Open Mind

Riddle me this …

December 7, 2009 · 198 Comments

Those who are in denial of global warming insist that the last decade of global temperature contradicts what was expected by mainstream climate scientists.

Here’s global temperature data from NASA GISS before the 21st century, for the time span 1975 to 2000:

I’ve computed and plotted a trend line using linear regression. In addition, I’ve plotted dashed lines two standard deviations above and below the trend line — we expect most of the data to fall within these dashed lines. Finally, I’ve projected those lines out to the present day.

That’s what mainstream climate scientists expected to happen.

Here’s what actually happened.

Gosh. What actually happened is exactly what was expected. Exactly. By mainstream climate scientists. You know, those folks who keep telling us that human activity is warming the planet and that it’s dangerous.

What’s that? You don’t trust NASA GISS? You think they faked it, or it’s all just urban heat islands? How about satellite data from RSS — there’s no urban heat island effect there!

Hmm. What actually happened is exactly what was expected. Exactly. By mainstream climate scientists.

Don’t trust the people at RSS? How about the denialists’ favorite data, satellite estimates from UAH?


We’re only 10 years into the 21st century, but so far, global temperature has done exactly what was expected by mainstream climate scientists. Exactly. You know — those folks who keep telling us that human activity is warming the planet, and that it’s very dangerous.

This is undeniable. Unless of course you’re in denial.

Yet people continue to deny it. They tell you it’s all a hoax, and to support that idea they repeatedly claim that the last decade of temperature data contradicts global warming.

Riddle me this: why can’t they get something this simple right?


Kevan Hashemi submitted this comment:

You picked the period 1980 to 2000 do you would get a positive slope. Why not fit a line to 2000 to 2010? Because it shows a negative slope and you don’t want that? Well, that’s precisely the point skeptics are making: if you look at the last 10 years, the trend is zero or down. You choose your own 20 years to show the trend is up. If you chose 1940 t0 1980 the trend would be down. In 1999 climate scientists told us the world would warm up over the next ten years. It didn’t.

You are wrong on all counts.

The trend lines from 2000 to 2010 (actually to the present since 2009 hasn’t ended yet) are all positive:

  • For GISS data, the trend from 2000 to the present is +0.0115 +/- 0.018 deg.C/yr.
  • For RSS data, the trend from 2000 to the present is +0.0017 +/- 0.030 deg.C/yr.
  • For UAH data, the trend from 2000 to the present is +0.0052 +/- 0.043 deg.C/yr.

    Notice all those plus signs.

    More to the point, the uncertainties in trend estimates using just data since 2000 are much larger than the trend estimates themselves. Attempting to delineate the climate trend using so little data is a fool’s exercise.

    That’s the point. They’re trying to fool you.

    As for “warming since 1999″:

  • Using GISS data the decadal average for the 1990s was its highest yet at 0.3176. For the 2000s it warmed considerably, averaging 0.5108.
  • Using RSS data the decadal average for the 1990s was its highest yet at 0.0833. For the 2000s it warmed considerably, averaging 0.2384.
  • Using UAH data the decadal average for the 1990s was its highest yet at 0.0587. For the 2000s it warmed considerably, averaging 0.2219.

    Finally — with GISS data I started at 1975 because that’s a natural turning point in the temperature trend, and data since 1975 actually does enable us to estimate a trend with sufficient precision to be useful (unlike your suggestion of starting with 2000). For RSS and UAH data I started at 1979 because that’s all the data there is.

    Riddle me this: why can’t Kevan Hashemi get something this simple right?

  • Categories: Global Warming

    198 responses so far ↓

    • Philippe Chantreau // December 7, 2009 at 2:45 pm | Reply

      Gee what a surprise! But it can’t possibly be true, those e-mails tell us exactly that it’s all a scayum. Plus, Al Gore is fat anyway, so who cares about the real numbers?

    • barry // December 7, 2009 at 2:47 pm | Reply

      “That’s what mainstream climate scientists expected to happen.”

      Is that right, Tamino? AR4 SPM recommends a trend of 0.2C per decade over the next two. That would seem to be a might steeper than the trend line extended in the above graphs.

      [Response: AR4 says "about" 0.2C per decade, which given only 1 significant digit means somewhere between 0.15 and 0.25. And that's what has happened.]

      Have you done a similar overlay after ginning up a nice spaghetti graph from the AR4 model archive? The smoothed runs in the AR4 document don’t show the degree of variability (obviously) and more than one denialist likes to argue that 2008 was the outlier that sucker-punched the models. One of these antagonists was good enough to do the graphics with a spaghetti graph put together in a post by Gavin Schmidt at RC, and 2008 just clipped inside the variability – to the chagrin of my honest contrarian.

      I tend to link to that RC graph whenever someone says models have failed because they didn’t project decadal flat/cooling trends.

      Have you done something similar – or am I on the wrong foot here?

    • Douglas Watts // December 7, 2009 at 3:04 pm | Reply

      Last week’s Nature editorial correctly used the noun “denialist” and did not use the word “skeptic.”

      This was for a reason, well illustrated by the graphs above.

      Thanks once again, Tamino.

      For recreational purposes, this USGS open file analysis of ice-out dates on Maine lakes from 1850-2000 is a nice palate cleanser.

    • t_p_hamilton // December 7, 2009 at 3:26 pm | Reply

      Simple, clear, direct. I wonder about the denial strategies that will be employed…

      Release all of the data! Did you delete any copies after you did your analysis? I want to see your code!
      It is the sun! Cosmic rays! If you don’t klive in a cave you aren’t being serious, and hence a hypocrite.

    • Nick Barnes // December 7, 2009 at 3:39 pm | Reply

      Excellently put. Any who remain doubtful of the surface temperature record should feel free to come over to the ClearClimateCode project to look at, and refine, the GISTEMP code.

    • Marcus // December 7, 2009 at 5:17 pm | Reply

      What do you think about the Fu/Washington U. corrections to the RSS/UAH trends? eg, a different way of removing stratospheric contamination from the channels, resulting, at least in the versions I’ve seen, in significantly higher trends? For some reason, no one seems to mention those much, but it seems to me like they would make a difference for all kinds of comparisons & tests…

      [Response: Indeed they would, and the Fu et al. method of correction seems very logical to me. It's at least possible that UAH and RSS take precedence simply because they came first (and UAH can't stand alone because of the history of problems with it).

      While I don't know enough about the correction methods or joining of different satellite records to endorse Fu et al. over other methods, I do think it shouldn't be neglected.]

    • george // December 7, 2009 at 5:23 pm | Reply

      Riddle me this: why can’t they get something this simple right?

      The obvious answer is that some people simply don’t want to get it right — ie, they are purposely getting it wrong.

      But among the “skeptics”, it is at least possible that there are some who are genuinely interested but are nonetheless asking the wrong question — or more likely, accepting the answer provided by those who are asking the wrong question.

      Instead of asking (as Tamino has done above) “Are the temperatures of the last 10 years what was expected based on (consistent with) the trend of the previous 25 years?” they ask “is the trend for the past 10 years the what was expected given the trend for the previous 25 years?”

      or, a different version of basically the same question: “is the trend of the past 10 years “the trend” [sic: there were actually many of them] “predicted” by the IPCC from 2001 onward?”

      There is considerably more uncertainty involved in answering the second question than the first because of the greater difficulty in estimating short term trends, including associated errors.

      Even if one attempts to account for the error bars on the (apparent) trend for the short term, the answer one gets for the error bars on the trend depend significantly on the noise model one uses.

      of course, most people simply compare the “value” of the short and long term trends without taking into account the error bars at all, which simply isn’t enough to answer the question.

      But even if one does consider the errors, “arguments” about whether a short term trend is consistent with a long term trend inevitably amount to “debates” about the actual value of the errors, which boils down to the choice of the “proper” noise model.

      in other words, it boils down to a fairly abstruse (or obtuse, in some cases) mathematical debate.

      So it’s really no surprise that many people “miss the simple” because when the question is posed as “is the short trend consistent with the long”, the answer is not “simple” at all.

      Arguments about which noise model to use can be answered by sophisticated statistics (still with some uncertainty), but most people have no clue which noise model is better (AR1, ARMA, PhRMA, ALARMA etc) , so it essentially all comes down to “whom do you trust on the statistics?”

      In my opinion, the graphical illustration above is far superior to comparing trends, not least of all because you can actually SEE what is going on.

      Of course, you still have to trust that the trend line and 2-sigma errors were calculated correctly, but in the latter case, the whole thing depends much less on the actual details of the noise model than for the case of comparing the short trend to the long.

    • Eli Rabett // December 7, 2009 at 5:23 pm | Reply

      FWIW, why is the 1998 El Nino so much larger in the satellite records than the surface temperature ones? Does this tell us something important?

      • thingsbreak // December 7, 2009 at 6:49 pm | Reply

        ENSO in general is amplified in the sat data vs. the instrumental record (e.g. 2008 La Niña), as to a lesser extent are major eruptions like Pinatubo and El Chichon.

