperature is very strong. The land/sea geometry (you pointed out) creates the greater warming when orbital characteristics say the world should be cooler. Geometry is obviously extremely important, and yet the IPCC claim that the cloud cover variation is unimportant (as it averages out). Only a slight variation in reflective geometry creates the same forcing as changes in pCO2, yet it can be dismissed. Only a slight change in how the thermal energy is being distributed would equal the power of pCO2 increases, but that, too, is dismissed. The TOA “natural” variation is so great that the idea that we are in equilibrium is bizarre: unless you think that running rapidly back and forth on a seesaw without falling off means the seesaw is in an equilibrium position. We are in a dynamic equilibrum, not static, situation. As such, any slight miscalculation in our speed of movement means the seesaw sags down at one end or rises precipitously at the other. This does not mean we are sensitive (though we can be to the main drivder, the sun) but that we live within a somewhat chaotic system. To say that we are responsible for such a tiny temperature rise is to assume a tremendous equilibrium that does not exist.

  12. Jim Cripwell says:

    Roy, Many thanks for taking note of my message. You write “Jim, what you suggest has yet to be done in the climate arena.”

    When I learned basics physics as an undergraduate, I was taught that physics is based on experimental data. A hypothesis is an idea with no experimental data to support it; a theory is an idea with some experimental data to support it; a law is an idea with overwhelming experimental data to support it. What I observe with AGW is an almost complete lack of experimental data. We cannot do a controlled experiment to add CO2 to the atmopshere, and measure how much the global temperature rises. AGW is a hypothesis. What I am waiting for is an experiment like Michelson/Morley on this issue. I was hoping your paper was a step in this direction.

    In the meanwhile I am left with my opinion that, when it comes to AGW, the physics is not good enough to tell what is happening. We just dont know.

  13. Stephen says:

    “I simply want to point out that you really don’t need to over egg the significance of your argument. If it stands it will stand if it doesn’t …well I suppose that’s just the way science works.”

    Sadly, that’s often not the way science works. As Dr Spencer says, scientists are affected by their biases and often don’t see good arguments and evidence right in front of their faces. History is filled with examples of majority scientific views hanging on long after they have been refuted.

    Part of our job as truth seekers, beyond presenting good argument and evidence, is the communication of those arguments and evidence in better and better ways to help them prevail.

  14. Anonymous says:

    Dr Spencer,

    I’ve read counters to the notion of a strong negative feedback that point to global temperature increases of 5 – 6C during the ~5k year glacial terminations – that insolation changes are too small to make such a large impact.

    How do you see the biosphere altering so significantly through ice age shifts under the provisions of the theory of strong negative feed backs to forcings? And how is it that the warming is global, rather than localised at the particular pole that receives more sunlight during Milankovitch cycles?

  15. barry says:

    Sorry, neglected to fill in the bars. That last post was from me.


  16. barry says:

    Thank you for the reply, Dr Spencer, but it doesn’t really answer my query. I’ll try to be clearer.

    There is ample evidence of a global temperature change of 5 – 6C during ice age cycles. The global climate clearly responded to some kind of forcing that warmed/cooled it. There must be a reason why the great ice sheets melted away and reformed.

    1) Does your calculation of a strong negative feed back to forcing permit such shifts?

    2) If so, what would be the magnitude of forcing for a 5C change in global temperature, taking into account your estimated climate sensitivity?

    3) Have you speculated on what mechanism(s) may have been strong enough to alter the global temperature so greatly in the face of strong negative feed backs?

  17. Dominic says:

    @ barry


    Roe, G. (2006), In defense of Milankovitch, Geophys. Res. Lett., 33, L24703, doi:10.1029/2006GL027817.

  18. barry says:

    Thanks Dominic. I’m not eager to delve too deeply into paleoclimate (and stray too far from the topic), but while the Roe paper posits summertime insolation being the primary driver of ice sheet reduction in the Northern Hemisphere, temperature change (and ice sheet reduction) is global through the ~100k deglaciations. Thus, I’m wondering how Dr. Spencer’s theory of strong negative feed backs fits within this regime.

    If we were to consider the NH only, then the polar temperature change from trough to peak is 10 – 12C during deglaciation, and ~3C in the tropics. Whether the radiative forcing comes from GHGs or solar variance, strong negative feed backs would be in play. My query would still apply here. If climate sensitivity is low, how does orbital variance, with a fairly weak insolation change in the NH (or SH), result in such profound shifts?

