UAH Global Temperature Update for Nov. 2011: +0.12 deg. C

December 15th, 2011 by Roy W. Spencer, Ph. D.

The global average lower tropospheric temperature anomaly for November, 2011 remained about the same as last month, at +0.12 deg. C (click on the image for the full-size version):

The 3rd order polynomial fit to the data (courtesy of Excel) is for entertainment purposes only, and should not be construed as having any predictive value whatsoever.

Here are this year’s monthly stats:

2011 1 -0.010 -0.055 0.036 -0.372
2011 2 -0.020 -0.042 0.002 -0.348
2011 3 -0.101 -0.073 -0.128 -0.342
2011 4 +0.117 +0.195 +0.039 -0.229
2011 5 +0.133 +0.145 +0.121 -0.043
2011 6 +0.315 +0.379 +0.250 +0.233
2011 7 +0.374 +0.344 +0.404 +0.204
2011 8 +0.327 +0.321 +0.332 +0.155
2011 9 +0.289 +0.304 +0.274 +0.178
2011 10 +0.116 +0.169 +0.062 -0.054
2011 11 +0.123 +0.075 +0.170 +0.024

Since last month I predicted another temperature fall for November, which did not occur, I will admit that I should have followed my own advice: don’t try predicting the future based upon the daily temperature updates posted at the Discover website.

FYI, I’m making progress on the Version 6 of the global temperature dataset, and it looks like the new diurnal drift correction method is working.

[Reminder: Since AMSR-E failed in early October, there will be no more sea surface temperature updates from that instrument.]

48 Responses to “UAH Global Temperature Update for Nov. 2011: +0.12 deg. C”

Toggle Trackbacks

  1. jacozz says:

    What’s up with all this failures?
    It’s a bit scary. There will soon be a day when all of the sensors stop working.
    And then what?
    Any plans for the future in the loop?

    • It’s true that the “golden age” of state-of-the-art Earth-observation science missions is gradually ending. There is currently a real problem that no government agency has been specifically funded to carry on climate observations into the future….

      NOAA has been charged with doing it, but with no additional funds to develop replacements for all of the NASA instruments now flying. So, they will have to use whatever weather satellite assets they have. As far as AMSR-E goes, it will be replaced with AMSR2 on a Japanese mission early next year.

      As of now, NASA has no plans to fund us to participate in that mission.

      • David Appell says:

        No funding for future observations, but at least the role of government is getting minimized.

        • Scott says:


          There’s a difference between minimizing government and completely eliminating its presence. I’ve seen no one here or elsewhere asking for a complete erasure of everything the government does (though I’m sure you can find some anarchists who do), including in the area of science. There are some areas that are basically a taxpayer’s responsibility to support via government because they won’t get accomplished any other way. Public roads are one. Climate satellites might well be another.

          Quote mining to get a few sentences for your blog leaves you making a red herring of Roy’s stance.

          By the way, going back to our conversation two months ago concerning your oddly-chosen metric of year-to-date sea ice extent, I was shocked that your metric didn’t turn against you when I predicted it would…but the reason wasn’t a lack of ice increase, it was the failure of ASMR-E, LOL. Using CT’s area metric however, you should note that the year-to-date average sea ice area is currently 32,639 km^2 higher than 2007’s. Given that we’re still more than a quarter million km^2 higher than 2007’s area, expect to see that number grow (I’m guessing it will finish the year with an average 40-45k km^2 above). Guess you’ll have to look for another magic metric instead.


          • David Appell says:

            My point was that people who advocate for minimal government, and private charity for the problems of the poor, seem to have no problem spending government money on *their* problems. Climate satellites or public roads aren’t the only needs that are not provided by the so-called “free market” — there is also health care for the uninsurable, for example, and a host of externalities that are currently handled by socialism. What’s good for the goose is good for the gander.

            Scott, you *COMPLETELY* missed the point of my 9/28 comment on YTD Arctic sea ice extent. It was simply a response to a cherry-picker showing him that two can play at that game, if that’s the game he thinks is valid (which it obviously isn’t for discussions of climate, of course — and I chose the example I did exactly because it makes that clear).