      • guthrie // December 7, 2009 at 8:05 pm | Reply

        I kind of wondered if it was due to the vast amounts of warm air going up high into the atmosphere from the nice hot ocean. Thus meaning the temperature at height is several degrees warmer than without an El Nino.

    • RW // December 7, 2009 at 5:24 pm | Reply

      You’ve made it abundantly clear here many times through extremely clear and accessible analysis that nothing of note changed in 1998. So why do some climate scientists even seem to be confused on this point? For example, in the infamous hacked e-mails, there was Keith Briffa saying “The fact is that we can’t account for the lack of warming at the moment and it is a travesty that we can’t”

      I do not perceive any lack of warming to account for. Your analysis shows there is no lack of warming to account for. Why does Keith Briffa think there is?

      • Ante // December 9, 2009 at 1:13 am | Reply

        No, not Briffa. That was Kevin Trenberth.

      • Harald Korneliussen // December 9, 2009 at 8:37 am | Reply

        For one thing, it wasn’t Briffa who said that, but Trendberth (there’s an appropriate name, eh?). And he was not referring to global temperature, but ocean temperature, complaining that it isn’t being measured enough (at enough places, depths etc.)
        The lack of sufficient ocean temperature data means we don’t know quite where all that heat associated with ENSO comes from and goes to. And that IS a shame. A more reliable heat budget for the oceans would be very valuable.

    • Deech56 // December 7, 2009 at 5:42 pm | Reply

      RW – that was Trenberth. Check out the “Crock of the Week” video for context.

    • Sekerob // December 7, 2009 at 5:42 pm | Reply

      RW, how about he´s talking about the dendrochronological information, Briffa´s field of expertise failing to show much at all in warming after ~1960s. Is this when vegetation is starting to show the signs of rapid change? Hey, we know that since then the nutritious value of wheat has dropped by 8%… due CO2

    • MapleLeaf // December 7, 2009 at 5:56 pm | Reply

      RW, Tremberth said that.

    • WAG // December 7, 2009 at 6:47 pm | Reply

      Great post. You might also repost this graphic, illustrating that temps are coming in at the upper end of the models:

    • GFW // December 7, 2009 at 7:22 pm | Reply

      Everybody knows that 1998 was an outlier because of a very strong El-Nino … but why is that so exaggerated on the two satellite measures compared to the GISS dataset? (Anybody who knows, please say, not just Tamino.)

    • Kevan Hashemi // December 7, 2009 at 7:24 pm | Reply

      You picked the period 1980 to 2000 do you would get a positive slope. Why not fit a line to 2000 to 2010? Because it shows a negative slope and you don’t want that? Well, that’s precisely the point skeptics are making: if you look at the last 10 years, the trend is zero or down. You choose your own 20 years to show the trend is up. If you chose 1940 t0 1980 the trend would be down. In 1999 climate scientists told us the world would warm up over the next ten years. It didn’t.

      [Response: Wrong on all counts. See the update to this post.]

    • GFW // December 7, 2009 at 7:33 pm | Reply

      Oops, I see Eli asked the same question – but the answer wasn’t enough. *Why* do satellite measure amplify ENSO and volcanic events? (So feel free to delete my redundant question and put the answer under that of the Bunny :-)

    • B Buckner // December 7, 2009 at 7:47 pm | Reply

      Slope of the GISS line about twice as steep as the UAH slope.

    • B Buckner // December 7, 2009 at 7:50 pm | Reply

      Never mind, my mistake, steeper but not close to twice.

    • Deech56 // December 7, 2009 at 8:57 pm | Reply

      Tamino – simple question. Is it OK to do a simple test for significance of the slope using the annual means? I know that when the monthly data are used, autocorrelation rears its head.

      [Response: Autocorrelation is still present in annual averages, but it's much weaker than for monthly data; the simple significance test will be approximately correct, but the actual significance will be less than that test indicates. So the simple test, applied to annual averages, is an indicator but not a precise one. If it's not significant, then neither will be a corrected test; if it's overwhelmingly significant, believe it; if it's borderline, don't believe it.]

    • Harold Brooks // December 7, 2009 at 9:56 pm | Reply

      ENSO and volcanoes are amplified in the troposphere because of water vapor feedback.
      From Chapter 5 of Temperature Trends in the Lower Atmosphere: Steps for Understanding and Reconciling Differences

      “When surface and lower tropospheric temperature changes are spatially averaged over the deep tropics, and when day-to-day tropical temperature changes are averaged over months, seasons, or years, it is evident that temperature changes aloft are larger than at the surface. This “amplification” behavior has been described in many observational and modeling studies, and is a consequence of the release of latent heat by moist convecting air (e.g., Manabe and Stouffer, 1980; Horel and Wallace, 1981; Pan and Oort, 1983; Yulaeva and Wallace,
      1994; Hurrell and Trenberth, 1998;
      Soden, 2000; Wentz and Schabel,
      2000; Hegerl and Wallace, 2002;
      Knutson and Tuleya, 2004)”

    • AndyL // December 7, 2009 at 10:25 pm | Reply

      I’m not quite sure what these charts show. They seem to show that if you extend the warming from the 25 years before 2000 in a linear fashion for 9 years, temperatures have not deviated from the trend by +/- 2SD. Correct?

      Isn’t that just extrapolation? Yes it shows warming has continued, but how is it “what was expected by mainstream climate scientists”? Eye-balling the trend lines, they appear to be at the bottom of your range .15 to .25 degrees per decade.

    • Tim // December 7, 2009 at 10:40 pm | Reply

      Hey Tamino, you are the best of the bunch (along with SkepScience).

      Keep up the stellar work.

      I’ve basically come up with a new term for skeptics. It’s based upon my assertion that science isn’t about “belief” (over here in Australia, people say “oh I don’t believe in it”). Science is what it is. Nature doesn’t lie.

      So, my position is, if you disbelieve the science, you’re not a skeptic, you’re actually a moron. Or an insidious being with something to gain by naysaying.

      I’m looking fwd to editorial reference to deniers as morons.

      Tim in Oz

    • naught101 // December 7, 2009 at 11:01 pm | Reply

      Hey Tamino,
      Great post, thanks.
      Would be good to see slope values on the graph, as well as the StdDev values.

      Also, when you say “most” data should fit within 2 stddev, it might be a good idea to specify that most means ~95% – for anyone without basic stats education, “most” in that context could mean anything from 51%-99%

    • Hank Roberts // December 7, 2009 at 11:10 pm | Reply

      Tim, it’s spelled differently over here:

    • Crooksmeister // December 7, 2009 at 11:13 pm | Reply

      Tim in Oz:

      “. . . deniers as morons”, eh? My brother (an environmental scientist, as you ask) was correct about the evangelicals being back in town.

      I gotta say that this sort of ridiculous, unfounded assertion merely serves to undermine any other argument that you (or the other evangelicals) may put forward

    • Tony O'Brien // December 7, 2009 at 11:27 pm | Reply

      There is a nice correlation between ENSO and temperature and there is a nice correlation between the solar cycle and temperature, but is there any correlation between ENSO and the solar cycle.

      Either way the answer does not affect the trend, just curious.

    • cce // December 7, 2009 at 11:34 pm | Reply

      “About 0.2 degrees per decade” is the expectation for the next two decades, which is different than the suspected AGW signal for the past three.

    • David B. Benson // December 7, 2009 at 11:53 pm | Reply

      Tony O’Brien // December 7, 2009 at 11:27 pm — With the available data there is no way to see any effect of the solar cycle on ENSO; of course that does not mean there is none, just too small. However, according to a recent paper by researchers at NCAR, the solar cycle gives rise to its own mini-ENSO; sorry I don’t have the reference.

    • Glenn Tamblyn // December 8, 2009 at 12:03 am | Reply


      Great post. The ‘Its cooled since 1998′ meme is perhaps the most destructive of the denialist lines out there. Certainly here in Australia it has influenced the political process.

      One general observation that surprises me a little is, how little, relatively speaking,emphasis is given by the Climate Science community on the observations of total heat content increase – over 2 * 10^23 Joules since 1950 according to Murphy et al. Thats around 3 Billion Hiroshima bombs. And graphs showing that 90% of this is occuring in the oceans, with the atmosphere almost a side show seems to me particularly visceral.

      Coming from an engineering background as I do, looking at AGW as principally an energy problem, ie too much of it, seems a natural thing to do. And the evidence for AGW seems to me to be something that can be very clearly encapsulated when looked at from an energy perspective.

      While predictions of temperature changes, sea level etc are far more complex, the energy balance perspective seems simpler to measure, demonstrate and convey to Joe Public. And a natural foil to the ‘its cooling’ meme. This might at least allow us to get past the ‘its not happening’ argument to move on to the ‘how serious will it be’ debate. And expressing this in visceral terms like X Atom Bombs may focus peoples attention.

      If we keep fighting the Denialosaurs on their ground of ‘whats the temperature?’ aren’t we yielding an advantage to them?

