Spencer & Braswell: Part III

by Judith Curry

The story surrounding Spencer & Braswell has gotten more interesting with the pre-publication of the rebuttal paper by Dessler.

The Dessler (2011) paper is in press at Geophysical Research Letters.  Dessler et al. is a critique of Spencer and Braswell (2011), and also Lindzen and Choi (2011).  Spencer and Braswell (2011)  is a critique of  Dessler (2010). Dessler (2010)  is a critique of Spencer and Braswell (2010).

Got that?

Here is the abstract and conclusion for Dessler (2011):

Cloud variations and the Earth’s energy budget
A.E. Dessler

Abstract: The question of whether clouds are the cause of surface temperature changes, rather than acting as a feedback in response to those temperature changes, is explored using data obtained between 2000 and 2010. An energy budget calculation shows that the energy trapped by clouds accounts for little of the observed climate variations. And observations of the lagged response of top-of-atmosphere (TOA) energy fluxes to surface temperature variations are not evidence that clouds are causing climate change.

Conclusions.  These calculations show that clouds did not cause significant climate change over the last decade (over the decades or centuries relevant for long-term climate change, on the other hand, clouds can indeed cause significant warming). Rather, the evolution of the surface and atmosphere during ENSO variations are dominated by oceanic heat transport. This means in turn that regressions of TOA fluxes vs. ΔTs can be used to accurately estimate climate sensitivity or the magnitude of climate feedbacks. In addition, observations presented by LC11 and SB11 are not in fundamental disagreement with mainstream climate models, nor do they provide evidence that clouds are causing climate change. Suggestions that significant revisions to mainstream climate science are required are therefore not supported.

See also the  video on the paper and the press release.

I just checked Roy Spencer’s blog, and I don’t see a response from him yet, but I suspect that one will be forthcoming.

Blogospheric analyses of the Dessler paper are starting to emerge.  Steve McIntyre provides an analysis of Dessler (2010) and (2011), which is well worth reading.  RealClimate has a new post that mentions Dessler (2011) but focuses on the broader brouhaha surrounding S&B.

JC comments 

I’ve done a quick read of Dessler (2011).  I have the same problem with Dessler’s paper that I had with S&B and LC.  These analyses don’t really tell us anything about cloud feedback, although they are interpreted as doing so.  The simple energy balance model upon which equation (1) is useful (at best) only for black box thermodynamics only analyses of the climate system.  Cloud feedback is frequency dependent, and the key issue S&B are trying to address is the role of natural internal variability in modulating clouds, which in turn modulates the radiative balance and surface temperature; it is in this sense that they then refer to clouds as a “forcing.”  This inconsistency arises from trying to use the black box energy balance feedback model to make inferences about the impact of internal dynamical variations on clouds and the energy balance.

So how to do a sensible observational analysis of cloud feedback remains elusive. The most significant element of all of these papers is the comparison of satellite data with models.  A key issue is which models to select.  S&B select models with highest and lowest sensitivity.  Dessler (2011) seems to select the models that best match the satellite observations.  IMO the best way to select the models would be to conduct an a priori analysis of the 20th century simulations for the period of interest, and compare the simulations with major circulation features of relevance to the large-scale cloud field, such as ENSO, the NAO, etc.   Then look at the relevant radiation and cloud fields for these models.

IMO, none of these papers are of particular scientific interest, although they have become  important in the public debate.  The broader issue is that we need more papers comparing satellite observations to climate model simulations, and need to design better ways of using these data to evaluate processes of relevance to cloud feedback.

Bottom line:  S&B and LC papers do have flaws, as discussed on previous Climate Etc. threads.  Dessler (2011) adds relatively little to this debate.  None of these papers are particularly useful in evaluating the sign or magnitude of the cloud feedback.

Nevertheless, all this stuff is like crack cocaine for the climate blogosphere :).

Moderation note:  this is a technical thread, moderated heavily for relevance.  Make your general comments on the Part II thread.

476 responses to “Spencer & Braswell: Part III

  1. Question: why only select a few models? Why not look at every model and how it compares to observation and then notice trends, such as 50% of the models are with in such-and-such distance from the observations?

    • That would be a possible strategy. But many of the models do a really poor job, and would clutter the interpretation.

      • Isn’t the point S&B are trying to make is that the models are doing a poor job? Why show six models are doing a poor job, when they can say something about all models used by the IPCC (or whatever criteria they wish the models to have).

        I understand the analysis will be more difficult, but how much so? It seems some what foolish to me to only select a few models for analysis and extrapolate to the whole, when one is able to analyze the whole, itself.

        This seems especially useful in this situation where the rebuttal uses the same method to achieve different results.

      • “But many of the models do a really poor job, and would clutter the interpretation”

        FAIL Judy, FAIL.

        I will bet that that when ever any model is tested against any observed data-set the ‘modelers’ will state ‘look we have these 10 with lovely fits’, and now doubt they would be correct. BUT, if those 10 sweet models were tested against a different observed data-set they would fail.
        i know, and you know, that these models have been trained like performing seals to match the historical temperature series.
        O.K., fine. They are not treated as individual hypotheses; they never not tested to destruction, but D11 allows us to do that.
        Every single one of the models presented in D11 is between execrable and awful.
        The three red ones are just awful; plot the model vs observed and show the correlation coefficient and rho, I bet you that p>0.05.
        In any biological simulation these models, including the red ones, would be discarded because they are so crap, no referee would allow ‘the shading represents the 2σ uncertainty of two of the data sets”, rather than using the normal 1.96σ 95% confidence levels. He fought for that 0.4499736%
        So Judy, for each of the three models D11 claims is a fit, we should know the r2/rho vs the global temperature series from 1960 to 2010 AND the r2/rho of the fit to the blue line in D11 Fig3.

      • That would be a possible strategy. But many of the models do a really poor job, and would clutter the interpretation.

        Would you agree that if, say, S&B tested 14 models, only showed the results of 6, and that those 6 were the worst at matching the satellite observations, that one might be inclined to raise a questioning eyebrow as to why they might do so?

      • Well IMO the important thing is that if you are going to subselect certain models, that you have an a priori reason for the selection. It seems that S&B did, based upon the sensitivity values. Not a fabulous choice. But cherry picking models based upon whether they agree or disagree with the obs and then drawing a conclusion based on this comparison is a no-no. The interesting thing is to understand why some models have better and worse agreement with the observations.

      • Judy, surely what you need to do is test all the models, and test them against all the observed tests you have.
        Every time a model fails a test, it should be discarded.
        Each model is a novel hypothesis, it is a mathematical description of a physical process.
        Each should be tested, to destruction.

        Compare and contracts drug design and testing.
        We design a drug in silico, based on a 3D crystal structure.
        We synthesize a compound.
        We test the compound on its target protein trying to get the highest binding constant/rate constant.
        Redesign occurs.
        We test the familial compounds on target protein.
        We pick the best 5 and test them in tissue culture.
        We use cells we know have got particular metabolic pathways up/down regulated.
        We go to mice, sometimes making genetically modified humanized mice.
        We test on paid young human volunteers.
        We test on a small population of patients with disease; double blind.
        We test on larger numbers at >4 different centers, double blind.
        Drug goes on market, GP’s have to do follow up on patients getting drug.
        Here about 15% are withdrawn from market due to unforeseen side effects.

        Would you describe how your models are tested?
        How do you whittle the numbers down until you have one that works; that is, describes the system you are studying?
        I suspect the answer is never.

      • Judith, he said that he analyzed all of the models and only showed the results for 6 of those that agreed with his hypothesis. These just happened to be those which had the poorest modeled ENSO even though he (Spencer) repeatedly emphasized that ENSO was important in the period analyzed. Spencer says this is what he did *in his paper*.

      • Braino in that last one. Spencer said that the analyzed all 14 models but only plotted the results for the 3 most sensitive and 3 least sensitive. This is not a priori reasoning.

      • Judy,
        Once again I wonder as to the problem with testing all the models, and analyzing all the results. Forgive me because my training is in pure mathematics with only the most basic knowledge on statistics. Also, in my work I do not deal with real world examples so don’t know how messy they can be. However, given that one spends months if not years crafting and submitting a paper (Dessler is a special case), it seems that taking the extra effort to analyze all models would be well worth it. Especially as it would not allow a rebuttal to do that which Dessler did, i.e.: take a different sample to analyze and conclude that you are wrong.

      • “The interesting thing is to understand why some models have better and worse agreement with the observations.”

        Simply by chance some models will outperform others. If 10 people guess what the climate will be like in 10 years, some will outperform the others. However, it doesn’t mean any one of them have the slightest skill at predicting climate beyond chance.

        Have 10 people predict past climate. Compare this against observations. Some will outperform others. Are these people then to be trusted to predict future climate?

        In Australia it was discovered that predicting today’s weather tomorrow was more accurate than the weather forecasters, so they laid off the weather forecasters and simply forecast what happened today for tomorrow. And the accuracy of forecasts improved.

      • However, given that one spends months if not years crafting and submitting a paper (Dessler is a special case), it seems that taking the extra effort to analyze all models would be well worth it.

        If they know how to run each model themselves with the same forcing inputs. The analysis is easy once you get to that point since it is all automated.

        I am interested in your motivation as a pure mathematician, and one that is I assume way out in the land of abstractions. Would you ever consider getting into applied math? And if not, why not? You evidently show some interest by commenting here. Its pretty interesting stuff, IMO.

      • Judith:

        But cherry picking models based upon whether they agree or disagree with the obs and then drawing a conclusion based on this comparison is a no-no.

        Thank you. S&B pulled that no-no. I’m surprised you missed it …

      • Spencer picked models with the greatest and least sensitivity according to another referenced paper. He didn’t make choices based on what fitted his hypothesis or anything to do with results of the analysis.

        Judith suggests this isn’t a fabulous choice. Why Judith? This is the ultimate question Spencer is addressing.

      • I think they chose highest and lowest sensitivity models because, bottom line, the sensitivity value is what is fundamentally at issue.

      • The short: 1 — Data was tossed that disproves the conclusion that a lower climate sensitivity model is closer to the actual data than is a model with higher climate sensitivity — a no-no. 2 — The methodology used here was flawed in the first place, so, even if the inconvenient data points had not existed, the conclusion would not have followed — bad science.

        Longer version:

        The feeling I get from reading blog analyses of this (I’m not a climatologist and haven’t read the actual papers) is that we have two types of climate sensitivity (100 year scale) models, those that do a decent job modeling short term effects such as ENSO and those that don’t. It seems the SB papers picked to graph from the bag of models that didn’t do a good job (when we ignore the error bars). It’s questionable why you would want to compare apple-ly data to orange-gy models.

        OK, it was then suggested that because the set of oranges that was a little bit more off the apple-ly data had a certain property — greater than average Vitamin C — then that means that property (greater than average Vitamin C) is likely problematic. The apple-ly data appears to be biased against greater than average Vitamin C! The logic doesn’t follow, especially because if we had the knowledge to know how to factor out (and define) the short term effects (ie, remove the apple-liness), we might end up with (orange-gy) data points that could be *below* both bags of oranges and hence actually closer to the greater than average Vitamin C oranges (“more sensitive” climate models). So, by comparing apples to oranges, correlations you find can easily be meaningless. This isn’t science. It’s speculation that may or may not be supported by science.

        Last, if we had listed all the curves, we would find:
        (A) Those with most sensitivity (high vitamin C) would tend to be off.
        (B) Many of those with sensitivity in the middle would have been very close to the data.
        (C) Those with the least sensitivity would be off but (on average) not as off as (A).

        To have these 3 results before you, which clearly show that higher sensitivity does NOT imply greater deviation from data, and ignore dealing with them looks like a “no-no” to me. You threw out data that would have rather easily disproved your conclusion (the conclusion being something along the lines that models with higher climate sensitivity correlates less with data than does models with lower climate sensitivity).

        To repeat the short conclusion: 1 — Data was tossed that disproves the conclusion that a lower climate sensitivity model is closer to the actual data than is a model with higher climate sensitivity — a no-no. AND 2 — The methodology used was flawed in the first place, so, even if the inconvenient data points had not existed, the conclusion would not have followed — bad science.

      • Yes,

        The same way if Mann or Briffa subselect certain trees or proxies.
        As McItntyre would argue you have an obligation to report adverse results. Glad to see you agree with him.

      • The same way if Mann or Briffa subselect certain trees or proxies.

        The opposite, actually. Mann and Briffa try to figure out how to select datasets which best proxy temperature, and whether or not you agree with the methodology used by Mann at various times, he does have a methodology.

        S&B explicitly picked those that did the worse job at modeling ENSO and therefore matching their observations.

        If you can’t see the difference, well … no surprise, is it?

      • Dhog,

        You clearly don’t understand what every statistician does about selecting data that correlates as a precondition and then trying to calculate a variance. You don’t understand, as those in the field do, as experiments with sythethic data demonstrate, that this method will, of necessity, reduce variance, flatten the blade.. necessarily and without doubt. I’d relitigate Yamal with you but you clearly lack background in the datasets at issue so it’s rather pointless.

        At this stage you should say Piltdown mann and scurry away

      • It’s one of those quantum zen philosophical questions: if you build a cherry-picking machine, and turn it loose, are you picking cherries?

      • I don’t know much about Mann and only some about Spencer, but I get the impression those two cases (Mann data picking vs Spencer data picking) may not be the same. Would the following be correct?

        In one case, you search to find a ruler that approaches data you know is correct.

        In the other, you search for a rule that identifies data with attributes you wish were correct.

      • On his blog he says he picked the 3 best and the 3 worst fits. With this he wanted to show that even the best fits are not all that good.

      • No he didn’t. He chose based on Sensitivity not best fit. Here is the relevent section of the paper

        “While we computed results for 14 of the models archived, here will present results for only the 3 most sensitive models (MIROC3.2-hires; IPSL-CM4; MIROC3.2-medres), and the 3 least sensitive models (FGOALS; NCAR PCM1; GISS-ER), where their sensitivity to transient carbon dioxide forcing was estimated by [7].”

        7. Forster, P.M.; Taylor, K.E. Climate forcings and climate sensitivities diagnosed from coupled climate model integrations. J. Climate 2006, 19, 6181-6194.

      • The fact that the IPCC averages all the models to come up with their conclusions should be an indication of how they should be used in this type of study. It probably wouldn’t make anyone happy, but, would at least be consistent!!

      • I think it is more fun they way they did it. Spencer averaged three that sucked and three that sucked less. Trenberth picked the least suckie. Then Dressler used the least suckie, one a little less suckie and one that almost sucked as bad as the Spencer six. Evidently the rest were so suckie, no one wanted to use them. So we now have a consensus agreement that three of the 14 models are better than the six of the 14 Spencer picked. That leaves 5 with roughly the same robustness as the UKMET hurricane model we all know and love. Luckily, all 14 models will be included in the ensemble to ensure that minimum uncertainty standards are maintained. Science moves forward.

      • The realclimate post was much funnier – the average of all models varied by about 1W/m^2/K with error bounds of +/- 3W/m^2/K. Everything fell within the envelope – so what’s the problem?

      • The fact that the IPCC averages all the models to come up with their conclusions should be an indication of how they should be used in this type of study. It probably wouldn’t make anyone happy, but, would at least be consistent!!

        That’s a pretty good approach. In oil depletion analysis circles, some people like to work out ensemble averages of the various models available. You end up potentially getting better results from a “wisdom of the crowds” approach. The following link is a good example of a composite average for peak oil:

        http://trendlines.ca/free/peakoil/Scenarios/scenarios.htm

      • The difference is that the oil companies know the oil is actually there.

      • When you pick a Model because it matched the data, you are not picking the best Model, You are picking the best Curve Fit. Matching Data, by itself, does not prove anything.

      • Dr. Curry:
        Your analysis is right on the money. With all due respect to Dr. Dessler and Dr. Spencer,
        The climate energy equation used in their papers is a form of the Stefan-Boltzmann law equation for small temperature change. While the equation applies for solid surfaces such as land and continents, it is not the controlling equation of the energy radiated from fluids such as surface water or ocean. Surface water outgoing radiations is controlled by convection heat transfer process, not by the Stefan-Boltzmann law equation. Therefore, using Stefan-Boltzmann law equation to calculate ocean temperature is a mistake and the model can never be right. There are no such things as climate sensitivity or feed backs.

    • Here is Spencer’s initial response. It appears to address the issue of why he didn’t look at every model:

      http://wattsupwiththat.com/2011/09/07/the-good-the-bad-and-the-ugly-my-initial-comments-on-the-new-dessler-2011-study/#more-46767

  2. Dessler seems to disagree with you that these correlations cannot be used to infer climate feedback when he says: “Rather, the evolution of the surface and atmosphere during ENSO variations are dominated by oceanic heat transport. This means in turn that regressions of TOA fluxes vs. ΔTs can be used to accurately estimate climate sensitivity or the magnitude of climate feedbacks.” Do you reject that argument?

    • For previous posts on this general topic, see

      http://judithcurry.com/2010/12/29/climate-feedbacks-part-i/

      http://judithcurry.com/2011/08/16/climate-sensitivity-to-ocean-heat-transport/

      the cloud feedback issue is much more complicated than portrayed by Dessler

      • Dear Dr. Curry:
        As a chemical engineer working extensively on process and thermodynamic control, there are no feed backs for the earth system. We are blessed to have an earth that is thermodynamically stable:The sensible heat exchanged and retained by the earth is virtually constant at all times even during climate change. The sensible heat retained by the earth can vary slightly during ENSO, AMO, or following volcanic events, but at the conclusion of these events, the sensible heat retained re-assumes the average, or constant, value. The earth will do whatever it takes to maintain the sensible heat exchanged and retained by the earth constant. Philosophically speaking, the earth will do whatever it takes maintain the precious gift of life, and it did for billions of years.

      • Thanks, Nabil, for a refreshing reminder of reality: “We are blessed to have an Earth that is thermodynamically stable.”

        Unfortunately “World leaders accepted Bilderberg’s decision that Earth’s heat source is thermodynamically stable” [1]. It is not!

        Experimental data and observations on the Sun were hidden, ignored or distorted for four decades (1971-2011).

        Climate change continued, impervious to efforts to direct the force that powers the Sun, the Earth, and the evolution of life [2,3].

        Current dilemma: The integrity of science was compromised for the benefit of society. Can the integrity of science be restored without damage to society?

        1. “The Bilderberg model of the photosphere and low chromosphere,” Solar Physics (1968) vol 3, 5-25

        http://adsabs.harvard.edu/full/1968SoPh….3….5G

        2. “Neutron repulsion,” The APEIRON Journal (2011) in press

        http://arxiv.org/pdf/1102.1499v1

        3. “Origin and evolution of life constraints on the solar model,” Journal of Modern Physics (2011) vol 2, 587-594

        http://dl.dropbox.com/u/10640850/JMP20112600007_31445079.pdf

      • curryja 9/6/11, 7:19 pm, Spencer Braswell III

        JAC: [C]loud feedback issue is much more complicated than portrayed by Dessler.

        The ISSUE of cloud feedback is horribly complex; but the EFFECTS of cloud feedback is understood on the simplest theoretical grounds . IPCC. As much as 100% error in climate sensitivity, a drastic error, can be caused by assumptions in cloud parameterization. IPCC found this problem, unresolved since the TAR, unsatisfactory.

        IPCC is not specific, but here is a probable translation of its simplest theoretical grounds . When SST rises, specific humidity rises (implemented in GCMs), and because the atmosphere has on average a surplus of CCNs (apparently not implemented in GCMs), cloud cover reliably rises (not implemented in GCMs), especially over the ocean, the primary shortwave absorber. When TSI increases, cloud burn-off increases, and cloud cover decreases (not implemented in GCMs). The increase in cloud cover with SST is a slow, negative feedback, mitigating warming from any cause (lower climate sensitivity). The burn-off effect is rapid (instantaneous on climate scales), and a positive feedback, amplifying TSI (not implemented in GCMs; see Stott, et al. (2003), Tung, et al. (2008)). And because cloud cover modulates TSI, it is the dominant and most powerful feedback in all of climate, and it is not modeled in the GCMs.

        To make matters worse, IPCC butchers cloud albedo as given in its baseline budget for RF due to Kiehl & Trenberth. It should be

        Cloud Albedo = Cloud Cover X Specific Cloud Reflectivity

        IPCC calls Specific Cloud Reflectivity (reflectivity per unit area) its cloud albedo, and then models cloud cover parametrically, apparently as a statistical constant. Without dynamic cloud cover, the GCMs estimate an open-loop climate sensitivity.

        IPCC has neither realistic models nor definitions for either feedback or cloud albedo. This is a result of trying to make the radiative forcing paradigm work. Feedback came into climatology (Hansen, Lacis, et al., 1984) from System Science (Bode, 1945). Feedback was, and still is, a physical signal (e.g., energy, force, displacement, material), or its representation as information, generated within any system and transported or communicated to alter the system inputs. IPCC approximates that definition in TAR and AR4 Glossaries, but doesn’t use it in the main body of either report.

        Instead IPCC defines feedback in two contradictory ways: as correlations between signals or processes, and as a parameter computed at run time. Dessler, Lindzen & Choi, Spencer & Braswell, and Trenberth, et al., each uses yet another definition of feedback. Lindzen & Choi’s representation is closest to the original, realistic definition, relying on forward and backward (feedback) transfer functions, which can’t be represented in radiative forcing. Unfortunately, L&C estimated feedback as a straight line, an unnecessarily restrictive assumption even in a linear model.

        The L&C and S&B models produced the expected, closed-loop reduction in climate sensitivity by fitting to satellite data. Their results are outside the limits claimed by IPCC, invalidating AGW. So Dessler (2011) came to the rescue, contradicting the simplest theoretical grounds. Dessler introduced another novel definition of feedback which showed that clouds don’t build up and burn-off as might seem obvious to laymen.

        L&C and S&B results are expected, but none of these sources implements feedback realistically. Because Dessler agreed with AGW, he could publish in one of the advocacy professional journals. Because L&C and S&B were different, they had to seek publication in nonconforming outlets.

        So goes post-modern science.

  3. Dr. Curry

    Thank you.

    I suspect your link to McIntyre is “well worth reading” mainly for the single line in the middle of it somewhere pointing out how little confidence emerges from the very limited data (which one presumes all three papers must share, though McIntyre only laments it of D11). I would hope to hear more statistical discussion, and discussion of how much more data would be needed to obtain meaninful results.

    If there’s anything else substantive there, please help me out, as sometimes McIntyre is not purely technical, so a chore to wade through.

    Your conclusion about the (low) technical importance of the three papers, the elusivity of how to do correctly what they seek to do badly, and the overly large attention to the papers compared the hype surrounding, hard to disagree with in any way, so long as the technical substance is all that’s at issue.

    Still, an utterly wrong conclusion of a weak paper may be sufficiently refuted by an only slightly stronger paper, which one suspects D11 will not be the last to do to SB11. That’s the relevant issue at hand, from my point of view.

    Hype + weak paper + refuted on rebuttal (sounds like Pons and Fleishmann all over again), and may signal a wider and deeper housecleaning for science journals and their peer reviewership is needed. Which Retraction Watch, I think, has been arguing for a while.

    • The McIntyre demolition of Dessler was amusing.

      “Given that the even the lagged relationship is weak, I’m reluctant to say that analysis using the methods of Dessler 2010 established a negative feedback, but it does seem to me that they cannot be said to have established the claimed positive feedback.

      Perhaps the editor of Science will send a written apology to Kevin Trenberth.”

      • Bruce

        So you couldn’t find anything else substantive in the McIntyre link, either?

        Thanks for confirming.

      • Bart , it was easy demolition for Steve to do …

      • Reanalyzing Dessler 2010 for lags, Steve McIntyre found:

        “Doing the same regression with 4-month lagged relationships (which both Dessler and SB agree to be more significant than the instantaneous relationship), the sign of the slope is reversed. Whereas Dessler 2010 had reported a slope of 0.54 +- 0.72 (2σ) W/m2/K, the regression with lagged variables is -0.90 +- 0.95 w/m2/K and has better diagnostics. . . .

        Dressler 2010 found r^2=0.01045. With 4 month lags,
        “Steve: Adjusted r^2 doubled :) to 0.02161″

        McIntyre’s 4 month lagged analysis of cloud feedback (per Spencer) versus Dressler’s (2010) instantaneous analysis finds:
        1) A 66% increase in slope.
        2) The OPPOSITE sign, and
        3) a doubled R^2 (yet still very low)!

        No wonder the American Statistical Association

        “recommends that more statisticians should become part of the IPCC process.”

      • correction: McIntyre’s quote ends at “has better diagnostics. . .”

      • Yes and wasnt SB’s main point that you cant measure feedback with this data because feedback and forcing effects are too intertwined?

        Doesnt the very weak R2 of Dressler’s positive slope, (and the fact it reverses using a four month lag) show that SB are right?
        Am I missing something here?

      • No, Bill. But the picture would be a lot clearer if the lag regression statistics started to at least look statistically meaningful.

    • Hi Bart,

      Such a moral quagmire you have fallen into. Proceeding from ignorance to calumny without a breath.

      Although I doubt that anything as fundamental as cold fusion is claimed. What is claimed is that that global warming cloud feedbacks can’t be determined this way because – shock horror – it is ENSO and clouds, water vapour, etc are involved. There is a corollary that the models aren’t perfect at ENSO in particular but also the NAO, AMO, NAM, SAM, IOD, THC, PDO, PNA – that is or should be no surprise. This is a big thing because we are told that the only way to understand climate is with supercomputers. Alternatively – that it is so simple that only the psychologically disordered could fail to believe.

      I suppose it must be hype then to suggest that the models are BS. ‘AOS models are therefore to be judged by their degree of plausibility, not whether they are correct or best. This perspective extends to the component discrete algorithms, parameterizations, and coupling breadth: There are better or worse choices (some seemingly satisfactory for their purpose or others needing repair) but not correct or best ones. The bases for judging are a priori formulation, representing the relevant natural processes and choosing the discrete algorithms, and a posteriori solution behavior.’ http://www.pnas.org/content/104/21/8709.full

      Well – first of all they are implausible as to formulation. ‘The accurate representation of this continuum of variability in numerical models is,
      consequently, a challenging and desirable goal. Fundamental barriers to advancing weather and climate prediction on time scales from days to years, as well as longstanding systematic errors in weather and climate models, are partly attributable to our limited understanding and capability to simulate the complex, multiscale interactions intrinsic to atmospheric and oceanic fluid motions.’ http://www.cgd.ucar.edu/cas/jhurrell/Docs/SeamlessModellingDraft03302007.pdf

      But that’s OK – as long as you can fall back on ‘a posteriori solution behavior’ and don’t have to tell anyone that it’s more like throwing non-linear darts at an almost infinite phase space.

      But where were we? Oh yes – there is nothing wrong with the SB11 data – CERES and HadCRUT3 – the treatment is simple and obvious and the results unremarkable. Other than in implying that the models are wrong of course.

      Rebuttal? Dessler makes a rookie error. He assumes that SST variability in ENSO is all ocean/atmosphere energy flux and uses a simple model to show that a, for instance, 3 W/m2 change in TOA in months makes absolutely no difference to climate. Clouds change a lot – but Dessler shows that this has absolutely no effect on surface temperature. Unless they are a global warming feedback that can be fed back into the models and justify a different and more calamitous a posteriori solution behavior.

      Global warming is not a scientific theory any more. It is a spaceship cult spread far and wide. From our perspective it has the usual characteristic of participants not being able to process non-conforming information. A frequent ending for spaceship cults involves cool-aid. Either that or the spaceship doesn’t come and most participants lose interest. The hard core simply change the schedule.

      Which will you be Bart – when the world does not warm?

      Robert I Ellison
      Chief Hydrologist

      • Chief

        Please. I clearly proceed from calumny to ignorance, as I always do.

        As AGW is the broadly accepted and unrefuted current hypothesis, your reasoning only works if you beg the question.

        McIntyre cut a wide swath, though a bit late, as everyone else including Dessler was already there, dismantling the statistical basis of D11, SB11, and LC too, equally due low R value.

        And then he droned on as he has all decade about how unfairly he’s been treated. Again. Still. Is being boringly iterative a Canadian thing?

        What he clearly missed was Dessler’s change-up, making the D11 case much less dependent on a weak trend in data and more on the reasoning used to interpret it.

        While you want to be arguing ENSO vs D10, the battlefield SB11 chose was ENSO vs Cloud; D11 merely made that plain. Every ENSO case you make, including arguing that the best ENSO-predicting _AGW_ models supports ENSO merely demotes the LC & SB cloud case more.

        I don’t set my watch by other’s spaceship schedules, or drink kool-aid. And since I don’t foresee the world warming but becoming less ergodic, and you know that about me, why won’t you ask which I will be when the world does not .. what is it a non-ergodic world does not?

      • Hi Bart,

        I was wondering if you had accidentally confused ergodic with ergotic and were actually walking backwards and only imagined that you proceeded from calumny to ignorance. Applying, in fact, the electric kool-aid acid test to the usual practice of proceeding from ignorance to calumny – as evidenced by the usual suspects around here.

        But as to your question. As far as I can figure out less ergodic is equivalent to more ergodic. Not in the sense of architecture – but in the way that the 2nd law of thermodynamics is only a statistical fact. The universe can be seen as a deck of cards that can be shuffled again and again until – just by chance – the original order is restored. So the entire universe must some day – by pure chance – return to the configuration it started in. I suspect by that time we will all be able to meet for tea and cake at a restaurant at the beginning of time – and watch the big bang happen again.

        I admit, however, that I am far more gone than that. Eschewing both the theory of evolution and entropy itself in the bigger picture of a 4 dimensional space/time continuum. You wonder then that I would dare to beg the AGW question? Although as I say – the latter seems more a spaceship cult than a viable scientific theory.

        As for entropy and evolution. ‘Since there exists in this four dimensional structure [space-time] no longer any sections which represent “now” objectively, the concepts of happening and becoming are indeed not completely suspended, but yet complicated. It appears therefore more natural to think of physical reality as a four dimensional existence, instead of, as hitherto, the evolution of a three dimensional existence.’ Albert Einstein

        Be that as it may – there are fundamental realities of ENSO that go beyond your alphabetic confusion. La Nina cool the planet and El Nino warms. They do this in 2 ways. With energy flux – to a cool ocean in a La Nina and from a warm ocean in an El Nino. The other way is as a modulation of radiative flux at TOA – up to 3W/m^2 in months.

        ENSO is technically a non-stationary and non-Gaussian phenomenon. It varies over decades to millennia. I will give you this one – http://www.clim-past.net/6/525/2010/cp-6-525-2010.html – because it’s a lot of fun. Have a look at Figure 5. The ‘red shift’ indicates El Nino. You can see many things in this graph. The drying of the Sahel 5000 years ago. The demise of the Minoan civilisation. Noahs flood – no only joking.

        I am afraid it is all over – bar the weeping and gnashing of teeth of the AGW spaceship culters. ENSO is ensuring that the world is not warming until at least the next Pacific climate shift due in a decade or three.

        Cheers
        Robert I Ellison
        Chief Hydrologist

      • Chief,
        thanks for this and your post above, for the links, and for not mentioning CO2 even once!

        ;-)

      • Chief

        Always amusing.

        When you decide to get back to Earth, will you apply that awe-inspiring mind of yours to issues nearer the problem at hand?

    • It’s not that big of a chore to wade through. What you see is a weird lack of statistical knowledge on the part of climate scientists involved. Thank god we didnt have to wait years to get the data to find out that there’s not much of interest here. Nothing worthy of a headline, or youtube video, much less a resignation.

      The salient points.
      1. These people have no discipline when it comes to ex ante reasoning about data selection. They pick datasets willy nilly to confirm or disconfirm what they want to. either limiting data to disconfirm what they desire or throwing a kitchen sink of data to spread the error bands and claim consistency.
      2. There is no care paid to standard metrics of significance. Mann’s legacy.

      • “Nothing worthy of a headline, or youtube video, much less a resignation.”

        This. And:

        “There is no care paid to standard metrics of significance. Mann’s legacy.”

        You have said it best, Mosher.

        It all makes me feel queasy watching this whole episode. But like a bad movie, sometimes you just can’t look away, as you can’t resist wanting to satisfying your curiosity over just how worst could it possibly get.

      • Mosh, I’m not sure which scientist you’re referring to exactly, but dont forget S&B’s abstract ended with

        “…It is concluded that atmospheric feedback diagnosis of the climate system remains an unsolved problem, due primarily to the inability to distinguish between radiative forcing and radiative feedback in satellite radiative budget observations.”

        Implicitely [in the period considered with the data we have]…which is an important result in itself.

  4. Dear Dr. Curry:
    Dr. Dessler’s paper agrees with my theoretical work-Clouds do not change climates. Please see lines 375 through 381, Article-12, Earth’s Magic, posted on my website http://www.global-heat.net.

  5. I think the peer review on these papers has been adequate. The back and forth in the journals is how this should play out. The story is the “hype” in the media and the apparent preferential treatment that “consensus” papers (usually) get in the review process (and which the S&B 2011 apparently got).

    • Dr. Curry

      You may be spot on.

      Overall, I’d much rather see a consistent, non-gatekeeping publishing standard that lets ideas duke it out on their merits in the wider world than private committees determining what I can or cannot see. In an ideal world.

      But I’ve also seen undertrained and unfit kids go to weigh-ins hoping for fights that would do them serious harm and give their sport a nasty reputation, so I see the value of rejecting unsuited candidates.

