Skip to content

Skepticism about skepticism of global warming skepticism skepticism

A group of University of California professors headed by physicist Richard Muller recently released a report confirming global warming. Then geophysicist Judith Curry, a coauthor on the papers produced by the Muller group, turned around and said that their data actually show that global warming has stopped. (Also see clarification here.)

Curry is described in the news article as the second author on the papers, but the authors are listed alphabetically so it’s probably more accurate to describe her as one of the ten authors. Muller’s one, Curry’s another, . . . now I want to know what 7 of the other 8 authors think! (One of the authors is Richard Muller’s daughter Elizabeth, so maybe we shouldn’t count her as an independent view.)

Some enterprising reporter should really interview the other 7 authors of that report. Just a quick question like, “Is there scientific basis for saying that warming hasn’t stopped?”

To add some fuel to the fire, let me repost what my physicist friend Phil Price wrote after seeing Richard Muller speak at Berkeley:

On Friday, I [Phil] saw a seminar by UC Berkeley physics professor Richard Muller. He is famously smart; he wrote the excellent book “Physics for Future Presidents”; and he and has been in the news recently for testifying to Congress that Yes, the earth really is warming, based on initial analysis of a bunch of temperature data (for some reason many people were surprised that he would say that).

Here’s the abstract of his talk: “Because of its huge economic and political implications, Climate Change is rarely presented without spin. This will be an attempt to do that. I’ll discuss the physics of the greenhouse effect, and the data that indicate global warming. Among key topics are: Copenhagen — why did we fail to get a major treaty? Climategate — what really happened? IPCC standards — and why they are undergoing major revisions. What are the top prospects among the many choices for alternative energy? What kind of example can the U.S. set that could be followed by the rest of the world? I’ll also report on new results of our “Berkeley Earth” project — a detailed re-analysis of the evidence for global warming; see"

I was not the only one to roll my eyes at the first lines of the abstract: does he really think nobody gives talks about climate change without “spin”? I’ve seen quite a few talks about the science. But maybe Muller hasn’t. Whatever.

But then…the talk was absolutely terrible. The abstract was presumably a mistake, since Muller made no effort at all to do any of the things he said he would do. (He did not discuss the physics of the greenhouse effect, did not discuss Copenhagen, did not discuss Climategate, did not discuss IPCC standards). He made one good point about the difference between predicting something before the fact and explaining it after the fact (and how easy it is to fool yourself that you are doing the following). And he did present a very cursory sketch of some of the data about global warming, including a list of difficulties in interpreting/analyzing land-surface data. But that’s it. Almost a content-free talk.

Almost the whole rest of the time, he took a page out of Charlie Sheen’s book: he bragged about how fair-minded and smart he is compared to everyone else, complained about a variety of behaviors by other people (some of which were indeed bad, but many of which weren’t), and just generally held himself up as an exemplar of all that is right and fair and true, while oozing ill will about everyone else. Lots of pontificating; no technical content. And through it all, he dropped the name of Louis Alvarez — a Nobel Laureate physicist with whom he worked decades ago — about every five or ten minutes.

It was really an astonishing talk, and left most of the audience in stunned silence. It did have a certain train-wreck fascination, though, and I (and many others) stayed for part of the Q&A, in part to see if anyone would take him on about his attitude. (Nope).

Some colleagues and I spent a while discussing the talk: how does something like this happen? I mentioned the old saw that the flaws that are most irritating on others are the ones that one has oneself, and that I saw some application here. I have sometimes started talks by pointing out ethical transgressions by others, or by showing plots and results from the literature and discussing the ways in which they’re wrong. And I’ve sometimes intentionally implied or stated that people were distorting scientific results, or the ways in which those results are presented (i.e. “spun”). . . .

(Phil’s report came in the midst of a heated discussion of Charlie Sheen.)

Muller being an obnoxious speaker doesn’t invalidate global warming (any more than the silliness of Daily Mail article is evidence in favor of it). For reasons I’ve discussed elsewhere on this blog, I go with the scientific consensus on climate change, and I’m speaking as someone who’s worked (a bit) with climate models and knows about their weaknesses.

In any case, this is an ideal Halloween post: climate change is scary!

P.S. Phil summarizes the immediate controversy well here:

The earth is somewhat warmer now than it was 50 years ago, and will be much warmer in 50 years than it is now. It would be very useful to understand the rates of increase on those timescales. It is not at all useful to understand it on a timescale of 10 years.


  1. Kevin says:

    Why would you say: “Muller’s one, Curry’s another, . . . now I want to know what 7 of the other 8 authors think! (One of the authors is Richard Muller’s daughter Elizabeth, so maybe we shouldn’t count her as an independent view.)”