          • Scott says:

            And how does that negate you making a straw man of Roy’s argue. He’s for minimalizing, but you’re still acting as if he’s for abolishing. I see this approach all the time. I had a conversation last Wednesday with some very hardcore conservatives…about as far right as you’ll see. Yet they were the ones to point out that the free market/pure capitalism will not work. People arguing against their viewpoints never seem to realize that they acknowledge the limitations of capitalism. Why? Probably because it’s much easier to argue a straw man argument than they’re real argument.

            And if the whole point of the ice argument was to show cherry picking, you did a good job of it. However, you didn’t indicate that that was your point until several comments into your argument, so I honestly don’t think that was your point until it was pointed out how obviously foolish your example was, twice. Obviously I can’t know motives so that’s purely speculation on my part.


          • David Appell says:

            Scott: Thanks for acknowledging that I did a good job of demonstrating the foibles of cherry-picking.

            I gather, then, that you also recognize the efficacy of not giving the game away too quickly. Obviously, that, too, was part of the strategy.

            Yes, Roy is for minimizing — but only (it seems) for what he does not consider important. The rest (like, he wrote, “poor people,” can rely on charity.) But that is a value judgement, and values vary. Note that he does not consider it unimportant that *his* particular project be cut…. That is the crux of the matter. I wonder how he feels about having all health benefits of UAH/state employees being cut and having them contend in the vaunted “free market,” which he otherwise champions. Perhaps he will tell us.

            Scott, if you really only believe that hardcore conservatives are the only ones who see the limitations of the so-called “free market,” then you have not been paying any attention at all and you seriously need to expand your reading circle.

          • Scott says:

            So David, you’re harping on Roy because he had different value judgments than you do? Well, as long as you admit that you’re talking about value judgments, I can leave that at that. Although I find it interesting that you’d expect people to want their own projects to be cut…or at least you find it odd that Roy doesn’t want his project to be cut.

            And you appear to be an expert at twisting what I say to build a straw man and knock it down. I never said that only hardcore conservatives see the limitations of a free market. I said:

            People arguing against their viewpoints never seem to realize that they acknowledge the limitations of capitalism.

            Completely different from what you imply. You then tell me:

            you have not been paying any attention at all and you seriously need to expand your reading circle.

            But you’re the one not paying attention. And I don’t know much about your reading circle, but you need to improve your reading skills and/or comprehension. The reason? I never said anything about reading anything and made it clear that I was referencing a conversation with hardcore conservatives. Again, you’re twisting what I say so you can knock it down. Why? Well, I can only guess, but I’m guessing it’s similar to the speculation I gave earlier concerning why people arguing against the conservatives won’t admit that they acknowledge the limitations of capitalism:

            Probably because it’s much easier to argue a straw man argument than they’re real argument.

            Thank you for demonstrating exactly how this works in a response to the very post where I speculated it.


  2. JohnD says:

    The weak correlation between the monthly UAH number and the Discover websiste daily data is distressing.

  3. I must confess to being totally baffled by the latest number. If one just compares November 2007 to November 2011 on ch05, it is clear that 2011 is a fair bit lower than 2007. Comparing the RSS for these two months, it went DOWN from 0.126 in 2007 to 0.033 in 2011. However with UAH, it went UP from 0.06 to 0.12. That is a difference of about 0.15. To put this into some sort of perspective, the difference in the Hadcrut3 data set between the 2nd warmest (0.482) and 12th warmest (0.352) is only 0.13.

  4. Walter Dnes says:

    One interesting item… if you plot the 12-month running means for UAH versus RSS since they started in the late 1970’s, the RSS 12-month running mean has *ALWAYS* *ALWAYS* *ALWAYS* been warmer than the corresponding UAH 12-month running mean. But, after 33 years, it looks like that streak is about to be broken next month. Not only the first annual temps RSS-cooler-than-UAH, but also the first ever 12-month running mean RSS-cooler-than-UAH. For the last 8 months (April-to-November) RSS has been cooler than UAH (or UAH has been warmer than RSS).
    The *ONLY* previous times there have been 2 or more consecutive RSS-cooler-than-UAH months was…
    July-September 1980 (3 consecutive months)
    March-June 1985 (4 consecutive months)

    This might conceivably be for real, But I suggest looking into it more deeply. Something seems amiss here.