      The candidates seeking glory and fame out of proportion with their desserts lead to a lot of issues. If only so much of the hype of SB11 could not be lain equally at the author’s feet as at the media’s door, this would be a simpler case, and not an ethical bog, for example.

      Spectators who want blood on the mats, or who are loyal to their side regardless of the obvious weakness of their proponent, do us and themselves no favor, to extend the sport metaphor.

      If the apparent preferential treatment problem ever were addressed fairly — whether real as in the case of pharmaceutical papers sped through for commercial reasons or merely the complaint of delicate egos rejected or delayed by ‘lesser minds’ — one would hardly expect Climatology to be the field that got it right first.

      Maybe science has something to learn from sports.

      • McIntyre makes a damning point:

        “It does seem to me that it’s been an awful lot easier for Dessler to publish this comment than it is to publish criticisms of Team articles. As CA readers are aware, important results of Santer et al 2008 did not hold up with updated data, but Team reviewers refused to permit publication. CA readers are also well aware of Steig’s concerted efforts to block publication of O’Donnell et al 2010 (which appeared only because of Ryan O’Donnell’s remarkable persistence.)”

      • Bart R., Sorry but I failed to find a point in your otherwise entertaining sports comment.

      • David

        Underperformers should stick to the junior leagues, or have a stronger buddy and extra protective gear when they’re out of their depth.

        See Dallas’ keyboard-endangering, but very insightful, suggestion below.

        I speak of the papers in this case.

        The promoters who overhype the papers like rabid hockey dads and soccer moms, perhaps ought be seated further from the press box in order to embarass their kids less.

      • This is just awful for your side Bart. A total humiliation.

        http://climateaudit.org/2011/09/08/more-on-dessler-2010/#comments

    • “and which the S&B 2011 paper failed to get”
      What is the evidence for that? Wagner’s hindsight complaint is effectively that it did get “unintentional” preferential treatment.

      • oops, my comment got mangled, it is fixed

      • I think you just proved their point.

      • Where is wagner’s proof of that? He claims the reviewers were “skeptics”
        Really? I’d like to see some proof. He claims major flaws in the paper? really? retract it and cite the error. There’s a normal process for handling this, let someone publish a criticism. Personally, I think there failure to show all the models demands at least a correction. Wagner should have pulled up his big boy pants and let science take it’s course. And knowing that a rebuttle was forthcoming his resignation is all the more stupid.

      • Wagner’s resignation issue is not that the quality of the paper. It is that the reviewers and the process left him in an impossible position.

        In the MDPI system, reviewers are selected by the Editorial Offices. They select the referees and conduct the subsequent interchange. Only when that process is deemed complete does the Editor-in-Chief (Wagner) get invited to look it over and decide whether to approve. And where do the Offices (in this case, Mr Elvis Wang) find referees? All we know is that he has a list of five suggestions from the author.

        Wagner finds after publication that all hell breaks loose, and people are saying “All this was refuted in Trenberth and Fasullo 2010″. So he checks and finds that neither the paper nor the reviewers mention that paper. I think he found that lending his name and reputation to that kind of snafu is just not what he wants to do.

        Papers are not withdrawn in these circumastances – that’s for issues of misconduct etc. A correction might be appropriate. But Wagner has had enough.

      • Good points.

        1) “Only when that process is deemed complete does the Editor-in-Chief (Wagner) get invited to look it over and decide whether to approve. ”

        And approve it he does.

        2) “Wagner finds after publication that all hell breaks loose”

        Obviously the problem was not the paper he approved, it was the “hell” raining down on him from Trenberth who can destroy his career.

        Spot on Nick. The paper was not the problem. It was Trenberth.

      • Sorry,

        I fail to see how this is refuted at all by Trenberth’s paper. I read it, I see
        no references whatsoever to this paper, its data, or it’s methodological approach. Dressler’s rebuttal at least advances the ball down the path
        of comparing models with observations.
        I fail to see how this put’s him in an impossible situation.
        I fail to see how a simple correction is so impossible.
        Impossible how? logically impossible? nope. physically impossible?
        nope? morally impossible? nope? uncomfortable, yes, difficult?
        sure? impossible? how? He had numerous avenues open to him, but he selected one which

        1. created a worse situation, not a better one
        2. created the kind of publicity he decried
        3. raised the appearance of impropriety and collusion
        4. gave more promanence to a suspect paper (SB11)
        5. doubled down the damage by apologizing to Trenberth who
        was not harmed.

        You would not have resigned. And one does not need to defend stupid human actions to believe that C02 warms the planet. Stupid, unecessary, theatrical stunts do more damage to the science than SB11. SB11 was already being answered. Why cloud the water with politics and personality. Makes no sense whatsoever. I can hear Wigly writing, as he did in the climategate mails .. we should just let this play out in the journals. His wise advice was almost never listened to.

      • Impossible position? You mean his day job? I would agree, hence the nuzzling at Trenberth’s heels. No question who the alpha is in that relationship.

      • Steven,
        It doesn’t matter whether you thought Trenberth’s paper is a refutation. Enough people thought that at least S&B had something to deal with there, and said so loudly. It certainly wasn’t just Trenberth. When Fox News takes it up, private messages from a scientist far away are a minor issue for him (if Trenberth did contact him – no-one has given any evidence for that).

        And Wagner’s problem was that he was caught flat-footed. The paper didn’t deal with it, and the referees hadn’t told him about it.

        I wouldn’t have resigned? Well, I wouldn’t have been there. I think the MPDI deal is impossible for E-i-C’s. They are sitting ducks.

      • Wagner specifically mentioned Spencer’s post-publication antics. (My word.) Spencer has stated that he sees muting political action over AGW as his job, and he over-spun the significance of the paper. Wagner’s inattention or lack of understanding of the game afoot gave a politician (Spencer) a stage for his propaganda and so he resigned.

        Did I miss Mosher’s condemnation of Spencer’s politicization of the science?

      • It’s pretty simple Jeffrey. I believe in AGW. I think the best way to get people to accept the reality is to speak the truth and keep your nose clean. That means share your data, share your code, and leave the political fights to others. I really don’t care to hold skeptics accountable for their bad behavior. I’d rather correct their science mistakes. Although I have been known to tear them a new one. Just ask. I spent most of of my time on WUWT explaining to people why gavin had made good counter points to Rays paper and what ray should do to answer those questions. Basically, I cannot convince skeptics to clean up their PR act. I can’t because they dont listen to someone who believes in AGW. Apparently I can’t convince anyone who believes in AGW that we should simply not engage in this sort of nonsense.

      • He didn’t claim they were skeptics Mosher. He said:

        “the editorial team unintentionally selected three reviewers who probably share some climate sceptic notions of the authors.”

        Notions? Was Trenberth annoyed some of them think the Sun has a role?

        Some? One “sceptic notion”? Two? How many is “some”.

        “Unintentionally selected” as in “OMG they didn’t have DENIER stamped on their cv’s and we picked them”?

        That sentence could have been written by The Onion.

        Maybe Wagner should have been honest and said the reviewers had skeptic cooties and it was all yucky.

    • Judith writes “The story is the “hype” in the media and the apparent preferential treatment that “consensus” papers (usually) get in the review process (and which the S&B 2011 apparently got).”

      Dont forget to mention the other aspect to the new weapon in the AGW arsenal of fast review …and that is fast review in a different journal with a new paper…and that, as I understand it, means they are guaranteed to have the last word on the subject against their own paper.

      I dont think I’ve seen this mentioned but I’ve sure as hell seen references to how S&B “played the system” to get their paper through. Well as far as I can see, S&B had no influence over who Remote Sensing chose as their reviewers. Perhaps they did somehow but if there is a link, nobody has spotted it yet.

      Compare this to the documented links between Wagner and Trenberth.

  6. “So how to do a sensible observational analysis of cloud feedback remains elusive.”

    Get a pair jet planes, equip them with SW/LW spectrophotometers, one looking up and one looking down, A couple of cameras would be nice too.
    Wait for a solar eclipse. Have the first planes fly ahead of the unbra and the second fly through the totality.
    You will get data for 35-40 minutes at mach 0.85.
    You will be able to measure the two photonic fluxes and the spectrum, from ground level up and from atmosphere down.
    During the flights you will know if there are clouds or not.
    It is not difficult.

  7. As a retired health professional, I cannot believe that ‘climate scientists’ are doing formal statistical regression on those data. One glance at the scattergram tells instantly that there is absolutely NO correlation (beyond pure chance) between the two variables. This is borne out by the pathetic value of the adjusted R squared metric = 0.01045 ? They are joking aren’t they?
    What sort of peer-review and editorial process allows this rubbish to be published in any science journal worthy of the name?

    • That’s climate science. If satellite data and balloon temperature data don’t agree with your model, estimate the wind speed with balloons and calculate your own temperatures from the wind speed. Simple.

  8. Norm Kalmanovitch

    Albedo measurements made by Project Earthshine show that the albedo decreased to 1998 and has increased since.
    Roy Spencer’s UAH MSU data shows that the global temperature increased to 1998 but has not increased since.
    Hadley HADCRUT3 data shows that the global temperature increased to 1998 but has not increased since and has actually decreased since 2002.
    The RSS MSU data also shopws that the temperature increased to 1998 but has not increased since and like the HadCRUT3 data this dataset also shows cooling since 2002.
    Since a reduced albedo lets in more energy causing an increase in global temperature and an increased albedo reduces the amount of energy coming in which causes a decrease in global temperature and these four datasets and even Hansen’s GISS dataset all show that warming ended after 1998 with no warming since in spite of the steady 2ppmv increase in atmospheric CO2 concentration the only intelligent conclusion is that the albedo manifested by cloud cover is having a far more dominant effect on global temperature than CO2.
    When theory disagrees with data scientists honour the data and discard the theory but politicians and lawyers by profession attempt to promote the theory by discarding the data or at least having it removed from the argument (usually by discreditingn the witness).
    Is this a science or political blog?

    • If albedo really has increasd since 1998, correlated with the stoppage of warming, it seems to be really important to know why.

      If it is mostly the increase in sulfate emissions from China, that would certainly show how important it is to consider rates of sulfate reduction going forward.

      If it is mostly some natural cycle causing increased clouds, or increased cloud albedo, we need to know why. This is what all these papers address, but they do so in the absence of discussing increased sulfate concentrations. Is that sensible?

      If it is a change in patter of cloudiness — more over dark oceans, less over land, for example — that would be important to know.

      So I find Norm’s contribution on point. Judy? Steve? Your thoughts?

  9. Thought I would post this again here – with a correction thanks to Nick.

    Dessler (2011) uses the simple model from SB11.

    Cp dΔT/dt = S(t) + N(t) − λΔ (1)

    ‘Equation (1) states that time-varying sources of non-radiative forcing S and radiative forcing N cause a climate system with bulk heat capacity Cp to undergo a temperature change with time away from its equilibrium state (dΔT/dt), but with a net radiative feedback ‘restoring force’ (−λΔT) acting to stabilize the system.’ SB11

    Dessler finds that σλΔ < 0.4 W/m2 and σN(t) = 0.5 W/m2. These are measures of magnitude.

    ‘To evaluate the magnitude of the first term, C(dTs/dt), I assume a heat capacity C of 168 W-month/m2/K, the same value used by LC11 (as discussed below, SB11’s heat capacity is too small). The time derivative is estimated by subtracting each month’s global average ocean surface temperature from the previous month’s value. Temperatures used in this calculation come from NASA’s Modern Era Retrospective analysis for Research and Application (MERRA) [Rienecker et al., 2011]. The standard deviation of the monthly anomaly time series, σ(C(dTs/dt)), is 9 W/m2′. (Dessler 2011) Thus:

    Cp dΔT/dt ≈ S(t)

    The σ(Cp dΔT/dt) term is large and dominates the equation. This suggests a predominant role for energy flux between the ocean and atmosphere in ENSO temperature changes.

    The problem emerges from the fact that sea surface temperature (SST) change is dominated by ENSO ocean dynamics and not by atmosphere/ocean coupling. The changes in SST are assumed by Dessler to be the result of non-radiative forcing only and as a result vastly over estimates this term.

    Equatorial Pacific Ocean sea surface temperatures have an origin in the abysmal depths. Deep oceanic currents are driven by thermohaline circulation and by the rotation of the planet. The deep currents interact with a sun warmed surface layer that is a hundred or more metres deep. Deep ocean currents occasionally push through the warm surface layer in the south eastern Pacific in one of the major areas for upwelling on the planet. Upwelling subsurface water is both frigid and rich in nutrients leading to booms and busts in biological activity affecting fisheries, mammals and birds off the Pacific coast of South America. This area is designated as Large Marine Ecosystem (LME) No. 13, is amongst the most productive environments in the world and is known as the Humboldt Current. A good introduction is provided by National Ocean and Atmospheric Administration (NOAA) at their LME13/Humboldt Current web page.

    The thermal evolution of the Humboldt Current is best understood in terms of ENSO. ENSO is an oscillation between El Niño and La Niña states over a 2 to 7 year cycle. An El Niño is defined as sustained SST anomalies greater than 0.5O C (in the Nino 3 region) over the central pacific. Conversely, a La Niña is defined as sustained SST anomalies less than -0.5O C. The oscillations are driven by complex interactions of cloud, wind, sea level pressure, sea surface temperature, planetary rotation and surface and subsurface currents. The short explanation is that the Pacific trade winds set up conditions for a La Niña. Trade winds, south-easterly in the Southern Hemisphere and north-easterly in the Northern Hemisphere, pile up warm surface water against Australia and Indonesia. Water vapour rises in the western Pacific creating low pressure cells that strengthen the trade winds piling yet more warm water up in the western Pacific. Cool, subsurface water rises in the eastern Pacific and spreads westward. At some point the trade winds falter and warm water spreads out eastward across the Pacific.

    The other error is less critical. Dessler manipulates the data to get cloud radiative anomalies. Bur because we are talking about ENSO dynamics involving wind, cloud, temperature, water vapour and cloud – the use of the all sky flux as used by SB11 is more appropriate when looking at a suite of changes and not simply cloud.

    ‘From these data, I extracted the part of ΔRall-sky caused by changing clouds, hereafter referred to as ΔRcloud. To do this, I took cloud radiative-forcing anomalies (ΔCRF) and adjusted those to account for the impact of changing temperature, water vapor, surface albedo, and radiative forcing, ultimately yielding ΔRcloud. ΔCRF is the change in TOA net flux anomaly if clouds were instantaneously removed, with everything else held fixed, and it is determined by subtracting ΔRall-sky, obtained from CERES measurements, from the clear-sky flux anomalies ΔRclear-sky, obtained from a reanalysis system.’ (Dessler 2010)

    ‘El Niño/Southern Oscillation (ENSO) is the most important coupled ocean-atmosphere phenomenon to cause global climate variability on interannual time scales. Here we attempt to monitor ENSO by basing the Multivariate ENSO Index (MEI) on the six main observed variables over the tropical Pacific. These six variables are: sea-level pressure (P), zonal (U) and meridional (V) components of the surface wind, sea surface temperature (S), surface air temperature (A), and total cloudiness fraction of the sky (C). These observations have been collected and published in ICOADS for many years.’ http://www.esrl.noaa.gov/psd/enso/mei/

    The change in SST in ENSO is not a result of energy moving between ocean and atmosphere – but primarily comes from frigid upwelling (or not) in the eastern Pacific and resultant changes in Hadley and Walker circulation.

    • Dear ChiefHydrologist:
      Equation (1) is incorrect and cannot calculate climates. It is a differential equation of the first order and its solution disagree with observations. The incorrectness of Equation (1) is clear when the equation is solved and examined. The equation suggests that, after carbon dioxide forcing levels off following a warming trend, surface temperature continues to rise and tends asymptotically to a higher equilibrium temperature. Antarctic Ice Core Data shows that this is not the case. Carbon dioxide leveled off twice between 14,000 BP and 11,000 BP. After carbon dioxide had leveled off, surface temperature did not continue to rise asymptotically as Equation (1) suggests, but decreased and leveled off at a lower average surface temperature. This is a fundamental disagreement between observations and what Equation (1) suggests, and demonstrates that Equation (1) is inadequate for calculating climates and earth’s energy balance.

    • Your “0.5O C” looks a bit confusing; if you want the degrees symbol, Alt-248 works: ° . So “0.5°”.

    • CH,

      I thought I’d repost my response to you again.

      Spencer’s S(t) is SUPPOSED to be the ENSO-dominated non-radiative energy addition. I share your concern however that Dessler has maximised the calculation of the amplitude of its flux contribution, and minimised the same for radiative forcing, but I think you have missed the how.
      Recall that the key issue for Dessler 2010 is actually whether or not there is a significant radiative forcing present or not – ANY radiative forcing. Spencer has shown, and this part is not controversial, that the presence of any radiative forcing decorrelates the flux response to temperature; more specifically any radiative forcing would make Dessler’s 2010 zero-lag flux-temperature regression less useful than a bag of horse manure. If Spencer’s N(t) term is significantly non-zero, then Dessler 2010 is wallpaper. So logic might suggest that, if he had a good case, Dessler should have been focused on proving that there is no evidence of ANY radiative-forcing in the data. If he succeeds in showing that there is negligible radiative forcing present, then Dessler 2010 is defended, and he can move on and easily address and dismiss the secondary question – whether or not there was a CLOUD forcing component involved.
      Dessler, however, chooses to focus on a different question, (misrepresenting Spencer’s argument in the process) – namely whether clouds cause temperature change (exclusively) or temperature changes cause clouds (exclusively). To test this question, Dessler starts with the UNSUPPORTED ASSUMPTION that there was zero radiative-forcing over the period, with the possible EXCEPTION of this absurd cloud forcing proposed by that fool Spencer.
      This allows him to misrepresent Spencer’s model by substituting the cloud forcing term (DeltaRCloud) for the total forcing term in the original model. This is important because this expression has meaning in its own right and it is not the same as the total forcing, the amplitude of variation of which must of necessity be larger.
      His next step then is to calculate the relative contributions of the total radiative forcing relative to the non-radiative flux addition whoops I mean the relative contribution of the DeltaRCloud to the non-radiative flux addition. This is now where the previous substitution comes in very handy. He substitutes an estimate of DeltaRCloud which comes directly from … Dessler 2010. It is abstracted directly as far as I can see from the net flux calculation with confounded feedback and forcing. Dessler estimates the standard deviation of this DeltaRCloud “radiative forcing” to be 0.5.
      He makes the sd of the LW feedback term approximately equal to DeltaRCloud by allowing the feedback parameter to vary as well as the temperature for reasons which are totally unexplained. This then means that all of the OHC gain – close to 100% – can be attributed to non-radiative forcing or ENSO in this case. Hence by assuming at the start that there is no radiative forcing component, he ends up proving that ..er.. there is no radiative forcing component.

      • Hi Paul,

        Spencer’s S(t) is SUPPOSED to be the ENSO-dominated non-radiative energy addition. I share your concern however that Dessler has maximised the calculation of the amplitude of its flux contribution, and minimised the same for radiative forcing, but I think you have missed the how.

        Yes S(t) is he so called non-radiative energy flux between the ocean and atmosphere.

        Recall that the key issue for Dessler 2010 is actually whether or not there is a significant radiative forcing present or not – ANY radiative forcing. Spencer has shown, and this part is not controversial, that the presence of any radiative forcing decorrelates the flux response to temperature; more specifically any radiative forcing would make Dessler’s 2010 zero-lag flux-temperature regression less useful than a bag of horse manure. If Spencer’s N(t) term is significantly non-zero, then Dessler 2010 is wallpaper. So logic might suggest that, if he had a good case, Dessler should have been focused on proving that there is no evidence of ANY radiative-forcing in the data. If he succeeds in showing that there is negligible radiative forcing present, then Dessler 2010 is defended, and he can move on and easily address and dismiss the secondary question – whether or not there was a CLOUD forcing component involved.

        Dessler is arguing that there is a cloud feedback – but that it is small when compared to the ocean/atmosphere flux and will become significant over time.

        Dessler, however, chooses to focus on a different question, (misrepresenting Spencer’s argument in the process) – namely whether clouds cause temperature change (exclusively) or temperature changes cause clouds (exclusively). To test this question, Dessler starts with the UNSUPPORTED ASSUMPTION that there was zero radiative-forcing over the period, with the possible EXCEPTION of this absurd cloud forcing proposed by that fool Spencer.

        This allows him to misrepresent Spencer’s model by substituting the cloud forcing term (DeltaRCloud) for the total forcing term in the original model. This is important because this expression has meaning in its own right and it is not the same as the total forcing, the amplitude of variation of which must of necessity be larger.

        Yes I think that the total net flux rather than the Dessler calculated CRF is more reasonable because we are looking at ENSO. That is a different problem to Dessler’s cloud feedback problem.

        His next step then is to calculate the relative contributions of the total radiative forcing relative to the non-radiative flux addition whoops I mean the relative contribution of the DeltaRCloud to the non-radiative flux addition. This is now where the previous substitution comes in very handy. He substitutes an estimate of DeltaRCloud which comes directly from … Dessler 2010. It is abstracted directly as far as I can see from the net flux calculation with confounded feedback and forcing. Dessler estimates the standard deviation of this DeltaRCloud “radiative forcing” to be 0.5.

        He makes the sd of the LW feedback term approximately equal to DeltaRCloud by allowing the feedback parameter to vary as well as the temperature for reasons which are totally unexplained. This then means that all of the OHC gain – close to 100% – can be attributed to non-radiative forcing or ENSO in this case. Hence by assuming at the start that there is no radiative forcing component, he ends up proving that ..er.. there is no radiative forcing component.

        The radiative change should include water vapour – and this would make the total radiative standard deviation about double.

        Thus the σ(Cp dΔT/dt) term still dominates by a long way – and there is no reason to doubt that calculation. It is so large because there are ocean circulations that drive SST and it not primarily a ocean atmosphere issue. So you would reduce the σ(Cp dΔT/dt) to a value that is the ocean/atmosphere flux and not the value that includes ocean circulation. How to do that is another question.

      • Hi Chief,
        “The radiative change should include water vapour – and this would make the total radiative standard deviation about double.

        Thus the σ(Cp dΔT/dt) term still dominates by a long way – and there is no reason to doubt that calculation.”
        Dessler just made a silly schoolboy mistake when he spoke about the water vapour “forcing” affecting this term. Radiative feedback from water vapour should have no impact on this term. On the other hand, any variation in solar, aerosols, GHG forcings, contrails, GCR-modulated cloud formation, all non-temperature dependent cloud variation – should all have been included in the radiative forcing term to make any sense in his calculation step for the non-radiative (ENSO) forcing. He assumes that all of the above are zero except for non-temperature dependent clouds, but he then estimates the magnitude of the cloud forcing from a NET FLUX term, one of the key things that Spencer is arguing cannot be done.

        I also note that he selected a large value of Cp, (despite correlative data to suggest that this is too large to include in Spencer’s model or the GCMs) in order to boost the value of σ(Cp dΔT/dt) and hence the value of the non-radiative contribution relative to the radiative forcing contribution.
        This part of Dessler’s analysis is just very very poor science.
        .

    • Chief writes,

      “Equatorial Pacific Ocean sea surface temperatures have an origin in the abysmal depths.”

      Just how abysmal are those depths Chief? (sorry couldn’t resist)

  10. Utilizing Argo data for ocean heat transport, his own previous data on radiative flux changes associated with cloud variation, and a plausible range of values for loss of heat to space, Dessler calculates that almost all surface temperature fluctuations can be attributed to ocean heat transport fluctuations. As a result, even if it were assumed that all the flux changes associated with clouds were a forcing rather than a feedback, the contribution to surface temperature changes would be very small. His terminology was confusing in that he referred to “energy trapping” by clouds, but it seems clear from the text and equations, as well as his treatment of cloud data in his 2010 paper that he included short-wave albedo-related flux changes as well as longwave IR absorption in his estimates of cloud effects.

    I find this reasoning plausible in the abstract, but also limited by uncertainties in all the above mechanisms – ocean heat transport, surface temperature, and cloud radiative effects as the difference between clear sky and all sky fluxes, etc. His argument thus reduces to saying that if cloud forcing exists, it’s small, and the counterargument would be that if it exists, its magnitude is still uncertain.

    Dessler then goes on to say, in essence, “Well, none of this matters anyway, because we already know that ENSO temperature fluctuations arise in the oceans and that excludes cloud forcing as a mechanism, and so we can use the cloud data to estimate feedbacks, as was done in the 2010 paper”. My concern about this latter point is that it is “refutation by fiat”, and is vulnerable to the criticism that maybe what “we already know” is wrong, as SB-11 claims.

    To some extent, my reservations about D-11 are similar to those about SB-11. It is not enough to claim that a different interpretation is wrong. To write a publishable paper that purports to document that conclusion, one must show that one’s own interpretation can explain observations better than the alternative. For reasons I’ve outlined over the past few days, I suggested that all the SB-11 data can be fully explained by our conventional interpretation of ENSO, with no need for cloud forcing, and that some of the data are better explained by the conventional interpretation (e.g., the magnitude of the temperature response over short intervals). I concluded that in the (perhaps unlikely) event that Spencer and Braswell are right in rejecting the standard interpretation, their paper failed to provide any evidence as to why it should be rejected, and they should not publish claims to have rejected it until they have adequate evidence.

    I believe Dessler has made a few valid points on the margins about the SB-11 paper (and also Lindzen/Choi, which I didn’t address here), but I don’t think he has shown SB-11′s theorizing to be wrong. I’m therefore unconvinced that either paper provided enough evidence for its conclusions to deserve publication, and even less convinced that either deserves much public attention.

    • Fred, you are pretty much right. Both papers are nothing but replies not deserving the status of “peer reviewed” publications. RP Sr. said as much then said since GRL doesn’t accept comments anymore, they should fast track a SB reply. There will be an unending stream of mediocrity. Dressler and Spencer should co-write a paper to get on the same page if they can find a stats guy with poor hearing.

    • “It is not enough to claim that a different interpretation is wrong. ”

      I disagree. Dessler 2010 claims an instantaneous response, SB 2011 says a 4 month lag is better and the Dessler 2011 data confirms.

      “Whatever view one might take on the differences between observations and models in the above data, the lagged relationship is more significant than the instantaneous relationship – a point shown in both the figures in Spencer and Braswell 2011 and Dessler 2011. This suggests that the original scatter plot in Dessler 2010 should be re-done using a lag of 4 months. I used the common HadCRUT3 data for the comparison – Dessler had observed that this accentuated the difference between models and observations, but it is nonetheless widely used and, if Dessler takes exception to SB’s failure to illustrate re-analysis temperature versions, one might make the same observation about the HadCRU3 omission in Dessler 2010. The results are shown below.

      Doing the same regression with 4-month lagged relationships (which both Dessler and SB agree to be more significant than the instantaneous relationship), the sign of the slope is reversed. Whereas Dessler 2010 had reported a slope of 0.54 +- 0.72 (2σ) W/m2/K, the regression with lagged variables is -0.90 +- 0.95 w/m2/K and has better diagnostics.

      Given that the even the lagged relationship is weak, I’m reluctant to say that analysis using the methods of Dessler 2010 established a negative feedback, but it does seem to me that they cannot be said to have established the claimed positive feedback”

      http://climateaudit.org/2011/09/06/the-stone-in-trenberths-shoe/#comments

      This advances science. It says Dessler 2010 was going down the wrong track.

      • Dessler-2010 stated that his results did not exclude a weak negative cloud feedback, although they were more consistent with a positive feedback.

        Regarding a lagged relationship rather than a contemporaneous one, if we are interested in the effect of temperature on cloud radiative flux perturbations as a measure of feedback, I don’t think a lag makes much sense. Clouds form and disappear in response to temperature changes within a period of days, and so an attempt at correlation would appropriately use contemporaneous relationships. If instead, we wanted to know how cloud-mediated radiative perturbations affected temperature, a lag would be necessary because of ocean thermal inertia. The problem, here, though, is that with ENSO constituting alternating El Nino and La Nina events and their corresponding fluctuations in ocean surface temperature, it’s almost impossible to interpret what is leading or lagging what else. A 4-month lag may show the greater statistical significance, but it will be hard to interpret it in terms of causality.

        I think that Dessler’s relatively tentative conclusions based on his own approach were not unreasonable.

      • I think S & B 2001 and Dellser 2011 has refuted Dessler 2010.

        The point of smearing SB 2011 was to keep Dessler 2010 in AR5. If it makes into AR5, AR5 will be an even bigger laughingstock.

      • A 4-month lag may show the greater statistical significance, but it will be hard to interpret it in terms of causality.

        Good point. Does a cosine wave lead or lag a sine wave? When signals start overlapping to a great extent, determining causality gets that much harder.

      • La Nina cool the planet and El Nino warm. They do this is 2 ways. By a simple energy flux – to a cool ocean in a La Nina and from a warm ocean in a El Nino. The other is by cloud variation – more low level stratiform cloud in the tropical Pacific in a La Nina and less in a El Nino. The TOA flux changes don’t have a lag – as one would expect. More cloud reflects more SW and vice versa.

        The temperature changes lag the SOI by 5 to 7 months – http://members.iinet.net.au/~glrmc/McLean_deFreitas_Carter_JGR_2009.pdf

        There can be no overlapping and the chain of causality is just fine thank you very much.

      • That McLean data is pretty impressive; no way will that kind of match occur unless there is an underlying common factor.

        The SOI is an obviously computed value. Is this computation based on temperature in any way? Let’s check:

        The SOI is used here is the Troup SOI, which is the standardized anomaly of the seasonal mean sea level pressure difference between Tahiti and Darwin, divided by the standard deviation of the difference and multiplied by 10.

        I guess not, as they use the barometric pressure. Gradients in temperature lead to gradients in pressure. Kind of neat and in accordance with gas fluid dynamics. Classic paper all around.

        Now all we need to do is find out the effect of CO2 on the mean temperature.

      • Fred writes “A 4-month lag may show the greater statistical significance, but it will be hard to interpret it in terms of causality.”

        But S&B are saying that its a bi-directional causal relationship and from what I can tell, that appears to fit the data much better than saying causality is in one direction only as Dessler is saying.

        Add thermal inertia, seasons, a rotating earth and ocean currents and suddenly you have a plausible theory.

      • Tim – There is no doubt that the temperature/cloud relationships are bidirectional – changing temperatures affect clouds which affect temperature. Each of these phases can be expected to require time to reach a peak. This is not evidence that clouds initiate the temperature changes rather than responding as a feedback, with its own effect on temperatures.

      • Thats because mainstream thinking hasn’t even considered the possibility that a lag exists that could enable it. Dessler simply doesn’t believe it and the rest of the climate establishment ignore Spencer (and his ideas) on principle.

        Are you really so sure that for example ocean currents that transport various amounts of cloud moderated energy from one location to another and hence change clouds in that location cant force climate?

        And then there is the whole Svensmark idea which is in its infancy.

        Answer yourself this. What is “natural variability” and how does it work over the timescales of decades and centuries such as in the MWP and LIA?

      • Robin Melville

        The “better diagnostics” double the correlation coefficient from 0.01 to 0.02. Neither paper demonstrates anything more than random noise. As does the debate I’m seeing here.

    • Fred
      Thanks for your observations. Here are my onlooker thoughts.
      1) I do not see where Dessler makes any mention of solar/cosmic ray cloud modulation which would provide an external forcing on lower altitude clouds, and thus modulating albedo that would cause a cooling not a warming. Consequently his conclusions are argumentum ignoratiam.
      2) Aside from giving some evidence that ENSO dominates, could not Dessler’s observation “that clouds did not cause significant change over the last decade” be equally valid for “CO2 did not cause significant change over the last decade”, seeing that CO2 increased but temperature did not?
      OR conversely:
      3) “The warming effect of CO2 has been countered by the cooling impact of clouds resulting in no change in temperature”?
      By ignoring both CO2 and external Solar/cosmic ray forcing, I find his arguments insufficient.
      (Though not mentioned, I presume CO2 may be included within the model calculations, since the larger water vapor of 0.5 W/m2 is mentioned.)
      The differences in Figure 2 between most models and the data is remarkable large at 4+ months and -12 months. All the models are outside the +4 month lag HadCRUT3 temperature used by S&B11. Dessler shows three ENSO good models that are within the uncertainty bounds of his selected temperature/flux variations, but are outside the temperature series used by S&B 2011. However the other ten models are outside (below) even the statistical boundaries of the temperatures Dessler uses at +4 months.

      That does not seem to give high confidence that “observations by LC11 and SB11 are not in fundamental disagreement with mainstream climate models”.

      Furthermore Steve McIntyre shows that using Spencer’s 4 month lag invert’s Dessler (2010)’s results, giving the opposite sign, twice the slope and twice the significance – which at 0.02 is still so small that McIntyre concludes:

      Given that the even the lagged relationship is weak, I’m reluctant to say that analysis using the methods of Dessler 2010 established a negative feedback, but it does seem to me that they cannot be said to have established the claimed positive feedback.

      Dessler opines: “over the decades or centuries relevant for long-term climate change, on the other hand, clouds can indeed cause significant warming” – with no evidence cited here. NIR Shaviv shows substantial evidence for the opposite conclusion of correlations between galactic cosmic rays and temperature over millenia to millions of years.
      Cosmic rays and climate etc.

      I agree that Dessler (2011) conclusions of ENSO over clouds are suggestive but “Not proven” over this short period.