    • oneuniverse // December 8, 2009 at 12:10 am | Reply

      I’d be grateful if anyone could point me to any comments or critiques in the peer-reviewed literature on Svensmark 2009 (on Forbush decreases and water-vapor), Shaviv 2004 (on GCRs and climate sensitivity) and Shaviv 2008 (on estimating solar forcing using the ocean as a calorimeter).

    • Joseph // December 8, 2009 at 12:10 am | Reply

      Keith Briffa saying “The fact is that we can’t account for the lack of warming at the moment and it is a travesty that we can’t”

      Keith Briffa didn’t say that.

    • Joseph // December 8, 2009 at 12:14 am | Reply

      Why not fit a line to 2000 to 2010?

      What you really want to say there is… Why not cherry-pick fit a line from 1998 to 2008?

    • David B. Benson // December 8, 2009 at 12:38 am | Reply

      oneuniverse // December 8, 2009 at 12:10 am — It’s not peer reviewed, but I showed why Shaviv (2008) is wrong in a thread on
      but back quite a few threads by now.

      For a better (by far) paper, see Tung & Cabin (2008).

    • MarkB // December 8, 2009 at 12:49 am | Reply

      The only way one can get a “cooling” trend in any dataset for anything close to 10 years is to start in the super el Nino year of 1998 – starting in a year that is more than 2 (or 3) standard deviations above the trendline and clearly unrepresentative. That’s called severe cherry-picking.

      Interestingly, UAH (the denialist’s favorite) is showing an anomaly of +0.50 (1979-1998 baseline) for November, a November record by 0.10. That crazy alarmist Roy Spencer must be fabricating his data ahead of Copenhagen! Actually, I suspect part of the anomaly is related to their very suspicious seasonal cycle, documented by DeepClimate.

    • Eli Rabett // December 8, 2009 at 12:59 am | Reply

      Thanks Harold

    • Deech56 // December 8, 2009 at 1:14 am | Reply

      RE Tamino’s in-line response December 7, 2009 at 8:57 pm:

      Thanks. So basically I’m not too far off battling the “cooling since…” claim.

    • Ray Ladbury // December 8, 2009 at 1:22 am | Reply

      Crooksmeister: Rather than castigate him, why not prove him wrong and actually LEARN SOMETHING?!!!

    • Scott A. Mandia // December 8, 2009 at 2:03 am | Reply

      Tamino, thnaks for the information. I have already added your work to my Global Cooling page.

      Isn’t it sad that we have to keep explaining that there has been no global cooling? Even my wife now gets pissed off when any of her Facebook friends brings this fallacy up! Gotta love it.

    • jyyh // December 8, 2009 at 2:38 am | Reply

      re: diff between sat vs. land based measurements… i think it’s from the water vapor feedback, i.e. el nino puts more of it higher in the sky skewing the results, but not sure of this.

    • jyyh // December 8, 2009 at 2:40 am | Reply

      no need for the prev, since a better answer already present…

    • Zeke Hausfather // December 8, 2009 at 3:55 am | Reply

      I wrote something fairly similar awhile back, though I only did the analysis with HadCRU.

      Funny enough, you actually get a -higher- trend adding in 1999-2009 than you would just looking at 1965 (or 1975) through the end of 1998 (ENSO notwithstanding). Lucia even lent a helping hand with autocorrelation corrections in the confidence intervals.

    • Oakden Wolf // December 8, 2009 at 4:51 am | Reply

      Too bad El Nino episodes happen over the winter — thus dividing their influence between years. 2010 will be an interesting year to watch.

    • Andy // December 8, 2009 at 5:41 am | Reply

      Dear Mr. Hashemi: I believe you are mistaken.

      The coldest temperature ever recorded (excluding Antarctica since its record is so short) was -90 F at Verkhoyansk, Russia on February 7, 1892. The hottest was in the Libyan desert, 136 F on September 13, 1922. Therefore, I conclusively conclude that the earth had heated up until 1922 and has been cooling ever since.

    • Kevan Hashemi // December 8, 2009 at 5:42 am | Reply

      You say I am wrong on all counts. Please clarify. I said the slope for the last ten years is zero or down. You say it’s +0.0017 +/- 0.030, which is consistent with “zero or down”, isn’t it? I say that the slope from 1940 to 1980 is down. We have surface data for that period (not that I trust it). And do you imagine that by insulting me, you are more likely to convince me? I have never found that to be the case in Physics. Perhaps it’s different in Climate Science.

      [Response: You just can't resist playing the fool.

      I'm the one who emphasized that "More to the point, the uncertainties in trend estimates using just data since 2000 are much larger than the trend estimates themselves. Attempting to delineate the climate trend using so little data is a fool’s exercise." The point of this post is that attempts to characterize temperature trends using such short time spans are foolish at best, deliberately deceptive at worst. You are the one who did so. The best we can think of you is that you're a fool.

      As for +.0017 +/- 0.03 being consistent with zero or down, it's also consistent with +.028 -- way faster warming than anyone believes -- in spite of your (and others) insistence on a cooling trend.

      As for the slope from 1940 to 1980, put some error bars on that and it's "consistent with zero or up." So you resort to the minimum error range for the 2000-2010 period (a fool's choice), but you make an unqualified claim about mid-century.

      As for insulting you being a bad way to convince you, nothing will convince until you realize what an idiot you've been.]

      • Layman // December 8, 2009 at 3:00 pm | Reply

        Hi, don’t mean to intrude here, just making my standard request for help for the uninitiated.

        You say that “As for +.0017 +/- 0.03 being consistent with zero or down, it’s also consistent with +.028″. I read +.0017 +/- 0.03 as stating a range from -0.0283 to +0.0317 (simply subtracting 0.03 from 0.0017 in the first case and then adding it in the second case). What am I doing wrong here? Please be kind, thanks!

        [Response: You're quite right; the arithmetic error was mine.]

    • John Mashey // December 8, 2009 at 6:45 am | Reply

      Fine exposition, as usual.

      But I think “Those who are in denial of global warming insist that the last decade of global temperature contradicts what was expected by mainstream climate scientists.” is imprecise.

      Really, who would say that? -)

      Well, here are a few, in a Letter to US Senate.

      (That is the APS Petition organizers + Lindzen).

    • Donald // December 8, 2009 at 8:32 am | Reply

      It’s HadCrut that sceptics usually mention, yet despite 10 years of cooling- Er…, 11 years of flat temperatures, temperatures now are exactly on the trend line up to and including 1998.

      Temperatures seemed to take a step up after the 1998 El Nino, then plateau for a while- although there is still warming evident- the 2008 La Nina being warmer than the 2000 one, according to the Met Office.

      Any chance of a more sophisticated analysis of the HadCrut data?

    • Scott A. Mandia // December 8, 2009 at 12:58 pm | Reply


      On my Global Cooling page, I added your link. Nice work!

    • oneuniverse // December 8, 2009 at 1:46 pm | Reply

      David B. Benson, thank you, that was an interesting paper.

      Tung and Camp 2009 find that the 11-year solar cycle is capable of varying global mean surface temperatures by 0.2 K, even though the radiative forcing from the solar cycle is only about 0.18 W/m2.

      They suggest postive feedbacks from changes in water vapor , lapse rate, cloud cover and ice albedo could provide the amplification needed to account for the 0.2 K temperature variations.

      However, the cloud behaviour is not that well understood, and GCRs are not ruled out.

      Tung and Camp also refers to the White et al. 1997 paper “Response of global upper ocean temperature to changing solar irradiance”, which demonstrates how changes in solar irradiance are driving upper ocean temperature changes (see figures 1, 3 and 5 – figure 3a is particularly striking, I thought).

      re: your comment on Shaviv
      Please let me get back to you.

    • Kevin McKinney // December 8, 2009 at 1:53 pm | Reply

      I agree that the “it’s been cooling” meme is destructive–to denialist credibility.

      It is admittedly tiresome to have to keep saying it so often, but everybody understands thermometers and everybody understands lying. The “it’s been cooling” brigade are showing themselves up as recidivists of falsehood.

    • Joseph // December 8, 2009 at 2:17 pm | Reply

      It’s HadCrut that sceptics usually mention…

      If you look closely at that graph, you’ll see that ’sceptics’ can fit this arbitrary flat trend every 10 years or so. For example, between 1988 and 1997 (roughly) – flat trend. Between 1978 and 1987 – flat trend again.

    • Ray Ladbury // December 8, 2009 at 2:59 pm | Reply

      Kevan Hashemi,
      You said: “Why not fit a line to 2000 to 2010? Because it shows a negative slope and you don’t want that?”

      This is simply flat wrong. If you have a quantity

      A+/-B, where B>A, you cannot say the quantity is less than zero, or even equal to zero. The most you could say is that the quantity is not inconsistent with zero to some level of confidence. Your inability to understand this bespeaks either a lack of competence in interpreting elementary error analysis or ideological motivation. Your attack on Tamino is quite simply wrong, and I would say you owe him an apology for your accusation of dishonesty.