  5. Doug says:

    Doesn’t make any sense to me looking at the discover data. The 2011 data is lower than 2007, 2008, and 2010, yet the UAH number doesn’t reflect that. What’s going on?

  6. An Inquirer says:

    Werner, good points. Those are valid questions. Perhaps part of the explanation is that the Channel 5 #s on Discover are not the tropospheric #s (14000 ft / 600 mb) that we are believed them to be. The Channel 5 chart reflects temperatures from a wide variety of altitude levels.
    However, I am not convinced that this feature is the explanation for all or even most of the puzzle in UAH temperatures. In the recent past, UAH helped RSS realize that the latter had been understating temperatures. Perhaps now for some reason, UAH is overstating temperatures. I do not know, but I am not convinced that recent UAH temperatures are reliable.

  7. MikeN says:

    Does NASA have the same problems you guys do? They haven’t reported November yet either.

  8. Jim says:

    Looking at the Discover web site, it’s evident that the drop in October arrested in November. However, at least to this point, it has resumed in December. I would be surprised if the current rate of drop continues, but it does look like December will be down.

  9. Ray says:


    Actually, I don’t think that the November UAH figure is strange, compared to previous AQUA CH5 & UAH values.
    Using the figures from 2002 to 2010 (limited by AQUA CH5 data), I estimated that the most likely November UAH figure was 0.101c.
    O.K. that is a bit lower than the actual figure, but it’s well within the range from past years.
    Actually, if anything, it is the 2007 UAH figure which looks odd, as it is about 0.107c lower than it should be based on other months.
    Incidentally, the latest AQUA CH5 temperature is now the lowest December figure since 2002, so it may be that the drop in temperature which Dr. Spencer anticipated in November, is happening in December.

  10. Ray says:

    Jim, sorry I hadn’t seen your pos when I posted mine.
    I agree that it seems unlikely that the latest fall in AQUA CH5 will continue, but if it does!!!

  11. Thank you for your comments Ray and An Inquirer.

    Ray, you say: “Incidentally, the latest AQUA CH5 temperature is now the lowest December figure since 2002, so it may be that the drop in temperature which Dr. Spencer anticipated in November, is happening in December.”

    The first 10 days of November also were the lowest since 2002. The La Nina did not change much, but the number of sunspots sure sky rocketed recently. Perhaps that is why the November trend did not continue.

    I think it is also worth noting that of the 5 data sets listed at the following, only UAH shows a positive trend for the past 10 years:

    • david says:

      Werner, you can’t draw a linear regression line through cyclical/seasonal data. You must adjust for seasonality first. Sorry, but this is a statistical/regression nightmare…

      • Bu J says:

        Luckily there is someone with statistical training who just did the analysis for us and published it on the web (and in a journal article)

        The most relevant graph to this discussion is the rate vs. start year for the different data sets. It clearly shows what most people who state linear regression coefficients forget about, the error bars (confidence intervals). Note that the longer the time series, the tighter the confidence interval.

        Werner should have stated that the confidence interval of 10 year regression coefficients of the raw data span an interval that contains zero and hence should not be used to determine if warming or cooling is indicated by simple linear regression.

        Computing these intervals is straightforward: even wikipedia has the formulas

        Quoting the oracle:

        Confidence intervals

        The formulas given in the previous section allow one to calculate the point estimates of ? and ? — that is, the coefficients of the regression line for the given set of data. However, those formulas do not tell us how precise the estimates are. That is, how much the estimators can deviate from the “true” values of ? and ?. The latter question is answered by the confidence intervals for the regression coefficients . . .

  12. Doug says:

    December 15, 2011 at 5:04 PM

    Doesn’t make any sense to me looking at the discover data. The 2011 data is lower than 2007, 2008, and 2010, yet the UAH number doesn’t reflect that. What’s going on?

    I checked out the linked Discover site. What I found was that the lowest altitude shown after 2008 is channel 5, said to be centered at or representing the 600 mb level. That is not lower troposphere, but middle troposphere – which has had less warming trend so far than lower troposphere. Also, that is only one satellite sensor, while lower troposphere trend indices appear to me based on multiple satellite sensors.

