A challenge to Dr. Roy Spencer

Show me one peer-reviewed paper that has ruled out leprechauns as the cause of most of the recent warming in the thermometer record.

83 responses to “A challenge to Dr. Roy Spencer

  1. Does Spencer, mean ruled out as in 100% ruled out or just very likely ruled out like greater than 90%?

    It would be disingenuous to ask the former. So let’s be generous to Roy and assume the latter.

    The paper Roy is seeking is SPM AR4.


  2. http://www.warmdebate.com/ozone-layer-its-recovery-cfcs

    This would seem more apt:

    I’ve been picking up a lot of chatter in the last few days about the ’settled science’ of ozone layer depletion. What most people don’t realize is that the vast majority of published research on the topic simply assumes that ozone depletion is manmade. It in no way “proves” it.

    If the science really is that settled, then this challenge should be easy:

    Show me one peer-reviewed paper that has ruled out natural, internal cycles as the cause of most of the decline in the ozone layer.

    Studies that have suggested that an increase in the total output of the sun cannot be blamed, do not count…the sun is an external driver. I’m talking about natural, internal variability.

    The fact is that the ‘null hypothesis’ of ozone layer depletion has never been rejected: That natural climate variability can explain everything we see in the climate system.

    Show us your stuff, Dr Roy Spencer.

    • Is “natural variability” anything more than a more sophisticated sounding way of saying “I have no idea what exactly I’m proposing as the alternative hypothesis?” I mean, I know that complex systems can have quasi-cycles with no external drivers, e.g. El Nino, but it seems to me that when people are talking about El Nino they call it “El Nino,” and when they have no specific idea of what they’re talking about they call it “natural variability.”

      With El Nino, we have a good idea of what the feedback effects that drive it are, and we can produce physical models that exhibit similar variation. So I guess my challenge would be, do you have a mechanism for your “natural, internal variability?” Can you show what kind of variability we should expect from your mechanism? If not, how could anything be tested against your hypothesis? Of course there are no peer-reviewed papers rejecting an idea that was never formed with any specificity.

      • Timothy Chase

        Eric L. wrote:

        Is “natural variability” anything more than a more sophisticated sounding way of saying “I have no idea what exactly I’m proposing as the alternative hypothesis?” …. Of course there are no peer-reviewed papers rejecting an idea that was never formed with any specificity.

        I believe this is a winner!

        Spencer goes on to call this non-specific causal hypothesis the null hypothesis:

        The fact is that the’ null hypothesis’ of global warming has never been rejected: That natural climate variability can explain everything we see in the climate system.

        See Kevin McKinney’s link below. Roughly equivilent to arguing that the theory of divine intervention is the null hypothesis for the theory of evolution. And yes, this is the guy who decided that intelligent design is science when it was still being called scientific creationism.

      • Thread is over, we have a winner.

  3. Which type of leprechaun are you referring to?

  4. Hey Tamino,
    I have a statistical question relating to an upcoming Skepticalscience post. I plotted up the following:

    And I am trying to decide whether similar to your extreme heat post that the distribution will not be normal and therefore will require me to use extreme value theory (or something of the sort) to determine the relative probability of recent events.

    My QQ plot is here:

    My Skewness is 0.32
    My Kurtosis is 0.04

    Any ideas?

    [Response: Visually, the Q-Q plot looks normal. But of course you want quantitative evidence.

    You can test whether the observed data follows any given theoretical distribution with the Kolmogorov-Smirnov test. Probably even better, a test tailored to comparing observed data with the normal distribution is the Shapiro-Wilk test. Both tests are incorporated in R.]

  5. Straw man fallacy rules!

  6. OK, I’m sorry, but this is precisely why I sometimes wonder if Roy really understands the scientific method. First, WTF is the fricking operational definition of a Leprechuan? What is the mechanism by which it would cause warming? Until we have some idea of these things, the question is meaningless. I would note, however, that it is just another instance of Roy’s usual schtick of “anything but CO2″…in fact, it’s an improvement as he is at least positing a named cause. Rule # 1 in science, Roy: We don’t try to explain the unknown in terms of the unknown!