    Elizabeth Muller is an educated adult, who is CEO of a consulting business. Do you think she’s incapable of independent thought?

    • Andrew says:

      “CEO of a consulting business” is not the first qualification I would think of to evaluate a climate science research claim! I’m sure she’s capable of independent thought, but it would be a bit much to ask her to adjudicate a dispute involving her relative.

  2. Frank says:

    I’ve been struck by the arrogance of the Muller quotes I’ve seen in the press (initially I thought they might have been taken out of context). I don’t recall the exact wording but they were all along the lines of: ‘We [Muller and co] are the first group of scientists to do a serious analysis of the data”…”No one else has been skeptical of this before” et cetera, et cetera. Pretty much in line with the way Phil described the talk. An odd approach considering a) their results have not been peer reviewed or published yet and b) their results are pretty much in accordance with all the former results. Sort like saying: “Even though we got the same answer everyone else did, we were the only ones who did it right”

  3. Fraunhoef says:

    … so we’re absolutely stuck with a ‘He Said, She Said’ situation that can not possibly be resolved by neutral outside review of the Muller/Curry/+8 data ??


    ” Professor Muller insisted … his claims… were {not} misleading because the project had made its raw data available on its website, enabling others to draw their own graphs.

    However, he admitted it was true that the BEST data suggested that world temperatures have not risen for about 13 years. But in his view, this might not be ‘statistically significant’, although, he added, it was equally possible that it was – a statement which left other scientists mystified.

    ‘I am baffled as to what he’s trying to do,’ Professor Curry said.

    Prof Ross McKittrick, a climate statistics expert from Guelph University in Ontario, added:

    ‘You don’t look for statistically significant evidence of a standstill. ‘You look for statistically significant evidence of change.’ “

    [ ]

    • Martin Vermeer says:

      > Prof Ross McKittrick, a climate statistics expert

      Perhaps the host of this blog might want to characterize this characterization ;-)

  4. Steven Sullivan says:

    Yes, you do, Prof McKitrick, but if it’s not there, you don’t say ‘it hasn’t warmed in ten years’. You say, ‘there’s no statistically significant evidence’, and if you want to expand on that for the layperson you say that it means it *may have warmed as much as X* , or it *may have cooled as much as Y* or it *may not have warmed or cooled at all*. Because that’s what UNCERTAINTY implies.

    And before you say any of that, you take account of outliers, and note the danger of parsing ‘meaning’ from a ten-year slice of a long and noisy record:

    • anna haynes says:

      Except I’m confused. What should the question be? Curry’s view (quoted at Tamino’s) seems to be that “There has been a lag/slowdown/whatever you want to call it in the rate of temperature increase since 1998″; but this could still be consistent with a Yes answer to the simple Q you posed for BEST authors above, “is there scientific basis for saying that warming hasn’t stopped”.

      I think I’ll ask both.

      • anna haynes says:

        Here’s what I asked the other 10 members of the team yesterday (by email, 1 indirectly):

        1. Regarding Dr. Curry’s having been quoted (perhaps inaccurately) by David Rose as saying “There is no scientific basis for saying that warming hasn’t stopped (since 1998)”
        In your view, _is_ there a scientific basis for saying that warming has continued since 1998?

        2. In your view, is there a scientific basis for saying that there has been a (meaningful) lag/slowdown/whatever you want to call it, in the _rate_ of warming since 1998? (compared to the rest of the historic record)

        No response yet, a little over 24 hours later. (if they’re all at the conference…)

        • Andrew says:

          Thanks! I’m curious how (and if) they respond.

          • Anna Haynes says:

            Well, there’ve been no responses to my emails asking the other team members; but their FAQ does make it pretty clear that they disagree with Curry, does it not?
            “Some people draw a line segment covering the period 1998 to 2010 and argue that we confirm no temperature change in that period. …[but] the decadal fluctuations are too large to allow us to make decisive conclusions about long term trends based on close examination of periods as short as 13 to 15 years.”

  5. Dan Pangburn says:

    There are five agencies that report temperature anomalies. I graph them all and average them to avoid bias. The average shows that there has been no average global temperature increase for a decade. They are graphed through August, 2011 in the pdf made public 9/24/11 at Links to the five agencies’ data are provided in the pdf made public 3/10/11.

    Also included in these pdfs is a simple equation (9/22/11 pdf) that calculates the average global temperatures (agt) since 1895 with 88.4% accuracy (87.9% if CO2 is assumed to have no influence). The future average global temperature trend that this equation calculates is down.