  13. Ray says:

    “Doesn’t make any sense to me looking at the discover data. The 2011 data is lower than 2007, 2008, and 2010, yet the UAH number doesn’t reflect that. What’s going on?”
    Sorry, the “reply to” facility doesn’t work for me.
    I don’t really understand what you are saying.
    The AQUA CH5 anomaly for November was the lowest since 2002.
    It is lower than 2008 and 2010 and so is the UAH anomaly.
    As I say above, if anything, it is the 2007 UAH figure which looks too low and last month’s figure is about where it should be.

  14. Copernicus64 says:

    Dear Spencer, about the data on discover, I don’t see big discrepancies with UAH…November since 2002 up to now was always over the 1980-2010 average and only november 2007 was colder than 2011 (+0.06 C)…probably, I suppose, because of the poor areas covered by marine ice in the arctic…

  15. James says:

    Does anyone know or can someone point me in the direction of an error range for the global anomaly estimate?

    i.e. the anomaly is 0.12 +/- ?? with ?? confidence



  16. nigel raymond says:

    To Bu J and to James

    “…an error range for the global anomaly estimate?”

    One of the embedded references of Bu J purports to
    have these or the equivalent.

    However, the analysis is obviously done with
    a “canned computer programme” (unspecified) and there
    does not seem to be any justification for whatever
    method was used.

    A calculation of “errors” relies on assumptions about
    the statistical distribution of residuals, and these
    assumptions have to be justified on theoretical grounds
    BEFORE starting. So far as I can see, there is no
    discussion at all of the nature of the “errors”.

    A bit of a waste of time, therefore.

    In any case, thirty years is but a drop in the ocean
    of time(to mix metaphors), for any natural phenomenon
    (e.g. climate) which involves the ocean system.

    When people are eagerly discussing monthly changes
    of some average expressed in hundredths of a degree C
    it seems more like a pride thing than a practical
    approach to life. The fact is that the Communist Party
    of China has a deliberate policy of doubling and redoubling
    their (1.3 billion people)use of fossil fuel, and there is nothing whatsoever that we in the West can do about it. Similar considerations apply to India (1.0 billion), Indonesia (200 million), etc. Perhaps in fifty years their
    descendants will say “Whoops, sorry about the mess we added
    to!” and perhaps they will say “Global warming? Wasn’t that some historical hysteria, designed to deny us our rightful place in the world?”

    • Bu J says:

      nigel raymond,

      You said

      “One of the embedded references of Bu J purports to
      have these or the equivalent.

      However, the analysis is obviously done with
      a “canned computer programme” (unspecified) and there
      does not seem to be any justification for whatever
      method was used.”

      The authors clearly state they use programs written in the statistical programming language of “R” and they even make the code freely available on the web for all to see. The methodology is clearly explained in a peer-reviewed article.

      So your comments are patently false, untrue, nonsense.

      There are well described techniques to test the statistical significance of linear regression coefficients (even if there is no error in the data). These techniques could care less about what the data set is, or how much coal the Chinese plan to burn.

      You can’t change the tenets of statistics based on geo-political ideology, liberal or conservative.

  17. nigel raymond says:

    I should have made it clearer that “errors”
    here mean specification errors of the model,
    and not instrumental measurement errors such
    as Gauss was concerned with.

    Further to China: consumption as a percentage
    of measured GDP is now just 36%. At some point
    in the next few years, they will have to stop
    making their useless physical capital investments.
    At that time their use of energy may well drop
    for a while.

  18. “Bu J says:
    December 16, 2011 at 8:38 AM

    Luckily there is someone with statistical training who just did the analysis for us and published it on the web (and in a journal article)

    See a rebuttal to this at:

  19. nigel raymond says:

    Bu J says:

    “The authors clearly state they use programmes
    written in the statistical programming language of “R”…”

    I am looking at,

    Grant Foster and Stephen Rahmstorf
    Global temperature evolution 1979-2010
    Environmental Research Letters 6 (2011),

    which I assume is the article being talked about.

    I see NO such statement. Perhaps you would be kind
    enough to direct my old eyes to the right paragraph
    in it? If they have said it somewhere else, they should
    have formally referenced it; but I do not see their
    own names in the References.