    Second, the temperature record makes a very strong case for anthropogenic causation all on its own, BUT in science, we do not consider any single piece of evidence in isolation. We don’t look for an explanation for global temperatures and a separate one for polar amplification and another one for stratospheric cooling and another one for expanding storm tracks. Ever hear of Occam’s Razor, Roy?

    Dare I say that this is precisely the sort of misunderstanding of the scientific method that I would expect from a man who thinks Intelligent Design is a scientific theory?

  7. TM,

    Are these Spencer’s words or yours? Does Spencer not believe in CFC ozone depletion?

  8. Wait til you see my next posting at WUWT. I compute a 100-year rolling average of the ALO (Atlantic Leprechaun Oscillation) and correlate it with a 100-year rolling average of sunspots. The r2 value of this correlation (0.833) shows that solar cycles drive the ALO, probably because an increase in the UV component of TSI interacts with GCRs to cause leprechauns to oscillate wildly between the northern and southern hemispheres. (Technically, there’s a lag of -6 months in the ALO data, meaning that it peaks before the solar does, but by eyeballing and common sense I’ve shifted that to a +6 months lag).

    Having done that, I look at histograms of the various global temperature measurements 1990-present (without rebaselining them, of course). Noting that the histogram of GISS is biased upward relative to UAH, I correct for this effect by subtracting monthly GISS measurements from monthly HADCRU measurements. The resulting data set shows a strong cooling trend.

    Returning to the smoothed ALO (Atlantic leprechaun oscillation) data, I calculate its anomaly relative to the 1814-1816 mean, and accumulate those anomalies over time. I then show that the accumulated ALO anomaly data can perfectly explain the cooling trend in my corrected surface temperature record.

    Ergo, solar cycles drive the ALO, and the ALO drives global cooling.

    I am hoping that some of the smart fellows over at WUWT can help me rework this into a manuscript to submit to E&E.

    [Response: Ladies and gentlemen, we have a winner.

    Seriously — I with I’d written that.]

    • I correct for this effect by subtracting monthly GISS measurements from monthly HADCRU measurements. The resulting data set shows a strong cooling trend.

      Er, I should have noted that this is based on the 1998-2010 data, of course. It’s very important for a proper analysis to start in 1998. Anything else would be cherry-picking.

    • But correlation is not causation, I’m often told.

      How do you know Atlantic Leprechaun Oscillation doesn’t drive sunspot numbers?

      • BTW, to save keyboard life, if you do post it at WUWT just call it ALO. Leave out the long verbage ;)

      • How do you know Atlantic Leprechaun Oscillation doesn’t drive sunspot numbers?

        Well, technically, the -6 month lag means that we should conclude that the ALO is driving sunspots. But that’s obviously unphysical. So I ignored that in my analysis.

    • Former Skeptic

      Ned, you funny bastard! I made the mistake of reading your analysis with my morning coffee. Send me a check for a new laptop computer, please!

    • Jeffrey Davis

      You must be right because you used so many words.

      • Jeffrey Davis:”You must be right because you used so many words.”

        Not convincing until I see lots of numbers and graphs. Inferno, King of the Blog Scientists, has now provided the definitive graph at denialdepot.

    • Thats a horrible straw man and tells more about you than R Spencer or any other climate skeptic and makes me not to want to read the blog you’re posting into.


      [Response: Duh. It’s Spencer’s challenge that’s the strawman, this is just a parody. As for no longer reading this blog, that’s up to you.]

    • Tamino writes:

      [Response: Ladies and gentlemen, we have a winner.
      Seriously — I wish I’d written that.]

      To be perfectly fair, you deserve the credit for pointing me in the right direction. My methods were based on previous work by some of the best skeptical scientists out there — guys like D’Aleo, Goodridge, Copeland, and of course the master, Watts himself. All guys who I discovered through your blog.

      My next goal is to read more of the work by Mr Goddard. I am very impressed by his work. My secret hope is that some day my ALO-sunspot-cooling hypothesis might find its proper place in the Palladium of True Skeptic Science alongside Mr Goddard’s work on CO2 snow and the triple-point of water.