      The decade you chose 2000-20010 is shaping up to be the warmest on record. The coolest year of that decade is the 8th warmest on record. Methinks perhaps you need to review the evidence as well as learning how to interpret it.

    • Kevan Hashemi // December 8, 2009 at 3:16 pm | Reply

      Also, where did you get your UAH data, and which variable did you plot? I plot their Global anomaly, downloaded from their website today, and I get this plot. Not the same as your plot: yours has an added upward trend.

      [Response: You've plotted the mid-troposphere channel, I was working with the lower-troposphere data. The mid-troposphere channel is known to be strongly contaminated by the rapid cooling of the stratosphere.]

    • Layman // December 8, 2009 at 4:01 pm | Reply

      From what I recall (and this is going waaaay back) there are ways of measuring how well a trend line fits the data. If I remember correctly, the most basic measure is R^2? Does it make sense to talk about this or any other kind of “fit” measure in this context?

      Also, there obviously are all sorts of different types of curves (linear, polynomial, etc. – I know this from Excel, which is not saying much, of course). You used linear here and I assume it’s because it’s simple and clear (which I appreciate!). Would any other type of curve make sense? If so, which, and what does it look like?

      Thanks for all the “basic” help over the past few days, it’s been great. I’m starting to get up to speed enough to take this info to friends now.

      [Response: Yes, quantities like R^2 are a valid way to measure the goodness of fit. But bear in mind that doesn't necessarily translate to statistical significance.

      Of course it's possible to fit other types of curves, low-order polynomials are a popular choice. But the fits so obtained are no better than by accident, so including too many degrees of freedom risks the hazard known as "overfitting."]

    • oneuniverse // December 8, 2009 at 4:15 pm | Reply

      Hi David,

      re: your post “Climate sensitivity, Shaviv and Tung et al.” on (which you linked to earlier):

      I agree with you that Shaviv’s figure of a 400m MLD seems too high, although this was the figure in use in the literature at the time. MLD apparently mostly varies from 25m – 200m, but can be over 1000m deep in some places. It’ll be interesting to redo his calculations using a smaller MLD. I’m not sure what the currently accepted figure is.

      I also agree with you that on first inspection, Tung and Camp’s treatment of the upper ocean layer in their Appendix A appears to be too simple.

      However, I haven’t been able to follow your treatment of SCCS. You first define the solar cycle climate sensitivity (SCCS) as “the change in temperature from solar minimum to solar maximum for the average solar cycle.”. After a reader spotted an order-of-magnitude numerical error in your calculations, you said: “Wasn’t that hard to fix. I’ll define SCCS as the 5 year response of a GCM to an instantaneous doubline of CO2.”

      But that is a completely different, ad hoc definition – I don’t see how it can be appropriate for SCCS.

      You also say that Tung et al. treat SCCS as if it is the same as TCS (transient climate sensitivity), but to me they just seem to note that there happens to be a similarity between them:

      “This solar RF turns out to be almost 1/20 that for the total change in RF due to a doubling of CO2 (RF 3.7 W/m2). Therefore the annual rate of increase in radiative forcing of the lower atmosphere during the five years from solar min to solar max happens to be equivalent to that from an average 1% per year increase in greenhouse gases, close to that used in TCR calculations.” (Tung et al. 2008)

    • george // December 8, 2009 at 5:08 pm | Reply


      Based on what he claims on his website, I don’t believe that Kevan Hashemi is going to be receptive to any argument based on the global temperature record (other than that “it is completely random”)

      Simply put, he does not believe it is even possible (at least not with the HADCRU data) to “know that the fluctuations in global temperature are not random”.

      From Hashemi’s website


      Let us suppose that Jones et al of the Climate Research Unit are correct, and their measurement of global temperature is indeed accurate to 0.1°C. Can we look at their graph and predict a trend? How do we know that the fluctuations in global temperature are not random?

      What we see in the Jones et al graph are variations from year to year of around 0.1 C, variations from decade to decade of 0.2 C, and variations from century to century of 0.4 C. This looks like the 1/f noise we see in many physical systems.

      The only way to prove that global temperature does not behave like 1/f noise is to present a long period of time, something like a thousand years, during which the global climate did not behave like 1/f noise. We have no such period of time.

      Before we can say that a rise in temperature is a trend caused by anything, we have to first prove that it is not random. To prove that it’s not random, we need a lot of accurate data. We don’t have that data. We cannot draw any conclusions about the future of the climate from the Jones et al graph.

      [Response: I guess he's another of the league who believes climate scientists never thought about that. The people who are most interested in global temperature, who have studied it most closely, who have intensely analyzed the factors which influence it -- never considered that it might be random!

      How arrogant.]

    • george // December 8, 2009 at 6:20 pm | Reply

      I wonder if I am right to assume that Hashemi accepts that all the things that have been happening simultaneously in recent decades to
      – global surface air temp (increase), arctic temperatures (increase), sea level (rise), glacier mass balance (increase), arctic sea ice cover (decrease), Greenland ice mass (loss), stratospheric temp (decrease), onset of spring (earlier), growing seasons (increased length), movement of plants and animals (into more northerly habitation zones), etc –
      are simply unrelated random occurrences (all 1/f noise).

      What’s the rule for finding the probability of multiple independent events happening simultaneously?

      Ah yes: multiply all the individual probabilities together.

    • Kevan Hashemi // December 8, 2009 at 6:28 pm | Reply


      [Response: Go away.]

    • Henk Lankamp // December 8, 2009 at 6:47 pm | Reply

      MetOffice says current decade warmest in 160 years:

      With 1 or 2 months to go, these are the differences between 2000-2009 and 1990-1999 (my calculation):
      GISS: +0.19°C
      NCDC: +0.17°C
      HadCRU: +0.17°C
      UAH: +0.16°C
      RSS: +0.17°C

    • Harold Brooks // December 8, 2009 at 7:27 pm | Reply

      Based on what he claims on his website, I don’t believe that Kevan Hashemi is going to be receptive to any argument based on the global temperature record (other than that “it is completely random”)

      Another odd thing on his website is that he cites a paper (the link doesn't actually) that appears to be from Giovanni Sturaro, who has been at various research groups in Padova, Italy, and refers to it as being from "Michigan State University", apparently confusing Microwave Sounding Unit with Michigan State University.

    • mspelto // December 8, 2009 at 7:41 pm | Reply

      George you meant mass balance decrease
      WGMS results

    • Ray Ladbury // December 8, 2009 at 8:00 pm | Reply

      Kevan seems to illustrate the old adage:

      You can always tell an engineer; you just can’t tell him much…

      • Harald Korneliussen // December 9, 2009 at 8:59 am | Reply

        Hey, we engineers know a lot!! Even to call myself a lowly “software engineer” I had to learn a good deal of physics, statistics, calculus and even economics!

        Actually, not enough to be useful, but probably enough to be a pompous arrogant fool… so yeah, the adage still holds true ;-)

        A slightly more serious point: There are many engineers compared to people with more specialised higher education. It may be that the many engineer denialists is more a reflection of that than our innate professional hubris.

    • george // December 8, 2009 at 8:10 pm | Reply

      mauri pelto says

      George you meant mass balance decrease
      WGMS results

      yes, thanks for correcting that.

      In light of all the other stuff that has been happening, a glacier mass balance increase would be surprising (though not impossible)

    • David B. Benson // December 8, 2009 at 8:24 pm | Reply

      oneuniverse // December 8, 2009 at 4:15 pm — Tung et al. use what Tung’s group has accomplished to infer a rather high value of Charney equilibrium climate sensetiity (ECS) by noting the connection between TCS and ECS. One of the climatologists commenting on my earlier attempt pointed out the connection is less obvious than Tung et al. seem to think; my point was that this short-range, solar cycle data, does not seriously constrain ECS (which takes many centruies to reach equilibrium).

      As for either Shaviv’s or Tung & Camp’s analysis of heat flux into the ocean, my point was than the former went too deep and the latter too shallow. A better starting point might be to consider the top 10 meters of the ocean as isothermal with the air just above. The reason for this is that wave action quickly mixes to about that depth.

      I recently learned that the situation is even more complex; some of the vertical mixing is due to the anti-diurnal movement of some small ocean animals which rise at night to feed and swim down during the day to hide. The PR suggests this adds as much energy to the oceans as winds and tides do.

      At this time I do not see how to proceed further with an analysis of the two approaches.

    • Didactylos // December 8, 2009 at 9:09 pm | Reply

      The Met Office is taking a leaf from Tamino’s book, by bringing the decadal averages front and centre, simplifying away all the weather noise. This has to be a good thing.

      Richard Black has written about it for the BBC: This decade ‘warmest on record’

      The CRU hack silliness has caused a lot of problems for many people, but one amusing side effect is that it has brought GISTEMP into higher prominence for Copenhagen, complete with interpolated Arctic temperatures. Toasty temperatures. Relatively speaking!