    “There are well described techniques to test the
    significance of linear regression coefficients…
    these techniques could care less (sic; I assume you
    mean could NOT care less) about what the data set is,
    or how much coal the Chinese plan to burn.”

    The techniques care very much what [the characteristics
    of] the data set is, once divided into model+residual.
    Elsevier Scientific Publishers has a “Reviewers’ quick guide to common statistical errors in scientific papers”
    from which I quote the following, as a type of serious error which they think vitiates papers:

    “…using parametric methods (e.g. t-test, ANOVA, or
    linear regression [which includes multiple, necessarily]
    when the outcome or residuals have not been VERIFIED
    as normally distributed…[my emphasis].”

    The paper in question does not seem to VERIFY this. The
    only discussion of residuals distribution that I see is a bald statement that all climate data sets when modelled
    are ARMA(1,1)in the residuals – and a correction has
    been applied for this putative state of affairs. This counts as some discussion of residuals, I agree, but does not seem to meet the, reasonable, standards of Elsevier.

    I agree the techniques do not care about the plans of the Chinese. That was hardly my contention.

    Incidentally, the paper describes their analysis as showing “strong influences”. “Influences” is just another word for “causes”. Regression analysis can hardly show cause and effect, by itself.

  20. nigel raymond says:

    On looking a little closer at the paper in question, I find
    the following in it:

    “Multiple regression can lead to misleading results
    when the independent variables are nearly collinear”


    “Hence we computed the correlation between the independent variables used during the time span under study. The strongest correlation was betwen TSI and the linear time trend with correlation coefficient -0.47…Hence the independent variables are certainly [sic] not nearly collinear.”

    It is not true that where there are more than two explanatory variables one can measure the degree of multicollinearity by looking at the pairwise correlation coefficients among the explanatory variables. There are only ad hoc criteria for this situation. As one textbook has it “…detecting serious multicollinearity is more art than science.” In other words,it isn’t science at all to use the word “certainly” in this context.

  21. Bu J says:

    Dear nigel,

    For your old eyes:

    “This post is only to provide access to the data and the code (all computations were done using R) used in Foster & Rahmstorf 2011 (blogged about here).”

    This is on the website I linked to originally, not in the paper. Your old eyes found the paper just fine, they couldn’t find the top post on the website?

    Let me remind your old eyes what you said that
    I was objecting to:

    “However, the analysis is obviously done with
    a “canned computer programme” (unspecified) and there
    does not seem to be any justification for whatever
    method was used.”

    So, you could have the decency to say you were actually wrong. It was not a canned computer program, and they provided plenty of justification for what they did. Feel free [sic] to blame your eyes if you don’t want to blame your brain. But [sic] be a decent human and say you were wrong on this one.

    Also, my original post was in response to Werner and david regarding short time linear regression estimates of global temperature indices. My point was that these are point estimates and one *should* include confidence intervals. Werner did not, Tamino did (although for two different things). David correctly pointed out that one should try to factor out known nonlinear effects. I agree. Werner did not do this, Tamino did. But notice I never said I think Tamino’s work is perfect. Were you assuming that because I was objecting to Werner’s inappropriate use of statistics, I must be an enemy? Did you bother to ask?

    In my opinion, the case against the IPCC mantra would be much more credible if people would stick to valid arguments (e.g. how much coal China is going to burn), and stop defending the use of bogus, cherry picked statistics claiming to “prove” there is no global warming. (I note here that Werner only cited bogus statistic, and did not go the usual extra mile)

    It’s like you go to buy a new car and the salesperson is wowing you with details on performance, fuel efficiency, reliability, comfort, etc. And then the salesman tells you that the hood (bonnet?) is configured to block the mind control rays from the aliens parked behind the moon. You can’t believe the valid arguments if the nonsense is offered with the same level of significance.

    Now you can try to say, “Well Tamino is using bogus statistics too”, but you aren’t going to convince anybody that knows anything that what Tamino is doing is at the same level of abuse as Werner’s ten year linear regression of raw data. If you know enough to attack Tamino, you know enough to say that Werner is not helping things.

    I guess what I am saying is that if you want to beat Tamino, you need to present arguments at his level of rigor. But clearly I am assuming a bias toward arguments that have some mathematical rigor. Otherwise, just click out of here and head for Mr. Watts.