    • Fred,
      A good comment.
      I was not impressed at all by the quality of Dessler’s main argument (see my response to Chief Hydrologist above), but I am left with a serious nagging question. If it is true (as you have suggested several times) that the shape of the lag regression does not per se demonstrate the presence of a radiative forcing component in the data, why could Dessler not have merely produced an AMIP model run over the relevant 2000 – 2010 period showing that such a shape can be achieved even with negligible radiative forcing? Why produce model comparisons which do not address one of the central planks of Spencer’s argument?

      • Paul – Is your point about Dessler’s Figure 2 that the models did not specifically cover 2000-2010? That would have been interesting, and Trenberth and Fasullo did address interdecadal variability. My interpretation is that models (all without “internal radiative forcing” as far as I know) do duplicate the general shape of the observations but tend to understate the magnitude of the variation in regression slopes, with some coming closer to the observations than others. ENSO is still one of the more difficult climate phenomena for models to simulate I think Trenberth or Dessler would argue that there are an enormous number of climate elements in ENSO that the models don’t yet have quite right but are getting closer, and that the existence of remaining disparities don’t validate an alternative mechanism that relies on an unidentified forcing mechanism, an unrealistically shallow ocean mixed layer and an unrealistically strong and rapid temperature response to a small flux change in order to eliminate the disparities. The intermodel variation is probably greater than the disparities between the better-performing models and ENSO observations, which makes it hard to where the remaining problems lie. I don’t know enough about the relevant parametrizations to be more specific, but a modeler’s comments on this would certainly be welcome.

        Finally, though, it’s my impression that the understanding of ENSO events as originating in changes in atmospheric and ocean currents that cause a shifting of ocean heat up or down as the main mediator of surface temperature changes, with subsequent TOA flux changes as responses rather than initiators, is not based on these models

      • Fred,

        You don’t do justice to ENSO – http://www.duke.edu/~ts24/ENSO/

        Here is an animation for August.

        http://www.ncdc.noaa.gov/oa/climate/research/sst/ani-weekly.html

        Note the cool water rising in both the north and south Pacific – in the typical V of a cool Pacific decadal mode.

        Current and archived satellite images can be found here – http://www.osdpd.noaa.gov/ml/ocean/sst/anomaly.html

        Why does it seem to have a solar signature? SAM was unleashed in July with a -1.52 SAM Index. The SAM responds to warming and cooling of ozone by solar UV in the stratosphere. A negative SAM pushes storms spinning off the polar front into lower latitudes and cold Southern Ocean water accumulates off the South American coastline.

      • @Fred,
        “Is your point about Dessler’s Figure 2 that the models did not specifically cover 2000-2010?”
        Yes. I find it very very curious. Spencer’s claim that the “form” of the lag regression proves the exisitence of a radiative forcing component could have been easily countered by (a) pick any single AMIP model which DOES have a good ability to match ENSO, (b) specify the Ts values, (c) run it out for 2000-2010 with no exogenous forcing, and (d) show that the “form” of lag regression plot was reasonably consistent with the lag regression plot obtained from the observational data. One model would suffice to challenge Spencer’s logic. Why did he not do this? My suspicion is that he may not be able to – and this leaves the main S&B11 conclusions untouched.

        Instead of this direct demonstration of the flaw in Spencer’s thinking, the logic he (Dessler) is applying is as follows:-
        Spencer says that there is a forcing present and that you cannot decouple forcing and feedback from (just) the net flux and temperature series. OK, then to prove him wrong let’s first assume (i) that there is zero radiative forcing present (except for Spencer’s cloud forcing) and (b) that we CAN estimate this cloud feedback term directly from the net flux term DRcloud.

        Curious logic. I liked your summary statement:-
        “His [Dessler's] argument thus reduces to saying that if cloud forcing exists, it’s small, and the counterargument would be that if it exists, its magnitude is still uncertain.” I agree, but why would Dessler completely ignore the larger issue, and the one of more importance to him for Dessler 2010, of whether there was ANY forcing present over this period. Unless he is truly stupid, and I assume not, then I must presume that he cannot furnish the necessary evidential support for his case.

      • Paul – SB-11 didn’t try to model 2000-2010 and so Dessler, if he attempted this (which would be an undertaking fraught with pitfalls) would have nothing to compare his results with. SB-11 cited observational data for 2000-2010 but ran their simple model for 500 years rather than with the observational data..

        Trenberth and Fasullo, using 100 years of twentieth century data analyzed decade by decade, found that models that simulated ENSO well reproduced the observations fairly well, and those that simulated ENSO less well did a poorer job, suggesting that it was the ability to capture the essential dynamics of ENSO, with no need to invoke radiative forcing, that determined model skill.

      • Fred, I don’t think T&F did that. They asserted that Echam5 replicated ENSO well and demonstrated that “it reproduced the observations fairly well”. Whether Echam5 is good fit to ENSO is controversial (see my other comment on this thread).

      • @Fred,
        Dessler ran the entire model suite with all cumulative forcings fixed to compare to his abstracted cloud feedback values in Dessler 2010. All he has to do to demonstrate that Spencer’s time-lag analysis is methodologically incorrect is to show that those models – or at least the ones which allegedly model ENSO well – show time-lag characteristics and flux magnitude similar to the 2000-2010 observations. (Or as I said earlier, even better would be to show reproduction of actual 2000-2010 data.)

        Instead, all of the comparisons shown to date by TF and by Dessler have included radiative forcings, unless I have seriously misunderstood what’s been presented to date. If Spencer’s analysis is correct, then those models SHOULD show a time lag in any event, so there is no test of Spencer’s hypothesis. Why is Dessler avoiding this?

        Since Dessler hasn’t published anything to show that that he can reproduce Spencer’s time-lag characteristics using a model that has all radiative forcing repressed, I must suspect that it is because he cannot do so.

      • “Instead, all of the comparisons shown to date by TF and by Dessler have included radiative forcings, unless I have seriously misunderstood what’s been presented to date”

        What radiative forcings are you referring to? As far as I can tell, the curves shown by Trenberth and Fasullo regress TOA flux changesagainst temperature, and so forcing moieties such as CO2, aerosols, or solar irradiance that are not changing perceptibly will have little effect, because they won’t differ more than minimally during the short intervals over which the curves are drawn. Furthermore, since the depicted curves are the averages of ten individual decades, variations within decades will be averaged out. The time lag characteristics seen in the curves can come only from the intradecadal flux variations and not the persistent forcings.

        None of the models simulates ENSO perfectly, but even if one did, it wouldn’t prove that the same results couldn’t occur with an intradecadal radiative forcing term – i.e., it couldn’t “disprove” SB-11. The burden, however, as I see it, is on Spencer/Braswell to document the SB-11 claim to have shown that most of the observed changes were radiatively forced. I don’t think the paper showed that, for reasons already discussed.

      • @Fred,

        Say whaaaa??
        “As far as I can tell, the curves shown by Trenberth and Fasullo regress TOA flux changesagainst temperature, and so forcing moieties such as CO2, aerosols, or solar irradiance that are not changing perceptibly will have little effect, because they won’t differ more than minimally during the short intervals over which the curves are drawn.”

        This is defying the law of additivity, Fred, as well as presuming the answer to the thing we are trying to test! I have modeled several GCM results using unmodified Runge-Kutta (RK4) integration of high order forms (feedback non-linear in temperature) of an energy balance (perturbation) model. For GISS-ER match to temperature and OHC, see the link here for example :-

        http://rankexploits.com/musings/2011/equilibrium-climate-sensitivity-and-mathturbation-part-2/

        GISS-ER results show no evidence of any ENSO-induced temperature change – all temperature and OHC changes are explained as a result of input forcings. At short time-frames, the models I have tested all asymptote to a Planck response (as I would expect), which means a linear form of the feedback equation becomes a good estimate. So let’s use the linear form for simplicity. If I modify my annualised data to monthly, regress the net flux term (F – lamda*DT) against DT for different timelags, andplot the results, I will see the same time-lag regression plot as Spencer produced. Now, if I fix the level of forcings for a period and repeat the experiment over that same period, what happens? I see a slow change in temperature in response to the previous forcing history. If I make a lag-regression plot, it shows me a Planck response value for the rate of change of flux with temperature occurring at close to zero time-lag. That’s because we are now plotting (-lamda*DT) against DT to obtain the regression value. (Cumulative F(t) = 0, or DF/DT = 0)

        This is what Spencer found. Now you can argue that this is a null result because GISS-ER does not model ENSO fluctuations. But you cannot then prove your case by showing me a lag regression plot from a model that does have non-radiative (ENSO) fluctuations while retaining the CRITICAL variation in fradiative forcings. This is just silly.

        Even less can you argue that one can ignore the monthly changes in forcings because they are small. The sum of those forcings over a year is 100% of what is driving the GISS-ER model, and all of the other models which have zero or poor ENSO modeling characteristics.

      • The TOA flux and temperature in the models is calculated on the basis of forcings and feedbacks. The models are compared to observations – that over the short CERES record – are mostly driven by ENSO.

        No one is claiming – I hope – that any model is good at ENSO. Just that one model is a bit better than others at replicating the CERES observations.

        ‘Consequently, our results suggest that there are good models and some not so good, but rather than stratifying them by climate sensitivity, one should, in this case, stratify them by ability to simulate ENSO. In the Figure, the model that replicates the observations better has high sensitivity while the other has low sensitivity. The net result is that the models agree within reasonable bounds with the observations.’ realclimate The bounds are I believe 2 sigma bounds.

        The closer accordance with CERES is loosely assumed to be an ability to model ENSO – but it isn’t true necessarily. No one is good at ENSO at all – and certainly through the (SH) spring prediction barrier.

        It is instructive to have a look of the ENSO predictions collected here – http://www.bom.gov.au/climate/ahead/ENSO-summary.shtml – to see the range of values generated from any one model over only 3 months.

      • @Fred,
        Correction.
        I meant of course to write:
        “(Cumulative F(t) = constant, or DF/DT = 0)”
        Paul

      • Paul – I don’t understand your point. The GISS-E model simulates ENSO poorly and is an example of what we see under those circumstances. The MPI.Echam5 model does a better job, and yields a curve similar to the observational data although with smaller up and down deviations. It does this without invoking any radiative forcing fluctuations, and so the dips and peaks that emerge can’t be explained by the varying cloud-initiated forcings claimed by SB-11, and are consistent simply with temperature changes originating in the oceans without prior changes in the TOA incoming and outgoing fluxes, and with clouds (and their own effects on temperature) responding to the original temperature changes rather than initiating them. SB-11 claims to have shown that the forcings go both up and down over a range that exceeds 1 W/m^2 during short intervals. Their data can be explained without that, and so they have failed to show it.

      • @Fred,

        “The MPI.Echam5 model does a better job, and yields a curve similar to the observational data although with smaller up and down deviations. It does this without invoking any radiative forcing fluctuations,…”

        The comparative results presented by Trenberth and Fusillo include ALL of the radiative forcing fluctuations in the GCMs. Read the article again.

        Dessler did run all of the GCMs with F(t) fixed (no radiative forcing fluctuations) for Dessler 2010 in order to abstract his RCF statistics. It should have been easy to examine comparisons of those runs against observations in his rebuttal, in order to show that even without radiative forcing, the lag regression plot for those models which are reputedly good at modeling ENSO could reproduce the lag time and feedback response magnitude of the ENSO-dominated observations – in other words, no need to invoke radiative forcing to explain the data. But he chose not to.

      • Paul – This thread format is an inefficient way to ensure that we’re not talking past each other. I’m not sure it’s worth continuing this over what is now agreed to be a paper of minor consequence. The “forcing fluctuations” I was referring to are those that go up and down in SB-11 Figure 2. SB-11 is where they are referred to as “radiative forcing”, whereas in the GCMs they are not, but are included as TOA flux changes secondary to changes in temperature. Flux changes are not forcings when they occur as responses to temperature changes, but are feedbacks. I expect that Trenbeth/Fasullo and Dessler would concur that the models were not looking at what they would call responses to forcings in their ENSO emulations.

        If you had something else in mind, please feel free to comment, and then I would be content to let other readers, if any are left, review the whole set of comments to see if they can make sense out of these exchanges.

      • @Fred,
        I too am puzzled, Fred, because I agree that we are talking past each other. I did not believe that there was any definitional problem between us – but from your last couple of responses I could be wrong.
        One last attempt on my part and I will leave it alone.
        You write:-
        “The “forcing fluctuations” I was referring to are those that go up and down in SB-11 Figure 2.”
        Maybe this is where the problem starts. In my book, these are not “forcing fluctuations”, they are net flux fluctuations. The observed net flux fluctuations combine the effects of
        – non-radiative forcing, which has a short term temperature affect and short-term radiative feedback,
        – radiative forcing from “conventional” sources,
        – plus direct and indirect temperature-dependent radiative feedbacks, including cloud feedbacks.
        I am presuming that this is non-controversial upto this point.

        Spencer argues that, in addition to the above components of flux variation, there may be a non-conventional radiative forcing, specifically brought about by cloud variations, mechanism unspecified but which are NOT due to temperature-dependent feedback.
        Are we good so far?
        He further argues that, if the changes in total radiative forcing – from both conventional and non-conventional sources – are small, then a lag-regression plot shows a close-to-zero lag time between temperature change and net flux response (since the individual regressions default to (-lamda*DT against DT), and peak response should be equal to short-term climate feedback, lamda. Conversely, if radiative forcing changes are significant (from whatever source), then the flux accumulation/loss precedes temperature gain/loss with a lag-time measurable in months.

        Are we still good? I am not asking you to accept Spencer’s argument – just asking you to confirm that we both understand it the same way.

        Next step – Spencer argues that the lag-time regression plots of the actual observed data are consistent with there being significant radiative forcing present. Dessler has previously argued, and continues to argue in Dessler 2011 that the radiative forcing component is negligible over this period; the observed net flux response can be fully explained in terms of temperature-dependent feedbacks to the non-radiative flux changes induced by ENSO events.

        Are we still on the same page? If not, where have we diverged in the above?

        Next step is the question I’ve been pounding on in all the posts above:- why doesn’t Dessler demonstrate that he can obtain the same net flux response (expressed in the form of the lag-time regression plot) from an AOCGCM or an AMIP with all changes in radiative forcings switched off i.e. just ENSO being modeled? And, for me he only has to do this with the model or models which he declares to be good at modeling ENSO. He clearly already has this data, as I have already stated.

        Instead, all of the lag-time regression comparisons we have seen from KF and Dessler have had changes in radiative forcing switched ON.

        This is my last attempt, Fred, honest.

      • I’m still lurking! Keep going – it’s fascinating!

      • Paul_K, the GCMs don’t run with unknown forcing fluctuations, just known changes, which are usually gradual except volcanoes. This is why Trenberth and Fasullo pointed out that echam5 which can produce El Ninos has a similar correlation behavior to Spencer’s observed one. This is huge because it shows you don’t have to invoke unknown forcing fluctuations to get Spencer’s pattern.

      • Same GCM hansen used to predict a Super El Nino this fall?

      • What are you talking about, Bruce? echam5 is a German GCM used in the IPCC projections. Spencer is getting the same results as some of them is all I am saying.

      • ECHAM5 does not have variable solar forcing

      • maksimovich, that would only strengthen the case by making the echam5 forcing even smoother.

      • Paul – We are mostly in agreement all the way through your comment. Where we diverge (rather than completely “disagree”) is in two respects.

        (1) The importance of eliminating “all forcing”, including the effects of CO2 changes over the twentieth century. To the best of my knowledge, the latter has led to rather small changes in net flux (incoming minus outgoing) over the course of 100 years (less than 1W/m^2), since the warming during that interval has reduced the flux imbalance while the increase in CO2 has operated to increase it – a typical estimate of the current imbalance is 0.9 W/m^2, and it probably started the twentieth century at a level greater than zero. During any ten year interval, the change in net flux due to CO2 is probably very much smaller than the flux changes in SB-11 – too small to significantly alter the shape of the curves that are shown.

        (2) The practicality of running a new ensemble of GCM model simulations with the elimination of those persistent forcings. If it could just be tossed off, it would be worth doing just to avoid the criticisms you’ve leveled, but I expect it would be a rather extensive undertaking to rerun model simulations for one hundred years with no CO2, solar irradiance, or aerosol changes, while maintaining ocean temperatures, the Walker and Hadley circulations, and other variables relevant to ENSO behavior at the observed levels, and getting everyone to agree that the inputs were the correct ones. It would probably be even more arduous to create new models designed specifically to test the SB-11 paradigm, and to get everyone to agree they are parametrized correctly, since it would certainly be easy to reproduce ENSO by arbitrarily tuning the parameters to achieve that goal. I think it could be done, but is it worth doing? Even if it could be done easily, and reproduce ENSO perfectly with no forcing at all, it wouldn’t “prove” that SB-11′s putative forcings couldn’t also do the same thing.

        Combining (1) and (2) above, I suspect that it isn’t worth the effort, and that it would be predicted that the undertaking would yield results only slightly different from the model outputs that have been depicted, after the expenditure of considerable time and cost. However, if those who do this type of modeling for a living tell me it would be easy, I wouldn’t be qualified to dispute them.

        My view of SB-11 is that it failed to substantiate the claim that most of the short term temperature variation (up to 0.4 C) and flux variation (up to 2 W/m^2) represented radiative forcing. If they haven’t made that case, I would be content to leave it at that and not try to prove them wrong.

        Finally, I’ve stated elsewhere in this thread my view that the significance of getting the cloud feedbacks in ENSO right has been exaggerated, because feedbacks and climate sensitivity to ENSO events will inevitably differ from those involved in long term responses to forcings originating in the atmosphere from CO2 or other moieties that impose atmospheric temperature changes on the ocean rather than the reverse. I don’t see these exercises as a useful way to refine our estimates of climate sensitivity to CO2.

      • @Fred,
        (1) I am not talking about eliminating CO2 forcing, I am talking about eliminating ALL radiative forcings.
        (2) Far from it being an extensive undertaking, it has already been done, and Dessler has the results, as I have stated several times above.

        “One obvious question is whether climate
        models also reproduce this cloud feedback in
        response to short-term climate variations. To test
        this, I analyzed control runs from fully coupled
        climatemodels, in which atmospheric greenhouse
        gas abundances and other forcings are held constant
        at either their preindustrial or present-day
        concentrations; thus, there are no long-term trends
        in the models’ climate, and the climate variations
        in the model runs are entirely due to internal variability.
        The control runs were obtained from the
        World Climate Research Programme’s (WCRP’s)
        Coupled Model Intercomparison Project phase 3
        (CMIP3) multimodel data set (31).”

        Hence my question: According to Dessler’s conceptual model (and yours, if I have followed your argument), the results from these runs – for the GCMs which better model ENSO – should reproduce the lag-regression characteristics seen in the observational dataset during the period 2000-2010. Game, set and match to Dessler. So why hasn’t he used them? Two possibilities:- he is not very bright or the results show something close to a zero-lag response, which of course would provide direct support for Spencer’s model.

      • Jim D,
        You’re not following the plot. The Echam5 results shown by K&F have input radiative forcing data ( not “unknown” forcing data), as do the results shown by Dessler – albeit from a different Echam5 run. These results are in each instance then compared with the satellite observations from 2000-2010. Spencer argues, on the one hand, there is a signature (in the lag-time regression plot from the observational data) of radiative forcing, and Dessler and K&F, on the other hand, argue that there isn’t. To dispute Spencer’s argument all they need to show is that they can reproduce the same lag-time regression characteristics WITHOUT any input radiative forcings. In other words, they prove that Spencer is wrong to conclude that the time-lag regression allows one to discriminate between having a significant radiative forcing component or just having feedbacks present. Dessler already has the data which would allow him to do this. (The quote above comes from Dessler 2010.)

      • Paul – I don’t think we’re going to resolve this. Dessler referred to “control runs”, but a control run is not the same as an ENSO simulation. Models can do the latter, but imperfectly, and they would have to be run with inputs that eliminated all forcings and then examined for short term flux/temperature relationships. As I mentioned, I think it could be done, but it would be an extensive undertaking, particularly if multiple models and multiple sets of initial conditions were compared, would provoke arguments about the correctness of inputs, and wouldn’t prove anything regardless of whether it came close to the results we’ve already discussed (as I suspect), or differed from them because the forcings were subtracted.

      • Just to add two points to my comment above.

        (1) 2000-2010 fluctuations were mainly ENSO related, but not all twentieth century internal variations were of that type. The reason why the curves we discuss could be reproduced by the 2000-2010 interval involves the alternation of warming and cooling intervals. Without that alternation, the curves would be different.

        (2) Much of the ENSO-related global temperature change starts with redistribution of heat within the ocean mixed layer, and can’t plausibly be attributed to a simple TOA flux change of the kind that would simply warm or cool but not redistribute. Without addressing this in more detail, I would suggest that SB-11 has produced no explanation of why observed surface temperature changes are accompanied by reverse changes at deeper levels. I don’t think one needs to “disprove” SB-11, particularly if considerable effort is involved with less than definitive results. SB-11 hasn’t substantiated its own claims, and it would be prudent to leave it at that, particularly if, as I suggest above, ENSO-related climate feedbacks and sensitivity tells us little about climate sensitivity to CO2.

      • Paul_K,

        I’m not sure I understand I understand what you mean by “no radiative forcing”. D11 shows the results from control runs in CMIP comparisons. By definition the control runs do not have any changes in forcings from the 20th century:

        …control run boundary conditions lack these forcing variations, we focus on means and other statistics that we judge to be largely unaffected by them. In the final part of this section we discuss the climate variability simulated by the CMIP control runs. This topic has also been addressed in more specialized studies (Barnett 1999; Bell et al. 2000; Duffy et al. 2000).

        From this. The CMIP runs shown in D11 were the control runs (SB11 used detrended 20th century runs, but it didn’t make much difference). But running the models w/o any radiative forcings (TSI gone? All CH gas forcings gone?) would quickly result in a frozen orb. This request makes not sense, so I think you are thinking of control runs.

        I also think you underestimate the resources necessary to “just rerun the models”. Each model run of a modern model takes considerable time on a rather large an expensive supercomputer. It is not a simple (or cheap!) matter of running the model on your PC (or even your department mainframe). For this reason model runs are typically pulled from the PCMDI database (link is in the paper ref’d above). If a paper refers to CMIP or AMIP, this is what they are referring to. In this case it is probably CMIP3, since CMIP5 (numbering scheme probably revised to bring things into sync with the IPCC assesment report numbers), are not done yet.

      • One more small point related to Dessler-2010, and relevant to the comment by Rattus N. as applied to Paul’s earlier comments. Dessler did not run models with forcings removed, but only with forcings held constant, which means they were continuing to mediate changes in TOA fluxes, albeit very slowly as I mentioned earlier. Removing the forcings would have produced significant climate alterations beyond their effect on internal climate dynamics such as ENSO.

      • @Rattus Norvegicus,
        Thanks for putting me out of my misery. I have rechecked Meehl et al 2007, the source for Dessler’s preindustrial control runs – and they do indeed have forcing switched off.

      • I shouldn’t have said that Dessler in D-10 “ran the models” – he analyzed the results of control runs from CMIP3 rather than rerunning them himself.

    • Fred writes “we already know that ENSO temperature fluctuations arise in the oceans and that excludes cloud forcing as a mechanism”

      Thats not valid logic until we properly understand the drivers of the ocean warming behind ENSO. If they are largely or even just significantly cloud related forcings and there is a long term trend in changed ENSO events then your logic fails.

      • Tim – If you review my original comment, you’ll see I was referring to Dessler’s logic, and not making an independent statement. I don’t think there is any doubt that this is his position. Most of this thread and the content of the various papers should be consulted for the reasoning behind the differing views on the subject, but I would say that the argument you make has already been addressed in many places here.

      • Fred, I assume you have been following the discussion and analysis over at CA and other sites. Everyone is waiting retraction of your earlier ad hoc analysis.

      • Rob – I addressed the CA comments earlier, at Comment 110441. I don’t think they contribute much to the topic. In my view, this is one occasion where Steve McIntyre, who often has useful things to say, drew conclusions that went beyond the evidence (e.g., implying that the Science reviewers had overlooked an obvious flaw), but readers can judge for themselves. In any case, there would be no reason to modify my ad hoc analysis on the basis of the CA commentary.

        I’ll make another point, though, Rob. Your statement that “everyone is waiting” for my “retraction” strikes me as reflecting the kind of ideological fervor that precludes any rational analysis and discussion of evidence . It is unhelpful in advancing our understanding of the climate dynamics involved in ENSO, feedbacks, and forcings.

      • Fred
        Please note that comment was made by “Rob” not me (RobB)

      • RobB – I did notice, and I appreciate the difference.

  11. Miskolczi shows that an infinite supply of water vapor– which is from the ocean so the supply is nearly infinite– controls the greenhouse effect. The atmosphere could hold more water vapor than it does but only a portion of the water vapor is actually taken up, thus maintaining energy balance. For an additional amount of CO2—whether it is from natural sources or from humans is irrelevant—the CO2 simply replaces an equivalent amount of water vapor. This is why the Earth’s greenhouse effect remains so amazingly constant.

    “There is a near infinite supply of greenhouse gases available to the atmosphere in the form of water vapor from the ocean to provide the greenhouse effect, but the relative humidity in the atmosphere is much less than one. Therefore, there must be some greenhouse equilibrium mechanism to control the strength of the greenhouse effect and the relative humidity. Otherwise, climate would be very unstable. The global average relative humidity at the surface is about 78%. It generally decreases with altitude and is about 37% at an altitude where the atmospheric pressure is 300 millibars (mb). Relative humidity is the fraction of water vapour in a small parcel of air relative to the total amount of water vapour the air could contain at the given temperature and pressure. So why isn’t the relative humidity 90%, or vary randomly? Relative humidity is at its current value because it is controlled by the laws of physics.” (Ken Gregory, The Saturated Greenhouse Effect)

    The IPCC’s runaway greenhouse problem assumes a large positive feedback from water vapour. Empirical shows just the opposite: changes in atmospheric water vapor is a large offset to CO2-caused warming.

    “Miskolczi used HARTCODE to compute the optical depth from 1948 to 2008 using the measured CO2 content at Mauna Loa, Hawaii and the global average water vapour content from the NOAA Earth System Research Laboratory. The optical depths are calculated for each greenhouse gas and summed line-by-line across the electromagnetic spectrum. The resulting optical depth curve is a measure of the total greenhouse gases by effect over the last 61 years … The results show that the total effective amount of greenhouse gasses in the atmosphere has not significantly increased over the last 60 years.” (Ken Gregory, Climate Change Science)

  12. You should read it for yourself (see below) and if you do, I am sure you will agree that we now have a complete and coherent explanation for all global warming, and that all fair-minded persons can only come to one single, clear and unequivocal conclusion about global warming, as follows:

    Energy in equals energy out. ‘Runaway greenhouse theories’ are a myth and violate basic laws of nature.

    The Earth is in a perpetual state of radiative balance due to an unlimited available amount of water vapor that that varies in concentration around the globe and at elevation. GCM assumptions of a constant relative humidity cannot work.

    Fractional changes in the concentration of water vapor in the atmosphere provide logarithmic radiative transfer effects that serve as a negative feedback to increases in other greenhouse gases including CO2. A change in atmospheric CO2 is irrelevant.

    The Earth is at all times in radiative equilibrium that is controlled by a special atmospheric transfer function – basic to all ‘Earth-type’ planetary atmospheres — that requires the continuity of the temperature at the ground surface. As a result, it is an inherent property of the Earth’s atmosphere that there always, “is a sustained planetary greenhouse effect with a stable ground surface temperature.”

    Greenhouse gas concentrations can never be the cause of global warming. A change in the greenhouse effect is not possible absent a change in the amount of energy that is input into the system from solar activity and any change in the amount of internally-generated thermal energy from within the Earth itself.

    ___________
    Miskolczi, F.M. (2007) Greenhouse effect in semi-transparent planetary
    atmospheres, Quarterly Journal of the Hungarian Meteorological Service
    Vol. 111, No. 1, January–March 2007, pp. 1–40

    http://met.hu/doc/idojaras/vol111001_01.pdf

    • Dear Wagathon:
      Mathematics show that the earth is thermodynamically stable and runaway greenhouse theories are a myth. So is the greenhouse gas effect, It is a myth too. The earth energy balance is inherently stable with and without water vapor and carbon dioxide.

  13. Judith, I don’t know how you could miss the obvious cherry pick in S&B. The analyzed all the models the listed (which included the ones which reproduced ENSO with reasonable fidelity) but chose only to graph the most and least sensitive. This is completely bogus when you know you have results which disprove this (and in retrospect because of the PR) finding of your paper.

    You need to close your mind a little bit, because your brains are leaking out.

    • “The analyzed all the models the listed (which included the ones which reproduced ENSO with reasonable fidelity) but chose only to graph the most and least sensitive”

      “reasonable fidelity” If I were comparing two data sets I would look at the correlation coefficient and the rho statistic, a p=>0.05 is reasonable. Unless of course I am comparing a large number of models, the more models I test, the more likely I am to get a false positive, i.e. the bigger the n number, the more chances I have of having a good fit. So for multiple models p has to be multiplied by n=model number/5.

    • Yes, Its rather like failing to use all the data that is available when comparing models to observations (Santer) or failing to report certain standard statistical metrics or claiming ( falsely) that you never computed them. Or leaving certain trees in a proxy record when NAS says they should be avoided, or using contaminated proxies. The list of bad science is long. SB11 joins a long list. But that wasnt your point was it

    • Rattus

      I’m glad you raised the issue of model selection. D11 says of its graph in Fig 2:

      Third, the models that do a good job simulating the observations (GFDL CM 2.1, MPI ECHAM5, and MRI CGCM 2.3.2A) are among those that have been identified as realistically reproducing ENSO [Lin, 2007]. And since most of the climate variations over this period were due to ENSO, this suggests that the ability to reproduce ENSO is what’s being tested here, not anything directly related to equilibrium climate sensitivity.

      The title alone of Lin2007 should have caused a peer reviewer a moment of thought – “Interdecadal variability of ENSO in 21 IPCC AR4 coupled GCMs”.

      Now a more recent piece of the literature is “ENSO Feedbacks and Associated Time Scales of Variability in a Multimodel Ensemble” Belmadani et al J of Climate 2010 and it takes the view (caveats accepted) that GFDL CM 2.1, MPI ECHAM5, and MRI CGCM 2.3.2A all are a bit lacking in terms of their ability to model ENSO. Other models “have ENSO periods closer to the observations on average and are considered the most reliable for climate projections under increasing concentrations of greenhouse gases.”

      I note that MPI ECHAM5 that Trenberth championed at his RealClimate critique of SB11 is the greatest outlier of the three.

      Now Dressler was obviously in a hurry, but you’d think he’d keep up with the literature, and as for the reviewers they did get to read the reference list titles?

  14. It appears that both S&B and Dressler are getting the concept and terminology of forcing and feedback wrong, or at least not quite right. Whether clouds or ENSO are a forcing of a feedback has nothing to do with their magnitude, it has to do with whether of not something else is driving them.

    It seems like Dressler’s point is particularly “interesting” in that he’s claiming that – due only to magnitude – that ENSO has to be a forcing. This is downright bizarre. What’s even more “interesting” though is what happens when you bring the greenhouse effect into this argument. Is Dressler claiming that the greenhouse effect isn’t a forcing because it’s much smaller than ENSO? Does the “team’ really want to go there?

  15. You’re a charming person Rattus

  16. Robert of Ottawa

    JC …. S&B’s conclusion was that the record at present could not allow a determination between positive and negative feedback. A non-contraversial position. Isn’t the Warmistas hyseria noteworthy?

  17. “These analyses don’t really tell us anything about cloud feedback, although they are interpreted as doing so.”

    Given that S&B claim there is no cloud feedback, that is a telling point.

    • Hi Eli,

      SB11 claim that we can say nothing about cloud feedback from global warming from the CERES and temperature record. This is a very different thing.

      During El Niño, the warming of the tropical eastern Pacific and associated
      changes in the Walker circulation, atmospheric stability, and winds lead to decreases in stratocumulus clouds, increased solar radiation at the surface, and an enhanced warming so that even models without ocean dynamics are capable of emulating some ENSO‐like variability [Kitoh et al., 1999]. Positive cloud feedbacks in observations have been shown to occur in association with ENSO and these variations are generally not well depicted in models [Kang et al., 2002; Clement et al., 2009], but challenges also exist for diagnosing these interactions in observations, as it is difficult to identify cause and effect in the context of multiple interactive variations.’ Trenberth et al 2009

      But clouds are as much part of the ENSO phenomenon as SST – we get as much as 3 W/m^2 between ENSO states and the world cools in a La Nina and warms in an El Nino. There are 2 contribution to warming or cooling. The transfer of energy between ocean and atmosphere – from a warm ocean in a El Nino and to a cool ocean in a La Nina – and the changes in cloud and water vapour. Lead and lag seem to be me only to confuse the issue. It is a simple matter of correlating the ENSO state to temperature. Hydrological variability is much more complex.

      Science suggests that the world is not warming (Mochizuki et al 2010, Swanson et al 2009, Tsonis et al 2007, Keenlyside et al 2008) – albeit with immense uncertainties surrounding the origins of decadal variability.

      My reading suggests that ENSO is variable on centennial to millennial timescales ( Moy 2002 – look for instance for the change from La Nina to El Nino conditions 5000 years ago that resulted in the drying of the Sahel. I am aware also of local work on the north Queensland coast line involving relic storm wrack. Periods of hundreds of years characterised by super La Nina cyclones – and hundreds of years of El Nino drought and a lack of storms. So really – what is the potential for longer term variability.