  9. To what confidence interval? Leprechauns are tricky buggers you know, leaping all over the place.

    • Rob Honeycutt

      That’s due to the effect of galactic cosmic rays on leprechauns. Leprechauns are highly activated as we pass through spiral arms of the Milky Way.

      Come on, it’s obvious!

      • My friend Rob is of course correct. I’m glad to see that someone’s been keeping up with the literature.

    • Thanks to my innovative use of a 100-year rolling average, the correlation between these series is very high. If I use enough polynomial terms in my model, I can get really narrow confidence intervals. That’s why I’m pretty certain this will be publishable in E&E.

  10. O/T.

    Tamino, a long while back you posted on the time it would take for the J-D global temperature record (1998) to be beaten in all the popular surface records, including HadCRU. IIRC, you postulated 2015 (or 2014?). Do you have a link for that post (can’t remember the title, and so don’t know if it’s in the list of returned-from-the-dead Open Mind posts at skeptical science).

    Just curious considering the near miss we had with 2010, and the unlikelihood of 2011 being warmer. Is it worth a new post?

  11. I find Spencer’s challenge troubling. I want to believe he’s honest and competent but simply wrong in his belief that warming is being caused by some unknown natural mechanism relating to clouds.

    However a challenge of this type creates a lot of pressure that can only be released by loosening either the “honest” valve or “competent” valve.

  12. Steve Metzler

    Ah, I found the context for this:


    So tamino is turning this around and posing a similar, unfalsifiable, challenge to Spencer to illustrate how ridiculous Spencer sounds.

  13. Presumably, Leprechaun-induced warming is a sure thing when you complete the analysis, but vanishes completely from the data once you send it off for review.

    Sounds familiar, come to think of it.

    Anyway, for those who may be puzzling over this post a bit, this link should render it quite clear.

    By the way, Dr. S. now has the January UAH anomaly up; it’s at -.1. His take: “Global Temperatures in Freefall. . . although this, too, shall pass, when La Nina goes away.”

  14. Oops, make that .01C!

    Maybe that was a “Leprechaun zero.”

  15. sharper00, it’s fine to believe that Dr. Spencer is honest and competent (but simply wrong in his beliefs…). However a perusal of his work over the last 20 years might lead us to question those attributes (i.e. honesty/competence, which we may generally “lump” together since we can’t always be sure of the basis for repeated technical error[*]).

    Even if we were to assign to him the role of “devil’s advocate” the facts simply show that he gets it wrong repeatedly. It’s pretty generous to give such people the presumption of good faith….

    [*] e.g.litany of errors in UAH temperature analysis

  16. But Tamino, leprechauns do cause climate change. This correlation proves it:

    • Good one Zeke!

    • How many irish leprechaun pirates are there?

    • Actually leprechauns cause a cooling effect on the climate, with the Pot Of Gold Albedo Effect. I’ll let Dr. Inferno explain:

      “As far as modern science can ascertain, Leprechauns influence climate by modulating the effects of Galactic Cosmic Rays (GCRs). It seems that an increase in GCRs prompts an increase in clouds. In turn this leads to more rainbows, each of which according to legend prompts a Leprechaun to place an extra pot of gold at it’s end. The albedo increase due to all these new shiny pots of gold being left around induces a small cooling effect on the planet.”

      He also has a good graphic of all the imaginary climate forcings:

    • Hi Zeke, all this looks really interesting, have you planned to correct this for immigration/emigration? Also the location of Conan o’Brien should be checked for correlation. :-).

    • A one-to-one correlation between Irish pop and Irish leprachauns? Are you sure there aren’t any amplifying (or suppressing) feedbacks?

    • There is an obvious divergence circa 1955 and the leprechaun proxy appears to be overly smoothed prior to 1955. I suspect siting issues.

  17. Is he joking? In the comments he brings up the MWP not caused by SUVs, which I’ve never thought of as being anything other than moronic. Is that a sign that he’s just kidding?

    I don’t get why he’d invoke “mystery thing we’ve not noticed yet” as the most likely explanation when it would also have to explain why CO2 is NOT behaving as the physics would dictate. He’s writing as though it’s a simpler explanation when to me it seems a more complex one.