    • Arthur Smith // December 8, 2009 at 9:51 pm | Reply

      Ok, anybody willing to take bets on how long before Lucia posts something scolding Tamino for his totally ridiculous errors in this post such as:

      * cherry-picking 2000 (uh-oh, Kevan already took that one)?
      * plotting trend lines with different slopes for GISS, RSS and UAH?
      * not realizing that what mainstream climate scientists predict is an acceleration in warming trend, not just continuing warming?
      * uh, something I can’t even think up right now?

      My guess: less than 24 hours from now…

    • cthulhu // December 8, 2009 at 11:28 pm | Reply

      I just saw this:

      I don’t understand it tbh, it claims a 0.6C (!) increase in tempreature from 2005 to 2015, but with lower increases from 2015 to 2025 and then 2025 to 2035. I guess I could rtfm but wondered if anyone else could figure this off hand

      • GFW // December 9, 2009 at 1:14 am | Reply

        I think it’s just one model run, or maybe an average over a few runs of the same model, but it’s probably not an ensemble mean over many models. The numbers are just a little wacky for that. Of course it’s also possible that the person who put together the web graphic transcribed the numbers incorrectly.

    • Dougetit // December 9, 2009 at 12:59 am | Reply

      This is bunk! Haven’t you heard? Search “Climategate” Nasa is in on it to! It’s all bogus! A SCAM!!! The biggest global SCAM in all of human history! Temperatures have been trending cooler since Jan 1998!
      You can fool some of the people most of the time, and most of the people some of the time…..

      But it’s not nice to fool mother nature!

    • Geoff Wexler // December 9, 2009 at 1:02 am | Reply

      “Leading scientists, including some senior IPCC representatives, acknowledge that today’s computer models cannot predict climate. Consistent with this, and despite computer projections of temperature rises, there has been no net global warming since 1998″

      [from Letter to Ban Ki-MoonSecretary-General UN ]

      This is from a document prepared by Marc Morano and Senator Inhofe? It is supposed to be a letter from 100 ’scientists’ including Nigel Lawson (the UK’s answer to Inhofe)

      The interesting point is that Professor Wegman signed it! Where did he crop up before? As the ‘impartial’ expert statistician who chaired the enquiry into Michael Mann’s stats. over the first hockey stick paper. The time is right for an enquiry into Wegman’s statistical judgment? I agree with Tamino’s earlier articles and think that Wegmann’s report was wrong over that issue as well. (There are lots of examples in physics where you have a choice of basis analagous to that faced by hockey people)

      Instead we are being subjected to more demands for enquiries based on trumped up charges against the UEA and New Zealand climatologists.

    • P. Lewis // December 9, 2009 at 1:22 am | Reply

      Arthur says about Lucia entering the fray “My guess: less than 24 hours from now…”

      Cue Jack [Tamino] Bauer, entering stage right: “Trust me, you don’t want to go down that road with me.”

    • Joseph // December 9, 2009 at 2:05 am | Reply

      Ok, anybody willing to take bets on how long before Lucia posts something scolding Tamino for his totally ridiculous errors in this post such as:

      If I were Lucia, my criticism would be that Tamino’s analysis cannot rule out a point of change in 1998. The data is consistent with a continued upward trend, but it doesn’t prove it.

      Of course, that’s a completely hypothetical scenario.

      I haven’t looked at her analyses trying to falsify the IPCC trend recently. I imagine 2009 is kind of a bummer when it comes to that, even within the limitations of those slope estimates.

    • Deep Climate // December 9, 2009 at 2:35 am | Reply

      Let’s see:
      - The warming trend since 1979 (or 1975) stands as high or higher today than in 2000
      - There has been 0.17 to 0.19C of warming in the surface record during the past decade relative to the previous decade
      - Even the central estimate of the short-term “trend” over the last decade is positive (although the error bars are so wide that the analysis is meaningless anyway)

      And yet some people still claim that “there’s been no warming in ten years … in fact, it’s been cooling.”

      … like the Calgary-based atroturf group Friends of Science in their
      recent Canadian radio ad campaign.
      (I’m in touch with Advertising Standards Canada about a possible false advertising complaint for that one).

      … or (surprise, surprise) a senior minister in the Canadian government.

      At a reception a few weeks ago, a senior minister in the Harper government stated as fact that the atmosphere was cooler today than in 1998, the inference being that climate change was a hoax, or at least not what it’s cracked up to be. It’s scary to have a senior minister believe such stuff, but this is the Harper government, after all.

      As the Copenhagen climate talks open, there’s lots of bogus science around…

      Yes, indeed, there is.

    • Steven M. // December 9, 2009 at 3:10 am | Reply

      Oakden Wolf: I believe there is about a 7 month lag between the occurence of El Nino and its maximum effect on global surface temperatures, so the 1997-1998 event was centered pretty well on the 1998 surface measurements, and the 2009-2010(?) event will be mainly seen in 2010. We may well get a record next year.

    • Sceptical Guy // December 9, 2009 at 3:43 am | Reply

      Nice post! And that’s from a sceptic! Well, I was an AGW-believer, then I became sceptical, now I admit this article has me wavering again.

      The only possible problem I can see is that your trend says nothing as to cause. Other than, “We predicted it therefore we’re right about the reason too”. However, if it stays within your prediction, I will change my point of view. If it falls outside, will you change yours?

      BTW – Even tho’ I’m a sceptic, I actually hope I’m wrong. Because if you guys are right and it IS us doing this, then we at least stand a chance of stopping it. If the sceptics are right and it’s all natural, then it really is a case of, “Adapt or die”.

      Please don’t be too harsh in your replies – I’m just trying to educate myself.

      [Response: The only thing this post shows is that, contrary to the claims of denialists, the last decade of temperature data does not contradict mainstream climate science. Those who make such claims do so out of ignorance or deceit or both.]

      • Tom Dayton // December 9, 2009 at 4:28 am | Reply

        Regarding causality, Sceptical Guy, I suggest you start by reading (or watching and listening to the slide version!) of cce’s The Global Warming Debate for a broad grounding. That will save you time in the long run, by giving you the knowledge to craft more pointed, better questions about specific issues.

        Then you should go poke around on, where you will find concise answers to those more pointed, better questions.

        • Sceptical Guy // December 9, 2009 at 8:07 am

          Thanks Tom. I will do both.
          Though you understand I must start out sceptical of a site claiming to get sceptical about global warming scepticism :)

          Looks like good info tho’. My journey continues…

      • Jim Galasyn // December 10, 2009 at 12:24 am | Reply

        Sceptical Guy, I’m glad to hear this. I always recommend reading a college-level text on climate science to solidify your understanding. Ray Pierrehumbert’s “Principles of Planetary Climate” is an excellent treatment and quite accessible. There’s a free draft online:

    • Tim Jones // December 9, 2009 at 3:44 am | Reply

      Certain aspects of natural variability such as cooler temperatures in the Northern Pacific Ocean due to the ‘07-’08 La Niña combined with the effects of a drop off in insolation due to the low ebb in the sunspot cycle last year provided for 2008 seeming to show a reversal of AGH warming. Climate change denialists are misleading everyone including themselves when they fail to recognize natural forcing have to be taken into account when calculating an overall trend. It’s quite likely the El Niño of 09-10 and the upswing in the solar cycle will put an end to their frenzied claims.

      • Kevin McKinney // December 9, 2009 at 5:09 am | Reply

        Yes, Tom they are misleading themselves at least–and the irony is that they frequently accuse the mainstream of failing to take natural forcings into account.

        Sometimes, on the other hand, they’ll tell you about the solar minimum (as if they just now discovered it personally) but fail to ask why, if the solar forcing is what it’s all about, was 2008 still one of the top ten warmest years?

        But it’s just these inconsistencies that let you know that it’s not about truth, it’s about “winning.”

    • Geoff Wexler // December 9, 2009 at 10:33 am | Reply

      Rattus Norvegicus

      You are right of course about Wegman’s remark about CO2 collecting near the ground. There are two categories however ; professors (experts) who display their ignorance of topics outside their expertise and those who do so within it. The first is easy to explain ; concentrating entirely on your day-time job increases your expertise but involves a price.

      The second category is a completely different matter and there is no need to spell out the explanation. Professors Wegman and Plimer are two examples outed by Tamino.
      Anyone more examples?

    • george // December 9, 2009 at 3:49 pm | Reply

      Arthur Smith says

      “how long before Lucia posts something scolding Tamino for his totally ridiculous errors in this post such as:

      * not realizing that what mainstream climate scientists predict is an acceleration in warming trend, not just continuing warming?

      I realize you are being sarcastic, but I think it is worth pointing out that the trend for the period 1975-2000 for GISS is 0.16+- 0.05 degC/decade, which actually encompasses the (rough) estimate indicated by the IPCC statement

      For the next two decades a warming of about 0.2°C per decade is projected for a range of SRES emissions scenarios.