      Fred has the peculiar idea of cloud being caused only by temperature. It seems unlikely – but even is that were the case it would be impossible to disentangle causality. So it seems more ideology than science and I suspect both you and Fred of belonging to the same spaceship cult. I would advise you to avoid the Kool-Aid – but I would just be accused of being a concern troll.

      Cheers

      • “Fred has the peculiar idea of cloud being caused only by temperature.”

        Robert – If you can quote a statement I made claiming that temperature is the only factor in cloud formation, I will be grateful for the opportunity to retract it. If you can’t, I hope you will retract your own statement.

        Of course, if you believe temperature has no role at all to play in cloud formation, then you and I will have to disagree, because I believe it plays an important role via its effects on evaporation and condensation.

      • ‘…this in contrast to the standard interpretation that the flux changes were simply responses to the temperature changes that characterize ENSO fluctuations between El Nino and La Nina intervals.

        they include reasons why flux changes are more likely a response than a major cause of ENSO temperature changes..

        No-one claims that ENSO is not associated with cloud changes, but Spencer is claiming that clouds, on their own, play a major role in initiating ENSO changes via an unidentified mechanism. The data, however, including Figures 2 and 3, do nothing to controvert the ordinary understanding that temperature changes occur first, and flux follows after a lag.’

        The claim about Spencer is utterly wrong. But you are looking for a positive feedback? Higher temps and less cloud? Not more moisture and cloud?

        Hydrology is complex, dynamic and non-linear – it may rain in the suburb next door and not in yours, it might not rain for years or you may get the biggest downfall ever seen – all by chance. When talking about rainfall you need to specify place – and then define the global standing ocean and atmospheric patterns that influence the likelihood of rain in that place. The ENSO, SAM, NAM, PNA, IOD, AMO – especially.

        I don’t want to get into a simple minded discussion based on simple physical conditions – because it is meaningless. We might get more storms because it is warmer is a concept looking for an application. Show me how it changes the global hydrological indices – and we may have something to talk about.

      • Chief,
        Did you have a chance to look at the limnology stuff I sent you?

  18. Dr. Curry,
    Your comments are always high quality and helpful but this one seems to be especially helpful. Your first paragraph puts the finger right on the issue and I wish you could expand that paragraph into an essay that explicates the matter. But I realize there is quite a bit of fog rising from this topic.

    You write:
    “This inconsistency arises from trying to use the black box energy balance feedback model to make inferences about the impact of internal dynamical variations on clouds and the energy balance.”

    Is it the case that Dessler, using the black box energy balance feedback model, treats clouds as GHGs for radiation leaving Earth but treats clouds as a black box for radiation arriving from the sun? Is that the explanation of Dessler’s claiming that clouds can “warm” Earth, as GHGs do, and denying that clouds can cool Earth through reflection of sunlight?

    You write:
    ‘Cloud feedback is frequency dependent, and the key issue S&B are trying to address is the role of natural internal variability in modulating clouds, which in turn modulates the radiative balance and surface temperature; it is in this sense that they then refer to clouds as a “forcing.”’

    It is commonsense to me that clouds modulate surface temperature by reflecting sunlight but I do not understand the effects that clouds might have on the radiative balance. When talking radiative balance, aren’t we talking black box assumptions and, for that reason, must accept Dessler’s claim that clouds cannot cool Earth?

    Do we have a primitive conundrum? Does someone have to take a step or two forward before we can clearly formulate a disagreement?

    Thanks once again for your fine work.

    I

  19. Not being a climate scientist, I am having some difficulty interpreting Figure 2 of S&B 2011. Am I right in thinking that this demonstrates/suggests
    (a) that clouds have BOTH radiative AND forcing properties with different temporal relationships? [Is there some fundamental law of 'climate science' which precludes this possibility?] and
    (b) that the IPCC models do not reconcile with empirical data? and
    (c) IPCC’s most sensitive models fare worse than the least sensitive models?

    • This is a key you are looking for to understand reality– much like deciphering Linear-B into English: “Relative humidity is the fraction of water vapour in a small parcel of air relative to the total amount of water vapour the air could contain at the given temperature and pressure. So why isn’t the relative humidity 90%, or vary randomly? Relative humidity is at its current value because it is controlled by the laws of physics.” (Ken Gregory)

      • Thanks for your post. My understanding is that the IPCC models were based on the assumption that relative humidity was constant with temperature – an assumption which has since been proved false by empirical data (sorry I cannot remember the reference).

  20. There are three notable points to be made. First, SB11 analyzed 14 models, but they plotted only six models and the particular observational data set that provided maximum support for their hypothesis. Plotting all of the models and all of the data provide a much different conclusion.

    http://geotest.tamu.edu/userfiles/216/Dessler2011.pdf

    It is like me saying to you: I know were you live; It is somewhere on earth.

    • Selection bias and poor handling of uncertainty? Where have we heard that before?
      What would be really impressive is if someone could explain why most of the models fail to come close to matching real world data and why one or two of the models come close to doing so. ‘Climate sensitivity’ does not seem to be an adequate explanatory factor. Where do the problems lie? Is it in the underlying assumptions? How do the various models differ in the assumptions they make?
      It is good that Dessler provides the confidence intervals (uncertainties) – these seem to be wide enough to cover both positive and negative correlations simultaneously. So what is the net effect integrated over time?
      Have the models incorporated data for solar magnetic and spectral variation, cosmic ray flux etc. What of the different cloud types, altitudes and microphysics?

  21. Lagged correlations are pretty useful for picking out causality. The one I have modeled quite extensively is the lag of global peak oil production with respect to peak oil discovery. This lag is around 40 years; in other words, the peak in production now is the result of a peak in the number of discoveries more than 40 years ago.

    Next what we do is a lagged correlation of atmospheric CO2 concentration after fossil fuel emissions. This one is almost as obvious.

    The one that is revealing the most uncertainty is the lag between atmospheric CO2 and average temperature. I don’t think we will know this one for awhile as it may be a short duration buried in the noise.

    If you notice, these are all a series of variable lags, all usefully modeled as compartments with flows between the stages. It goes from discovery -> production -> increased concentration -> temperature response (?)
    IMO, all the uncertainty is in the last lag.

  22. How is it some people here are ignoring some recently, yet, should be commonly known, science. First of all, Spencer clearly states that clouds are bi-directional, meaning, clouds are feedbacks and forcing. Obviously, this was misrepresented by a few here and worse, it was misrepresented in Dressler’s drivel. Dressler’s paper was a strawman argument, and he did a very poor job of knocking down his own strawman. Secondly, given the results of the CLOUD experiment, not only does this make it entirely plausible, it makes it likely. Then, Dressler uses a time period that had no change in temp, to prove clouds didn’t change the temps? Well, nothing did, including the imutable hypothesis that CO2 is hotting us up. It cannot hold that CO2 holds that much power over our climate and that something didn’t mitigate its effects! It is really rather Boolean. If something didn’t mitigate the warming, CO2 doesn’t hold the power as advertised. Does someone have a better idea than clouds? Did SB11 get it all right? Probably not. But people shouting their certitude about clouds is beyond believable. Up to and including that scientific imposter at Texas A&M. Some people want to give Spencer grief about not including all of the models on the graphic? He stated he didn’t! It’s not like he was hiding anything and deleting e-mails. I know it may not occur to people here, but graphics are used for clarity. I want to look at 8 more inept models? It doesn’t matter how many or what he put on the paper, the models are programs used by psuedo-scientists to generate psuedo-facts. Or is someone here prepared to show one that is accurate?

    Lastly, try an empirical test yourself. On a warm and partly cloudy day, have a cloud pass over between you and the sun. Feel the coolness? Yes, the cloud is effecting the temps. Next, on a cool partly cloudy night, have a cloud pass overhead. Do you feel the radiation burns? No? What’s the net effect of those types of clouds? Yes, there are other types of clouds with different behaviors. They need to be distinguished, identified, and then quantified as to their effects on our climate. When we’ve done that, then we can start pretending we know something about our climate. Until then……this stationary science will not progress.

    • I forgot to add, the team has fragmented. Is Hansen on the outs?

      Earth’s Energy Imbalance and Implications
      James Hansen, Makiko Sato, Pushker Kharecha

      Aerosol climate forcing today is inferred to be ‒1.6 ± 0.3 W/m^2, implying substantial aerosol indirect climate forcing via cloud changes.

      http://www.columbia.edu/~jeh1/mailings/2011/20110415_EnergyImbalancePaper.pdf

      So, warmista, pick an hero! They aren’t both right.

      (My thanks to Steve Goddard.)

      • “Is Hansen on the outs?”
        No. The AR4 SPM gives aerosol forcing at -1.2 W/m2, of which -0.7 W/m2 is due to cloud albedo effect. The small difference with Hansen’s figure is well within the SPM error ranges.

      • Yeh, thanks Nick, I was referring to the verbiage. See how Hansen specifically calls it a forcing, more he calls it substantial.

      • SPM calls it a forcing – it’s in their table. And 0.7 out of 1.2 is substantial.

      • From Trenberth at RC:

        Clouds are not a forcing of the climate system (except for the small portion related to human related aerosol effects, which have a small effect on clouds). Clouds mainly occur because of weather systems (e.g., warm air rises and produces convection, and so on); they do not cause the weather systems. Clouds may provide feedbacks on the weather systems.

      • Yes, and that is in direct odds with Hansen’s statement.

      • Ok, It seems to me that here’s what we have (and please excuse my lack of background in the science):

        Hansen says that there is a significant aerosol forcing due to cloud changes. Could that not mean that clouds represent a vehicle for feedback that occurs in response to a forcing from aerosols?

        Nick points out that the IPCC indicates -0.7 W/m2 aerosol forcing is due to cloud effect (out of -1.2 W/m2 aerosol forcing).

        That leaves -0.5W/m2 aerosol forcing that is not not a product of cloud effect.

        Trenberth says that -0.7 W/m2 forcing (manifest as a feedback through cloud changes) is a small portion of forcing related to the human influence.

        There seems to be some disagreement there – but who do you think is correct? Is -0.7 W/m2 a small forcing relative to the human influence, or is (potentially) -1.9 W/m2 (some unspecified portion of which that is manifest through a feedback of clout changes) a substantial forcing?

        Further, do you think it possible that Hansen meant that the forcing agent is aerosols, and that “clouds changes” serve as vehicle for feedback mechanisms (as opposed to a forcing) that manifests that forcing? Note that Trenberth has said clouds are a feedback that amplifies and ameliorates climate.

        So – is it not possible that they both thing that aerosols are a significant forcing agent, and that clouds are a feedback mechanism that manifests that forcing?

      • Joshua,
        Yes, many people are getting tripped on the semantics of the conversation. I really don’t think Hansen is making that distinction, that would be akin to stating aerosolized H2O, comprising a cloud is a forcing but a cloud is a feedback.

        The thing that gets me, is that with recent work in the area, it seems apparent to me that clouds are an unknown variable. We can call it whatever we want, but to pretend clouds don’t effect the climate, in my view, is a silly position to take. This seems to be Dressler’s position. At least Hansen acknowledges that we need to come to a better understanding of what they’re doing.

        As far as pinpointing the w/m2. I’ve got my own thoughts, but, they wouldn’t be on any more solid ground than anyone else’. I think the clouds we typically think of, have a net negative effect on the earth’s temps, but there are also other types of clouds that may or may not have a net negative.

        This is why I believe Dressler was entirely out of line. Spencer’s paper didn’t say much more than “we don’t know”. Dressler jumps up and down and says we do too know!!! But his paper obviously shows that he doesn’t. Clouds are an avenue climatology will have to seriously address if it is going to move forward. Lindzen and Spencer not only advocate this, but, apparently, so too does Dr. Hansen. If those guys agree on something like that……it probably means that’s an avenue that needs pursued.

      • Yes, many people are getting tripped on the semantics of the conversation.

        Agreed. It is annoying. Instead of trying to really understand each other, and in doing so clarify the semantics, people are trying to tear each other down based on semantics.

        that would be akin to stating aerosolized H2O, comprising a cloud is a forcing but a cloud is a feedback.

        But that is what Trenberth said, specifically

        We can call it whatever we want, but to pretend clouds don’t effect the climate, in my view, is a silly position to take. This seems to be Dressler’s position.

        I don’t know if that’s Dessler’s position – but it clearly isn’t Trenberth’s (he said that clouds amileriorate and amplify changes in climate forced by other factors). I would say that if you think that Dessler and Trenberth are in disagreement about that issue, then you’re getting caught up in semantics. Perhaps Hansen and Trenberth disagree – it seems far less likely that Dessler and Trenberth do.

      • Josh, I think Dessler rushed his paper. He seemed more interested in getting something out that stated that he disagreed more than he was interested in putting together a coherent argument. His conclusion pretty much shows it,

        “These calculations show that clouds did not cause significant climate change over the last decade (over the decades or centuries relevant for long-term climate change, on the other hand, clouds can indeed cause significant warming).”

        Obviously, this last statement screams for clarity, but, I guess that would have slowed him down.

      • suyts –

        The statement is unclear in one sense, and clear in another.

        It clearly is focused on refuting S & B’s findings – which were based on analyzing the role of clouds as a forcing agent over the last decade. The debate is about whether S & B’s findings valid in re correlations of recent temperate trends and CO2 emissions.

        This is an earlier statement from Trenberth:

        “The interannual global temperature variations were not radiatively forced, as claimed for the 2000s, and therefore cannot be used to say anything about climate sensitivity. Clouds are not a forcing of the climate system”

        It seems his statement about long-term effects of clouds is not clarified – which is appropriate if you don’t know how to quantify that effect precisely.

        That said – I wouldn’t doubt that the paper was rushed. Unfortunately, this battle is being played out by partisans on both sides and the science suffers as a result of tribal influences. Of that, I have little doubt.

      • Clouds are an effect of the earth’s energy balance, not a cause. In order for clouds to have a feed back there must be a correlation between clouds coverage and the controlled variable in the equation used by Dr. Spencer, which is ocean temperature. A change in ocean temperature does not necessarily induces a change in cloud coverage, for this coverage is directly related to the amount of energy absorbed by the earth, which constant. Therefore there can be no feedbacks from clouds for climate considerations. For more, please see Article-12, Earth’s Magic on http://www.global-heat.net.

        Furthermore, the climate energy equation used by Dr. Dessler and Dr. Spencer in their papers is a form of the Stefan-Boltzmann law equation for small temperature change. While the equation applies for solid surfaces such as land and continents, it is not the controlling equation of the energy radiated from fluids such as surface water or ocean. Surface water outgoing radiations is controlled by convection heat transfer process, not by the Stefan-Boltzmann law equation. Therefore, using Stefan-Boltzmann law equation to calculate ocean temperature is a mistake and the model can never be right. There are no such things as climate sensitivity or feed backs.

      • So if something caused there to be a lot more clouds and fewer sunny days, then the climate wouldn’t change? In UK at least cloudy days tend to be cooler and damper than sunny ones. And if this carried on for a long time, then surely the climate would change….

        Please explain.

      • Dear Peter:
        Math shows that cloud coverage over a year or more must be constant for the sensible heat absorbed by the earth is constant. While clouds can affect daily temperatures, in the long run they have no impact on climates.

      • Nabil,
        Reality trumps your math.

      • Dear Hunter, please show me the different realty. Dr. Dessler’s work is in agreement with my math. Clouds are an effect of the earth’s energy balance, not a cause.

      • Sorry, error “reality.”

      • Ha-ha, nice work, Joshua. Shining a spotlight on Trenberth Logic (what used to be called pretzel logic) is a valuable public service. Lovely.

      • I have no trouble with shining the spotlight on the reasoning of people on both sides of the debate. It’s interesting that you think that I’d have trouble with that. Why would that be, Ken?

      • Be gracious when someone pays you a compliment…

      • touché

      • Really, Josh? Show me where you shine a spotlight on the reasoning of people on the “AGW” side of the debate, which isn’t adamant, unthinking agreement with. Show me posts on RC starting arguments against their logic as you love to do here and elsewhere.

        You are very contrarian: saying things just for the sake of starting arguments.

      • Ged -

        Re-post that on a non-technical thread where I’m active, and I’ll take it up with you.

      • Josh-

        That is very reasonable. You have my word, I will take you up on that if I see and feel you are acting contrarian in the future.

      • intrepid_wanders

        Joshua,

        Some of us were in your shoes, a “few” years ago. The inconvenient physics are that cloud shapes are not “well-mixed” GHGs. The variable geometries would be a modeling hell (Gavin, let me know if I a wrong) Until the anticyckonic Red Spot on Jupiter can be adequately “modeled”, there is going to be a some doubt on these forcings and feedbacks.

        H/T to you on not sounding like a “bunny” ;) I image Eli is back in his comfy hole, studying his ENSO for a ‘deflated’ shot on CH.

  23. According to Pinker et al., 2005, surface solar irradiance increased by an average 0.16 W/m^2/year over the 18 year period 1983 – 2001 or 2.9 W/m^2 over the entire period.

    This change in surface solar irradiance over 1983 – 2001 is almost exactly 1.2% of the mean total surface solar irradiance of the more recent 2000 – 2004 CERES period of 239.6 W/m^2 for which the mean Bond albedo has been claimed to be 0.298 and mean surface albedo to be 0.067 (Trenberth, Fasullo and Kiehl, 2009).

    The ISCCP/GISS/NASA record for satellite-based cloud cover determinations suggests a mean global cloud cover over the 2000 – 2004 CERES period of about 65.6% and over the entire 1983 – 2008 27-year period a mean of about 66.4±1.5% (±1 sigma).

    ISCCP/FD and Earthshine albedo data for the 2000 – 2004 period enables estimation of the relationship between albedo and total cloud cover and it is best described by the simple relationship:

    Bond albedo (A) ~ 0.353C + 0.067 where C = cloud cover. The 0.067 term represents the surface SW reflection (albedo). For example, for all of 2000 – 2004; A = 0.298 = 0.353 x 0.654 + 0.067

    According to ISCCP/GISS/NASA mean global cloud cover declined from about 0.677 (67.7%) in 1983 to about 0.649 (64.9%) in 2001 or a decline of 0.028 (2.8%).

    This means that in 1983; A ~ 0.353 x 0.677 + 0.067 = 0.305

    and in 2001; A = 0.353 x 0.649 + 0.067 = 0.296

    Thus in 1983; 1 – A = 1 – 0.305 = 0.695

    and in 2001; 1 – A = 1 – 0.296 = 0.704

    Therefore, between 1983 and 2001, the known reduction in the Earth’s albedo A as measured by ISCCP/GISS/NASA should have increased total surface solar irradiance by 200 x [(0.704 - 0.695)/(0.704 + 0.695)]% = 200 x (0.009/1.399)% = 1.3%

    This estimate of ~1.3% increase in solar irradiance from cloud cover reduction over the 18 year period 1983 – 2001 is very close to the ~1.2% increase in solar irradiance measured by Pinker et al (2005) for the same period.

    The period 1983 – 2001 was a period of claimed significant global (surface) warming.

    However, within the likely precision of the available data for the above exercise (perhaps of the order of say ±0.5% at ± 2 sigma?), it may be concluded that it is easily possible that the finding of Pinker et al (2005) regarding the increase in surface solar irradiance over that period was due to an almost exactly equivalent decrease in Earth’s Bond albedo resulting from mean global cloud cover reduction.

    • There is an interesting paper from the EBRO Observatory in Spain.

      They measured bright sunshine and the graph of changes looks remarkably like the temperature record.

      “There is an overall increasing trend in the number of
      bright sunshine hours, amounting to about 100 h in
      the last 100 years (0.96 h/year), which represents an
      increase in bright sunshine hours of about 4% in a
      century.”

      In relation to clouds though they say this:

      “It is surprising that a statistically significant increase
      in cloudiness is not accompanied by a simultaneous
      decrease in sunshine at Ebro Observatory over the past
      century. The explanation may lie in a change in the
      proportions of the cloud types. We have shown how high
      clouds, less dense and optically more transparent than
      low clouds, have increased during the last part of the
      century, with perhaps little effect on the sunshine records.”

      At Ebro, cloudiness is not a replacement for sunshine.

      http://www.iac.es/folleto/research/preprints/files/PP08038.pdf

      • That is why I used the ISCCP/FD satellite data with cross check against the Earthshine. Palle et al have lots of (good) papers on this (surface) estimation problem. Remember I was primarily considering albedo.

  24. I’d like to echo Doc Martin’s post WAY above (didn’t reply there as the threading seemed wonky).

    It’s maddening that the models are not evaluated in an appropriate way- Doc Martin outlines the API testing procedure, can ANYONE outline the climate model testing procedure??

  25. Judith says:
    “IMO, none of these papers are of particular scientific interest”

    I respectfully submit that Spencer and Braswell’s 2010 paper, with its supporting study of phase space and the relationships revealed in time series linked points on the scatterplots is streets ahead of anything produced by the ‘Team’.

    Dessler’s 2011 ‘rebuttal’ is a joke. Bad logic, self contradiction, stawman arguments, and a fig2 which pretty much confirms what S&B 2011 says anyway.

    McIntyre points out that as Dessler now accepts the four month lag, his 2010 positive cloud feedback now becomes negative using his own data.

    :)

    • Tallbloke,
      Agreed.
      Your last point, though, is very misleading. Dessler only accepts that the four month timelag improves the significance of his regression fit – not the same as saying that he accepts that it is correct to apply this time-lag to his regression to abstract feedback.

  26. Judith says:
    “IMO, none of these papers are of particular scientific interest”

    A further thought is that Spencer’s demonstration that the data doesn’t tell us much of anything and the models don’t match the data is actually the important scientific result.

    It wouldn’t be of great interest if the Team hadn’t bulled the models up to make them sound better than they are, and wasted enourmous sums of public money in the process.

    • Let’s try to stick to the facts. Given Spencer’s analysis is correct, we can conclude that there are models which are really bad. Really bad in simulatiing ENSO variability, that’s all.
      I miss one point in the discussion here: Does this necessarily means that these models are bad in predicting a large scale climate change mostly caused by GHGs?
      Are these models made for predicting natural ENSO variability? If a modeler claims his model is made for this purpose, then you could make a point, tallbloke. Otherwise it’s silly.

      • Capo,
        No, it is silly to pretend climate is driven only by changes in GHGs. Climate scientists have ignored the first step of attribution, which is fully understanding natural climate variation. Until we know the range and velocity of natural climate variation, how can we ever determine what changes are anthropogenic and which are natural?

        ENSO variability is really only part of the issue. It seems ENSO is strengthened or weakened depending on the PDO. The interplay of a long-term oscillation with a shorter term ENSO may be important. And, of course, the role of clouds. We know Svensmark’s theory has gained credence from the CERN group, but there are other theories about clouds which may play a role – including dimethyl sulfide, which grows in the oceans at a greater rate when atmospheric CO2 is elevated. http://www.agu.org/pubs/crossref/2011/2011GL047069.shtml

      • Hi Ron,

        CO2 is not a limiting nutrient usually – so adding more doesn’t change phytoplankton abundance. What does change abundance is upwelling of nutrient rich water in the Humboldt Current off South America and in the California Current. ENSO and the PFO respectively.

        Cheers

      • Not forgetting the iron supply from volcanos. Another limiting nutrient made abundant by eruptions, which seem to have been more frequent at times of low solar activity. Plankton create their own UV shade by chucking CCN forming chemicals into the atmosphere above them.

        Solar feedbacks take many forms.

      • Lots of good papers on field microcosm experiments confirming increased cyanbacterial growth rates with increasing CO2 – even in places like the Ross Sea. Easy to find by Googling. Regards Steve

      • ditto.

  27. Dr. Curry,

    I think these papers and the related ‘excitement’ concerning them have provided a wonderful insight into the quality of the science and the analytical abilities of the people who have had such a dominant role in determining the current “understanding” of the science. Very revealing.

    We may not have learned much science, but we sure learned a lot about the science.

    • I think that recent events betray the deep insecurities of The Team and its camp followers

      Scientists truly confident in their science and in their beliefs would rise insouciantly above these relatively minor squabbles..their place is on the World Stage, not grubbing around in the trenches.

      And perhaps, pre-Climategate, pre-Copenhagen, they would have done so.
      But now their self-confidence has gone and they are having to make futile gestures to satisfy their own support that they are ‘doing something’..such as the needless resignation of Wolfgang Who?

      Peter Cook got it spot on years ago: see

      And the panicked fast track response to Spencer and Braswell only gives legitimacy to that paper and brings uncomfortable interpretations of the data front and centre once more. Will Dressler’s paper make many new converts to (C)AGW alarmism? Unlikely. But the fact of its existence and the rapidity of its gestation will draw attention to the controversy and cast further doubts on ‘conventional’ climatology.

      It may be that the Team is composed of some of the finest scientists the world has ever seen (though I’d need a lot of new evidence to be convinced). But as strategic thinkers in a drawn out conflict they have very little credibility.

      • latimer,

        Will Dressler’s paper make many new converts to (C)AGW alarmism? Unlikely. But the fact of its existence and the rapidity of its gestation will draw attention to the controversy and cast further doubts on ‘conventional’ climatology.

        The other day you were complaining that criticism of S&B’s methodology wasn’t published in the peer reviewed literature. Now you are criticising Dessler for publishing criticism of S&B in the peer reviewed literature. ISTM that the “Team” can’t win whatever they do.

      • Andrew, actually the “Team” can win if they present the most compelling science. A thoroughgoing refutation of SB11 is still possible, but Dessler’s paper is not it.

        In the past Dessler has behaved better than some on the Team. For example, he archives and/or shares his data and he supported McIntyre in the effort to get review comments of AR4 available online. And he was cordial in his early communications with Roy Spencer, even though they disagreed.

        However, in this paper Dessler uses phrases that make it sound like he doesn’t think Roy is in touch with reality. Dessler’s science and statistics are not sound while he is making claims for his results which are not even close to being warranted. I’m still shocked by Dessler’s claim in his video that Spencer did not use real data. I think Roy will be very surprised to hear that.

        I cannot follow all of the maths, but I can follow the physical arguments and the behavior of the scientists. If you are confident in your results, you don’t have to get emotional, use unscientific language or falsely accuse others of “serial mistakes.” (Between Dessler and Trenberth, they did all of these.)

      • Ron,

        There are some people for whom nothing climate scientists could ever do will be enough, they will never be persuaded.

        I don’t see that Dessler has done anything wrong whatsoever. He makes pretty damning arguments against Spencer’s (and Lindzen’s) claims (as have Trenberth and Barry Bickmore and people such as Fred here) and if his paper is strongly worded then that’s what S&B deserved. I find his video much less shocking than Spencer’s blatant cherry picking of the model results which best suited the argument he wanted to make.

        I made the point over at Bart’s that Trenberth’s article was a mistake (which is not to say it was factually incorrect) but he is no more guilty than Spencer of politicising the whole episode.

      • I wasn’t ‘complaining’ at all about them publishing their reply in a peer-reviewed journal. It is the correct thing to do. No complaints from me there.

        But I was pointing out that the very unusual fasttracking of the reply drew far more attention and betrayed their insecurities. They really must be s**t scared of S&B. That’s all

        But I was pointing out

      • aa,
        Criticism of the team is slowly published if at all.
        The team has a growing scalp count of editors who dared crossed them.
        Papers critical of the AGW consensus get published slowly or worse.
        Dessler’s clearly flawed paper was scooted through to publication like rice through a goose.
        Ignoring this does not make you look more engaged.

  28. The joy of black box analysis is that you can ignore whatever is in the black box. In fact, there are many different ways to create a system with the same response.

    On the other hand, sometimes you can find clues if you look beyond the obvious data that the system provides. Here’s an electrical example:

    Any linear (resistive) circuit can be modeled as either a voltage source in series with a resistance (Thevenin equivalent circuit) or as a current source in parallel with a resistance (Norton equivalent circuit). There is no way, using voltage and current measurements, to tell what you have in the black box.

    If you are allowed to measure the temperature of the black box, you get a clue as to what is in the box. A voltage source in series with a resistor will generate no heat unless it flows current into an external load. A current source in parallel with a resistor will generate maximum heat if it can not flow current into an external load. If you can measure heat, you can distinguish voltage source from current source.

    So, can one use a black box analysis to distinguish a system with negative feedback from one with positive feedback? I would have to say that you can not rule the possibility out. Some clever person might find an convincing analog of the above example. ;-)

    • commieBob, You mean like ‘Pong’? It would measure both positive & negative feedback & if you could ‘score’ high enough, you would even be peer reviewed. And it cost just a quarter too. This black box was digital… you could use ‘High Speed’ by Williams though, if you are set on using analog. It is more fun I think.

    • Electrophysics is a much easier problem than climate because it’s linear. Having said that, the consequences of positive and negative feedback are so dramatically different, one should, in principle, be able to distinguish the two, even in a nonlinear system, through black box analysis. How to do that in reality is another question entirely.

  29. Now that the Dessler Spencer exchange has advance scientific and public understanding, the general public seeing that poor communication and statistical math skills ensure job security for climatologists, perhaps we can move on some of the more interesting aspects of modeling weather/climate.
    Katia really wants to be a major storm, she tries and tries, but she just can’t sustain major status. When she gets all the energy input she needs to be major, she just screws it up by blocking the paths she needs to get the energy to remain major. That should be a simple feed back problem that I am sure that Dessler is capable of solving since he has such a strong grasp of control systems.

    • Since Katia is weather, that analogy may be lost. How about a simple climate related example of a system with a region of instability (not to be confused with chaos).

  30. Look at the temperature of earth over the past ten thousand years that was recorded in the Antarctic and Greenland ice core data…

    The temperature has been within plus or minus two degrees during all of the past ten thousand years. The temperature has been within plus or minus one degree for most of the past ten thousand years.
    There is powerful negative feedback to temperature.
    When we are warm, it snows more. When we are cool it snows less.

    http://www.ncdc.noaa.gov/paleo/pubs/alley2000/alley2000.html

    This chart does show that snow accumulation is changed by earth temperature. This is what makes earth temperature extremely stable. This is the powerful negative feedback.

    • Extremely stable is relative. Look at the system over the past 100K years and there are two relatively stable operating points. Since there are two potential operating points, there is a region of instability in the performance curve. At the upper point you would have stronger negative feed back and the lower you would have stronger positive feed back. Both points have sufficient opposite signed feed backs to maintain relative stability. To me this implies a non-linear sensitivity or a region of instability.

      • To go a little further with Herman Pope’s comment, the first part of the Holocene had more temperature fluctuation, because there was a larger area of glaciers/snow coverage. The system becomes more stable when there is a minimum of snow/glacier coverage.

        The there is the paradox, when there are major snow events with a minimal snow albedo, they can produce more climate impact. That impact is more regional than global because of the thermal mass of the oceans. The southerly shift of the snow cover shifts the cloud cover average further south where it has a greater impact on solar reflection. A delay in feedback/forcing that would be more pronounced during a cool climate oscillation.

        So in my opinion, you can’t confidently model global climate until you can confidently model regional climate.

    • herman,

      Yes, temperatures have been quite stable for the last 10K years but we know that hasn’t always been the case – for example we know that the temperature variation between ice ages and interglacials is about 5C. Yet the forcing effect of Milankovitch cycles which are responsible for the ice ages is pretty small, about 0.5 Wm2 IIRC, so how is that compatible with the strong negative feedback which you mention?

      • my powerful negative feedback is based on ice and water.
        Your are right that Milankovitch cycles are small changes to the heat balance of earth.
        Milankovitch cycles are unlikely causes or major contributors to global warming or ice age cycles.
        The theory expressed by Maurice Ewing and William Donn is much more easy for an engineer, like me, to believe.

    • GHGs have been very stable, probably within 5% during that stable temperature period. Do you ever wonder whether a 40% bump-up in GHGs in a century (only the last 1 per cent of those 10 k years) might do something to that stability, with maybe 100% or more to follow? Some people think it might, and that is based on paleo evidence when it was many degrees warmer with more CO2 only a few tens of millions of years ago. Worth thinking about, I would suggest.

      • Open a warm and a cold carbonated drink. The vapor pressure of CO2 is changed a large amount by temperature changes of the the liquid. The oceans on earth are a huge carbonated drink. CO2 does go up and down with temperature. That is simple physics. They use this relationship to prove that CO2 is driving temperature. That is not something that an engineer, like me, can believe. Temperature does drive the Vapor Pressure of the CO2. CO2 is a trace gas and may have a trace effect on temperature, but only a trace.

      • 50 million years ago your trace gas was making things a lot warmer on earth, or do you have an alternative hypothesis that science has missed somehow?

      • Earth started out warmer than now. Nothing to do with CO2. Earth cooled and warmed many times. Nothing to do with CO2. Earth cooled and warmed when CO2 was much higher than now. Earth does know how to the grab CO2 from the atmosphere and store it. The current level of CO2 is closer to the lowest value that life as we know it can exist, by order of magnitude, than it is to the highest. Below 150 ppm life as we know it cannot survive. Above 8000 ppm we have a problem. we are below 400.

      • Science always misses a lot. When do you think the scientific progress will stop?

        To be so convinced and certain what the “trace gas” was doing 50 million years ago is very unscientific. It’s a little knowledge. It’s pseudo-science.

      • Low CO2 has been associated with the previous spell of icy conditions at the end of the Permian 250 million years ago too. It is correlated well.