    • Pough,

      Re the SUVs, he has used that line several times on his blog and elsewhere. You know, Denialst Chum.

      Whether or not he is actually serious we’ll never know. One thing I do know, saying that in public is a sure way to undermine your credibility as a scientists, never mind a climate scientist.

  18. It astounds me that so many here have ignored the obvious negative feedbacks associated with leprechauns, most notably the high albedo of gold. Any leprechaunologist worth their salt will know the impact of gold filled pots located exactly where the incoming light energy is most focused. When these negative feedbacks are taken into account we will see that climate sensitivity to leprechaun doubling will be a fraction of a degree at most.

    • Hang on. Are you taking rainbow spectral refraction for when the light enters the pot of gold into account? According to THIS PUBLICATION there may also be an effect on cloud albedo.

    • “It astounds me that so many here have ignored the obvious negative feedbacks associated with leprechauns, most notably the high albedo of gold”

      Who are you so wise in the ways of blog science?

  19. Tamino
    The “Shapiro-Wilk test” link probably should be

    [Response: Right you are, thanks. I simply omitted the link.]

  20. Dikran Marsupial

    I have posted on Dr Spencer’s blog to ask what he would consider would constitute the falsification of the hypothesis “natural, internal climate cycles as the cause of most of the recent warming in the thermometer record”. If the hypothesis is unfalsifiable it is unscientific (according to Popper) and hence not overly surprising that it hasn’t been addressed in the peer reviewed litterature.

    • Dikran Marsupial

      It seems my posts on Dr Spencers blog have suddenly started being subject to moderation… No answer yet to my question though.

      • Dikran Marsupial

        Oh well, experiment over, looks like my posts are not going to make it past moderation (not that they were offensive or off-topic). Shame, my opinion of Dr Spencer has gone down a notch or two.

  21. OK, despite of the hilarity of leprechauns or Dinner with Judy, you have a lot of work ahead of you to outdo the life science crowd.

    Two reasons for their success: better props and more women.


  22. David B. Benson

    Oh, but one also has to consider the
    and the

    I’d say Dr. Spencer has his work cut out for him!

  23. Inferno at DenialDepot schools Tamino:


    Time for me to school Inferno.

    Like the alarmists, Inferno neglected some key forcing components:

    Station Drop Out *: 90%
    Remote Area Upward Adjustment **: 70%
    Black Carbon ***: 100%
    Roy Spencer Fraudulent Adjustments ****: 110%

    * Tamino vehemently denying this fact is proof that it’s a fact.

    ** As been proven, the frauds adjust stations in remote areas because they know that no one is there to call BS. Proof is in their claim that the Arctic is warming rapidly, and this winter the alarmists claimed Greenland was warm. This hypothesis was put forth by blog commenter “JS” and this person has been vindicated.


    *** It’s important to acknowledge black carbon because it means we don’t have to create a one-world communist government to reduce CO2. Instead, we can pin more of the blame on the communist Chinese. I understand other climate realists like me respectively disagree, and see such acknowledgment as a slippery slope, but we can agree to disagree. The great U.S. Senator Inhofe has conceded black carbon and he’s clearly not on the government dime. The important thing is that it’s not CO2. How dumb is it to blame the air we expel?

    **** All of the warming in the UAH record is the result of upward corrections, and this proves Roy Spencer is fabricating the evidence to hide the decline. More evidence of Spencer’s alarmism is his arguing with realists that increased atmospheric CO2 can warm the planet.

    – MarkBClimateRealist

  24. From Merriam-Webster:

    “Definition of NATURAL
    1: one born without the usual powers of reason and understanding
    2a : a sign ♮ placed on any degree of the musical staff to nullify the effect of a preceding sharp or flat b : a note or tone affected by the natural sign
    3: a result or combination that immediately wins the stake in a game; specifically : a throw of 7 or 11 on the first cast in craps
    4a : one having natural skills, talents, or abilities b : something that is likely to become an immediate success c : one that is obviously suitable for a specific purpose
    5: afro
    See natural defined for English-language learners »”


    “Definition of INTERNAL
    1: existing or situated within the limits or surface of something: as a (1) : situated near the inside of the body (2) : situated on the side toward the median plane of the body b : of, relating to, or occurring on the inside of an organized structure (as a club, company, or state)
    2: relating or belonging to or existing within the mind
    3: intrinsic, inherent
    4: present or arising within an organism or one of its parts
    5: applied or intended for application through the stomach by being swallowed
    — in·ter·nal·i·ty \ˌin-ˌtər-ˈna-lə-tē\ noun
    — in·ter·nal·ly \in-ˈtər-nəl-ē\ adverb”

    and then on to guess what the expression ‘internal natural’ in various contexts means.