    The huge effective thermal capacitance of the oceans (about 30 times everything else) will cause the decline to be only about 0.13°C per decade. The decline may be as much as 0.22°C per decade if the sun goes really quiet.

    • Clark Andersen says:

      I don’t doubt that the equation can provide a reasonably good prediction of global temperature, but I am concerned that ESST (effective sea surface temperature) may well be confounded (correlated) with CO2 and other greenhouse gases, such that it is not possible to isolate the effect of one from the other. Also, since ocean temperature moderates air temperature, a strong relation between ESST and global temperature is to be expected. Perhaps I’m missing something.

      • PI says:

        The equation is meaningless for the reasons you described. It’s circular to use temperatures to predict temperatures (even if it’s using SSTs to predict global temperatures). And the model makes no physical sense, despite claiming to be physically based. It has terms that are supposed to vaguely represent physical processes, but it doesn’t have a consistent treatment of them. It’s kind of a mangled “cargo cult” version of a zero-dimensional energy balance model with no transient dynamics. It tries to integrate the solar forcing in an ad-hoc summation of forcings instead of letting a real ocean model do the integration of heat fluxes; simply adding a heat capacity coefficient does not introduce physical dynamics. It separates out the CO2 forcing from the solar forcing and treats them differently, applying a heat capacity and ocean integration to one but not the other. It treats SST as a forcing instead of a response. It ignores known significant forcings such as non-CO2 greenhouse gases and industrial aerosols. It’s just nonsensical from a physical standpoint. From a statistical standpoint, it has the confounding problems you describe, and is useless for attributing the relative human and natural influences on the climate.

        A more useful physical model would be a real energy balance model, with a better ocean component such as a two-layer ocean model or upwelling-diffusion model that acts to integrate forcings. If you want to include solar activity you can put it in as a radiative forcing at the top of the atmosphere, perhaps with its own coefficient to modulate its strength if you want to postulate an additional solar-cosmic-cloud link or whatnot. Of course models like this, and much more sophisticated ones, have been fit in the climate science literature for decades. They produce entirely different predictions.

        • John Mashey says:

          PI: see Dan Pangburn’s earlier work.
          I am afraid your (reasonable) explanation may fall on deaf ears.
          His key conclusion was highlighted:
          “The conclusion from all this is that carbon dioxide change does NOT cause significant climate change.”

          Sadly, his groundbreaking result, which overturns much of modern physics, seems restricted to his website, ClimateRealists and similar venues, rather than being published in Science, Nature, etc.

          • Nick Cox says:

            Pauli used the put-down “not even wrong” about highly confused physics.

            Looking at this site I came across this claim:

            “Tens of billions of dollars have been wasted in futile attempts to prove that added greenhouse gases are a major cause of Global Warming and to predict what might happen if the planet continued to warm.”

            Setting aside the judgments of waste and futility, where does “tens of billions of dollars” come from? Assuming here a context of the United States, and fixing conservatively on USD 30 billion, that could be factored in several ways, e.g.

            a million climate researchers on average receiving USD 30,000 each
            a thousand … on average USD 30,000,000 each

            Clearly a total has to be consistent with a sample size and an average, regardless of the fact that some grants are much bigger than others.

            Looks to me like an exaggeration by some powers of 10 whichever way you think about it. I’d welcome a refutation with real data if Pangborn has one.

          • Dan Pangburn says:

            Nick Cox,
            A quick look at the GAO report at reveals that spending in the US alone since 1993 adds up to about $90 billion.

        • Dan Pangburn says:

          The pdf made public 9/24/11 at shows an equation that, when fitted to the average global temperatures prior to 1990, accurately predicted the average global temperatures since then (including the latest decade) using only the sunspot numbers.

          • Nick Cox says:

            Thanks for the reply. I don’t doubt that responses to mitigate climate change are very, very expensive. I read your claim as being by implication about research funding. What else does “attempts to prove” mean?

          • PI says:


            One can construct infinitely many physically ridiculous models that validate well but make nonsensical long range predictions, especially if you choose a relatively short and quasi-linear period for validation. Yours is among them. It is simply wrong and self-inconsistent, from a physical standpoint. Validation is necessary but not sufficient for prediction. Furthermore, your validation procedure is illegitimate because you, circularly, use SSTs as covariates for temperature “prediction”.

            In short, both your climate physics and your statistics are wrong. That’s pretty much all I have to say on this subject, other than you should spend some time reading the scientific literature about how to construct and fit a physical climate model. (And also reflect on the question of why more physical models also fit the data and validate well on withheld data, but produce very different predictions.)

  6. Pete Dunkelberg says:

    Shoot out at the Santa Fe Corral:

    Check confirmed speakers and also the schedule. Muller, Curry and others even more infamous will be there.