  31. Dallas | September 7, 2011 at 10:07 am | Reply
    Extremely stable is relative. Look at the system over the past 100K years and there are two relatively stable operating points. Since there are two potential operating points, there is a region of instability in the performance curve.

    The upper and lower bounds like correspond to a phase change. Similar to the curve you get when heating ice to water and then to steam, there are plateaus at 0C and 100C.

    In the case of the earth there are three obvious phase changes. Water to water vapor largely towards the equator, and water to ice largely towards the poles, and water vapor to water and ice, largely towards the sky.

  32. Very true ferd, a big reason Venus is not a great example to compare to Earth. That and a couple dozen other reasons.

  33. Richard Saumarez

    I posted in the last S&B thread about the definition of feedback being imprecise.

    In the feedback equation used by both S&B and D, the feedback term is NOT a function of time. Yet they used lagged correlations. D10 used instantaneous regression, which is at least consistment with the equation he used.

    It really strikes me that, at least, the equations should be written in a way that conforms to their analysis. Since the feedback is delayed and is presumably a function of several, unknown, and presumably linear, compartments, it should be expressed as a series of convolutions. Since the system is assumed to be linear, it could be better expressed in terms of Laplace transforms and the analysis performed in the frequency domain.

    I am highly sceptical of authors who write an equation that purports to be a lumped physical model and use methods of analysis that do not reflect that equation.

    Perhaps I am being irredemiably stupid, and in Climate Science, the term DeltaT is implicitly a time series, but I really think these authors should use mathematics that is at least consistent with their analysis.

    I still stand by my comment that the issue of feedback needs to be looked at more rigorously from a systems/control theory point of view.

    • Richard Saumarez

      What am saying is this:

      The thrust of the SB thesis is that the is delay between forcing and feedback, as shown clearly in Figure 3 of their paper. If this is due to clouds,
      their basic equation could be written:

      CpdDT/dt=Sigma(fluxes)- lambda.Cl

      where Cl describes the effect of clouds. If we are dealing with a simple model, we could write:

      dCl/dDT=F(DT)

      so the basic equation becomes:
      cpdDT/dt=sigma(fluxes)-lambda.cl(DT,t)

      This is a very different equation to that used by SB and D because delay is explict in the Cl(DT,t) term.

      This should give rise to, in my opinion, more powerful methods of data analysis. Expressing the equation in Laplace transform terms. the pole of the “Cloud term”, and hence the delay can be extracted or the delay can be extracted from cross-phase spectra between T and Flux.

      I do think, however, that the feedback term should be expressed as a function of time because this makes the basic equation consistent with the way in which S&B have analysed their data.

      I am NOT saying that S&B are wrong – I am saying that at first sight, their equation does not appear to be consistent with their analysis. This could be a source of attack, although it does not detract from their arguement.

    • That is interesting. Since Spencer was criticizing Dessler, he was using the equation to illustrate the need for a time lag component, but never bothered attempting to rewrite the equation with that term. I guess that is the cliff hanger for the next petite paper.

    • I agreThe four main papers discussed here have reduced the thermodynamics of the system to thermo statics. The assumption of the psuedo equilibrium condition allows for great lattitude and error. Yet because of this, S&B with the lag times may well indicate that transforms or frequency analysis could be fruitful area to search. However, I would not assume that the system is linear, nor that the equations that should be used are tractable.

      I think the phase space presents the systems/control theory with the number of degrees of freedom problem. Perhaps a more rigourous set of simplications that could be agreed or examined with uninformative priors would indicate a better system definition that the authors could then examine within the nomenclature so an inordinate amount of time would not be spent looking at how the differences in the definitions defines the arguments.

      • Richard Saumarez

        I have no idea whether the equations concerning feedback are linear or not. The point is that the term as I have written it Cl(DT,t) involves lag.

        Linearity is the best approach of engineering mathematics (viz Gelerkin) or the last refuge of mathematical scoundrels!

        I am slightly dismayed by an analysis that does not conform to the root eqauation descibing the system.

  34. Has anyone done an experiment comparing the cooling rate of two identical objects exposed to the night sky?

    Since glass reflects IR, it would seem that one could make a glass periscope to cover both objects in such a way that the radiation received by both from the periscope would be identical, but that the periscope would reflect the DLR from the sky onto one object and shield the other object. Thus, if DLR is significant. there should be an observable difference in the cooling rate of both objects.

    These results could then be compared to using the device in the daytime, and also with and without clouds, to answer in a direct fashion the effects of DLR without the confusion that results from attempting to answer these questions based on theory.

    Has there been such a study conducted and the results published? I’ve seen some studies, but in the main the experimental design did not provide the necessary rigor to compare identical objects with and without DLR.

    • well not with a periscope, but quite a few actually. The results in all are complicated by inconsistent R values and/or convection control attempts. The easiest experiment is a vacuum bottle with and without reflective lining. This of course doesn’t let you measure the direct infrared radiation from the sky, but it does show the impact of change in radiant loss, which is the point anyway.

    • What you are suggesting may not work. The shielding for one object would need to be at a very low temperature, around 3K.
      I am working to improve this experiment -

      http://tallbloke.wordpress.com/2011/08/25/konrad-empirical-test-of-ocean-cooling-and-back-radiation-theory/

      - with a cold aluminium “sky” at around -5C. However this experiment works because most outgoing LWIR is reflected back to the sample, not just the small amount that would be backscattered by GHGs.

  35. Richard Saumarez

    I should say that the feedback, while it is clearly a function of time as an instantaeous function of temperature, it is not a function of time extending away from the specific instant as is the case of the convolution integral.
    Apologies

  36. “Latimer Alder | September 7, 2011 at 10:17 am | Reply
    Scientists truly confident in their science and in their beliefs would rise insouciantly above these relatively minor squabbles..their place is on the World Stage, not grubbing around in the trenches.”

    Current events have very little to do with science. It is politics using science to justify one point of view above another. A quick review of the history of science shows that many times in the past we have been lead down the wrong path by what appears at the time to be a very persuasive argument, only to find out decades or centuries later that it is garbage, with many lives and fortunes wasted along the way.

    Study the history of Cancer treatment, and how surgery rose to prominence as the treatment of choice, largely as a result of personalities, with little or any science to justify. The mistaken idea that we could cure cancer if we simply cut a bit deeper, to make sure we “got it all out”.

    Now look at climate science. Same problem. Personalities and their beliefs, not science, are driving the conclusions.

    • “Study the history of Cancer treatment, and how surgery rose to prominence as the treatment of choice, largely as a result of personalities, with little or any science to justify. The mistaken idea that we could cure cancer if we simply cut a bit deeper, to make sure we “got it all out”.

      I rarely respond to false statements that are off-topic in a particular thread – even statements as egregiously wrong as the above – but since it might influence decisions made by readers with consequence for their personal health, I thought it worthwhile to point out that the statement is garbage.

      Probably about half of all cancers are cured by surgery (far more than half if we include non-melanoma skin cancers). No other treatment modality even comes close, with only occasional cures by radiation therapy alone and even fewer with chemotherapy. Surgery is rarely curative once a cancer has spread, but rather than a reason to dismiss surgery, it is a reason to detect cancers early, when they are curable in the majority of instances.

      The point that “cutting deeper” isn’t always a guarantee of a greater cure rate is true in general, but in many situations, the extent of removal is critically important – these are decisions that must be made on a case by case basis and not through generalizations. If that were the only point Ferd was making, I wouldn’t have bothered to respond, but the implication that there is little scientific basis for considering surgery to be an effective curative treatment could not be allowed to go unanswered.

      On a more general level, I deplore the tendency of some individuals to give unwarranted medical advice on the Internet. That is particularly true of people like Ferd whose knowledge the matter appears to be seriously deficient, but it also applies to experts. I would not have made any statement unsolicited, but I couldn’t let Ferd’s remark stand uncorrected because it had the potential to cause harm if taken seriously.

  37. The hijacking of the peer review process by the climate cabal led by Trenberth has reached ridiculous extents. He prevented LC and SB from being published in more mainstram journals and when they manage to publish anyway, very late, he takes a fit. How did this zealot become so important? Dessler by contrast, like Santer before him, and indeed anyone who claims thermageddon is around the corner, gets a free ride in GRL and Science with palpably bad papers. Regardless of the merits or demerits of SB and LC, climate science publishing stinks! Good riddance to Wagner, but they should all resign, every man jack of them until the smell dissipates.

    As for Trenberth wrongly saying Spencer has a track record of errors, well Trenberths attribution of the Russian drought and Pakistan flood to global warming was refuted by virtually everyone. Zealots such as he, who don’t need data or models just their own arrogant shamanistic certainty, should not even be reviewers, never mind apparently being gatekeepers of climate science publishing…

    Now obviously AR5 is the big issue and it’s a top down approach. They want carbon taxes, therefore they need strong conclusions in AR5, therefore they don’t want any alternative narrative to appear anywhere, lest they might have to write about it, therefore alternative views are ruthlessly suppressed.

  38. “As for Trenberth wrongly saying Spencer has a track record of errors, well Trenberths attribution of the Russian drought and Pakistan flood to global warming was refuted by virtually everyone.”

    And the flooding in Australia.

  39. “JC comments

    I’ve done a quick read of Dessler (2011). I have the same problem with Dessler’s paper that I had with S&B and LC. These analyses don’t really tell us anything about cloud feedback, although they are interpreted as doing so. ”

    So I am sure we will see the editor of GRL resign really soon.

  40. Judith Curry

    I believe Climate Etc. has run many excellent threads covering interesting topics related to the ongoing scientific and policy debate related to climate change (aka global warming) and I thank you for this.

    I also agree with Paul K. regarding this thread (Part III):

    I think it is the worst thread I have ever seen on Climate Etc.

    S+B 2011 has been published.

    The Editor-in-chief of the publication resigned for some unknown reason, attempting to discredit the publication in the process by naming it the reason for his resignation.

    S+B 2011 has been bad-mouthed by the “insiders” (as have its authors), but it has NOT been scientifically refuted.

    Until it is scientifically refuted, it stands.

    Dessler’s paper has not refuted the basic conclusion of S+B 2011, but simply skirted around it with another hypothesis and a statement that clouds per se have not impacted our climate (duh?).

    The S+B 2011 conclusion seems so non-controversial to me that I have a hard time figuring out what all the brouhaha is all about.

    Let me re-quote it here, with the kind request if YOU see anything controversial, please tell me what it is. If not, I will assume that you are in basic agreement with the S+B 2011 conclusion quoted below:

    It is concluded that atmospheric feedback diagnosis of the climate system remains an unsolved problem, due primarily to the inability to distinguish between radiative forcing and radiative feedback in satellite radiative budget observations.

    Seems like a no-brainer to me, but I am not a climate scientist.

    Thanks.

    Max

    • Max,
      My comment on the quality of the thread was about Part II.

    • Judith Curry

      Judith, I see you already answered my question above regarding the validity of the S+B2011 conclusion on cloud feedbacks versus forcing (in an earlier post to Robert of Ottawa):

      It is concluded that atmospheric feedback diagnosis of the climate system remains an unsolved problem, due primarily to the inability to distinguish between radiative forcing and radiative feedback in satellite radiative budget observations.

      Your response:

      this is a totally sensible conclusion

      That answers it for me.

      Max

    • SB11 based their conclusion on seeing that cloud effects lead El Nino, and can appear as part of the forcing, which would obviously confuse any evaluation of sensitivity, being an unknown variable ‘external’ forcing. The dispute to that comes in several forms. Firstly, it is too weak to be a significant forcing for El Nino as Dessler shows. Secondly, it is likely a statistical artifact of SB11′s compositing method that clouds may precede El Ninos and appear like a forcing for El Nino when they are actually a response to the previous la Nina. Thirdly, the idea that clouds can be considered an ‘external’ forcing implies that they have behaviors that don’t depend on the climate that they are in, which is on its face implausible.

      • Hi Jim,

        This is not true at all. ENSO involves contemporaneous effects on wind, ocean circulation and cloud. ‘El Niño/Southern Oscillation (ENSO) is the most important coupled ocean-atmosphere phenomenon to cause global climate variability on interannual time scales. Here we attempt to monitor ENSO by basing the Multivariate ENSO Index (MEI) on the six main observed variables over the tropical Pacific. These six variables are: sea-level pressure (P), zonal (U) and meridional (V) components of the surface wind, sea surface temperature (S), surface air temperature (A), and total cloudiness fraction of the sky (C). These observations have been collected and published in ICOADS for many years.’ http://www.esrl.noaa.gov/psd/enso/mei/

        Temperature changes lag – presumably because of thermal inertia of the oceans especially. In the much maligned McLean 2009 – http://members.iinet.net.au/~glrmc/McLean_deFreitas_Carter_JGR_2009.pdf – temperature lags the Southern Oscillation Index by 5 to 7 months. The SOI is of course an ENSO indicator involving sea level pressure in Darwin and Tahiti.

        ENSO varies on interannular to millennial scales – along with cloud and therefore cloud radiative forcing. The cloud changes are fast – temperature lags. The changes in radiative flux can be 3W/m^2 in months. Clouds don’t cause ENSO and no one has ever claimed that they do.

        Cheers

      • On a previous thread it was worked out that radiative forcing changes near 1 W/m2 (which SB11 deal with) can only change a 500 m ocean temperature by .01 degrees in a year. I only make the point that these forcings have no impact on El Ninos.

      • Year to year changes are much greater – look for instance at Dessler 2010 – Figure 1 for 2009 – 2010.

        Or over a longer time – http://isccp.giss.nasa.gov/zFD/an9090_TOTnet_toa.gif

        Considering the Pacific Ocean as a slab for simple energy exchange doesn’t work – the SST are more a function of ocean dynamics. SST can change by 5 degrees C in months in the central Pacific.

        I simply make the point that no one ever claimed that ENSO was driven by driven by surface temperature change.

      • CH, it is unclear to me whether you think Spencer’s small oscillations in forcing are real or fictional, and if real, whether they are important, and what are they? The numbers I gave suggest, real or not, they are not important.

      • Jim,

        You make up some numbers – and then say that they are unimportant – I agree.

      • OK Chief, I’m confused now. Yesterday you said ENSO *is* clouds. What did you mean?

      • Just that ENSO involves not just SST – but wind, clouds and ocean currents. It is all the El Nino Southern Oscillation. I have linked above to Claus Wolter’s multivariate ENSO index above – which is the proper way to look at ENSO.

        There is a good audio/visual ENSO presentation here – http://judithcurry.com/2011/09/06/spencer-braswell-part-iii/#comment-110072 – and a funny doggy.

      • Brian H

        Indeed, the projections that are fooling you are Chief’s.

        ENSO is just temperature change of 0.5 °C or more at the surface averaged over the east-central tropical Pacific Ocean, not counting accompanying teleconnections over the entirel globe that Chief projects confusedly as all the same thing, having lost his sense of causality.

        The actual region of ENSO is a tiny part of the Pacific Ocean. You may as well argue that the Atlantic and Pacific are just spurs of the Arctic.

        See, all this is implicit in my question where I mentioned ‘equator’, and emphasized where I ask about teleconnections.

        If you want to get me to ask ludicrous questions about the entire Pacific Ocean, please do it in a way less likely to introduce a straw man to our already confunded Hydrologist.

        Right now, I’m just asking ludicrous questions about a zit on the Pacific’s nethers near Peru.

      • Chief

        What a kind invitation. I’ve always wanted to visit New Zealand.

        So, you say the Arctic cold is being transported to the Gulf of the St. Lawrence, where it forms a baffle that channels what once would have been Arctic winter snows into European snowfalls instead, like a failed souffle flopping over off the North Pole to eastern North America and Europe during the winter?

        That seems a conventional enough answer, especially for you.

        Do you believe this jauntily flipped winter chef’s hat, this temperate-zone liripipe of snow, will be a long term mode?

        What would be the best policy response to this new regime, should it last, in your view?

        I mean, can New Zealand really absorb the entire Northern Hemisphere into its sheep pastures?

        And.. while we’re on the topic of two significant “coupled ocean-atmosphere phenomenon to cause global climate variability on interannual time scales,” do you still contend the deep ocean is the cause of (per Occam) both?

        Do you have a proposed mechanism for this cause?

        The AGW hypothesis appears to work well enough, absent a working theory.

      • The sea surface temperature in the Nino regions – http://ioc-goos-oopc.org/state_of_the_ocean/sur/pac/ – is but one aspect of ENSO.

        http://www.duke.edu/~ts24/ENSO/

        The actual area impacted by a cool PDO and a La Nina is immense – http://www.osdpd.noaa.gov/data/sst/anomaly/2011/anomnight.1.24.2011.gif

        It is obviously why I quoted the world’s leading ENSO expert – Claus Wolter.

        ‘El Niño/Southern Oscillation (ENSO) is the most important coupled ocean-atmosphere phenomenon to cause global climate variability on interannual time scales. Here we attempt to monitor ENSO by basing the Multivariate ENSO Index (MEI) on the six main observed variables over the tropical Pacific. These six variables are: sea-level pressure (P), zonal (U) and meridional (V) components of the surface wind, sea surface temperature (S), surface air temperature (A), and total cloudiness fraction of the sky (C). These observations have been collected and published in ICOADS for many years.’ http://www.esrl.noaa.gov/psd/enso/mei/

        BTW – here are Andy Dessler’s feedbacks for global warming –

        f = 0.6 (water vapour) – 0.3 (moist adiabatic) + 0.1 (ice albedo) + 0.15 (cloud)

        It is a separate process happening in parallel to Pacific variability.

        It is a bit difficult talking to someone who doesn’t share the same reality as everyone else – i.e. fails to process information that doesn’t agree with preconceived notions – and responds with wilder and sillier notions. This is why I refer to it as a spaceship cult.

      • Now you have overstepped the mark by a great deal. I rescind my offer and you can go to New Zealand for all I care. We don’t have any sheep in Queensland so I am sure you would be lonely anyway. We have bulls with horns and horses – just like Texas. My local pub has an inside bull ring. It is a place where a real man can ride broncos, drink, spit and swear at the same time. You would be well out of order riding sheep around the place.

        As the name implies – the Atlantic Multidecadal Oscillation is likely to last more than a week and half. I can only imagine that your electric kool-aid addled brain can no longer remember a decade past last alone 3 or 4.

        As you have been so addled by kool-aid that you no longer recognise the ordinary meaning of words – I have linked to an animation. If that is still too complex for you – there is one of a doggy somewhere. You won’t get the analogy – just laugh at the doggy.

        ENSO begins with wind driven deep ocean upwelling on the western coast of South America. This water is frigid, nutrient rich and super saturated with carbon dioxide at surface pressures. Upwelling is explained in the AV. I know it is difficult – but try to concentrate.

      • Seems to me that there is always a risk of ending up in a muddle when talking about the causes of ENSO because of the inter-related nature of and the feedback loops between all the factors the multivariate index is based on.

        Ultimately it is solar variation which drives ENSO. It’s not a coincidence that the big el nino’s occur near or soon after solar minimum during positive PDO phase particularly. I believe this to be the oceanic reaction to increased cloud fraction. Excess energy sequestered when the sun is strong and cloud is reduced, wells up out of the ocean when insolation is reduced.

        Another part of the cycle is the cold upwelling at the start of La Nina bringing up the nutrients which cause plankton blooms. These protect themselves from UV at the high part of the solar cycle by generating the airborne chemicals which seed CCN’s. The resulting cloudiness in the east Pacific starts the pressure differentials and winds which are the precursor to el nino later on.

        Make sense?

      • Alas

        And I so looked forward to the twenty hour plane ride each way.

        So, the wind drives ENSO and cloud formation. One expects an atmospheric source also drives the Arctic melt.

        And that makes these events feedbacks for whatever forcing drives the wind to pick up, then?

        Say, I heard a promising hypothesis about what forcing it could be..

      • Bart,

        The mechanisms of ENSO as they are generally understood are mostly in in the Duke University animation.

        I have no objection to clever and amusing interactions – but you have descended into pointless distraction for the sake of trivialising the discourse for some blinkered ideological purpose.

        You are either sowing confusion for the sake of it – or are actually confused and can’t process simple notions put to you in cartoon format.

        Cheers

      • K. G. Pavlakis: ENSO surface shortwave radiation over the tropical Pacific 2008

        SST anomalies in the Ni˜no 6 region tend to be out of phase
        with those over the central and east equatorial Pacific. Therefore,
        low (high) DSR values over the off-equatorial western
        Pacific region precede the low (high) SSTs in the same region.
        Thus, negative (positive) DSR-As in the off-equatorial
        west Pacific might be contributing to the decrease (increase)
        in SST there. Sensitivity tests with atmospheric models show
        that off-equatorial west Pacific SST anomalies of only modest
        amplitude are sufficient to produce equatorial easterly or
        westerly winds over the western Pacific and consequently
        provide a negative feedback for the ocean-atmosphere system
        so as to oscillate between warm and cold ENSO phases
        (Wang et al., 1999)…..

        ….We investigated the covariability between the time-series of the
        mean regional DSR anomaly in the western Pacific (10 S–
        5 N, 120–140 E) and the Ni˜no-3.4 index, and found a significant
        correlation (coefficient equal to 0.78). The DSR-A
        changes between +20Wm−2 and −20Wm−2 between warm
        and cold ENSO events, respectively.

        We further investigated the covariability between the timeseries
        of mean regional anomalies of DSR (DSR-A) in the
        off-equatorial western Pacific region (7–15 N 150–170 E)
        and the Ni˜no-3.4 index. The DSR-A in this region leads the
        corresponding Ni˜no-3.4 index by about 7 months and it is
        indicative of ENSO operating through the mechanism of the
        western Pacific oscillator.

        http://www.atmos-chem-phys.net/8/5565/2008/

      • tallbloke

        Atmospheric feedbacks magnify the solar signal some order of magnitude, you say?

        These positive feedbacks might include clouds?

        Why, if you can produce another two orders of magnitude, you might be onto something big!

        Might I ask why you call me a scientist? Do I seem so sciency to you? Is that your basis for evaluating what is and isn’t science? The labcoat it wears?

        I’m nothing of the sort of a scientist. I’m barely coherent, ill-read, scantly competent in the fields of logic and reason, mathematically limited and admit freely to a number of biases that prejudice my viewpoint.

        That I still come to a cleaner and clearer perspective and understanding than you show here indicates perhaps who has and has not failed.

        Scientists have quite clearly shown the quite commonplace and unremarkable fact that clouds don’t change without cause. From there, that cause, that forcing, has been shown to have a demonstrable large anthropogenic component yet to be determined to much precision or with much accuracy.

        The details being quibbled over do not touch this logic.

      • Chief

        Oh Chief.

        I know you mostly mean to be sincere, but you forget you’re dealing with mere mortals.

        And, it appears, you often forget to get a decent night’s sleep, which worries me.

        Harumph, you actually believed my little diversions sincere? You do need a good night’s sleep.

        Mental gymnastics and the gyrations you undertake are diversionary, intentional or no. Divide et impera, man. One intractable at a time. Strong medicine was needed to wring sense out of you finally, so I administered it cleanly, and might I add my success endorses the diagnosis and treatment.

        “There is too much natural variability to distinguish a greenhouse gas signal – although I agree it should be there. It is one factor in many..”

        See, you ought have started there.

        Everything that precedes is rehash, too long, off topic, however interesting or true, or of dubious value. Why go there?

      • Bart,

        You were saying that the melting Arctic Ice made a difference to albedo. Mad www theories abound – and you certainly shows clear signs of incoherence – dressed in stained pj’s, drooling, ill mannered, bullying, resistant to change, surly, mumbling, obnoxious, belligerent, hectoring, little able to do more than repeat a few simple phrases – but I am normally patient for mad theorists explaining in language simple enough for anyone. Science communication is the prime responsibility of an environmental scientist. So I find your antics and pleasure at attempting to abuse my good nature obnoxious – something you pursue for some odd political point scoring exercise.

        Everything I said was not to you however – have you not understood
        my comment in relation to the pointlessness of discourse with the perpetually addled – waiting for AGW with the faithfulness of a spaceship cultist.

        I was correcting your misdirections in case anyone believed you. I was talking past you and not to you. You are a troll and a bully – and I can only suppose that it is coming out of a millennialist inspired desperation.

        You have lost – the world is not warming and there will never be global or even very widespread cap and trade. Loser.

        Cheers
        Robert I Ellison
        Chief Hydrologist

      • Chief

        So hard to tell the inmates from the staff.

        The delusion you have that anyone you care about might want to listen to me is a dead giveaway.

        I mean, clearly you don’t want to listen to me.

        If your care for others were genuine, you wouldn’t be so predisposed to antics and tantrums when your toys are taken away.

      • Chief

        You play games and then complain when you lose. I reply humorously and you complain that I don’t take it seriously, I reply with detail tand reference to your trifling comment and you complain that it is too long and off topic, you insult and denigrate and I reply with patience and science.

        And then you complain that I don’t want to listen? Well that is true at least.

        I go back to my original point – you are a ‘spaceship cultist’ and the spaceship isn’t coming any time soon. Have fun with at the kool-aid party.

      • Chief

        And I thought we might be making progress.

        I’ve been here for the whole thread, and it’s still right there. No one needs a summary, nor will your transparently self-serving perspective convince anyone who reads the whole of the exchange objectively.

        I understand, it’s a tough habit to break, that agenda manipulating slant toward your own personal bias that pervades all your posts. But until you can cleanly and objectively state a case, you’re likely doomed to remain blind to the world as it is, chained by your own hallucinatory terrain.

        Really, what cult would have the likes of thee or me?

        And what cult that would have me or thee would anyone, including we, tolerate the company of?

      • Chief

        Here’s a puzzle for you.

        When will http://nsidc.org/arcticseaicenews/ hit the point that it overtakes ENSO as the most important coupled ocean-atmosphere phenomenon to cause global climate variability on interannual time scales?

        It takes 80 times the energy to convert ice at freezing to water at freezing as it does to move water one single degree Celsius.

        Doesn’t this argue that ice is 80 times as important per unit volume as water?

        Combining albedo, vaporization and other near surface effect differences, isn’t ice similarly much more significant per unit area than water?

        By these measures, ENSO is a pimple on the butt of the equator, while the loss of Arctic sea ice is a virtual decapitation of the ocean-atmosphere system.

        Where is all that effect going? Have you seen a decent correlation demonstrated for the interannual effects of the Arctic sea ice calamity, in clouds or temperatures, droughts or floods?

        Has ENSO changed one fiftieth so dramatically in the past four decades as the Arctic?

        If we see the Arctic sea ice melting, there must be some effect of it, proportional to many times, at least an order of magnitude, the effects of ENSO.

        Where are these missing teleconnections?

      • Fooled by projections, again. (Mercator?) The Arctic is tiny compared to the tropics, and the Pacific is massively larger than the Arctic sea ice. Further, temperature changes of far more than 1K are involved, and evaporation is over 3X more energy-intensive than melting.
        Big, volcanic, “pimple”!

      • ice and water is what stabilizes the temperature of earth.

      • Hi Bart,

        See isn’t it better when you play nice?

        Your question can be answered by considering global energy dynamics.

        Ein/s – Eout/s = d(GES)/dt

        Ein/s and Eout/s are the average unit energies at TOA respectively for a period. Ein/s depends on solar activity. Eout depends on greenhouse gases and albedo. GES – global energy storage – depends on heat content of oceans and atmosphere, enthalpy and net carbon flux to geological storage.

        The big ticket item is ocean heat content. I did the enthalpy calculation at one time – it is a minor item in the energy budget. Net carbon flux (energies of reduction or oxidation) is likewise negligible.

        Albedo has changed. ‘The overall slow decrease of upwelling SW flux from the mid-1980′s until the end of the 1990′s and subsequent increase from 2000 onwards appear to caused, primarily, by changes in global cloud cover (although there is a small increase of cloud optical thickness after 2000) and is confirmed by the ERBS measurements.’

        Ice albedo at the poles is relatively unimportant. At any rate – the Arctic temperatures are both amplified and have varied from cool to warm and back again in the instrumental record – correlating more to the AMO than anything else – http://www.lanl.gov/source/orgs/ees/ees14/pdfs/09Chlylek.pdf.

        http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=265

        The effect on thermohaline circulation is perhaps more important. Warmer temps lead to less ice which translates to warmer water in the north Atlantic and less ‘meridional overturning circulation’. At the same time less MOC leads to cooler temps and more ice – so who knows.

        ‘A final clue emerged. Analyzing satellite and in-situ ocean data, the researchers said a large amount of pack ice and fresh water was exported into the northwest Labrador Sea in the summer of 2007. This froze the following winter, significantly extending the ice edge farther offshore. As a consequence, cold air from the North American continent traveled farther over ice, instead of warmer ocean waters, remaining cold until it hit warmer open water in the middle of Labrador Sea. The resulting temperature contrast helped trigger the sinking process.’

        http://www.whoi.edu/page.do?pid=12455&tid=282&cid=54347

        The positive AMO since 1995 has reduced MOC leading to cooler temps in North America and Europe – with a blip in MOC in 2008. At some point this seems likely to trigger runaway ice sheet growth in Europe and North America and a descent into another glacial within a decade.

        Should that happen anytime soon – I have a spare room in north Queensland you can use. I am sure we can get you into the country as a Canadian climate refugee.

        Cheers
        Robert I Ellison
        Chief Hydrologist

      • Chief

        It appears you have caught on to my transparent use of diversionary tactics as anodyne to your diversionary tactics. Took you long enough.

        Cutting out the funny stuff.

        Forcings are not a political agenda. Feedbacks are not a religious conviction. The data is not an ideological motivation. Predicate logic is neither a kool-aid nor rye-smut induced hallucination.

        Solar variability does — despite tallbloke’s claims — not produce enough change to account for the forcings seen in the data. It’s possible there’s something going on with solar variability at scales of half a millennium or more, but then we’re seeing this effect at much less than half of that scale and less.

        Cloud effects are feedbacks. That is a mathematical conclusion that falls out of the data, not a semantic game or a play on the beliefs of readers by excluding pertinent information from arguments.

        GHG levels are spiked over 44% above the mean of the past 800,000 years, peaked more than 20% above the maximum amplitude of this previously ergodically stable span of time, and they’ve gotten to this peak in under 250 years on what is normally a somewhat smooth 100,000 year cycle.

        This perturbation is bound to disrupt climate systems, as a mathematical certainty. +/-AGW is the theory space that best fits the observations, and +AGW appears by far to be the most likely reasonable explanation for the utterly inadequate but still compelling data collected to date.

        However much misattributing what SB11 claims and doesn’t claim you want to do, Chief, you’re whitewashing a turd.

      • Bart R asserts:
        Solar variability does — despite tallbloke’s claims — not produce enough change to account for the forcings seen in the data.

        As Nir Shaviv demonstrates in his JGR paper on using the oceans as a calorimeter, the solar signal is amplified 7-10 times by a terrestrial factor – probably clouds. This is sufficient to act as the underlying forcing for the nonstationary nonlinear dynamics of ENSO.

        Cloud effects are feedbacks. That is a mathematical conclusion that falls out of the data

        better scientists than you have tried to prove that cloud amount changes only as a feedback to changes in temperature and failed. It’s a remarkable claim, which requires remarkable proof, which is not forthcoming.

      • Bart,

        I was playing fair and correcting you’re deliberate misdirection – an asymmetric game in that only one person was playing. The ‘humour’ was of the most juvenile kind – but at least you amused yourself. My essential complaint is that you and you’re fellow travellers have always been too narrowly focused – and now seem incapable of processing information that doesn’t accord with the world view. Science suggests that the world is not warming (Mochizuki et al 2010, Swanson et al 2009, Tsonis et al 2007, Keenlyside et al 2008) – albeit with immense uncertainties surrounding the origins of decadal and longer variability.

        Solar irradiance in the visible and near IR spectrum changes too little to influence climate by much. More recent work is identifying both a solar signature in global climate and a mechanism for this influence on standing waves in Earth systems in solar UV drift.

        ‘During the descent into the recent exceptionally low solar minimum, observations have revealed a larger change in solar UV emissions than seen at the same phase of previous solar cycles. This is particularly true at wavelengths responsible for stratospheric ozone production and heating. This implies that ‘top-down’ solar modulation could be a larger factor in long-term tropospheric change than previously believed, many climate models allowing only for the ‘bottom-up’ effect of the less-variable visible and infrared solar emissions. We present evidence for long-term drift in solar UV irradiance, which is not found in its commonly used proxies.’ Lockwood, M., Bell, C., Woollings, T., Harrison, R., Gray. L. and Haigh, J. (2010), Top-down solar modulation of climate: evidence for centennial-scale change, Environ. Res. Lett. 5 (July-September 2010) 034008 doi:10.1088/1748-9326/5/3/034008

        Judith Lean (2008) commented that ‘ongoing studies are beginning to decipher the empirical Sun-climate connections as a combination of responses to direct solar heating of the surface and lower atmosphere, and indirect heating via solar UV irradiance impacts on the ozone layer and middle atmospheric, with subsequent communication to the surface and climate. The associated physical pathways appear to involve the modulation of existing dynamical and circulation atmosphere-ocean couplings, including the ENSO and the Quasi-Biennial Oscillation. Comparisons of the empirical results with model simulations suggest that models are deficient in accounting for these pathways.’ Lean, J., (2008) How Variable Is the Sun, and What Are the Links Between This Variability and Climate?, Search and Discovery Article #110055

        My specific interest – for decades – has been the causes of drought dominated and flood dominated regimes identified in north eastern Australia. These emerge from variability in the Pacific and the question naturally arises as to drives this variability – as well as to the limits of variability. If we view ENSO as a complex and dynamic system – we would expect that there is a control variable that initiates shifts in states. My thinking is that this control variable is an impulse at the right time from as a result variable sea level pressure in the Antarctic. An associated problem is the Pacific Decadal Oscillation and what drives the change in upwelling here over such long periods. The periods are to long to be due to an internal cause such as Kelvin waves. We have also the millennial variability that is shown in the Moys (2002) study and as discussed in the Tsonis (2009) paper on the Minoan civilisation.