  25. or maybe ‘natural’ is used as an adjective in this context?
    “Definition of NATURAL
    1: based on an inherent sense of right and wrong
    2a : being in accordance with or determined by nature b : having or constituting a classification based on features existing in nature
    3a (1) : begotten as distinguished from adopted; also : legitimate (2) : being a relation by actual consanguinity as distinguished from adoption b : illegitimate
    4: having an essential relation with someone or something : following from the nature of the one in question
    5: implanted or being as if implanted by nature : seemingly inborn
    6: of or relating to nature as an object of study and research
    7: having a specified character by nature
    8a : occurring in conformity with the ordinary course of nature : not marvelous or supernatural b : formulated by human reason alone rather than revelation c : having a normal or usual character
    9: possessing or exhibiting the higher qualities (as kindliness and affection) of human nature
    10a : growing without human care; also : not cultivated b : existing in or produced by nature : not artificial c : relating to or being natural food
    11a : being in a state of nature without spiritual enlightenment : unregenerate b : living in or as if in a state of nature untouched by the influences of civilization and society
    12a : having a physical or real existence as contrasted with one that is spiritual, intellectual, or fictitious b : of, relating to, or operating in the physical as opposed to the spiritual world
    13a : closely resembling an original : true to nature b : marked by easy simplicity and freedom from artificiality, affectation, or constraint c : having a form or appearance found in nature
    14a : having neither flats nor sharps b : being neither sharp nor flat c : having the pitch modified by the natural sign
    15: of an off-white or beige color
    — nat·u·ral·ness \-nəs\ noun”

  26. Why won’t Spencer mention the Loch Ness monster or yowie? What is he trying to hide?

  27. g2-b31f1590b0e74a6d1af4639162aa7f3f

    Just a quick note from a layman-engineer’s perspective regarding the surface temperature record. I recently spent some time playing around with the GHCN data — wrote a program to crunch the data and compute simple “dumb averages” of the temperature station anomalies. My results closely matched NASA’s “Northern Latitudes” temperature index (available at the NASA/GISTEMP web-site). Not surprising, because the GHCN stations tend to be concentrated in the temperate latitudes of the NH.

    Also found minimal differences between the GHCN “raw” vs. “adjusted” data results.

    I then added a bit of code to sift through the GHCN V3 metadata files and crunch the rural and urban stations separately. Found very little difference in the warming trends produced by urban vs. rural stations.

    Then added gridding/geospatial-weighting functionality to my code. My gridded results ended up matching NASA’s Land Station Index very closely (surprise, surprise). Also, raw vs adjusted data and rural vs. urban station data all produced very similar results.

    I also added a bit of code to pick at random small subsets of the temperature stations. I found that I could throw away 90% or more of the station data and still get results very similar to what I got from processing data from all of the stations.

    Now, what struck me about this little “spare time” programming project is how darned *easy* it was! The most important thing I came away with from this little project is not an appreciation of the robustness of the surface temperature record (it really is quite robust), but of the breathtaking idiocy of the attacks on CRU/NASA/etc, and of the incredible collective laziness and incompetence of the denier crowd.

    [Response: And you even did the analysis *before* coming to your conclusions about the result.

    Science. It works.]

    • You ought to post that code somewhere, just another example of how various laypeople have been able to quite easily build temperature reconstructions that surprise! surprise! show results that closely match those from “fraudulent” scientists at CRU and NASA.