      So, adjust the central trend line in the GISS graphic so that it has the slightly different slope — 0.2C/decade — ( and slightly different intercept) from the current one and adjust the bracketing error lines appropriately.

      It does not change the basic conclusion that the temperature values for the years 2001-present are consistent with the expectation.

      Perhaps Tamino would be so kind as to show that graphic as well.

    • Layman // December 9, 2009 at 4:43 pm | Reply

      ScepticalGuy, if I may ask, what moved you from “AGW-believer” to sceptic?

      • Sceptical Guy // December 9, 2009 at 11:28 pm | Reply

        Layman – It wasn’t really any one thing, more a collection of doubts that (to me) seemed reasonable. At the risk of getting my head chopped off (considering the site we’re on), the work of McIntyre and McKitrick got me questioning, as did Viscount Monckton, the errors in Al Gore’s movie, and things like the polar bears seeming to be doing OK now that we’ve stopped hunting them. Then I started reading sites like Icecap, the Resilient Earth, Watts Up With That, etc, and learning about the work of Fred Singer, Richard Lindzen, Nils Axel-Morner, etc, etc. There certainly seemed to be enough doubt to continue my search. And even Gareth Morgan in his book “Poles Apart” admits the science is far from settled (not that I’ve actually read the whole thing – being honest here). If I _had_ to pick a clincher, it would probably have to be the missing hot-spot in the tropical troposphere. Followed closely by the so-called ’saturation’ of the 15um wavelength of the earth’s surface radiation spectrum. I do always try to look for rebuttal of every article I read (from BOTH sides), as that is the true definition of ’sceptical’, and my favourite resource is – which is how I came here.
        So my journey continues. It may well end up being circular, but that doesn’t mean I shouldn’t take it :)

        [Response: OK let's be serious. If you can mention Monckton with a straight face and you don't know how horribly incompetent Anthony Watts is, then you've been seriously hoodwinked. And "hoodwinked" is the reason you think there's a "missing hot spot" and that "so-called saturation" is an issue.]

        • cce // December 10, 2009 at 4:39 am

          Here are five papers identifying enhanced tropical tropospheric warming in observational datasets:

          “In the tropical upper troposphere, where the predicted amplification of surface trends is largest, there is no significant discrepancy between trends from RICH–RAOBCORE version 1.4 and the range of temperature trends from climate models. This result directly contradicts the conclusions of a recent paper by Douglass et al. (2007).”

          “Insofar as the vertical distributions shown in Fig. 3 are very close to moist adiabatic, as for example predicted by GCMs (Fig. 6), this suggests a systematic bias in at least one MSU channel that has not been fully removed by either group [RSS & UAH].”

          “The observations at the surface and in the troposphere are consistent with climate model simulations. At middle and high latitudes in the Northern Hemisphere, the zonally averaged temperature at the surface increased faster than in the troposphere while at low latitudes of both hemispheres the temperature increased more slowly at the surface than in the troposphere.”

          “We find that tropospheric temperature trends in the tropics are greater than the surface warming and increase with height.”

          “At middle and high latitudes in the Northern Hemisphere, the zonally averaged temperature at the surface increased faster than in the troposphere while at low latitudes of both hemispheres the temperature increased more slowly at the surface than in the troposphere.”

          Re: Saturation

          Read anything from Gilbert Plass in the ’50s. e.g.

          “Skeptics” are only about 50 years out of date.

        • Gavin's Pussycat // December 10, 2009 at 11:18 am

          Hmmm… is an “interesting” site… set up by two philosophers.

          (Mental image of a site providing a balanced set of links relevant to the real existence, or non-existence, of gravity)

          No wonder you’re confused.

        • Sceptical Guy // December 15, 2009 at 12:23 am

          Guess I walked into that one. In my country, Layman, that would be called a “hospital pass” – Cheers ;)
          Gavin’s Pussycat – Of course, I see it now. That site must be rubbish for 2 reasons:
          1. The authors made the rookie mistake of thinking they actually had brains of their own and were capable of thinking for themselves
          2. It led me here

          Good-natured jibing aside – Of course you can debate the existence of gravity. People can debate anything. I would take the “it exists” side for 2 reasons:
          1. The evidence for this stacks up under scrutiny
          2. The opposing view (that everything sucks) does not
          Both are valid approaches to the argument. A hypothesis cannot be proven, it can only be falsified. Therefore, just as important as having a hypothesis survive falsification, so is having competing views not survive. To this end, you actually need the contrarians. You need their conclusions to be placed alongside yours to see what survives the light of scrutiny and what does not. I agree you should not have to refute over and over the same argument presented with slightly different wording, or deliberately deceitful and malicious ones (tho’ this is difficult to prove). But to dismiss all opposing views under these banners, or to simply describe the authors as foolish or incompetent, does nothing to advance your position or weaken the contrarians. It requires clear, concise reasoning, backed up with sound mathematics – as shown by Tamino in this article (tho’ IMHO it could’ve done with a bit less attitude, but I’m learning how to filter that out and cut to the chase – on BOTH sides).
          Climate science may not be in it’s infancy anymore, but public understanding of it is. But the public WANT to understand. They want to know what’s going on. They want to ask questions like, “Why ISN’T it the sun?” – and they want real answers. Not scorn or ridicule just for asking. This is where sites like and books like “Poles Apart” (which I am reading) come in. They amass all the evidence together, with no agenda, and let the laypeople see for themselves what stacks up, and what does not. Nothing wrong with that.
          Anyway, I appear to have gone off-topic. I do appreciate your patience with me, and I’m grateful to those who have genuinely tried to be helpful. Could I possibly trouble you with one more question – may I ask your opinion of the work of the Pielkes? Mainly Senior, but Jr as well. Cheers.

    • csa // December 9, 2009 at 4:46 pm | Reply

      I mostly liked that you are ready to answer the crticis’ remarks in detail. I cannot comment on what you edited out. I wonder if you have ever come across with Mr. Mickolczi. He wrote some computer program during a grant about I believe convections (hartcode). During another grant he let it run through two big satellite databases. What he found made him check out present global climate models and found that they use equations from 1922 (among others). This equation appeared to have boundary conditions that may be acceptable for stars but not for planets. He claims to have corrected it and found that the earth seems to be in a stable state concerning green-house gases. If we increase CO2, water vapor will decrease, and the balanced state will be restored. CO2 has mid to long term no impact on global warming. Please note, he does not claim that there is no global warming, he actually says it may be due to solar activities and stuff. He also says I believe that solar activities are reduced in the last decade so what appears to be a lack of warming since 2001-2002 may be a sign for that. Obviously too little data for statistics. He also does not claim that CO2 is benefitial or harmless. He wrote his paper while working for NASA but they did not like it. He had it peer-reviewed in a less-known paper. You may also find info about it under the name of Zagoni who was a big Kyoto1 supporter till he burried himself into Miskolczi’s paper. Too much controversy and too little constructive critics so far about it though. But worth dealing with it.

      [Response: Miskolczi has been shown to be not just wrong but foolish, in so many ways, by so many people, it no longer merits discussion. If you're really interested look here:

    • pough // December 9, 2009 at 5:10 pm | Reply

      I’m beginning to wonder if the confusion on what’s expected from CO2 – in terms of warming – comes from an authoritarian mindset. It’s as if they think that if CO2 is responsible for warming, then all other factors will stand aside and let CO2 completely run the show. If that’s not happening, then CO2 has no authority and can be dismissed as irrelevant.

    • Didactylos // December 9, 2009 at 5:28 pm | Reply

      There is nothing wrong with scepticism – it is the cornerstone of science. Without questioning and testing everything, we would get nowhere. Or, more likely, arrive at any conclusion – no matter how wrong.

      “Climate scepticism”, though, is just dressing up denial in a mantle of plausible pseudo-respectable science, using weasel words to mask the true nature of quackery.

      There are a few genuine sceptics hidden among the ranks of the deniers. They tend not to be heard, but this is largely because they have no significant contribution to add, not the result of any conspiracy. Their timid concerns are drowned out by the lies and manipulations from their own camp.

      It’s sad when words get twisted.

      Another word that has been overworked lately is “believer”. There should be no AGW believers. How can you “believe” in something that is there clearly to be seen? It’s like saying “I really believe in this desk”. It’s nonsense. Science isn’t a matter of faith. Again, this is one of the most fundamental scientific principles. Religion is a matter of faith. Science, famously, isn’t.

      Politics? Now that’s where my faith wavers…. Can our elected leaders see this through? I know that the UK hasn’t taken half of the promises to Copenhagen that I would have liked.

    • dhogaza // December 9, 2009 at 5:40 pm | Reply

      What he found made him check out present global climate models and found that they use equations from 1922 (among others).

      People use equations dating back to Newton all the time, to compute trajectories and the like.

      Why is using 1922-era physics a problem?

      • Rattus Norvegicus // December 10, 2009 at 5:38 am | Reply

        Because we obviously know more about radiative physics than we did way back when. Obviously, Fourier and Tyndall, who worked almost two centuries ago must have been wrong. [/sarcasm]

    • george // December 9, 2009 at 6:52 pm | Reply

      People use equations dating back to Newton all the time, to compute trajectories and the like. Why is using 1922-era physics a problem?”