        As for global warming theory – recent warming all occurred between 1977 and 1998. Most of that warming occurred in 1976/77 and 1997/98. The first period is the “Great Pacific Climate Shift’ – the second is the great El Nino ‘outlier’. Most of the rest involves a change in cloud cover. ‘The overall slow decrease of upwelling SW flux from the mid-1980′s until the end of the 1990′s and subsequent increase from 2000 onwards appear to caused, primarily, by changes in global cloud cover (although there is a small increase of cloud optical thickness after 2000) and is confirmed by the ERBS measurements.’ http://isccp.giss.nasa.gov/projects/browse_fc.html

        The reasons for cloud change in ENSO involve changes in convection. ‘‘During the 1997–1998 El Niño, observations indicate that the SST increase in the eastern tropical Pacific enhances the atmospheric convection, which shifts the upward motion to further south and breaks down low stratiform clouds, leading to a decrease in low cloud amount in this region. Taking into account the obscuring effects of high cloud, it was found that thick low clouds decreased by more than 20% in the eastern tropical Pacific.’ Zhu, P., Hack, J., Keilh, J and Zhu, P, Bretherton, C. 2007, Climate sensitivity of tropical and subtropical marine low cloud amount to ENSO and global warming due to doubled CO2 – JGR, VOL. 112, 2007

        Clouds are not an ENSO feedback – they are an intrinsic part of the ENSO and the PDO systems that vary in observations (e.g. Clement et al 2009, Burgmann et al 2008 and Zhu et al op. cit.).

        SB11 claimed that there was no possibility of determining cloud feedback to global warming from ENSO variability – because of multiple interacting factors – and that models had difficulty in simulating these ENSO factors.

        It is not controversial to state that climate models are deficient in terms of tropical variability in the atmosphere on many timescales [Lin et al., 2006; Lin, 2007] and a more realistic simulation of ENSO events in coupled simulations remains a high priority for model developers. During El Niño, the warming of the tropical eastern Pacific and associated changes in the Walker circulation, atmospheric stability, and
        winds lead to decreases in stratocumulus clouds, increased solar radiation at the surface, and an enhanced warming so that even models without ocean dynamics are capable of emulating some ENSO‐like variability [Kitoh et al., 1999]. Positive cloud feedbacks in observations have been shown to occur in association with ENSO and these variations are generally not well depicted in models [Kang et al., 2002;Clement et al., 2009], but challenges also exist for diagnosing
        these interactions in observations, as it is difficult to identify cause and effect in the context of multiple interactive variations.’ Trenberth et al 2010.

        There is something very strange happening – and I think it is down to an AGW millenialist cult. You are either in the cult and right or out of the cult and wrong. Even if saying exactly the same things.

        There is too much natural variability to distinguish a greenhouse gas signal – although I agree it should be there. It is one factor in many – but I suppose we shall have to wait for another few years of no warming. The AGW spaceship just ain’t coming.

        Cheers
        Robert I Ellison
        Chief Hydrologist

      • Chief,
        If you already saw this question, please accpet my apology in advance.
        Did you have a chance to look at the limnology stuff I sent you?
        The topic is fascinating, and the implications are tantalizing.
        regards,

  41. Dessler has been fighting a rear guard action for over two years now.
    His political and academic standing depends on contiuing this fight long enough so that the failure of his armageddon predictions are lost inthe mists of time, or he retires.
    This is effectively the same strategy that Paul Ehrlich and his pals have done since the their failed attempt to predict a population/natural resource crisis from the late 60′s / early 70′s..
    that this strategy is acceptable to academics is a seperate issue.

  42. “As for Trenberth wrongly saying Spencer has a track record of errors, well Trenberths attribution of the Russian drought and Pakistan flood to global warming was refuted by virtually everyone.”
    And the flooding in Australia.”

    And yet such mistakes, as we have seen, do not matter. When it became apparent that snow was in fact not going to become but a nostalgic memory for those of a certain age, at least not any time soon, they turned it on its head and called it PROOF OF COOLING. Meanwhile, the most fundamental mistake of all, the lack of warming since 1998 (woops) contrary to all warmist predictions, does not even merit a mention in the NYT’s et al.

    A delusion is defined as “a false belief held with absolute conviction despite superior evidence.” I can’t think of a better term for the belief in CAGW.

    • Dear Ron:
      I looked at the response and have the following comment:The climate energy equation used is a form of the Stefan-Boltzmann law equation for small temperature change. While it is the controlling equation of the energy radiated from solid surfaces such as land and continents, it is not the controlling equation of the energy radiated from fluids such as surface water or ocean. Surface water outgoing radiations is controlled by convection heat transfer process, not by the Stefan-Boltzmann law equation. Therefore, using Stefan-Boltzmann law equation to calculate ocean temperature is a mistake and the model can never be right. There are no such things as climate sensitivity or feed backs.

    • I must disagree vehemently, Judy. Briggs starts out with an intemperate rant. In an update, he cools down, but then focuses on “gossip” he heard about Trenberth. His analysis of the science in SB-11 strikes me as peripheral to many of the important issues, probably because it deals mainly with the statistics, which I presume is his area of expertise.

      To my mind, this is not the type of post that advances our understanding of anything except the visceral reactions of strong partisans.

      • Fred, go ahead and ignore the first sentence if it offends you,but the second one is correct.

        “In one of the most asinine, self-promoting, sniveling, absurd, nakedly political moves Wolfgang Wagner has resigned, with trumpets blazing, his editorship of Remote Sensing.

        Why? Because the journal under his command dared follow its editorial guidelines, and follow them properly.”

      • Fred,
        Oddly, you consider anything that disputes the AGW apocalyptic consensus as ‘not advaning understanding”.
        Dessler’s piece, and its immediate publication, along with the transparently false excuse Wagner contrived, is a great example of the reality of cliamte science in its out-of-control reality.

      • You should read his resume. He is statistician to the stars. And an objective Bayesian. Someone to be taken seriously, not dismissed

      • Fred will dimiss your point.

      • At interminable length….

      • cwon14 9/7/11, 7:19 pm, Spencer Braswell III

        c: Fred, are you trying to say Trenberth isn’t a strong partisan?

        Your post may be caught in the spam filter, but navigating around your likely problematic link led to this 2007 statement by Trenberth:

        KT: In fact there are no predictions by IPCC at all. And there never have been. The IPCC instead proffers “what if” projections of future climate that correspond to certain emissions scenarios. There are a number of assumptions that go into these emissions scenarios. They are intended to cover a range of possible self consistent “story lines” that then provide decision makers with information about which paths might be more desirable. But they do not consider many things like the recovery of the ozone layer, for instance, or observed trends in forcing agents. There is no estimate, even probabilistically, as to the likelihood of any emissions scenario and no best guess.

        Trenberth should know. He was a Draft Contributing Author to that Summary for Policymakers. He was a Contributing Author to AR4, Chapters 1 and 7, and to the Technical Summary; he along with Phil Jones were the Coordinating Lead Authors for AR4, Chapter 3; he participated in the drafting of the TAR SPM; he was a Lead Author for TAR Technical Summary; he was a Contributing Author for TAR Chapters 2 and 8; he was a Lead Author for TAR Chapter 7. His publications appear in AR4 References 49 times. Hardly a partisan; he ranks as a leader.

        Nevertheless, and contradicting Trenberth’s statement, IPCC said with respect to the subject of this thread:

        “The equilibrium climate sensitivity is a measure of the climate system response to sustained radiative forcing. It is not a projection but is defined as the global average surface warming following a doubling of carbon dioxide concentrations. It is likely to be in the range 2°C to 4.5°C with a best estimate of about 3°C, and is very unlikely to be less than 1.5°C. Values substantially higher than 4.5°C cannot be excluded, but agreement of models with observations is not as good for those values.” Bold added, IPCC, AR4, SPM, p. 12

        In IPCC parlance, likely means the Likelihood of the occurrence/outcome is >66% probability, and very unlikely is <10% probability. And

        Unless noted otherwise, values given in this report are assessed best estimates and their uncertainty ranges are 90% confidence intervals (i.e., there is an estimated 5% likelihood of the value being below the lower end of the range or above the upper end of the range). AR4, Box 1.1, p. 121.

        Contrary to Trenberth’s urging, IPCC reports do include predictions. Equilibrium climate sensitivity (ECS) is an IPCC prediction in ordinary English and in science, but not a projection in IPCC parlance. Trenberth doesn’t understand science. Models are mappings from observations to future observations (predictions). They are of the form {hypothesis, conclusion}, where the hypothesis is the experimental set-up, the what-if as he puts it, and the conclusions are predictions, notwithstanding any semantic games.

        Furthermore, ECS corresponds to a doubling of CO2 valid for any scenario. IPCC quantified ECS probabilistically, even though it didn’t bother with the pointless task of assigning probabilities to emission scenarios per se. With regard to the S&B paper, and the L&C paper previously, the prediction above is failing, tending to invalidate AGW.

      • Just get a logical bearing;

        http://nofrakkingconsensus.com/2011/07/05/landsea-the-ipcc-the-union-of-concerned-scientists/

        Trenberth, Union of Concerned Scientists, IPCC, AAAS;

        Is there any person of group here that aren’t left-wing in posture?

      • “I wrote the “Politics” in the heat of the moment, bubbling mad. The review of the paper, I had started weeks before. My criticisms of Wagner, Black, etc. would have had more effect if I had written them coolly.”
        -William Briggs

        —-

        Climate science papers (more pointedly their conclusions) seems to depend largely on the use, or misuse of statistical methods in its current form.

        After so many criticisms on both sides appear to involve inadequate applications of statistical methods, I don’t know why more authors do not consult statisticians while writing their papers.

    • Marlowe Johnson

      what’s ‘good’ about his post Judith?

    • Briggs makes the important point that there is nothing exciting about SB11, with which I agree. It has been blown out of all proportion, especially as skeptics think his model criticism is in some way fatal to models, when actually it is not particularly important, and already known, that model El Ninos are not always well represented. SB10 made a similar point about sensitivity statistics, and was also blown out of proportion with some reading a negative feedback into it, even though his main point was that it was not a good way to get feedbacks in the first place. I never understood the disconnect between its press and its contents in that case either, as the paper itself was again routine and unexciting.

      • Maybe you think it is unexciting to suggests clouds can change temperature.

        But the AGW crowd blew a gasket and made Wagnet commit career suicide. They were excited.

      • It was the publicity by those that failed to understand the insignificance of this paper. Wagner’s decision was affected by the publicity, not by the contents of the paper. Spencer gets good marketing in return for his Heartland appearances, though I suspect even he is embarrassed at the misguided spin his papers are given.

      • Wagner said he wanted lots of publicity when he started the job.

        Steve McIntyre is redoing his evisceration of Dessler and its worse than we thought.

        “Dessler 2010 made the curious decision to combine ERA clear-sky with CERES all-sky to get a CLD forcing series. This obviously invites the question about the impact of using CERES clear-sky in combination with CERES all-sky to calculate the CLD forcing series. One would have thought that this is the sort of thing that any objective peer reviewer would ask almost immediately. Unfortunately, as we’ve seen, climate science articles are too often reviewed by pals.”

        http://climateaudit.org/2011/09/08/more-on-dessler-2010/#comments

        Talk about an “own goal”. The climate community tried to destroy Spencer and it ends up destroying the whole warmenista side.

    • curryja 9/7/11, 3:10 pm, Spencer Braswell III

      JAC: William Briggs has a good post … ..

      William Briggs: If the paper is flawed and its conclusions are genuinely refuted by other papers that Spencer and Braswell, with malice aforethought, ignored, then Wagner would have retracted the paper. That would have been the “honourable” thing to do.

      Briggs recommends retraction for non-conformity, which is hideous science. If a model is refuted, and the refutation is shown invalid, the original model is strengthened. All these transactions should remain viable as publications and open to discussion as a matter of best practices in science.

      Retraction might be better reserved for such abuses as Jan Hendrik Schön’s fraudulent graphs and data on nanotechnology. Other candidates might include IPCC’s manufacture of a family of hockey stick graphs by software blending of smoothed data from different sensors, absent raw, or at least less-cooked, data; or, its manufacture of false fingerprints of human activity by paralleling divergent records with chartjunk; or its concealment of a graph showing that the Revelle Factor was no more than Henry’s law of solubility.

  43. Fred

    Judith said the Briggs post was ‘good.’ That is a different thing to accurate, considered, balanced etc which is what you seemed to want out of it. It was indeed ‘good’ probably because it was a little intemperate if still pertinent.

    tonyb

  44. Someone up thread expressed doubt that clouds could affect climate-that is to say they could have an influence to warm/cool the atmosphere over a climatically significant period.

    Whilst reading the modern day research of Dessler and Spencer hopefully denizens might be interested in Luke Howard, who several hundred years ago was responsible for classifying clouds and through observation did much to examine and quantify their effect on weather and climate. This link provides details of the man himself as well as several articles that relate to his research on clouds.

    http://www.cloudman.com/luke_howard.htm

    Clouds were fairly widely interpreted in as much Howard was interested also in those created by volcanic eruptions which had a long lasting effect of months and even arguably years.
    tonyb

  45. During the last major warming, coming out of the last major ice age, large amounts of water from melted ice sheets, which had been trapped for thousands of years, dumped into the ocean. The most significant of these events was Younger Dryas. This new ocean level helped the Arctic, the thermostat of earth; regulate temperature in an extremely stable narrow range. Look at this data for the last eight hundred thousand years. The current ten thousand year warm period is more stable in a more narrow range than any warm period in ice core history.

    http://upload.wikimedia.org/wikipedia/commons/1/1e/EPICA_delta_D_plot.svg

    Before, it would snow enough to lower the oceans enough to cut off flow to the Arctic. Then it would stay cold for a hundred thousand years while the ice melted and raised the oceans enough to let water back into the Arctic. Now, with higher ocean levels, that does not happen.
    Now, it warms until melts Arctic Sea Ice. Then it snows until it cools enough to allow the Arctic Water to Freeze. That stops the Arctic Ocean Effect Snow and it warms again.

  46. JUst found Peter Gleick’s nauseating article in which he writes…. “In his editorial resignation, Professor Wagner says the paper was reviewed by scientific experts that in hindsight had a predetermined bias in their views on climate that led them to miss the serious scientific flaws in the paper, including “ignoring all other observational data sets,” inappropriate influence from the “political views of the authors,” and the fact that comparable studies had already been refuted by the scientific community but were ignored by the authors.”

    Can someone tell me if these supposedly “biased” reviewers (as if warmists aren’t biased!) have made any public statements?

    • Peter “Photoshop for Science” Gleick is simply parroting a parrot. I haven’t heard any of the reviewers identified, much less making statements. But even to the most casual observer, this is simply another argumentative fallacy. Neither Gleick nor Wagner offer any evidence of these accusations. Of course, this is part and parcel of the team’s M.O. I’m still looking for the “serious scientific flaws” and laughing about the political influence the authors brought to bear. It’s just more people spending credibility they don’t have.

      • Except for one thing. Wagner knows who they were, so I think that he can reasonably say what he said about the reviewers.

      • Again, no evidence. He may know their names, but he knows how much political influence Spencer brought to bear? Without proof, that’s just more libelous blatherings from a failed gatekeeper.

        Rattus, this is what I don’t get. Wagner states, “serious scientific flaws”. There were none. Some flaws, perhaps, but serious scientific ones? Ignoring data sets because what? They used HadSST? These have shown to be erroneous and misleading statements at best. But when he says “predetermined bias” and “political influence”, you’d believe him? Your reasoning is because he knew their names……. which, of course, are still anonymous to the rest of us.

        What? You think because they agreed to be reviewers you think they talked politics or shared drinks, or what?

      • He said really, really nice things about them.

        “three senior scientists from
        renowned US universities, each of them having an impressive publication record. Their reviews had an
        apparently good technical standard and suggested one “major revision”, one “minor revision” and one
        “accept as is”. The authors revised their paper according to the comments made by the reviewers and,
        consequently, the editorial board member who handled this paper accepted the paper (and could in fact
        not have done otherwise). Therefore, from a purely formal point of view, there were no errors with the
        review process.”

      • Really. So now he’s an expert on who is a real skeptic. Really?
        not in my book.

  47. In Global Atmospheric Trends: Dessler, Spencer & Braswell David Stockwell compares (and provides his analysis code):

    . . . the scatter plot of monthly average values of ∆R_cloud (eradr) versus ∆T_s (erats) using CERES and ECMWF interim data. There is extremely little correlation as noted by Steve. In fact, it is not statistically significant in the conventional sense, . . .
    The points in red are the sequential difference of temperature against the cloud radiance. While these have a lower slope, unlike the former, they are conventionally significant, almost to the 99%CL. . . .

    The Adjusted R-squared: increases four fold from 0.01045 for Dessler’s method to 0.04236 for Stockwell’s method.

  48. Alexander Harvey

    Dessler gives us a value for the standard deviation of the month to month variation in oceanic heat content of 9W/m^2.

    Which is derived in part from a heat capacity of 168 W-month/m^2/K and and a temperature history and partly from Argo data.

    I think that the 168 figure corresponds to a slab ocean of around 100m which is not beyond reason where it not for its use at the monthly time scale where it infers that the ocean fully couples with the surface to that depth within one month globally.

    I cannot speak for the Argo data and if it supports this it does, but it would again imply that the top 100m of the ocean is coupled in the above fashion or the heat is bypassing the surface mixed layer.

    The 9W/m^2 figure may be problematic for a global average for it is rather large and it is not entirely clear what it refers to but if it is random fluctations then at best this would seem to have implications for the steadiness of the global temperature at the decadal timescale.

    As a first approximation one could try and scale this value out to find the standard deviation of the mean at the 30 year horizon (360months) by dividing by the squrare of 360 (~ 19) giving 9/19 ~0.45W/m^2.

    So a 2-sigma ~0.9W/m^2.

    Now I am not suggesting that is the case, far from it I really don’t. It is just an illustration of just how big a 9W/m^2 globally averaged fluctuation at the one month timescale is.

    As far as figures go, this 9W/M^2 is one I do find surprising. Especially when I considered that this is averaged both temporally and spatially. It seems large compared to the TOA variation, which I think is central to Dressler’s argument.

    I have tried to stay away from the S&B paper and I hope I continue. The Dessler paper is of a different character, a rebuttal which I think carries the heavier burden. It is also a championing of the orthodox position which carries risks if it errs in any one detail.

    To those that are more informed on these matters may I pose a simple four part question:

    Is the 9W/m^2 value correct,
    what type of fluctuation is it (time and space correlation etc.),
    whence comes it and whither goes it if not via TOA,
    what if any are the implications for global temperature variation at the 30 year timescale?

    Alex

    • Alex,
      I don’t claim to be more informed, but I’ll have a go at answering your questions anyway!
      Is the 9W/m^2 value correct?
      Very unlikely. The data used by willis, levitus etc (the ARGO data) yields a direct measurement of OHC (variation), which should equate roughly with CpDT/dt. Spencer here ( http://www.drroyspencer.com/2011/09/the-good-the-bad-and-the-ugly-my-initial-comments-on-the-new-dessler-2011-study/ )abstracts a sd value of 2.3W/m2.

      what type of fluctuation is it (time and space correlation etc.)?
      Over the long term global OHC gain/loss should reflect the integral of the total net flux difference for the planet. Over the short-term, the mixed layer also responds to the injection/removal of energy via ENSO events. For the specific period here, Dessler believes that variation in OHC was almost 100% dominated by ENSO events.

      whence comes it and whither goes it if not via TOA?
      It comes from solar heating, precipitation and El Ninos, and it goes via evaporative losses, conductive losses, La Ninas and LW emission. (Ignoring latent heat changes, and losses from sensible heat redistribution etc)

      what if any are the implications for global temperature variation at the 30 year timescale?
      If S&B is correct, then we may have an explanation for why GCMs are so bad at matching SW and LW series even when they get close on net flux. This may even produce a valid explanation for the strong SW heating through cloud albedo reduction between 1980 and 1999, which was missed by all of the GCMs. This would necessitate some serious rethinking of the basic physics, but it is difficult to predict the impact.

      • Alexander Harvey

        Paul,

        Thanks for replying in detail.

        I agree that th 9W/m^2 value seems large and I think problematic given that a rebuttal should be robust.

        He gives that value as if it represents a real flux in and out of the oceans. He caluclates it as the product of temperature and heat capacity, which is all well and good but it has the wrong sign if he wishes to produce El Nino warming from a redistribution of heat.

        As I recall the big El Nino 1997/8 was characteristed by an initial loss of OHC, as would be expected, which was regained by the end of the event. The situation with respect to ENSO since 2000 is I think not so clear.

        I think that there is a potential issue whenever it is suggested that the heat flow in or out of the ocean is much greater that the change in TOA flux. The combined thermal capacity of the atmosphere and land is known to be low, perhaps the equivent of ~5 metres of water. Acting alone as a reservoir for heat a 1W/m^2 flux difference between the two boundaries, TOA and ocean surface would heat 5 metres of water by ~0.5C. It would seem to me that only a change in the latent heat content could surfice to mop up all those extra Watts, and in the case of water vapour that requires continued evaporation without condensation, not even as clouds. I think this is problematic, I think his 9W/m^2 is a stick he has made to beat him with.

        If he had suggested that OHC releases were sufficient to supply the heat required to initiate the ENSO effects that would be well and good but he has sought to belittle the effect of clouds by suggesting that it is swamped by his 9W/m^2 and I think he has muddied pools unnecessarily.

        I should be a little surprised if no one has picked up on that value and used to to suggest that it is so large that it belittles the GHG effect not just monthly but is also comparable on the decadal scale. It is precisely flux noise values of that order that are required to show that observed multidecadal warming has no statitstical significance.

        I think if forced to choose I would find Spencer’s values more acceptable which if Dessler is putting forward the concensus view would mean rejecting that view. I think this is known as collateral damage. Win the fight but disenchant the audience.

        Alex

      • Alexander Harvey

        Oops, my error, one part should have read:

        Acting alone as a reservoir for heat a 9W/m^2 flux difference between the two boundaries, TOA and ocean surface would heat 5 metres of water by ~1.0ºC.

        Alex

  49. Stephen Pruett

    Why is it that the lack of agreement between models and satellite data is not scientifically interesting? Is it just that it was already known that models do not handle ENSO well or is the selection of the three least sensitive and three most sensitive models a real problem? Based on the graph shown in Spencer’s response, none of the models seem to be in particularly good agreement with the satellite data.

    If Dessler is making an issue of the fact that the confidence intervals of some model results overlap with the confidence intervals of the satellite data, this is not an appropriate comparison. The difference between two lines can still be significant at the 95% confidence level if the 95% confidence intervals overlap, because the overlap does not take into account that only one of the confidence bounds is involved in overlap, whereas the formulas leading to the 95% confidence intervals assume that either bound could be violated. It has been shown that overlap of the 83.7% confidence intervals reveals lack of significance at 95% (Journal of Quality Technology 21:140, 1989; Journal of Quality Technology 1:256, 1969).

    Finally, the implications of the need for multiple models does not seem to have received due attention in the climate science community. Doesn’t this directly indicate our ignorance of important aspects of the climate system? In fact, if a few models happen to be closer to the observed data than others, this would not be surprising due to chance alone. The total range of the models is quite large. it would be difficult to find an observational data set that is not close to at least one or two of the models. Thus, it seems particularly inappropriate to use model output as though it was real data, a practice that seems popular in climate science. If everyone in climate science already knows this, then the Spencer 2011 paper is perhaps uninteresting. However, if it is a rare demonstration of the empirical inadequacy of most models, this seems rather worthwhile to me. Whether inferences regarding clouds are justified, I do not know, but comparing models to real data seems extremely worthwhile.

    • After watching the Desser/Lindzen debate video I am thinking “Merchant of Doubt Phobia” is the cause. It is a new physiological defense mechanism, not an actual phobia. The ultimate form of denial-ism. “I cannot be wrong because I am uncertain. Uncertainty is the new certainty, therefore I am certain I am not wrong because I am uncertain.” It is a spherical reasoning process.

      Dessler in his debate referenced “Merchants of Doubt” then explained how the instrumental data sets did not agree with the models requiring the use of a third data set based on balloon thermometrics which has an R^2 statistical significance of 0.0012 which Lindzen had avoided in an attempt to deflect sound reasoning with manufactured uncertainty. This was shortly after Dessler reminded the audience that climate scientist always look for multiple data source to confirm their work. Therefore, Dessler won the debate because Lindzen was confused.

      And people wonder why Judith posts the psych stuff.

      • Lindzen is incredibly boring, but if you can make it half way through, you will see what I mean.

      • I’m suprised you thought Dessler won. He spent half his time attacking his own credibility. Don’t believe anyone that talks about conspiracies but anyone that disagrees with me is a merchant of doubt? Scientists look for agreement of data sets except in this case where we take wind speeds to determine temperature because it agrees with our hypothesis? Lindzen pointing out the laughing meteorologist at the wind speed to temperature conversion was a classic. I think Lindzen won and he probably could have won without saying much at all.

      • No Dessler won since it was a popularity contest not a meaningful scientific debate. Dessler selected the original version of Lindzen’s paper without corrections that he was fully aware of, played the doubt card, justified a statistically poor replacement data set and even cast doubt on Lindzen’s competence for not using that data set. It was a slam dunk since Lindzen is not what I would call a dynamic personality.

      • Forgot to add, Dessler even pointed out Lindzen was a smoker.

      • It is not a conspiracy theory to say that in the entire world, there exist a handful of scientists who are willing to be “a merchant of doubt”. It is simply common sense, backed up by past experience.

        What is a conspiracy theory is to suggest that essentially the entire scientific community has been hijacked so that one can no longer trust the conclusions of the National Academy of Sciences and the analogous academies in all of the major G8+5 nations, etc., etc.

        Do you see the difference?

      • I don’t think it’s a conspiracy theory to suggest that most people would be slow to admit that they may have backed the wrong horse – especially if they had bet the farm on it.

      • Exactly Peter!

      • Peter317,

        “…especially if they had bet the farm on it.”

        …especially if they had bet everyone else’s farms on it. 8>)

  50. Hello
    Assuming cosmic rays has a significant effect on cloud cover, would that be relevant to the debate over SB11 and its implications for climate change?

    Hope that is not a silly question.

    klee12

    • klee12,

      Is the question relevant? Yes, it is, but it is not central to the issue. If GCR’s were modulating cloud variation then this would appear as an exogenous radiative forcing rather than a temperature-dependent feedback.

      Spencer claims that the data exhibit evidence of the presence of significant radiative forcing. He attributes this to variation in clouds which cannot be explained as a temperature-dependent feedback (from a previous forcing), without specifying the mechanism. The mechanism could include GCR variation.

      • Thanks Paul_K for your reply

        If I understand you correctly, Spencer thinks that there is evidence that variations in clouds, not due to previous (CO2 ?) forcings, are radiative forcings which could include GCR variation. If one believes that GCR variation is a significant forcing, would that not lead one to believe that Spencer is correct. Put another way, if Spencer is wrong, would that imply that GCR is not a significant forcing?

        I realize that logically Spencer may be correct and there may be other forcings we are unaware of. Hence Spencer may still be correct and GCR is not a significant forcing.

        klee12

      • Hi again, Klee,
        “If one believes that GCR variation is a significant forcing, would that not lead one to believe that Spencer is correct. Put another way, if Spencer is wrong, would that imply that GCR is not a significant forcing?”

        I don’t think that these questions are answerable without a bunch of qualifications. In the first instance, if Spencer is right, it wouldn’t hurt the case for GCR modulation of clouds, but it is a separate step to prove the case. On the other hand, If Spencer is proved to be wrong, the answer to your question is dependent on why he was wrong. If his case is thrown out on methodological grounds, then the GCR hypothesis probably remains untested. If it is thrown out because it can be shown unequivocally that changes in radiative forcing were negligible over the period, then that might weaken the case for GCR modulation of clouds given the variation in GCR over the same period.

        I would restate my view, though, that this is secondary to the main debate going on.

      • Hello Paul_K

        “I would restate my view, though, that this (GCR) is secondary to the main debate going on.”

        I understand you position. I looked at all this from the point of view of how significant is AGW. All one can do, IMHO, is come to the conclusion that the preponderance of evidence is on the side of AGW or isn’t. In my undereducated view, the CLOUD experiment, together with the historical record, moved the needle against AGW. The CLOUD experiment is not conclusive since, among other reasons, there are still more steps from condensation nuclei to increased cloud albedo. From what I gather from this discussion, the Spencer paper does not address the missing steps, so even if the paper is correct, it does not add much to evidence for or against AGW.

        Thanks for you explanations

        klee12

  51. It looks like Spencer’s opening broadside is enough to sink Dessler’s ship. The most telling points are that : –
    However own chooses to calculate the fluctuations (expressed as a standard deviation by Dessler) in the Forcing term (F) and the feedback term lamda*DT, the difference term, which is the net flux term (F – lamda*dT) must reflect the fluctuations in the actual observed net flux data. (Note also that when I was a lad the variance of A-B was always the sum of the variances of the two elements. I think that this has been changed by Dessler in post-normal science.)
    There are direct measurements of OHC fluctuations from ARGO, which show Dessler’s calculation by difference likely to be wildly in error.
    Dessler is sunk and hoist by his own petard, to mix metaphors.

    It is a shame that the Dessler rebuttal was so poor. A better response might have stimulated Spencer to address the more important questions surrounding his paper. Can it be shown that the alternative model (feedback dominated) CANNOT explain the observations (Fred Moolten’s point)? What impact does the choice of the 3 parameters – Cp, lamda and the relative contribution of radiative to non-radiative forcing – have on the simple model behaviour? What does a comparison of model results look like for the specific 2000-2010 period, especially for the models deemed to be good at modeling ENSO, and what level of radiative forcing do such models include to match the data?

  52. “Chief Hydrologist | September 7, 2011 at 7:42 pm |

    Your question can be answered by considering global energy dynamics.

    Ein/s – Eout/s = d(GES)/dt

    Ein/s and Eout/s are the average unit energies at TOA respectively for a period. Ein/s depends on solar activity. Eout depends on greenhouse gases and albedo. GES – global energy storage – depends on heat content of oceans and atmosphere, enthalpy and net carbon flux to geological storage. ”

    Quite possibly, but can you a very simple question.

    Why arnt the oceans isothermal?

    Seriously, why is there a temperature gradient where the vast majority of water at death is colder than the top ?
    We know that heat can only be radiated from the surface, but the surface is hotter.
    Now I have been told that the bottom is colder due to cold water from the poles sinking. Possible, but the problem is that minimum salinity is at 750m and increases as one goes lower.
    Now oddly enough, the brine density is constant below 750 meters. Density is a function of both temperature and salinity, as you drop in depth temperature drops, salinity increase and density remains the same.
    Now here is a thought CH, perhaps the oceans are isodensile, rather then isothermal, and changes in the heat in the system are not manifest in terms of temperature, but in changes in brine density?
    What if the changes in heat in the system are buffered, not in a change in the temperature of the upper 750 meters of brine, but in the density of the 750-4,000 meters of density of saline?

    • Sounds interesting. What mechanism do you propose for the conversion of sunlight to salt?

      • Not quite direct conversion of sunlight into salt, but the conversion of sunlight into salt ionic bond length. The bond length of the water:cation bonds in the hydration sphere that surrounds each cation is dependent on the (attractive) hydration energy which is due to the bond energy of the oxygen lone pair interaction with the cations electron cloud and (repulsive) antibonding energy from the positive zeta-potential of the hydrogen interaction with cationic nucleus.
        In an ocean, the average M+-O(H2) bond length will increase with temperature and decrease with pressure.
        I suggest that the heat content of a volume of water taken a 1km down is identical to one take at 4 km down. Although the pressure at 1 km is lower and the temperature is high, in thermodynamic terms, these (I suggest) cancel.
        If this is the case then a change of heat intput into the oceans will manifest itself in a change ionic bonding distance, which will lead to a change in the average density and average volume.

      • Doubly interesting. So how is the energy stored and released? Not just as gravitational potential energy as a function of the thermosteric component of sea level, but as also as a gravitational potential in the sea level as a function of the bond length affecting density?

        If so, it heat released as a consequence of the bonds shortening again?

    • Doc,

      Doesn’t evaporation at the sea surface increase the salinity of the upper ocean?

      • Yes, Edim, the first 750 m of the ocean surface is noted for a change in both salinity and temperature, at the very top salinity and temperature are high; they both fall in step to about 750 m. Then after this point temperature falls and salinity increase; but density remains constant.
        here is the salinity profile.

        http://www.windows2universe.org/earth/Water/images/salinity_depth_jpg_image.html

        temperature profile;

        http://www.windows2universe.org/earth/Water/images/temperature_depth_jpg_image.html

        density

        http://www.windows2universe.org/earth/Water/images/density_depth_jpg_image.html

      • Doc,
        This density plot looks like nonsense. Reduction of temp, increase of pressure and increase in salinity all tend to increase density. I have checked several sources to see whether there is any inversion of the first two relationships reported at high pressure, but, no, most textbooks say the same thing. You may be being misled by bum blog data.