  7. Michael Margolis says:

    If we want to assess the statistical significance of a claim like “warming stopped” we shouldn’t ask whether there was significant warming in some recent past. We should ask whether there is significant difference between the recent observed temperature record and that predicted by previous trend. That is, test the null hypothesis that the old trend continued. It seems to me that is the only foundation acceptable for claims of the sort Curry made, at least among classical significance tests.

  8. Eli Rabett says:

    One of the issues here is that the papers have been submitted to journals (JGR?). That means that all of the authors agree with all of the claims in the papers. Maybe not in this case, but were Eli an editor, he would surely try and find out which pig is in which poke.

  9. @Nick Cox’s comment above (which for some reason has no “reply to this comment” link for me:

    I think it’s fair to say that the money spent in dollars is not purely in the form of grants to scientists. If we increase taxes on carbon emissions then we increase the costs of goods with the goal of decreasing consumption of carbon, if it were true that carbon emissions had no effect on global climate, this would be a significant source of waste.

    That being said, I agree that a purely statistical model averaging various other peoples models is not a physically based model. However, it might very well give us some insight into the uncertainties pertaining to the original models. It’s quite plausible to me that depending on what physical effects you choose to include in your model, and how you choose to model those effects you could get different signs for future trends in temperature in a totally valid and non-intentional way. Chaotic dynamical systems have a tendency to go off in weird directions.

    • Nick Cox says:

      A minor detail first: I don’t know who “we” is precisely. Contributors to this blog often seem to assume that “we” means the people of the United States. I don’t deny anyone their right to write from their personal perspective, but I am not in that “we” == “USA” and I suggest simply that some threads would be clearer if more perspectives were made explicit. So, what else is new?

      That said, your first point is a good one. But it’s the same issue: What goes into the accounting? Higher prices through taxes on certain goods mean income directly or indirectly to some in linked industries, the taxes can be used for various purposes, and the money feeds into the economy (national and global) through various feedback loops. So higher prices are just not costs in a pejorative sense.

      Your second point is not aimed at my comment but I think I agree.

    • Nick Cox says:

      Also, the claim by Pangborn was that the dollars were being spent in an “attempt to prove”. Higher taxes are not part of any such attempt, are they? I took the reference to imply the research being done. Again, if Pangborn says one thing but the money cited is for another thing, then that is par for the course for his confused and contorted argument.

  10. Phil says:

    The business about whether temperatures have stabilized over the past ten years reminds me of an issue that came up earlier in my professional career, which is the dose-response relationship for low levels of radiation exposure. In both cases there was a standard model (called Linear No-Threshold) that predicted an upward slope, and a counterproposal (Threshold) that predicted a slope of zero. If you look at data that cover a small range of dose, you can easily be in the situation that the data do not deviate “significantly” from the model. That appears to be the same situation with the past ten years of temperature data. If you just look at the past ten years, the over-land temperature data are consistent with no change…and are also consistent with the expected increase. (Take a look at for example).

    We know a lot about the physics of the earth-atmosphere system — certainly we understand it much better than we understand the physiology of low-dose radiation — so basing assessments of anything on very short segments of a time series seems stupid to me. The earth is somewhat warmer now than it was 50 years ago, and will be much warmer in 50 years than it is now. It would be very useful to understand the rates of increase on those timescales. It is not at all useful to understand it on a timescale of 10 years.

    As others on this comment string have suggested implicitly, looking at just the data from meteorological stations is not ideal from the standpoint of addressing energy balance. A major complication is the role of the oceans: the amount of heat exchanged between the atmosphere and the oceans varies by year (due mostly to changes in ocean currents), so the air can be cooler this year even if the oceans are warmer (or vice versa).

  11. Dan Pangburn says:

    You need to get out more.

    What part of “…calculates the average global temperatures (agt) since 1895 with 88.4% accuracy (87.9% if CO2 is assumed to have no influence).” and “…accurately predicted the average global temperatures since then [1990] (including the latest decade) using only the sunspot numbers.” did you not grasp?

    • PI says:

      I grasped it completely. What part of “hindcast and short term validation skill do not imply predictive validity” did you not grasp? For that matter, what part of “physically realistic models also have hindcast and short term validation skill, but entirely different predictions” did you not grasp?

      • Dan Pangburn says:

        Your use of the word ‘hindcast’ (there is/was none) demonstrates that you still don’t get it.
        “None are so blind as those who will not see”

        • PI says:

          Add “misunderstanding of the word ‘hindcast'” to your list of mistakes, then. A model fit to past data is a hindcast, and so is validation by predicting (withheld) past data.