      They may use the equations of Newton, but to be most accurate, they should really use the equations of special and general relativity, which were developed later…

      in 1905 and 1915.

    • dhogaza // December 9, 2009 at 8:02 pm | Reply

      Good one, george! :)

    • Geoff Wexler // December 9, 2009 at 8:18 pm | Reply

      According to Joseph’s point // December 8, at 2:17 pm, the BBC were wrong in repeatedly asserting that CRU might have been trying to exaggerate the effect of global warming. In the highly unlikely event that the British data has to be thrown away the effect would be the opposite of what the BBC headlines suggested, i.e to make the recent warming (estimated incorrectly, of course, by means of short term regression lines) slightly worse. More here:

    • David B. Benson // December 9, 2009 at 10:15 pm | Reply

      Sceptical Guy // December 9, 2009 at 3:43 am — For a historical perspective, please read “The Discovery of Global Warming” by Spencer Weart:
      after reading Andy Revkin’s review:

      • Sceptical Guy // December 10, 2009 at 12:01 am | Reply

        Thanks David – I will. I had already come across the chapter about Roger Revelle’s Discovery, but will go deeper.

    • Arthur Smith // December 9, 2009 at 11:58 pm | Reply

      Took her a little longer, and she was much more subtle about it than I expected:

      • Layman // December 10, 2009 at 3:02 am | Reply

        OK, could someone please clue the newbie in to the inside joke here? Help! Thanks.

      • Ray Ladbury // December 10, 2009 at 10:59 am | Reply

        Hmm, how to put this charitably. Let’s say, Lucia and Tamino have differences in how they interpret predictions of climate models vs. data. Some of her analyses might be termed “creative”, but they lack much physical insight, IMHO.

        I do not think the Lucia is so much ideologically motivated as perhaps a true contrarian. Let’s face it, the most exciting times in science occur when a longstanding model is overturned. At some point, though, you have to accept that a theory is so successful, that it is bloody unlikely to fail completely. So it is with the consensus model of Earth’s climate.

        I would add, that my opinion of Lucia’s motives is more charitable than that of some others.

    • Barton Paul Levenson // December 10, 2009 at 1:02 am | Reply

      Hey, Tamino, thanks for citing my Miskolczi essay! I always feel good when somebody takes one of those seriously. : )

    • Hank Roberts // December 10, 2009 at 4:02 am | Reply

      L., just compare the picture to the ones above and imagine what you’d think if you were the kind of person who just liked to look at the pictures to decide what to think.

      As it says over there,

      “This method minimized the appearance of divergence…. The reason this is useful is that people can say “Hmm… ”
      That’s the in joke, that it’s so predictable.

    • jyyh // December 10, 2009 at 3:29 pm | Reply

      good point , the authoritarian one, if someone still has an issue with various forcings, the old metaphore (on a class, way back) goes like this: “Basically, think of it like a tug-of-war with multiple ropes attached to each side. usually there’s the one with endurance (the sun) on one rope, and then on the other one there is the a skinny one (CO2) who has taken some pills (fossils), third rope pulling the same direction might be a strong man (methane) who doesn’t last too long, yet there are the animals and the decaying vegetation. On the other side of the flag that indicates the temperature there are the enduring living plants, strong man (sulfates), and the skinny albedo who has been on bad food (soot). Where do you think the flag goes?”

    • george // December 10, 2009 at 5:17 pm | Reply


      Lucia likes to focus on the short term (2001 onward), which is dominated by uncertainty due to weather noise (the year to year ups and downs in the temperature due to things like el nino and la nina).

      Weather noise can (and does) have a significant effect on temperatures (and “trends” based on them) over the short term (a trend is line fit to the data that gauges whether temperatures are increasing, decreasing or “flat” over some period and where they might be headed in the future (and by how much for each passing year)

      Weather noise makes a comparison of short term trends (2001-onward) in the observed temperature data to the IPCC’s AR4 trend estimate for the first two decades (“about 0.2C/decade”) of dubious value.

      Weather noise also makes a comparison of observed temperatures to the temperatures projected by various model runs over the short term (or even the average output of a number of such runs) of dubious value.

      The models are climate models (not weather models), which means they are not even intended to tell you what will happen over the short term (at least not with any confidence).

      In science, it is critically important to ask the “right” questions — ie, the ones that can be answered with a fairly high level of confidence.

      Tamino has asked — and answered one such question above:
      “Are the temperatures of the last 10 years what was expected based on (consistent with) the trend of the previous 25 years?”

      They are.

      The lines around the central trend line on Tamino’s graphic are meant to “bracket” the weather noise, which goes up and down essentially randomly but usually stays within some distance of (above or below) the center line over the longer term(decades).

      When one makes a prediction, the expectation is that the weather noise will behave in a similar manner. It will bounce around, but the bounces will usually not be higher than some amount.

      That is what is shown by Tamino’s graphic.

      What Lucia has “shown” is anyone’s guess.

    • Christopher // December 10, 2009 at 6:49 pm | Reply


      I think what prompted Kevan’s “thoughts” is best expressed here:

      I like your graphs better tho (longer time series, multiple data sources, two snapshots).

      A final quibble. The graphs do not show that the increase in temperature is anthropogenic, i.e., “… that human activity is warming the planet”. Unless you assume time is a proxy for human activity. This is merely a semantic quibble tho :)

    • luminous beauty // December 10, 2009 at 8:34 pm | Reply

      I notice that Lucia’s A1B average is based on modeled surface air temps, which she is comparing to HADCRU3 and GISSTEMP Surf+SST temps.


    • Chad // December 10, 2009 at 8:42 pm | Reply


      Why the “Hmmm?”

    • jg // December 10, 2009 at 8:48 pm | Reply

      Thank you for this topic. With your permission, I’d like to post your second GISS graphic on my blog where I’m inviting local denialists harping on the CRU hack bandwagon to discuss the science. As they are also calling for integrity in the peer review process, and since you are the only scientist I know of who referees papers, I want them to see your site. I’ve posted your graph and linked it to this topic.
      John G

      [Response: Feel free to copy the graphic.

      Most scientists end up refereeing papers eventually.]

    • Deep Climate // December 10, 2009 at 8:56 pm | Reply

      I think the Rahmstorf approach (which was discussed a couple of times here wrt to TAR projections) of comparing smoothed obs to smoothed projections makes more sense. In the short and medium term all the scenarios are virtually identical anyway. The interannual wiggles in the model projections are not relevant – the model mean ensemble had a local peak in 2008, but the obs were well down, of course.

      Even doing it the Rahmstorf way, though, it’s too early to assign much meaning to the analysis.

      Nevertheless, I gave it a shot earlier this year:

      So for 2000-2008, the IPCC smoothed projection was an average of 0.33 deg +/- 0.13 deg (90% confidence interval) above the 1980-99 baseline. Both NASA GISS (0.26 deg) and HADCrut (o.25 deg) were within that range, albeit in the lower part.

      Lots of caveats etc. One important one is that if solar forcings and ENSO fluctuations were accounted for, the obs would come even closer to the central estimate, IMHO. The projections, of course, assume neutral natural forcings.

    • Deep Climate // December 10, 2009 at 8:59 pm | Reply

      I think both model prj and obs are global (land+ocean surface), so the comparison data sets look right to me.

    • luminous beauty // December 10, 2009 at 11:46 pm | Reply

      Deep climate,

      I don’t know, but it looks to me like the model tas (air surface temps) over oceans is about 1C higher than the sst’s of models that have an sst output field. There is something called a land/ocean mask available for some data sets, but I haven’t found out how to apply it or what it does, yet.

      I can’t say for sure what difference that makes, and I don’t know exactly what buttons Lucia punched, but I’m skeptical, to coin a phrase.

    • Maarten Nieber // December 11, 2009 at 12:38 am | Reply


      I have skimmed through the comments (there are too many to read all), and it seems to me no experts on statistics have answered. Have you tried to get your ideas verified with an expert?

      [Response: I am an expert.]

      I think that if you really want to make a point, you should go the official way: formulate a hypothesis, explain which tests you will use to test the hypothesis, do the actual analysis and then draw conclusions from that.
      The “eye-balling” approach is good to get a first impression, but not enough to make a clear point in my opinion. Indeed you may be right that the more recent data is not in contradiction with a model that predicts a positive slope; then again, the same data might not contradict a model of zero slope either. A more convincing approach for me would be if you could show that the hypothesis of having a zero slope over a reasonable period of time can be refuted (leaving the question open of which period is reasonable).

      [Response: It's clear that the data for brief time spans is too sparse to establish the direction of the trend. That's the point of this post. The only claim I've made about the trend based on such brief time spans is that the time span is too short.