  22. nigel raymond says:

    Dear Bu J

    Originally, I followed your link that you posted on December 16 and read the piece by ‘Tamino’ down to the beginning of the comments. I THEN clicked to the paper.

    I have looked again, using your link of Dec 16, and in the piece by ‘Tamino’ I can see NO such statement as you aver.
    In any case, if it isn’t in the paper, or referenced in the paper, it doesn’t exist by the conventions of science.
    After all, you wrote “The AUTHORS state…” not “some bloke somewhere says about them…”

    According to my employees I am not even a human being, let alone a decent one.

    It is not polite for the authors not to credit R Stats.
    It is normal practice to mention the programme that has
    done all the real work.

    Using the routines of R Stats is, in fact, pretty much
    what I meant by using a ‘canned’ programme, in the sense
    3a of the American Heritage Dictionary:

    Used repeatedly with litle or no change.

    I have no idea who Tamino is, except that anyone who
    adopts the moniker of the hero from ‘The Magic Flute’
    has a good opinion of himself.

    Your excited tone, and talk of “beat Tamino”, tends to
    confirm that the reality is being ignored as usual –
    namely that we in the West can do nothing whatsoever
    to stop the inevitable – whatever that inevitable turns
    out to be.

    • Bu J says:


      I gave you a direct link to the web page from which I quoted. I don’t know how I can make it easier.

      Here it is again. Use that little mouse thing and click on it

      How am I supposed to reply to somebody who thinks if they can’t find information, it doesn’t exist?

      Did you notice anything about that link? The first part is


      Wait, the http thing is distracting. How about


      Wait, too long. How about


      nigel, do you know who tamino is now?

      It is probably too hard for you still. tamino is the web name of the guy who writes the page on which the graphs and R code are posted. He is also one of the authors of the paper you refer too. I’ll let you figure out which one since you seem to be such a whiz at sleuthing out information on the web in your spare time.

      I really hope you get well soon.

  23. nigel raymond says:

    A clarification:

    When I write “I do not know who ‘Tamino’ is”,
    I am not just saying that I do not know the
    real name or affiliation of him (or her).
    I am saying I do not follow “the top site on
    the web”. I generally have better things to do,
    but am laid-up at present. Not for much longer!

  24. Right now it is wait and see, to see what might be in store later this decade.

    Solar activity has been moderate to high since Aug. The question is ,is this the maximum for this cycle, or is there more to come.

    My scale for solar activity versus atmospheirc circulation AO,NAO for example, and for geological activity and temperatures over the long run is as follows:

    Solar flux 150 or higher +AO, low geological activity,warm temp

    Solar flux 90 or less,likey a -AO,NAO,high gelogical activity if spurts of activity occur within the low solar activity, and colder temperatures in the long run.

  25. chris brandow says:

    Instead of the admittedly meaningless 3rd order fit, why not add in a 5 or 10 year moving average to complement the 13 month average that nicely removes monthly variations? 5 or 10 year moving average would decently minimize annual variations.

  26. antbev says:

    You are getting into the general area of smoothing signals.

    Moving averages are smoothers.

    a 5 year moving average would remove any five year cycle.

    a 10 year moving average would remove any ten year cycle.

    But in my opinion 30 years is far too short a period for

    any sort of time series analysis of global earth science

    data. That is presumably why the 3rd order fit is called

    “for entertainment”.

  27. antbevca says:

    Of course the subject is much more complicated
    than my couple of lines indicates. But I am sure
    you know that.

    One problem with moving averages is that an extreme
    in the data is converted into a bump or crater in
    the average – the Slutzky-Yule effect. A few chance
    jumps or falls in the data can look awfully like a
    meaningful cycle, in the average.

  28. antbevca says:

    I rather get the impression that nigel raymond is
    not interested in web wars!

    Two minutes on the web for me turned up the following
    amusing exchange :

    “Who the f**k is ‘Tamino’?’

    “Grant Foster!”

    “Who the f**k is ‘Grant Foster’?”

    IF, in fact, G. Foster is puffing his own work by
    pretending to be (deliberately, or by not making the connection clear enough) an independent commentator,
    then IMO that is good enough reason never to speak
    of him or his work, or his friends’ work, ever again.