      Unfortunately the satellite temperature reconstructions are far more difficult, one of the things that made it difficult to refute Christy and Spencer’s first efforts which were so loudly proclaimed to be “the wooden stake through the heart of global warming” (the denialsphere has had to switch to coffin nails since there are so many of them that have been shown to be crap – there’s really no reason to put multiple wooden stakes through a vampire’s heart if your aim is true and your stake truly wooden).

      • You ought to post that code somewhere, just another example of how various laypeople have been able to quite easily build temperature reconstructions that surprise! surprise!

        Heh, heh…

        A few weeks ago, I posted some preliminary results on the climate-change forum over at my hometown newspaper’s web-site (linky: http://forums.signonsandiego.com/showthread.php?t=111201).

        Here’s one of the more amusing loonball responses:

        Here’s a suiggestion for you- since you’ve always done little other than to appeal to authority, and scream out words such as “REKNOWNED”,”ACCREDITED”, and “PEER REVIEWED” regarding your eco-scientist menagerie, along with perjoritives directed toward skeptics on a daily basis, why not publush your C++ source code here via text or link (prefereably with comments and perhaps YOUR algorithm) along with your open office spreadsheet template, so that it too can be reviewed? there are many people here who are software savvy, and in some cases, have access to professionals who are system architects, analysts, or IT auditors that can give it a once or twice over “smell test” for valdity and lack of self serving massaging…

        When I replied to the loonball who posted the above drivel, I included a link to my code (an early, crude version that still produced surprisingly good results) — and the crickets have been chirping over there ever since.

        caerbannog aka g2-b31f1590b0e74a6d1af4639162aa7f3f (my wordpress userid must have gotten corrupted somehow).

  28. Challenge accepted: Why Leprechauns can’t explain the recent warming:


  29. Reading all of this, I am surprised there is nary a mention of the correlation between the number of Pirates and Gloabl Warming. It’s not CO2, it’s pirates!


  30. g2-b31f1590b0e74a6d1af4639162aa7f3f

    Folks, when I claimed that I was able to replicate NASA’s results very closely with a very simple “gridded average” program that I whipped up in some spare time, I wasn’t blowing smoke.

    Here’s a plot of my GHCN raw data gridded-average results vs. NASA’s official Meteorological Stations (GHCN) results: http://img689.imageshack.us/img689/6175/caerbannogvnasalandinde.jpg

    I was able to whip this up in my spare time over just a few days (would have been even quicker if not for the beer). Now, why couldn’t Anthony Watts and his followers do something similar in the *years* since the establishment of their silly “surfacestations” project? (rhetorical question here).

    • I can see the headline:

      Warmist replicates temperatures while boozed up! Deniers break out in cold sweat at thought of trying.
      “…would have been even quicker if not for the beer…”

  31. g2-b31f1590b0e74a6d1af4639162aa7f3f

    Just one additional note: My results are plotted in blue, NASA’s in orange.

  32. “I was able to whip this up in my spare time over just a few days (would have been even quicker if not for the beer).”

    Completeness in discussing methodology is undoubtedly a virtue. . .

  33. caerbannog aka g2-b31f1590b0e74a6d1af4639162aa7f3f

    Yeesh. I’ve wondered for some time how to pronounce “caerbannog” (guess I’ve run across your comments on a bunch of climate-related sites). But now how in the heck am I supposed to pronounce your new username?

    Nice work, by the way. That’s just great.

  34. Ned,

    “g2-b31f1590b0e74a6d1af4639162aa7f3f” is pronounced “Go-to guy.”

  35. Kevin Stanley


    Wow. What is that, the key to Julian Assange’s encrypted “insurance” file?

  36. g2-b31f1590b0e74a6d1af4639162aa7f3f

    Wow. What is that, the key to Julian Assange’s encrypted “insurance” file?

    That’s my current wordpress userid after an apparent data corruption incident. With my “new” userid, I don’t dare log out because I’m pretty sure that I would never manage to get myself logged back in again!

    And as for the pronunciation, I think that it goes something like this:

    “The user whose login name was formerly caerbannog.”

    • TUFKACUADCIFUHLN may be easier:
      The User Formerly Known As Caerbannog Until A Data Corruption Incident F’ed Up His Login Name

  37. “cearbannedogain” maybe MrPete or Mosher hacked wordpress and corrupted your personal data :)