      • The source is a site of the National Earth Science Teachers Association, but there seems to be some confusion in the plots.

        One source of confusion appears to be that the density given is not the real density but the potential density, i.e. the density that the water would have at surface pressure keeping the temperature and salinity as they are. The real density at the depth of 4500 m is significantly higher (perhaps 1.04 – 1.05, but I haven’t checked the precise value).

        The other plots may be approximately right for some location, but similar plots at different locations are very different. A salinity minimum around 700 m is a real phenomenon in the South Atlantic, but a plot in the textbook of Wallace and Hobbs indicates that the salinity profile may be more complex for deep Atlantic (a maximum around 2000 m and lower again at greater depths).

      • @Pekka,
        I get a density of 1.046 for 2 deg C, 35000ppm and 4500 m water depth, so we are in exactly the same ballpark. If you go back to the original plot shown by Doc, this represents a very significant gradient on the density vs depth plot. VERY misleading.

    • The surface is warmer because warmer water is less dense and is therefore buoyant. It also heats at the surface with solar SW.

    • Alexander Harvey

      Doc,

      It was noted in the 1950/60s that the temperatures of deep water between ~1km and ~2km not only decline but tend to do so exponentially (in approximation) and that this was a stable feature. This was explained by Walter Munk as being due to diffusion down through an upwelling column. He determined that the upwelling rate throughout much of the oceans was around 4m/yr, he also determined a value for the “effective” or eddy diffusion that is about 1000 times that of the standard bench test value for unperturbed water.

      This model has a stable state were the flux of heat down across a horizon of fixed depth is balanced by the return of heat across that horizon due to the slow bulk vertical motion. This stable state has an exponential temperature form with a characteristic height of ~1000m.

      In that sense it is not simply that cold water sinks at the poles but the existence of a steady upwelling thoughout much of the ocean area with a period of the order of 1000 years to rise from the bottom to the top.

      For many years a diffusive upwelling ocean presented the best available thermal box model for the ocean but is simplimistic and no longer much used.

      I believe that below Munk’s exponential region, a lapse rate predominates which is not apparent in some diagrams due to the use of potential temperatures not thermometer temperatures.

      I believe that above Munk’s region there are many confounding factors but the existence of the underlying rising region acts to inhibit the vertical transport of heat down to the abyss.

      It is worth noting that the scientific understanding of the ocean has evolved much in the last decade or two but mine hasn’t, so some or much of this understanding may have been heavily revised.

      Alex

      • Alexander Harvey, 9/9/11, 10:26 am, Spencer Braswell III

        AH: In that sense it is not simply that cold water sinks at the poles but the existence of a steady upwelling throughout much of the ocean area with a period of the order of 1000 years to rise from the bottom to the top.

        I have been relying on a different model. Why do you or Munk distribute the upwelling over the ocean? Do you have any evidence for such unfocused upwelling?

        Because of the period, you seem to be describing the ThermoHaline Circulation, which oceanographers give a period of about one millennium. The THC needs some kind of first order outlet, which the Ekman pump seems to satisfy. Might the Ekman pump also account for the upwelling in your model, focusing the return at the Equator?

        Furthermore, the THC model from poles to Equator and the conventional estimates for air-sea CO2 flux corroborate one another.

  53. bart said:
    “GHG levels are spiked over 44% above the mean of the past 800,000 years, peaked more than 20% above the maximum amplitude of this previously ergodically stable span of time, and they’ve gotten to this peak in under 250 years on what is normally a somewhat smooth 100,000 year cycle.”

    um… water vapor has spiked 44% over the mean of the past 8000000 years, has it?
    how do you figure that?

    • gnomish

      Oh what a cute little creature of myth you are.

      Water vapor is a GHG. GHG’s have spiked. Therefore water vapor has spiked.

      Socrates is a man. Men wear blue jeans. Therefore Socrates wore blue jeans.

      • Here’s another one for you:

        CO2 is a GHG
        GHGs can cause warming
        We have observed warming
        Therefore, CO2 caused observed warming

      • manacker

        Surely you mean:

        GHG’s have the property that they must cause warming in some circumstances.
        CO2 is a GHG.
        We have observed the circumstances.
        Therefore, CO2 must be responsible for warming.

      • Socrates is a man. Men wear blue jeans. We have observed the circumstances. Therefore Socrates wore blue jeans.

      • Edim

        Which circumstances are those?

        Men since the 19th century in the Western world might wear blue jeans where available.

        Socrates lacks the 19th century circumstance.

        GHG’s must increase warming when increased in concentration where conditions include broad band incoming radiation and OLR in the spectra including the non-overlapped absorbtion wavelength of the GHG at levels below the GHG’s saturation point.

        Increased concentration? Check.
        Inbound broad spectrum light? Check.
        OLR? Check.
        Nonoverlapping absorbtion wavelength? Check.
        Nonsaturating? Check.

        Socrates is warmed by increasing CO2.

        How is logic so hard for you? It’s only predicate logic. Chimpanzees can do this.

      • Bart,

        I mean the Brasilian football player Socrates (20th century).

        “Increased concentration? Check.”

        Caused by climatic factors. There’s no correlation between increased concentration and human emissions.

        Other points are irrelevant.

      • No, Bart, I really did NOT mean what you wrote, but rather what I wrote.

        I will make it even clearer:

        CO2 is a GHG
        GHGs can cause warming
        We have observed warming
        Therefore, CO2 caused the observed warming

        or

        GHGs cause global warming
        There was no increase in GHGs during the Middle Ages
        Therefore the historically recorded global warming during the Middle Ages was not real

        or

        GHGs cause global warming
        There was very little increase in GHGs from 1910 to 1944
        Therefore the observed warming from 1910-1944 did not really happen

        or

        GHGs cause global warming
        There has been a major increase in GHGs since 2001 but no warming, but sight cooling instead
        Therefore the observed cooling since 2001 did not really happen

        ad nauseam

      • manacker, surely you mean
        Increased GHGs cause global warming
        GHGs are increasing
        …the rest follows from logic

      • Edim

        Caused by climatic factors. There’s no correlation between increased concentration and human emissions.

        Other points are irrelevant.

        “Caused by climatic factors?”

        Okay, so we have a clue here what’s going wrong with your approach to logic.

        You don’t have one.

        You introduce a whole range of assumptions needlessly without logical basis, conflate ambiguity, and intentionally change elements of the problem space.

        http://dilbert.com/strips/comic/2011-06-03/

        manacker

        It appears both logic and rhetoric are outside your experience, else you might be expected to recognize a rhetorical device when presented with one.

        In the face of logic you produce a flimsy cascade of surmise and obfuscation that is nothing but a tissue of disconnected assertions, mainly ones that are inconveniently difficult to establish and dependent on false reasoning.

        With reasoning skills like this, why aren’t you in politics?

      • Ged

        Thanks.

        Did you also look up Argumentum ad Absurdum?

  54. doc martyn:
    “Why arnt the oceans isothermal?
    Seriously, why is there a temperature gradient where the vast majority of water at death is colder than the top ?”

    ice floats. rejected brine sinks.

  55. yah, i see your abuse of logic, there.
    so how about the evidence that water vapor has increased 44%?
    unless the number comes from a place sun doesn’t shine, eh.

    • gnomish

      If you see the speck in my eye, how is it you still miss the beam in your own?

      I have little to say about water vapor (a subset of GHGs) levels, with regard to the spiking of other GHG’s: CO2 measured at 390 ppm, 150 ppm above the 800,000 year mean of 230 ppm; methane levels tripled, per HIPPO, exotic industrial products that nature does not produce remaining or increasing in the atmosphere, again as revealed by HIPPO.

      Water vapor is clearly increasing, too. Perhaps not so much, but then, why quibble about it?

      I’m perfectly satisfied to amend to, “virtually all GHG’s have increased measurably in the atmosphere and a large number of GHG’s have increased over 44% from levels since before the human race evolved.”

      See how easy that is?

      • You’re the one who’s a bit short on comprehension.
        Water vapour is a greenhouse gas.
        There’s more than 100X more water vapour than CO2
        Therefore CO2 constitutes less than 1% of greenhouse gases
        So a 44% increase in CO2 is less than a 0.44% increase in greenhouse gases.
        Therefore a 44% increase in greenhouse gases would necessarily mean a ~44% increase in water vapour.

        Now pick holes in that logic.

      • Peter317

        Interesting choice of words.

        The holes in H2O’s absorbtion bands, ie the non-overlapping parts of the spectrum, fit nicely where CO2′s absorbtion peaks.

        Further, with GHG absorbtivity, the effects are logarithmic. Simplifying, a 100x increase in concentration is only on the scale of a 10x increase in effect. (Well, it’s more than 10x still, but for our purposes, as we can see later, hardly important.)

        Further, water vapor has always been around the levels it is today, give or take. Except slightly higher now, due positive feedbacks, so we’re looking at net marginal increase, not totals.

        Further, though most GHG’s are incredibly short term, some have more impact than others. The nameless GHG’s in the category of methane etc. are among the most intensely effective. H2O too is more powerful than CO2, if only its feedback were negative, which we know it isn’t, and if only its residency time weren’t so pitiably small relative to CO2, and if only CO2 weren’t accumulating with an almost exponential growth in emission, and if only we weren’t concerned about biological feedbacks that might independently run away regardless of what we do once soil microbes reach a critical density.

        Further, CO2 is nonsaturating, in addition to being non-overlapping, meaning it keeps absorbing regardless of concentration.

        Further, the mean path length of OLR is determined not by concentration (a volume dependence), nor so much by optical density (though that is a dramatic level of effect, and we’re altering that in the IR bands a jaw-dropping amount), but by simple linear frequency of incidence, equating to a one-dimensional, linear amount.

        So there’s your holes plugged.

        44% increase in CO2. Tripled methane. Significant and growing levels of industrial — some banned — GHGs resident. Increase in NOX’s and other N-’s due agricultural activities.

        Positive feedback in H2O.

        I stand by my number, however poorly a single number may reflect a complex situation.

      • All of that’s beside the point.
        The argument was about a 44% increase in GHG’s That argument has been shown to be false.
        If the argument was about the relative effect then you should have said so at the start.
        Don’t start an argument and then move the goalposts.

      • water varies, of course, but over the 70% of the planet misnamed ‘earth’ instead of ‘ocean’, can we call it 40,000 ppm?
        CO2 is all the way up to 390 ppm? that’s about 0.0000039%, right?
        doesn’t sound like somebody splitting hairs but rather somebody zealously inflating the significance of a hair.
        not just that, but exaggerating an almost invisible hair, attributing to it supernatural powers and evangelizing its divinity. you seem to be preaching.
        after all, this thread’s topic is, nominally, about that neglected but dominant ‘ghg’ with albedo.
        co2 hasn’t got much to do with that.
        clouds make shade, you see.
        now, if you wish to contend that co2 makes shade in the infra-red region, cool. :)

      • Peter317

        Your argument may have been about a 44% increase in GHG’s, with the fine point being the inference of ‘all GHG’ you insist on reading into what was clearly intended in context otherwise.

        Which is the argument of a hairsplitting nuisance.

        Sure, I’m quite happy to exploit your nuisance argument goalpost-movingly to emphasize the merit of my case, but I’m equally happy to stipulate to the amendment, “almost irrelevant water vapor concentrations aside, an increase of 44% of CO2 over the mean CO2 level (230+/-50 ppm compared 390 ppm) our species is adapted to, and a much higher change of virtually all other GHGs.”

        For hair-splitters who think any single number captures the intricasies of complex, dynamic, nonlinear systems. :)

      • No, you said GHG’s when you really should have said CO2.
        Then when others pointed out the logical flaw, instead of simply correcting yourself, you went into full-on attack mode.

      • gnomish

        Perhaps your quibble about the name of the planet ought contemplate depth, and assert the use of ‘Magma’ in place of ‘Earth’ instead.

        However, depth doesn’t seem to be your thing.

        You include a lot of zeroes (five of them after the decimal point) ahead of your little 39. Those zeroes of Nitrogen and Oxygen are understood to not be GHGs (at least not very powerfully, certainly enough that we can for diverse reasons dismiss them).

        And while you think CO2 is small by concentration at 1% of the concentration of the GHG H20 vapor, (Peter317′s already ventured a 0.44% increase of atmospheric GHGs above the historic mean of the past 0.8 million years) I’ve already thrown down the gauntlet, which you do not answer, of the narrow band of IR not absorbed by other GHGs but only by CO2 (narrowing the aperture through which IR escapes), of optical density, and linear mean path length of OLR encountering GHG molecules.

        Which it seems you lack the depth to apprehend.

        So, to make the point as flat as a sheet of paper, we’re adding as much optical density of IR absorbtion every decade to the sky above your head as would happen by filling 8.5″x11″ pages with pencil until the pencil were used up.

        We know inbound solar radiation contains a tiny fraction of IR, compared to the fraction contained by outbound radiation that had been absorbed at the surface. CO2 isn’t a magical shade to keep the IR off our heads, but to warm the atmosphere from the energy of the light it absorbs.

        We’re making IR encounter so many more molecules of CO2, so much more often, that some significant effect is inevitable.

        And I’m not even a particular AGW-convinced proponent.

        To me, at the surface, the outcome might be cooling, heating, increased wind, or more frequent change in wind direction, something malignant, something benign, but whatever else, something unplanned and large in an ill-understood complex system we depend on.

        This is poking a hornet nest with a sharp stick.

        It’s past time to take the sharp stick away from those without the reasoning ability to understand this simple, straightforward logical Risk-abatement case.

      • i’ll give all your analogies a pass – they are fanciful.
        you’ll have to importune someone else for troll food, now. so sorry.

      • Peter317′s already ventured a 0.44% increase of atmospheric GHGs above the historic mean of the past 0.8 million years
        I’ve done no such thing.

      • Meant to end that with, “I’ve done no such thing”, but messed up the html tags

      • Peter317

        “There’s more than 100X more water vapour than CO2
        Therefore CO2 constitutes less than 1% of greenhouse gases
        So a 44% increase in CO2 is less than a 0.44% increase in greenhouse gases.”

        Would be the thing you did such a thing as, to disambiguate. Compared to a gnomish 0.0000039, much closer to the concentrations at issue.

        Were it concentrations that mattered, and not holes in the IR absorbtion spectrum that CO2 fills, nor the logarithmic relationship of IR absorbtion to concentration of gases (H2O concentration changes thereby being much less influential than CO2 changes, due H2O already being so concentrated in so many places historically), nor positive feedbacks, nor the much longer residence time of CO2 compared H2O, nor the problems of optical density and mean path length of OLR, nor the biochemical impacts of CO2 heretofor unmentioned.

        It appears gnomish cannot cope with reason by logic, nor by analogy, nor by classical allusion, nor by dialectic, but only by resorting to name-calling and foot-stomping, and so is a lost cause. I sincerely wish you better fortune with your intellectual capacity.

      • Peter317

        “No, you said GHG’s when you really should have said CO2.
        Then when others pointed out the logical flaw,..”

        Indeed, I said “GHG’s” when I ought have said “all known GHGs except water vapor have increased at least 44% above their pre-Industrial mean of the last 0.8 million years, and water vapor concentrations have increased as a feedback.”

        Because it was simpler, and anyone following the discussion would know what it meant or could figure it out for themself with a little common sense.

        You took a mere gloss and equated it with a logical flaw, which itself is a logical flaw.

        You did it knowing what the gloss represented, indicating you’d rather leap in and commit a logical error hypocritically to go on the full attack than offer a clarification or seek one.

        I have nothing to answer to a hypocrit for.

      • GISP2 suggests CO2 was 324.8 ppm during the Holocene.

        Which is about what Mauna Loa was in 1969.

        http://theinconvenientskeptic.com/2011/08/why-the-co2-ice-core-reconstructions-matter/

      • This GISP2 CO2 record is not reliable:

        The CO2 concentration of air trapped in GISP2 ice formed during periods of rapid climate change
        Smith, H.J., M. Wahlen, D. Mastoianni, D. Taylor, P.A. Mayewski.

        The CO2 content of air occluded in GISP2, Greenland, ice formed across two separate stadials, centered at approximately 46 kyBP and 69 kyBP, is as much as 90 ppm more during warm intervals than during cold. These CO2 variations are superimposed on changes in annual layer thickness and 18O of the ice and do not show the 200-700 year offsets which would be expected for concurrent variations in the atmosphere and the ice. The CO2 concentration during the stadials are similar to the atmospheric values recorded by Antarctic ice of the same age, so processes occurring in the ice after bubble enclosure must be enriching the air trapped in GISP2 ice formed during the interstadials. This conclusion is supported by ca content and ECM data for the ice, which show that adequate carbonate is present to produce these enrichments and that CO2 content is high only when the electrical conductivity (a proxy for H+ concentration) is high. High resolution mapping of one 4 cm section of ice reveals a 200 ppm increase in the CO2 content of the trapped air, from approximately 275 ppm to 475 ppm. Analyses of the TIC of ice from both the LGM and Holocene show that the 13C approaches that of soil and marine carbonates and imply that most of the Ca in the ice is from CaCO3. These results suggest hat the CO2 record preserved in ice can be altered by in- situ decarbonation reactions and that only ice containing either abundant or essentially no carbonate contains a reliable record of paleoatmospheric CO2.

      • Different time frame Rattus. That paper is dicussing 46,000 to 63,000 years ago.

      • Bruce, I think the take home message here is that you need very clean ice, or ice which does not show evidence of carbonate decomposition to feel certain about the measurements. There is also the possibility of surface melt contamination, yet another process which is known to contaminate CO2 records in ice cores. The GISP2 record shows multiple (more than I though possible, really) portions of the core which show melt contamination.

        See the GISP2 page where I learned all of this. It is not as complete as you and I (certainly I) might like, but it does provide some interesting information.

      • Plant Stomata suggests even higher levels.

        However, one of the papers I was reading points out that CO2 levels tend to be lower in Greenland, so that 300+ ppm in Greenland may mean even higher levels in the rest of the world.

        Consider GOSAT shows a 30ppm different between regions, I think the Greenland CO2 levels may possibly be under reporting.

      • Bruce

        GISP2 among ice cores very interesting and important, and a long way remains to go in advances on CO2 level measurement, I’m sure.

        This isn’t the first, nor even only present, challenge to the most acceptable CO2 level models, and I’ll be happy to hear an analysis that improves current understanding of the ice core.

        Nothing, however, in what the skeptical analysis says to definitively alter the indications in the wider data available.

        Which remains 230+/-50 ppm to six sigma for 0.8 million years, until 250 years ago.

        The GISP2 values remain outliers that don’t invalidate that statistic.

      • Bart – The GISP2 data didn’t show what Bruce claimed. That one value was the highest value in the group, with the majority from that era (and the average) less than 300 – GISP2 data.

      • Fred: “The GISP2 data didn’t show what Bruce claimed”

        Yes it did. There were 19 measurements above 300ppm

        1664.02 327.1
        1637.03 324.8
        1716.025 323.5
        1640.03 322.4
        1634.02 320.9
        1682.02 318.6
        1662.02 315.7
        1670.02 311.6
        1722.025 308
        1728.03 307.3
        1664.06 306.9
        1631.13 306.8
        1672.025 302.7
        1646.05 302
        1616.02 301.4
        1625.02 301.3
        1650.02 301.3
        1746.02 301.2
        1622.02 300.5

      • “Temperature changes recorded in the GISP2 ice core from the Greenland Ice Sheet (Figure 1) (Cuffy and Clow, 1997) show that the global warming experienced during the past century pales into insignificance when compared to the magnitude of profound climate reversals over the past 25,000 years. In addition, small temperature changes of up to a degree or so, similar to those observed in the 20th century record, occur persistently throughout the ancient climate record.”

        http://wattsupwiththat.com/2011/01/24/easterbrook-on-the-magnitude-of-greenland-gisp2-ice-core-data/

      • Bruce cited a value GISP2 value of 324 as indicative of CO2 concentrations in the early Holocene. That was cherry-picked because it was the highest among the 45 or so values in the time range he cited between 1622 and 1746. The values average out below 300 ppm. Here is the link again for readers to visit for the entire set:

        GISP2 CO2.

      • Fred, 19 out of the 220 CO2 measurements were above 300. I wouldn’t call that cherry picking.

        However, the Dye 3 ice core shows an average CO2 level of 331 ppmv +/-17 during the Preboreal (~11,500 years ago).

        GRIP shows 295ppm +/-16

        ftp://ftp.cricyt.edu.ar/pub/jaranibar/ice%20cores/97JC00182.pdf

        Page 26,541

        Admittedly these higher values are claimed to be because of chemical reactions.

        OTOH:

        “Plant stomata suggest that the pre-industrial CO2 levels were commonly in the 360 to 390ppmv range.”

        http://debunkhouse.wordpress.com/2010/12/25/co2-ice-cores-vs-plant-stomata-wuwt/

      • Here again is the link to the GISP2 Data in case readers want to confirm that Bruce’s value of 324 was cherry-picked, and that the average values are below 300, as are the large majority of values.

      • Both GRIP and Dye 3 suffer the same problem pointed out in the abstract from the paper I posted above. Indeed, your own link to the GRIP (not Dye 3) paper points this out.

      • Fred, GOSAT shows 30ppm differences from region to region today. To call the 300+ ppm readings cherry picking when there were 19 of them is quite silly and very desperate on your part.

        May I remind you Dye 3 shows a 331ppm average.

      • Rattus, reading the papers, one of the reasons they claim GISP2 and Dye 3 are of is because they claim CO2 is well-mixed.

        http://onlinelibrary.wiley.com/doi/10.1034/j.1600-0889.1991.t01-1-00003.x/pdf

        GOSAT shows otherwise.

      • “Studies of plant stomata show that the currently-held view of predominantly stable CO2 levels (260-280 ppm) before the Industrial Revolution (1750 AD, i.e. 200 years B.P.) may be an inaccurate view. CO2 levels appear to have regularly exceeded 280 ppm– the average of CO2 concentrations across the Holocene interglacial period (last 11,000 years) appears to have been approximately 305 ppm (see ref. 10-20).

        Contrary to the prevailing notion of CO2 stability, CO2 swings of 20-50 ppm or more over timespans of 500-1000 years appear to be the norm– not the exception.”

        http://www.geocraft.com/WVFossils/stomata.html

      • Chief, the problem with your statement is that you are assuming that a proxy for CO2 is a better estimate of levels than a direct measurement of CO2. Do you care to justify why a proxy is better when you have a direct measurement?

      • Rattus, unlike treemometers which we all know quit working in 1960:

        “A standardized way of counting stomata– called the stomatal index ( SI [%] )– has been found to be a good way to estimate the CO2 content of the atmosphere when the plant was alive. The SI-CO2 relationship varies according to plant species, habitat altitude, and other factors.

        Correlation charts are constructed using modern plant specimens by determining their SI numbers and corresponding CO2 concentrations. When SI and CO2 ranges are fully characterized for a plant species, the charts are used as to estimate CO2 levels for related species in the geologic past.”

      • 800,000 years. If we had daily measures for every one of those years, similar to Mauna Loa, we’d have approaching 300 million observations.

        If three of these observations lay outside the trend, we’d still have eight sigma confidence.

        We don’t, of course, have daily observations, however by the plurality of sites, goodness of fit, and various statistical and curve smoothing methods, and our knowledge of the workings of CO2 in the atmosphere, even a few score paleo outliers do not reduce us below six sigma in the observed record.

        Greenland is more subject to volcanic and biological sources of unusual CO2 concentrations than Antarctica.

        A few random higher measures in Greenland or Hawaii do not dislodge the ice core record.

        It merely makes it more interesting.

        Chief’s stomata nonsense.. leaves me nonplussed. I’d think he’s smarter than to pull off a graph like that.

      • GOSAT CO2 measurements fluctuate all over the place with a range of 30ppm.

        from http://www.gosat.nies.go.jp/index_e.html

        The myth of stable pre-fossil fuel CO2 is a just another big practical joke the consensus has played on people.

      • There are many other reference to variability seen in stomata and not ice – for reasons involving gas diffusion leading to broad sampling intervals.

        http://www.pnas.org/content/105/41/15815.full.pdf

        Do some research before making a bigger idiot of yourself.

      • Gack, Bruce, you do know about the seasonal cycle, don’t you? This is on the order of 10 ppm annually, on a global basis. Most of this is due to plant respiration in the NH, and indeed this is what is seen in the GOSAT images. It would be nice if these were not animated GIF’s so that detailed comparison of month to month variations could be evaluated, but they aren’t.

      • 10ppm 30ppm

  56. McIntyre continues: http://climateaudit.org/2011/09/08/more-on-dessler-2010/

    Which is then blog hyped (travesty!) here: http://www.c3headlines.com/2011/09/the-dessler-2010-travesty-its-now-obvious-why-he-avoided-using-hadcrut-data-the-gold-standard.html

    Anyone care to claim this is not a scientific debate, albeit wrapped in hyperbolic rhetoric?

    • David – I’ll weigh in on this, mainly to make a larger point. First, though, it’s worth mentioning that Dessler referenced earlier work by Sohn and Bennartz who reported a bias in CERES satellite clear sky measurements due to water vapor differences between clear and cloudy regions. His choice not to use those measurements was not unreasonable.

      My larger point, though, relates to all of the similar studies – by Lindzen/Choi, Dessler, and Spencer/Braswell – in their various incarnations, and it’s a point I’ve made previously but is relevant to current debates among the combatants. All of these studies attempt to evaluate feedbacks from climate fluctuations that are mainly ENSO events. In that sense, I don’t believe that whether the calculated cloud feedback is negative (as suggested by LC and SB), or positive (as suggested by Dessler) is at all informative about cloud feedbacks to long term forcing by CO2 or other greenhouse gases. ENSO involves temperature changes that (a) originate regionally in the tropical Pacific, (b) are short term, and (c) involve the imposition of a changing sea surface temperature on a previously unwarmed or uncooled atmosphere. In contrast, CO2 forcing is persistent, operates globally, and imposes temperature changes first on the atmosphere with only secondary responses by the ocean over time. This must inevitably result in differences in humidity, cloud distributions, ice/albedo changes, and a number of other relevant climate variables.

      I’m agnostic about the sign of cloud feedbacks on ENSO phenomena. I doubt very much that it tells us much about long term feedbacks to forcings that cause temperature changes originating in the atmosphere.

      • I agree with Fred, and am also very skeptical that this approach is going to tell us about doubling CO2. We see the biggest response to CO2 in the Arctic, and this approach isn’t even considering that. I have also stated this kind of doubt in the past. The better approaches look at global temperature and CO2 increases directly over long periods,

      • Fred- Why do you think that “His choice not to use those measurements (CERES satellite clear sky measurements) was not unreasonable.” When I read the concerns expressed in the study by Sohn, it appears that concern would not have been germane to the Dessler study.

        At a larger level- when we discuss ENSO, isn’t it reasonably probable that there will be other climatic events/processes that will happen in the future, or have happened in the past that we do not yet know of, much less understand?

      • Rob – If there’s a bias in the difference between clear and cloudy skies due to water vapor differences, and if that bias changes with temperature, as is likely, it’s reasonable to use a different method to arrive at unbiased estimates. I’m too unfamiliar with the details to comment further, but that’s probably true of the bloggers who raised the issue as well.

        On your other point, there are obviously things we don’t understand well, but also some we do. The latter include the radiative properties of CO2 (with confirmatory observational data), as well as some constraints on the range of climate sensitivity to CO2. Learning more about the things we don’t yet understand won’t make it possible to eliminate CO2 as an important contributor to global temperature change..

      • There are a couple of points to be made.

        Spencer and Braswell do not say that it is not possible to determine cloud feedback from global warming from these ENSO effects on TOA radiative flux and temperature.

        Most recent warming occurred in 1976/77 and 1997/98 – a simple observation. Most of the rest occurred as a result of cloud change. ‘The overall slow decrease of upwelling SW flux from the mid-1980′s until the end of the 1990′s and subsequent increase from 2000 onwards appear to caused, primarily, by changes in global cloud cover (although there is a small increase of cloud optical thickness after 2000) and is confirmed by the ERBS measurements.’

        Clouds are not a feedback to ENSO – but are an intrinsic part of the complex and dynamic system. Observations show that cloud is negatively correlated to SST (e.g. Zhu et al 2007, Clement et al 2009, Burgmann et al 2008).

        ‘During the 1997–1998 El Niño, observations indicate that the SST increase in the eastern tropical Pacific enhances the atmospheric convection, which shifts the upward motion to further south and breaks down low stratiform clouds, leading to a decrease in low cloud amount in this region. Taking into account the obscuring effects of high cloud, it was found that thick low clouds decreased by more than 20% in the eastern tropical Pacific. Zhu, P., Hack, J., Keilh, J and Zhu, P, Bretherton, C. 2007, Climate sensitivity of tropical and subtropical marine low cloud amount to ENSO and global warming due to doubled CO2 – JGR, VOL. 112, 2007

        Science suggests that the world is not warming as a result especially of decadal changes in the Pacific (Mochizuki et al 2010, Swanson et al 2009, Tsonis et al 2007, Keenlyside et al 2008) – albeit with immense uncertainties surrounding the origins of decadal variability.

        Decadal variability can be seen in the multivariate ENSO index of Claus Wolter – http://www.esrl.noaa.gov/psd/enso/mei/ – a La Niña bias (blue) to 1976, an El Niño bias (red) to 1998 and a return to an La Niña bias since.

        Anastasios Tsonis, of the Atmospheric Sciences Group at University of Wisconsin, Milwaukee, and colleagues used a mathematical network approach to analyse abrupt climate change on decadal timescales. Ocean and atmospheric indices – in this case the El Niño Southern Oscillation, the Pacific Decadal Oscillation, the North Atlantic Oscillation and the North Pacific Oscillation – can be thought of as chaotic oscillators that capture the major modes of climate variability. Tsonis and colleagues calculated the ‘distance’ between the indices. It was found that they would synchronise at certain times and then shift into a new state.

        It is no coincidence that shifts in ocean and atmospheric indices occur at the same time as changes in the trajectory of global surface temperature. Our ‘interest is to understand – first the natural variability of climate – and then take it from there. So we were very excited when we realized a lot of changes in the past century from warmer to cooler and then back to warmer were all natural,’ Tsonis said.

        Four multi-decadal climate shifts were identified in the last century coinciding with changes in the surface temperature trajectory. Warming from 1909 to the mid 1940’s, cooling to the late 1970’s, warming to 1998 and declining since. The shifts are punctuated by extreme El Niño Southern Oscillation events. Fluctuations between La Niña and El Niño peak at these times and climate then settles into a damped oscillation. Until the next critical climate threshold – due perhaps in a decade or two if the recent past is any indication.

        The shifts in the instrumental record are the tip of the iceberg – past shifts in ENSO resulting in major changes in climate over the holocene.

        This seems fairly well supported by peer reviewed science – I am surprised that something so well established can be so poorly understood by some. It just doesn’t seem to register. I guess we will have to wait for the world to continue not warming for a bit yet.

      • Whoops – SB say it is impossible to calculate feedback – whether positive or negative.

      • Fred, I do not see the relevance of your comment to mine. Nor that of the comments below. Your speculation does not erase the debate.

      • Fred, there is no ideological fervor on my part. I have been lurking for months and paying attention to your often well thought-out and sage comments, but it became apparent that you might be the one with the ideological fervor, but you disguise it surreptitiously in a passive aggressive manner. I have examined McIntyre’s comments thoroughly and agree with him. Your only substantive comment is that he overreached and you previously stated that SP should not have been published. Specifically, how do you find fault with McI. No more overreaching pronouncements on your part – get into the math and stats if you will Fred.

      • Good luck with that, Rob.

      • Rob – Thanks for your reference to my “well thought-out and sage comments”. Hunter, who responded to you above this, may not agree.

        Readers who see these comments here may not realize that you’re referring to comments upthread, where you emerged from lurking to say that “everyone is waiting” for me to retract earlier statements I made, based on observations reported in CA by Steve McIntyre. Briefly, Rob, my earlier comments were critical of SB-11 for reasons unrelated to this, and also expressed reservations about D-11 for unrelated reasons. McIntyre’s points about D-10 wouldn’t change any of this. However, although it’s not terribly germane to the evaluation of SB-11, I thought Steve M. overreached by implying that the Science reviewers should have found D-10 to be flawed in its choice of datasets, implying that they overlooked those flaws. My only point was that D-10 provided a reason for the choice and the reviewers may well have considered the issue and agreed. Steve M. would have been better off merely reporting his observations without implications about deficiencies in either the paper or the review process – but again, this is a small point unrelated to the main topic of this thread.

        I don’t agree that everyone is waiting for me to retract comments, but I’ll leave the issue of ideological fervor for others to judge.

      • Fred , thanks for your reply. Your answer, though, was completely unresponsive. You were the one who stated un-categorically that SB was minor and insignificant enough not to be published. You are hand-waving again. I asked for specific mathematical and statistical criticism of McIntyre by you since it was you that broadly dismissed him. You may deny this, but you did. If you choose not to respond( preferably at CA), I will just assume you were hand-waving again. Time to get specific Fred.