  29. antbev says:

    Re-reading my comment it strikes me that it probably comes over as being a bit highanded (and maybe it is).

    But there is a HUGE problem, in the whole system called
    Academia, which we all know sub-consciously, but usually
    try not to think about, because the implications are

    I quote some figures from

    The Growth and Number of Journals
    Michael Mabe
    Serials, Vol 16, No 2, July 2003:

    “Surveys of the number of researchers in the world conducted
    by UNESCO in 2001 estimate that there are about five to six million…they [produce and publish] one million papers
    a year.

    “The number of active journals has been [steadily]
    doubling every 20 years…”

    Allowing for the passage of time since 2001, the “academic
    conversation” now is actually a “roar” of some ten million
    people, who are writing some one and a half million papers
    a year of formal claims (say 4,000 million words), and some even more astronomic number of pages of informal communications. Even if you are a specialist in a tiny
    field, you can’t possibly hold the necessary knowledge
    in your mind.

    There has to be some means of sifting what comes to one.

    The VERY LEAST one can do is to ignore everything which
    has even the FAINTEST taint of (probable*)intellectual shenanigans.

    * Need to have real evidence, of course.

  30. david says:

    GISS V3 temps posted for Nov: 0.48. Much lower than my expected value of 0.60 (which was simply based on averaging the last 10yrs worth of Nov data). Note that V3 of GISS revises temps post-1975 upwards by between 0.00 and 0.02C, etc. That’s also a way of making AGW… more here:

  31. Bu J says:
    December 18, 2011 at 11:21 PM

    “(I note here that Werner only cited bogus statistic, and did not go the usual extra mile)”

    Fair enough. But I did not see any acknowledgement that you saw my earlier post at
    Werner Brozek says:
    December 18, 2011 at 12:19 AM
    See a rebuttal to this at:

    This rebuttal is by someone whose statistical knowledge far surpasses mine.

    Phil Jones, July 5, 2005:
    “The scientific community would come down on me in no uncertain terms if I said the world had cooled from 1998. Okay it has but it is only seven years of data and it isn’t statistically significant.”

    So to keep up with Jones, I will say that according to some data sets, there has been cooling for over 14 years but it is not statistically significant.

  32. antbev says:

    Hang on a minute!!

    Aren’t we all (starting with G&F)simply committing the ecological fallacy here?

    The dependent variable is an aggregate of
    individual temperatures, and the variability
    of these individuals is enormously greater
    than the variability of the average (in Lindzen’s
    evidence before Congress on Nov 17, 2010, there
    is a rather startling graphical illustration).

    Hence, we are set up for the fallacy.

  33. antbev says:

    “(starting with G&F)…” is a typo (aka blunder)
    for (“starting with F&R)…”

  34. J. Seifert says:

    Hi Dr. Spencer,

    what I would prefer is 1.)

    a 4-year mean (which compensates Ninas with Ninos), this would be a meaningful graph compensating opposite short term events
    and 2.) a 10 year mean —this is the decadal trend graph what we need especially for the coming
    years to demonstrate the flat temp-plateau on which we are
    now for the coming 32 years…..also since AR3 and AR4 refer
    to decadal trends in their forecasts and you could show the
    real decadal trend compared to the IPCC SRES AR3/4-trends….it would also show the tipping point of 2001…

    3. the 13 months line does not explain trends sufficiently, to much up and down with short term processes….., also the binomial line does not provide much insight…..

    I wish you could give it a thought…
    Best wishes for all festivities…

  35. Dikran Marsupial says:

    Dear Dr Spencer,

    please can you remove polynomial fit from the plot. I know that you explicitly state that the fit is for “entertainment purposes only”, however that doesn’t mean that others will not use your plot without this caveat and claim that the fit has genuine predictive power, for example, see the end of this blog article

    There will be others who read that article and use it as evidence that temperatures will fall.

    The signal to noise ratio of the climate debate is already low enough and posting statistical models of no predictive value for entertainment purposes merely contributes to the noise and detracts from the value of your scientiic contributions. What entertainment it provides simply isn’t worth it IMHO.

  36. hey huff what happened to all my 50 fans buy wow gold

Leave a Reply