    I don’t know if Dr Spencer has considered his theory WRT to ice age shifts, but it is one of the primary rebuttals to the notion of a low climate sensitivity. Hence my curiosity here.

  19. barry says:


    should be;

    100k year

  20. Dominic says:


    You say:

    ‘the Roe paper posits summertime insolation being the primary driver of ice sheet reduction in the Northern Hemisphere, temperature change (and ice sheet reduction) is global through the ~100k deglaciations.’

    June insolation at 65N is indeed very much an NH phenomenon(!) BUT, to paraphrase Roe, ~85% of the ice during the last glacial maximum ~21ka ago was in the NH [Peltier 2004].

    NH ice melt introduces fresh water into the high latitude oceans and global-scale climate change gets under way. The NH versus global climate change problem is resolved.

    The large, rapid deglaciation that occurred at the end of some ice age cycles may be attributable to the internal dynamics of ice sheets. Rapid collapse may occur if a threshold size is exceeded [Imbrie & Imbrie, 1980; Paillard, 2001]. [I’m still just paraphrasing Roe here BTW].

    Speaking for myself, I suspect that climate sensitivity itself varies depending on the forcings involved and the dominant climate state (ie ice age or temperate).

    Excursions from temperate to ice-age conditions and back again are obviously the defining characteristic of the Quaternary. Powerful positive feedbacks associated with the EXTENT of NH ICE COVER utterly dominate.

    But does this mean that temperate climates are protected from significant WARMING by negative feedbacks?

    Put another way – is it misleading to treat the dynamics of climate sensitivity that control the onset and end of ice ages with the way negative feedbacks might inhibit warming caused by CO2 in a temperate climate?

    I’m not sure that there is a definitive answer to this question as yet. So I’m equally unsure that it is safe to assume that Dr Spencer’s argument is invalid on the grounds of ‘conflict of sensitivities’.

    You raise an interesting point and this is merely my tuppence-worth. I do not claim to be right or that you are wrong.

    • Dominic says:

      Apologies for the gibberish in my comment above. It’s obviously been a long week. Perhaps what I meant to say was:

      Speaking for myself, I suspect that climate sensitivity may vary dependent on the forcings involved and the dominant climate state (ice age/temperate).

      Excursions from temperate to ice-age conditions and back again are the defining characteristic of the Quaternary. Powerful positive feedbacks associated with the EXTENT of NH ice cover appear to drive major climate change.

      But are the dynamics of climate sensitivity that control the onset and end of ice ages distinct from negative feedbacks that may inhibit warming caused by CO2 in a temperate climate?

      I’m not sure that we actually know. A ‘conflict of sensitivities’ may not exist between Dr Spencer’s position and paleoclimatic change.

  21. Andrew says:

    With regard to the glaciation cycle arguments: The cases of Milankovitch forcing and CO2 forcing are NOT analogous. Many people are mislead by the focus on a particular latitude at a particular time of year insolation figures into think that this is the quantitative, global effect, that’s the same everywhere. If we were talking about CO2, then the forcing would be spatially uniform, almost entirely so in fact. But Milankovitch “forcing” varies strongly with latitude and season. Indeed, there is a reversal of many of the effects in the Southern Hemisphere. From an energy balance point of view of an Earth which is just a watery rock ball, Milankovitch being a global forcing is nonsense. Fortunately the real Earth has all kinds of asymmetries of terrain, it has oceans and continents, ice sheets deserts and forests all over the place. But crucially, there is also heat transport from one location to another, and with Milankovitch, differential solar heating of the atmosphere/ ocean system by latitude/season changes from it’s normal behavior. So the atmospheric and ocean circulations also change. This in turn might cause changes in cloudiness, which, well, you’ve been reading this site, right? But it turns out, also, that if the equator to pole heat transport is altered, and a strong negative feedback occurs in the tropics, tightly constraining them, then a large change in mean temperature can occur from that irregular forcing, even if it’s TOA average was small (Lindzen and Pan, 1994). Also, no one, to my knowledge, has seriously questioned the idea of a positive ice albedo feedback, which would further amplify such changes in the high latitudes. That’s how, I think, you can get these large changes with low sensitivity.