      • Rob – My criticisms of SB-11 have been presented in these recent thread in extensive and quantitative detail in many long commentaries, with specific reference to Figures 2, 3, 4, and the text conclusions. McIntyre’s observations about Dessler’s 2010 paper do not touch on any of the points I made, but to appreciate that, you will have to revisit my many comments.

        My criticism of McIntyre is exactly as I stated above – it’s not his mathematics but the unwarranted implications he makes. It’s also very minor, for all the reasons I’ve stated – see, for example Comment 110441, and shouldn’t be lumped together with the evaluation of SB-11. Whether you agree or disagree with McIntyre should have no relevance to the quality of SB-11.

        If you revisit my earlier comments in this thread, you’ll find that I judged neither SB-11 nor D-11 to have been worth publishing, and that I gave reasons for that conclusion..

  57. i find that honesty is the easiest thing in the world.
    it really should be easy to correct a misstatement.
    it was an oopsie, then.

    for a 44% increase in ‘ghg’, if you forget to H2O – and it seems that the CO2 fixated do constantly disregard it – you’d need more than 44% increase in the others to make your statement true.

    and now i’ll correct your other misapprehensions-
    i’m neither mythical nor cute.
    neither your personality, nor your biblical quotations, nor your thoughts on socrates’ wardrobe have any relevance to ‘ghg’.
    sophomoric wisecracking only merits a rebuke.

    • gnomish

      One would think freshman sophistry merits the greater rebuke.

      If you insist on misreading to the point of ridiculous hairsplitting, then ought you not be ready to admit you’re no more than a hairsplitting nuisance, not to be taken seriously?

      But thank you for ascribing to me personality in so ambiguous terms as someone might think you believe I possess one worth mentioning.

      And you call me wise, say my thought relevant, and I have merit, to boot.

      How you flatter me.

  58. Ah, internet science. Dessler has contacted Spencer saying that he can change a few things that he noted in Spencer’s first take evaluation. It appears the Dessler 2011 has not be officially published. Its listed as an update on WUWT.

    • It sounds like Dessler, is making some changes based on errors pointed out by Spencer. Things that should have been obvious to a competent peer reviewer. Will anyone at GRL be resigning over this review failure? Perhaps a letter of apology should be sent to Watts?

      Cross post at RC

    • I personally am glad that Dessler is giving himself permission to make some changes in the galleys. I say quite seriously that it would be better for science if Dessler were to voluntarily retract this paper to give himself time for a complete rewrite. Unfortunately, I suspect that this is not going to happen because of the PR implications. (With all the public screaming that S&B11 is scientific nonsense, one would imagine that a rebuttal should be easy.)

      Without a searching rebuttal, S&B11 will ride through on default , instead of being tempered in the fire of good quality scientific debate, and that is not a good thing in my view. Or IPCC AR5 authors will declare by steamroller that the Dessler rebuttal is brilliant, irrespective of its quality, and argue that Spencer has been thoroughly discredited. Not good choices.

  59. “Dessler (2011) seems to select the models that best match the satellite observations.”

    Dessler is of the Church of ENSO; that’s his real point on the “cherry-picking” allegation –that the models that best represent ENSO also best match the satellite obs. Spencer is not of that church (at least regarding looking for evidence of cloud feedbacks). There is a degree to which from their own context they are just talking past each other on that point.

    • Oopsie, make that “cloud forcings” re Spencer and ENSO.

    • Spencers point is that IPCC considers an average of the 14 models in arriving at its sensitivity calculations. It is therefore not undreasonable to take the three most sensitve and three least sensitive models to be bounding the relevant curve, which is a long way from matching satellite data combined with HADcru temperature data.

      • tallbloke 9/9/11, 8:06 am, Spencer Braswell III

        t: Spencer’s point is that IPCC considers an average of the 14 models in arriving at its sensitivity calculations. It is therefore not unreasonable to take the three most sensitive and three least sensitive models to be bounding the relevant curve, which is a long way from matching satellite data combined with HADcru temperature data.

        Once upon a time if you bought a 22 ohm, 10%, discrete carbon resistor, which was promised to be between 20 and 24.5 ohms, the chances that it measured between 21 and 23 ohms was approximately zero! Super fat tails – all tails, no body. That was because those between 21 and 23 ohms went into the 22 ohm bins at 5% or less. If you checked the most accurate resistors offered by a vendor, say 1% or 2%, the distribution was usually uniform. Now no fat tails.

        IPCC reports radiative forcing, F_2x, from 19 models in AR4, ¶8.SM, Table S8.1, p. 8-73, and using the same set of models, it reports climate sensitivity, T_2x, in Table 8.2, p. 631. F_2x varies between 3.09 and 4.06 Wm^-2, and is approximately uniformly distributed (P = F – 3.12) in that range. T_2x varies between 2.1 and 4.4ºC, and is approximately uniformly distributed there (P = 0.40*(T-1.88)). These data support three scenarios.

        (1) IPCC already threw out the highs and lows to come up with the 19 models in the desired band.

        (2) The AGW algorithm is a random number generator. Or,

        (3) Both of the above.

      • You fell for it.

        Look at the models that Dessler plots.

        See the ones that fit best?

        Guess what? They are the models that fall in between the extremes
        Their ECR is 3.4 and TCR is 2.2

        So Spencer MAY have felt that showing the high and low was a fair approach. but those in the middle perform differently. Thats because this is not about sensitivity. Its about the ability to get ENSO right

      • On the basis of what principle can the “models that fall in between the extremes” be selected. In absence of a principled reason for their selection, you are arguing in a circle. The fact that they fit the data best is a result of hindsight not something inherent to these models. The fact that they fit ENSO is also a result of hindsight. Circular – circular – circular.

      • Mosh, Spencer is right in that it’s the average of the models that is at issue, since it is the average of the models which is used by IPCC to determine sensitivity. If a couple of the models happen to bear a passing resemblance to reality but are immersed in with a load that don’t, that’s tough.

        You don’t get to cherry pick the ones you like.

      • Just to repeat, it isn’t about “ability to get ENSO right”.

        D11 says “.. the models that do a good job simulating the observations (GFDL CM 2.1, MPI ECHAM5, and MRI CGCM 2.3.2A) are among those that have been identified as realistically reproducing ENSO [Lin, 2007].”

        First off Lin 2007 does not include MRI CGCM 2.3.2A as one of its models that realistically reproduce ENSO. It is in the group that “.shows an oscillation with a constant period shorter than the observed ENSO period, sometimes also with a constant amplitude.”

        Second, LIN2007 uses wavelet spectra from the models to identify its preferred group of models. “The third group of models [which does include GFDL CM 2.1 and MPI ECHAM5] generally displays significant interdecadal variability of ENSO in both the amplitude and period … Therefore, we do have a number of CGCMs that can produce interdecadal variability of ENSO.” So LIN2007′s test is reasonably weak, although it does find “Among these models, only the MPI model reproduces the observed eastward shift of the westerly anomalies in the low-frequency regime.”

        Third, as I noted earlier in this thread “ENSO Feedbacks and Associated Time Scales of Variability in a Multimodel Ensemble” Belmadi et al (2010) offers a closer look at the GCMs classifying them “into three groups that account for the dominant feedback process of the ENSO cycle: horizontal advection (mainly in the western Pacific), vertical advection (mainly in the eastern Pacific), and the combination of both mechanisms.”

        Belmadi et al do an initial assessment of the models based on absence of any significant peak in the interannual ENSO frequency band. In so doing they eliminate GISS-EH that LIN2007 had included in preferred group of models. Belmadi note that this exclusion is consistent with earlier work. Belmadi at this stage also continue their analysis with the models LIN2007 had excluded because they displayed “a pronounced spectral peak with period shorter than the observed ENSO period”. Belmadi is explicit about their testing regime whereas LIN2007 is unclear.

        Belmadi then classifies the various models according to the dominant feedback mechanism, and in so doing eliminates GFDL CM2.1 and MRI ECHAM5 from their preferred group (INM-CM3.0, IPSL CM4, UKMO HadCM3, and UKMO HadGEM1). Both are eliminated on the basis that they are overly dominated by the thermocline feedback. MPI CGCM 2.3.2A is also eliminated because of it’s dominated by zonal advection feedback.

        The point of all this is that what is being measured in GCM’s fit to the SB11 data is likely something other than ENSO performance, and that ENSO ability in GCMs is not a settled science.

      • Could you provide a link to a preprint or other online copy of Lin 2007?

      • Using google you can get it from biosfera.dea.ufv.br

      • No go, the only copy Google scholar found was on an apparently dead server at Ohio State. Same for standard Google. Could you post your link?

      • HAS – In addition to posting the Lin link, would you also post the Belmadi link?

      • Lin2007: ftp://ftp.biosfera.dea.ufv.br/users/francisca/Franciz/papers/Lin%20GRL%202007.pdf however being a ftp this won’t get you it. If you do a search in standard google for the title name and with filetype:pdf it will be around number 2 on the list and you can get it by clicking on the link (or by using Quik View)

        Belmadi et al: http://web.yonsei.ac.kr/climate/board/4/20100826034622_2010_JC_Belmadani_etal.pdf

      • OK, Got it. I have FileZilla and I know how to use it :)

        And thanks for the link, er, pointer.

  60. ‘(2) Much of the ENSO-related global temperature change starts with redistribution of heat within the ocean mixed layer, and can’t plausibly be attributed to a simple TOA flux change of the kind that would simply warm or cool but not redistribute. Without addressing this in more detail, I would suggest that SB-11 has produced no explanation of why observed surface temperature changes are accompanied by reverse changes at deeper levels.’ Fred M

    This seems typical of a confused narrative that emerges from trying to fit facts to theories rather than the other way around. I think it is just meaningless verbiage.

    The TOA flux variation with ENSO as a result of cloud changes is abundantly clear in the SW record. Nothing but cloud changes could cause such a large and rapid change. But clearly there is energy being transferred between ocean and atmosphere as well. Because we are talking ENSO – we are talking low level cloud and SW changes dominating. This is also apparent in the CERES record.


    Finally, since much of the temperature variability during 2000–2010 was due to ENSO [Dessler 2010], we conclude that ENSO-related temperature variations are partly radiatively forced. We hypothesize that changes in the coupled ocean-atmosphere circulation during the El Niño and La Niña phases of ENSO cause differing changes in cloud cover, which then modulate the radiative balance of the climate system.’ (SB11) No one is suggesting anything different – including Trenberth et al 2010.

    The second sentence seems to suggest warming (cooling) at the surface and cooling (warming) at depth. SB11 certainly did not address this. As far as I am aware there nothing in the ENSO literature that suggests this. The depth of the thermocline varies over time – but this is a result of wind and current dynamics.

    • Robert – Much of your comment repeats earlier comments you’ve made, but I don’t see it as addressing the issue of whether the flux changes are radiatively forced – the fact that clouds can mediate flux changes doesn’t answer this and therefor does not serve as substantiation for SB-11. I won’t go into more detail here because I’ve done so extensively in earlier comments.

      You state: “The second sentence seems to suggest warming (cooling) at the surface and cooling (warming) at depth. SB11 certainly did not address this. As far as I am aware there nothing in the ENSO literature that suggests this.”

      During an El Nino or La Nina, surface temperature changes out of proportion to changes in total ocean heat content – i.e., an El Nino is not an overall ocean warming but more a heat redistribution phenomenon, with an actual loss of OHC during early stages. If the surface warms compared with neutral conditions, but OHC doesn’t increase, then lower depths must experience a net cooling. Even if OHC increases at some point (later in an El Nino), if it’s not commensurate with the surface warming, there will be a relative cooling at depth compared with the surface.

      A similar principle applies to La Nina, where cold water upwelling from below the thermocline results in surface cooling. This water must be replaced by warmer water, again resulting in relative warming at depths compared with cooling at the surface. Only if the surface cooling were matched by a commensurate loss of OHC would we not find this change in the temperature ratios between surface and depth. A description of upwelling cold water from below the thermocline is given in an ENSO overview from the University of Washington .

      Isaac Held has an informative blog item on the comparison between OHC gains from external radiative forcings and OHC losses associated with surface temperature increases due to internal ocean dynamics – see Heat Uptake and Internal Variability.

  61. Fred,

    I didn’t repeat myself at all. The specific comment was in relation to SW radiative flux changes in CERES especially as irrefutable evidence that cloud changed in the period. The change in SW flux is direct evidence of change in cloud. The timing of flux change links it unmistakenly to ENSO. Clouds change in response to ENSO and have a significant effect on global energy dynamics. That is the reality in the data – and your inability to process this is the crux of the problem.

    I supplied earlier a link to an ENSO animation from Duke University in response to your simple and incorrect ENSO description. Take it that I understand the essential ENSO dynamics.

    http://judithcurry.com/2011/09/06/spencer-braswell-part-iii/#comment-109863

    Held discusses surface temperature changes in response to a mix of ocean SST changes and external forcing. It doesn’t seem all that relevant or compelling.

    I said that I was unaware of any discussion of this in the literature. The Pacific gains heat in an El Nino from the SW changes – and loses it through enhanced losses to the atmosphere. In a La Nina heat is transferred from the atmosphere to a cool sea surface and receives less SW. So unless you have some data – that is non-existent – your laboured and simplistic reasoning doesn’t interest me at all.

    I think yours is a confused narrative that arises from cognitive dissonance – as I have said. So a rational discussion with you is quite impossible. I will leave it to the reader to decide.

    Robert I Ellison
    Chief Hydrologist

    • You don’t understand how ENSO works – look at the animation for super elevation of the water surface in the western Pacific at the peak of an El Nino.

    • Robert – I’m surprised you don’t understand ENSO well. During an El Nino, the surface warms without an increase in OHC (at least initially), and so deeper layers cool. The reverse is true for La Nina. If you have reason to believe otherwise, you should state the evidence.

      I’m also surprised that you found Isaac Held’s analysis irrelevant since it deals directly with ocean heat loss in association with ocean surface temperature rise (and vice versa for La Nina coolings).

      Finally you state that clouds change in response to ENSO (with consequences for flux and temperature). I have never heard anyone claim otherwise.

      • Robert – A minor observation on blog dynamics. Our recent comments crossed in passing, each with a “you don’t understand” phrase in them. I know that you have a good understanding of many details underlying ENSO, but on this particular point, you seem to have misunderstood what is a simple principle – if the surface warms but ocean heat doesn’t increase, some deeper layer must cool (and the reverse for surface cooling).

  62. Fred,

    At the height of a La Nina the trade winds blow strong and true and pile warm water up against Australia and Indonesia. The water level in the western Pacific is elevated by up to 0.5m over neutral conditions. At some stage the trade winds falter – it involves Kelvin waves – and warm water moves eastward across the Pacific. It is warm water moving from west to east and not up and down.

    The thermal evolution of La Nina depends on upwelling of cold water in the region of the Humboldt Current. It doesn’t change the heat content above the thermocline – just piles it up against Australia and Indonesia.

    I specifically suggested that you look at the A/V with respect to superelevation of water in the western Pacific.

    There are of course feedbacks – energy transfer between ocean and atmosphere and cloud changes – that are largely unquantified.

    Held suggested that there was a split between surface warming by energy transfer to and from the ocean and by radiative forcing – certainly I have never suggested otherwise.

    T = Tf + Ti – OK but trivial

    ‘Much of your comment repeats earlier comments you’ve made, but I don’t see it as addressing the issue of whether the flux changes are radiatively forced – the fact that clouds can mediate flux changes doesn’t answer this and therefore does not serve as substantiation for SB-11… Finally you state that clouds change in response to ENSO (with consequences for flux and temperature). I have never heard anyone claim otherwise.’

    ‘flux changes are radiatively forced’??? You mean temperature changes are partly forced by ENSO cloud changes – which is what SB11 claimed.

    Nothing about this discourse encourages me to think that rationality will prevail over your dissembling – unconsciousness or otherwise.

    • I’ve noticed, Robert, that when someone corrects a mistake you make, your responses tend to be (a) long, (b) irrelevant, and (c) accompanied by personal insults against the person who corrected you.

      Just for the record, I pointed out that during ENSO events, surface and deeper layer temperatures tend to move in opposite directions, at least initially. You seemed to imply that was wrong, but it isn’t. You were simply in error in disputing that point.

      Second, you have responded with long descriptions of ENSO mechanisms, but that didn’t address the points I made. No-one claims that the vertical temperature disparities I cite are all there is to ENSO, and in fact horizontal (east/west) changes dominate over vertical ones. Those vertical disparities are nevertheless real, despite the fact that ENSO is much more complicated than just that.

      Finally, the problem with the SB-11 paper is that it didn’t substantiate its claim that most of the surface temperature changes associated with 2000-2010 (predominantly from ENSO fluctuations) were initiated by radiative forcing mediated mainly by clouds. Despite numerous comments from me and others that the existence of cloud/radiative flux/temperature relationships, while real, doesn’t prove SB-11′s point, you have continued to insist that clouds can affect fluxes and temperature as well as respond to them, as though that were disputed. It isn’t.

      You should probably step back a bit, and consider whether you want to apply your knowledge and skills to mutual attempts to understand climate better, or just to picking arguments. You have a poor success record with the latter, but you probably could do well if you chose the former instead.

      • Fred,

        “Finally, the problem with the SB-11 paper is that it didn’t substantiate its claim that most of the surface temperature changes associated with 2000-2010 (predominantly from ENSO fluctuations) were initiated by radiative forcing mediated mainly by clouds.”

        SB11:

        “Finally, since much of the temperature variability during 2000–2010 was due to ENSO [9], we
        conclude that ENSO-related temperature variations are partly radiatively forced.”

        There is a difference of opinion as to what SB11 claimed and showed.

        In case you don’t have a copy:

        http://www.mdpi.com/2072-4292/3/8/1603/pdf

      • Kuhkat – I’m familiar with the paper.

      • Fred,

        Your unwelcome and false characterisations notwithstanding – you are well known for bringing in red herrings and confused abstractions.

        ‘No-one claims that the vertical temperature disparities I cite are all there is to ENSO, and in fact horizontal (east/west) changes dominate over vertical ones. Those vertical disparities are nevertheless real, despite the fact that ENSO is much more complicated than just that.’

        You have have no citations or evidence for what you mean. Just a laboured and simplistic reasoning. Surface and deeper temperature do not move in different directions in ENSO – unless you mean in terms shifting depth of the thermocline. But this is related to the east/west dynamic. This is in fact a meaningless distraction – wrong as well – and as usual you defend your nonsense to the death.

        ‘Finally, the problem with the SB-11 paper is that it didn’t substantiate its claim that most of the surface temperature changes associated with 2000-2010 (predominantly from ENSO fluctuations) were initiated by radiative forcing mediated mainly by clouds. Despite numerous comments from me and others that the existence of cloud/radiative flux/temperature relationships, while real, doesn’t prove SB-11′s point, you have continued to insist that clouds can affect fluxes and temperature as well as respond to them, as though that were disputed. It isn’t.’

        You continue to misrepresent SB11 – first of all. What they concluded was that cloud changes in the period – mostly associated with ENSO – were partly responsible for surface temperature change. This is precisely how it was expressed in the paper – as I have quoted. It is obvious that this is the case.

        You obfuscate – whether deliberate or unconscious – and the above is an example of what I believe is your cognitive dissonance expressing as a confused narrative. A psychological commitment to a world view – an AGW equivalent to a ‘spaceship cult’ – and an inability to process information that is not in accord. It goes beyond confirmation bias in actively rejecting certain information.

        I can never win a argument with you in your estimation at least – because of your complete acceptance of the AGW world view. Any information that is not part of the world view is reflexively rejected. Thus a pattern of mental confusion emerges from the constant struggle with anomalies.

        For those of us who can recognise this – it is a fundamental constraint to effective communication. I am simply pointing this out as it is a relevant dynamic – a reason for intractable disagreement.

      • Robert – You continue to make multiple mistakes (along with continued insults). I would ignore both the mistakes and the insults except that the principles are fairly important for understanding both ENSO and the SB-11 paper. I’ve tried to provide an accurate perspective, and that isn’t helped by your inaccuracies about both ENSO and my comments.

        You quote me:
        No-one claims that the vertical temperature disparities I cite are all there is to ENSO, and in fact horizontal (east/west) changes dominate over vertical ones. Those vertical disparities are nevertheless real, despite the fact that ENSO is much more complicated than just that.’

        Then you state:
        “You have have no citations or evidence for what you mean. Just a laboured and simplistic reasoning. Surface and deeper temperature do not move in different directions in ENSO… and as usual you defend your nonsense to the death”.

        Robert – A relevant citation is the 2011 GRL paper on The Global Ocean Imprint of ENSO. From the paper (the bolding is mine): “The ENSO-related vertical redistribution of globally-averaged heat content between surface and subsurface layers, occurring throughout the record, is due primarily to changes in the east-west tilting of the equatorial Pacific thermocline. The large temperature changes in the individual layers mask the smaller vertically-averaged temperature change, in which the ocean loses heat when the surface layer is anomalously warm and gains heat when the surface layer is cool.” A warmed surface in an ocean that has lost heat inevitably requires cooling at greater depths, confirming the point I made and refuting your contrary claim that “surface and deeper temperature do not move in different directions”.

        You quote me on SB-11:
        “Finally, the problem with the SB-11 paper is that it didn’t substantiate its claim that most of the surface temperature changes associated with 2000-2010 (predominantly from ENSO fluctuations) were initiated by radiative forcing mediated mainly by clouds. Despite numerous comments from me and others that the existence of cloud/radiative flux/temperature relationships, while real, doesn’t prove SB-11′s point, you have continued to insist that clouds can affect fluxes and temperature as well as respond to them, as though that were disputed. It isn’t.”

        Then you say:
        “You continue to misrepresent SB11 – first of all. What they concluded was that cloud changes in the period – mostly associated with ENSO – were partly responsible for surface temperature change. This is precisely how it was expressed in the paper.”

        The paper went beyond “partly”, as follows (again my bolding): “We have shown clear evidence from the CERES instrument that global temperature variations during 2000-2010 were largely radiatively forced.” This is what I stated, and so I didn’t misrepresent what SB-11 claimed.

        I believe, Robert, that if you stop making one of your goals trying to prove me wrong, and instead focus on the goal of reaching an accurate understanding of the climate dynamics we’re talking about, everyone will be better off. As of now, you’re not accomplishing either goal.

      • More from the same GRL paper on quantitation, regarding the La Nina preceding the 2009 El Nino:

        “For the two years prior to the 2009 El Niño, the upper 100 dbar of the ocean gained 3.3 × 10^22 J yr−1 of heat, while the 100–500 dbar layer lost a similar amount.” That is a substantial divergence in the direction of heat flow.

      • An interesting example of the heat redistribution phenomenon is seen in this NOAA graph, with evidence of a biphasic nature. For example, at the time of the 1998 El Nino, a pronounced upward spike in SST is accompanied by a dramatic reduction in tropical Pacific OHC (where WWV = warm water volume above the 20 deg isotherm = a surrogate for OHC). The dip is followed quickly by a recovery upward.

  63. Fred,

    The Global Ocean Imprint of ENSO – ftp://ftp.astr.ucl.ac.be/publi/2011_08_03-08h21-hugues.goosse-12.pdf

    Let’s deconstruct it shall we?

    The ‘The ENSO-related vertical redistribution of globally-averaged heat content between surface and subsurface layers, occurring throughout the record, is due primarily to changes in the east-west tilting of the equatorial Pacific thermocline.’

    That is what I said. This is yet another example of reaching a false interpretation of a scientific discussion. Reading things the way you want them to be rather than as they are.

    ‘The large temperature changes in the individual layers mask the smaller vertically-averaged temperature change, in which the ocean loses heat when the surface layer is anomalously warm and gains heat when the surface layer is cool.’

    They find net warming of the ocean of 1W/m^2 in a La Nina and net cooling of the oceans of about the same in an El Nino. There are ‘vertically averaged’ changes in temperature – i.e. not warming (cooling) at the top and cooling (warming) at the bottom – that are masked by the changes in the layers.

    Previous studies have shown the reverse. ‘It differs from previous findings specific to the 1997 El Niño episode, of 1.5 W m−2 maxima in downward net planetary radiation and ocean heat gain in late 1997 [Wong et al., 2006, Figure 7]. The issue is of considerable interest for understanding interannual and longer variability in the planetary energy budget, but will need careful study of a longer instrumental record than the present Argo dataset. Argo’s value will continue to grow as its global coverage is extended in time to a decade and longer.’

    Here is the Wong (2006) – Figure 7 – http://i1114.photobucket.com/albums/k538/Chief_Hydrologist/Wong2006figure7.gif – a reference I have shared with you previously.

    As far as SB11 is concerned – is the discussion now about ‘largely’ or ‘partly’? There is no essential semantic difference. The simple model suggests 70/30. There is still no absolute quantitative certainty possible – both effects occur.

    I am far from insulting you Fred. I note only the passive/aggressive mode you indulge in and the confusion of your interpretation and expression. The latter emerges from cognitive dissonance as I say – the ongoing struggle to incorporate/reject anomalies.

    The latter is – as I say – not specifically aimed at you but in increasing understanding of the asymmetric nature of the discussion and of the participants. There is absolutely no common ground possible – because of the inability of warmists to process certain types of information.

    If what I think is what you perceive as an insult – so be it.

    Cheers
    Robert I Ellison
    Chief Hydrologist

    • Robert – My reply to this was below in the thread, and not nested here – I’m sorry if that leaves any confusion. In referring again to the various data sources, I would particularly emphasize the 2011 GRL paper, because it is only in the past few years that ocean heat data have become reasonably reliable. I hope you revisit this and other data sources to correct some misconceptions you appear to harbor.

      • To be more specific, significant earlier technical problems with the ARGO floats were addressed by Willis et al in 2007, leading to improved ocean heat data during the past several years.

        A La Nina episode started in mid-2007 and lasted into 2009, and is relevant to the GRL paper. You stated, in regard to the paper:
        “There are ‘vertically averaged’ changes in temperature – i.e. not warming (cooling) at the top and cooling (warming) at the bottom”.

        However, the paper states:
        “For the two years prior to the 2009 El Niño, the upper 100 dbar of the ocean gained 3.3 × 10^22 J yr−1 of heat, while the 100–500 dbar layer lost a similar amount.”

        Again, please see my more general comments downthread about what I see as misunderstandings on your part that you are reluctant to concede and which frustrate attempts to arrive at an accurate understanding of how the climate is behaving.

      • Global Ocean Heat Content Is Still Flat

        (And slightly down from 2003)

        http://wattsupwiththat.com/2011/09/08/global-ocean-heat-content-is-still-flat/

    • It would be helpful if you included some explanation or discussion with these links to your blog. We should not have to click through to find out what it is about.

  64. You’ve repeated yourself, Robert, but you haven’t yet found the moral courage to concede that you were wrong in denying that heat (along with temperature) move in opposite directions between the surface and deeper layers during the early stages of an ENSO event, as described in the links I provided and the text I quoted – you can go back and review them, and so can anyone else who is interested. You also repeat comments you’ve made before about SB-11, and so I don’t see any value in addressing them again.

    I don’t know about you, but I don’t see these things as a form of combat in which winning is more important than getting things right. Whatever self-image you try to preserve by not admitting error is going to result in the opposite effect on your image in the eyes of other observers.

    Regarding ENSO, I would have thought that you, as someone interested in acquiring a comprehensive knowledge of the topic, would be eager to correct your misconceptions, and so it’s regrettable that you don’t seem to be, at least in public.

  65. Fred,

    You are impossible.

    ‘Surface and deeper temperature do not move in different directions in ENSO – unless you mean in terms shifting depth of the thermocline. But this is related to the east/west dynamic.’ Chief Hydrologist | September 10, 2011 at 10:27 pm | Reply

    ‘‘The ENSO-related vertical redistribution of globally-averaged heat content between surface and subsurface layers, occurring throughout the record, is due primarily to changes in the east-west tilting of the equatorial Pacific thermocline.’ Your reference – Roemmich and Gilson 2011

    The other link was to a NOAA graph that shows heat content to the thermocline (sometimes taken to be the 20oC isotherm) in the equatorial Pacific vs. SST in the central Pacific. They co-vary. But this is an example of a Moolten red herring.

    So that exhausts your links – yet you persist in pretending that there is something more here than meets the eye.

    This is the extent of my repetition on SB11.

    ‘As far as SB11 is concerned – is the discussion now about ‘largely’ or ‘partly’? There is no essential semantic difference. The simple model suggests 70/30. There is still no absolute quantitative certainty possible – both effects occur.’

    It is impossible to talk to you on a rational basis. You simply continue to insist you are right about any trifle that comes into your head. And the smarmy concern about my moral welfare is concern trolling – I am not in the least interested.

    Cheers
    Robert I Ellison
    Chief Hydrologist

    • Robert (Chief Hydrologist):“Surface and deeper temperature do not move in different directions in ENSO”

      Roemmich and Gilson – GRL 2011 regarding the 2007-2009 La Nina years: “For the two years prior to the 2009 El Niño, the upper 100 dbar of the ocean gained 3.3 × 10^22 J yr−1 of heat, while the 100–500 dbar layer lost a similar amount.

      You have a conflict, Robert, between two desires. One part of you wants to understand ENSO accurately. The other part wants to believe that you weren’t wrong in insisting that surface and deeper temperature don’t move in different directions during certain ENSO phases.

      The two desires are irreconcilable. You will have to decide how to resolve the conflict, although you can do it in the privacy of your own conscience rather than in this public forum. I urge you, though, to consider my earlier point that self deception you engage in to preserve your self-image is likely to tarnish your image more in the eyes of others than simply admitting you were wrong. Based on my own experiences with mistakes I’ve made in the past, I also urge you not to rush impulsively into replying here, but to step back and contemplate your choices before saying anything further – or not saying anything.

      • Fred,
        Perhaps reflecting on what you do- simply repeating your assertions as if they are true and ignoring counter evidence- does not exactly brighten your image.
        But as I point out downt thread, the unintended irony you provide is of the richest sort.
        I look forward to your continued ignoring of the lack OA, and development of the new frontier of AGW apologia, that AMO is a statistical illusion, like the MWP and LIA.

  66. Fred,

    You stoop to quoting me out of context.

    ‘During an El Nino or La Nina, surface temperature changes out of proportion to changes in total ocean heat content – i.e., an El Nino is not an overall ocean warming but more a heat redistribution phenomenon, with an actual loss of OHC during early stages. If the surface warms compared with neutral conditions, but OHC doesn’t increase, then lower depths must experience a net cooling.’ FM

    I pointed out that this was clumsy reasoning with no basis in ENSO dynamics.

    ‘Surface and deeper temperature do not move in different directions in ENSO – unless you mean in terms of shifting depth of the thermocline. But this is related to the east/west dynamic.’ Chief Hydrologist | September 10, 2011 at 10:27 pm | Reply

    This is really just a simple observation that you failed to think about and entered into another lengthy, confused, pointless and quite shockingly misleading argument.

    This is the entire passage from Roemmich and Gilson 2011 – which again you quote out of context to support your rookie notion about ENSO energy dynamics.

    ‘For the two years prior to the 2009 El Niño, the upper 100 dbar of the ocean gained 3.3 × 10^22 J yr−1 of heat, while the 100–500 dbar layer lost a similar amount. The ENSO‐related vertical redistribution of globally‐averaged heat content between surface and subsurface layers, occurring throughout the record, is due primarily to changes in the east‐west tilting of the equatorial Pacific thermocline.’

    You then indulge in smarmy moralising based on the quotes taken out of context. You are beneath my contempt.

    Robert I Ellison
    Chief Hydrologist

    • I’ve tormented you enough, Robert. I’ll let your conscience do the rest and leave you in peace.

      • I called Fred on a confused and rambling narrative.

        ‘(2) Much of the ENSO-related global temperature change starts with redistribution of heat within the ocean mixed layer, and can’t plausibly be attributed to a simple TOA flux change of the kind that would simply warm or cool but not redistribute. Without addressing this in more detail, I would suggest that SB-11 has produced no explanation of why observed surface temperature changes are accompanied by reverse changes at deeper levels.’ Fred M

        ‘This seems typical of a confused narrative that emerges from trying to fit facts to theories rather than the other way around.’ me

        The confusion emerges from a struggle to reject or incorporate anomalies into his world view – which remains essentially unchanged whatever the subject. There is a pattern of obfuscation, convoluted reasoning, distractions, narrative devoid of any reference to science – and all in a passive/aggressive mode .

        I think that this typical of warministas rather than the exception. A cognitive dissonance that makes it impossible to communicate from outside of the mindset.

      • Fred,
        You demonstrate that acccidental irony is the best irony.

  67. The bottom line in regard to Spencer-Dessler is that fitting linear regressions to phase planes is a ridiculous way to try to measure feedback. Better tools reveal the truth of the matter: the feedback is decidedly negative.

  68. Richard Pinder

    I think that the Greenhouse Effect is as Miskolczi says it is. The speed of warming relative to the speed of cooling. That would explain the Greenhouse effect on Venus. As for the Earth, Miskolczi shows that the Oceans are part of the Atmosphere.

  69. Christopher Game

    Dr Curry writes: ” This inconsistency arises from trying to use the black box energy balance feedback model to make inferences about the impact of internal dynamical variations on clouds and the energy balance.
    “So how to do a sensible observational analysis of cloud feedback remains elusive.”

    Christopher opines: Of course Dr Curry is right here. They have to move to a second order model with two internal state variables, temperature and cloud extent, and they have to include information derived from the effects of an external driver variable, such as the annual cycle at least. No escape from this. Christopher Game