      It's those in denial of global warming who have attempted to use the last 10 years (or 12 years or 9 years or 8 years or 7 years) of data to conclude that the observed behavior contradicts expectation. The only theme here is that they're wrong.]

      As an aside, from your remark about the “slopes since 2000″ being positive, I get the impression that you are not applying the concept of confidence intervals, and from what little I know of statistics, using the CI is essential in interpreting statistical results.

      [Response: Again: I did not claim that the data since 2000 establish an upward trend. I was merely showing that Kevan Hashemi was wrong to state that those data establish a downward trend.

      As for confidence intervals, note that they're right next to the slope estimates (after the "+/-"). That is done to drive home the point that using only the last 10 (or 12 or 9 or 8 or 7) years of data to make statements about warming trend (which is what those in denial of global warming do) is foolish and wrong.

      It's the denialists who have consistently drawn outlandish conclusions from time spans that are way too short to do so.]

      Best regards
      Maarten (neither a denier nor a believer of the effect of human actions on the climate, I really don’t know)

    • luminous beauty // December 11, 2009 at 1:05 am | Reply


      I think I got it backwards. sst’s are higher than ocean tas’s. I guess the land/ocean mask is just the land points only and ocean points only buttons.

    • Steven M. // December 11, 2009 at 2:58 am | Reply

      It looks like my suggestion above of a possible record in 2010 beat the UK Met Office by a day (I admit that it’s quicker to crunch the numbers if you skip the step of crunching the numbers). It’s a bit depressing though that the BBC photo caption says “A weakening El Nino event will lower global temperatures” when the Met Office is saying that if El Nino weakens before expectation the warming will not be as great as their projection. Are there no reporters who know the difference between a first and second derivative?

      • Kevin McKinney // December 11, 2009 at 5:09 am | Reply

        It’s also odd that their last sentence says that 2009 was “pushed into” the top ten warmest years by a “burgeoning El Nino” when the entire decade* has made it into the top ten warmest years.

        (Though 2008 will be forced down into 11th this year in the Hadley ranking (but not the American rankings.))

        *Since the decade properly began with 2001.

    • Jim Eager // December 11, 2009 at 6:02 am | Reply

      Have you tried to get your ideas verified with an expert?

      You must be new here, Maarten.

      Tamino is an expert time series statistician.

    • David Horton // December 11, 2009 at 9:17 am | Reply

      Tamino, you might want to take a look at the particularly egregious use of graphs here

      • Glenn Tamblyn // December 11, 2009 at 1:00 pm | Reply

        Yep. But this is the IPA after all. Bastion of middle of the road politics and balanced analysis.

      • Layman // December 11, 2009 at 9:17 pm | Reply

        So here is my standard appeal for some basic observations as to why this article is no good. Your insights are appreciated!

        • David Horton // December 11, 2009 at 11:11 pm

          Because, among other things, he does a plot of temperatures in the last thirty years. Or rather, temperature departures from An Average of Temps over 1979-1998. Now, how do you imagine that comes out? Well, for a start, it means that there can automatically be no temp rise for that 20 year period – do you see? And then, ponder the question of why choose 98 as the cut off point for the average? Got it? Ah yes. makes it likely that even the next few years will be pulled down. And even with all that, the temperature increases.

          And his two graphs suggesting that urban heat islands explain all the global warming, don’t show that, both town and rural temps rising.

        • Deep Climate // December 12, 2009 at 7:48 pm

          The graph shown is of UAH, which shows the least warming of all the data sets. Use of 1979-1998 baseline is fine as it is the standard one for both satellite sets (UAH and RSS). Anyway, the choice of baseline will not affect overall trend.

          The graph is actually from Roy Spencer’s website, so it is semi-official I suppose. He used to show a polynomial curve (with a misleading downward end segment), something I blogged about a while back.

          He really should be showing the linear trend (which is at 0.13C/decade IIRC), as the official data set does. The rolling 13-month curve cuts out the monthly variation, but doesn’t really add any useful information. Still you can see the trend is upward in general since, the recent temperatures are all above the baseline average.

          I also suspect if we have a moderately strong El Nino as predicted the curve may well rise close to or above the 1998 maximum by the end of next year. I wonder what he’ll do then.

          The UHI Australia graphs are only up to 1991, so the study is 17 years out of date and counting.

          There are likely other problems with it, but when it’s so far out of date, why not turn to the latest research? Just a guess, but perhaps because it shows inconvenient temperature rise everywhere in Australia over the last two decades.

          At least, ABC Unleashed identified the IPA connection, although a description of its mandate, stance and funding would have been more transparent. But it’s an improvemnet on the Plimer’s bio, which left out his affiliations with coal mining companies, and with the skeptic disinformation group International Climate Science Coalition.

          Still, it continues ABC’s tradition of inviting comment from those with no discernable qualifications or expertise in climate science. Perhaps that’s because there are no Australian “skeptics” with such expertise. If so, the obvious course of action is not to run such tripe at all.

        • Layman // December 13, 2009 at 12:27 am

          Thanks for the comments! I did not even catch that the graphs at the end only went to 1991 – what the heck is that??

          I think I understand the point about using the average for 1979-1998 as the baseline. I guess my follow up question then is with respect to the graphs presented at the top of this page. They have a baseline? If so, what is it? Or is it a completely different type of graph/analysis? Sorry if my questions are not the most well thought out, I am still working on the basics!

    • Barton Paul Levenson // December 11, 2009 at 10:41 am | Reply

      cce–I can’t get the Sherwood or Vinnikov links to work. Do you have a citation?

    • george // December 11, 2009 at 3:17 pm | Reply

      Ray says “I do not think the Lucia is so much ideologically motivated as perhaps a true contrarian. ”

      I think you have it right, Ray.

      Why else would she still be pushing the “short term” bean around after everyone and his brother has pointed out that short periods of time (decade or less) are not reliable indicators of what is happening with the climate?

      Of course, a contrarian takes the opposite view no matter what you say, like the guy in the Monty Python argument sketch.

      And, as in the sketch, the arguments that result are usually not particularly insightful, helpful or even meaningful.

    • Ray Ladbury // December 11, 2009 at 4:22 pm | Reply

      Yes, and we all know how hard it is not to say,
      “Don’t give me that, you snotty-nosed heap of parrot droppings…”

      I really think that Lucia’s problem is a complete lack of understanding of the physics. To her, it is all moving numbers around–a sort of Soduku in front of a live audience.

    • Hank Roberts // December 11, 2009 at 6:25 pm | Reply

      Comments found at Chad’s site
      on Chad’s statistical charting work:

      > Chad,
      > Great work….
      > Comment by Chip Knappenberger

      > Very interesting! This really will help …
      > Comment by lucia


    • Chad // December 11, 2009 at 7:47 pm | Reply


      What may I ask is the point with the quote from Lucia?

    • Deep Climate // December 11, 2009 at 8:52 pm | Reply


      What may I ask is the point with the quote from Lucia?

      … since the point with the quote from Chip is obvious to all, he said helpfully.

    • Hank Roberts // December 11, 2009 at 9:30 pm | Reply

      Just — hmmmmm.

    • Chad // December 11, 2009 at 10:34 pm | Reply

      Just — hmmmmm.

      That clears it up.

    • FredT34 // December 11, 2009 at 11:02 pm | Reply

      Nice again ! I wondered if such graph had been made with US temperatures (and even with “urban-heat-free” temps) ? Might be interesting if even those ones were climbing, too…

      (I started fighting with R for a week-end 6 months ago and could get time to go further so far…)

    • dhogaza // December 13, 2009 at 1:15 am | Reply

      I think I understand the point about using the average for 1979-1998 as the baseline.

      Just to make it clear, the first UAH satellite reconstruction showed up right around the end of the millennium, so this baseline represents the span of years in the first analysis.

      Then ongoing anomalies are given relative to the original baseline given by that first analysis …

      RSS didn’t start doing their own reconstructions until they were certain that UAH’s had serious errors and I think it’s safe to assume they adopted the same baseline for convenience.

    • Hank Roberts // December 13, 2009 at 3:15 am | Reply

      hmmmm — Tamino, have you access to the full text of this one?

      First author Perlwitz was mentioned here last year, re an earlier paper.

    • Hank Roberts // December 13, 2009 at 3:19 am | Reply

      Oh, and two of those authors and others were a few years ago, if I understand the abstract, arguing that”NAO plus noise explains the trend” — I”m not at all sure I understand that right though.

    • Chad // December 13, 2009 at 8:00 am | Reply

      The full text is freely available. Here’s the original submitted version:
      and here’s the revised version:

    • mspelto // December 13, 2009 at 11:32 am | Reply

      Hank-”commingled with a decade-long fall in global mean temperatures”- How do they come up with that in the Perlwitz et al., paper. In fact they will take some heat on that statement from colleagues. The second group is from authors that I have cited. They are really explaining that NAO trends are being influenced by increased SST and they cannot explain the recent enhanced winter NAO pattern without conisidering the impact of the warming of the tropical waters.

    • Sekerob // December 13, 2009 at 1:35 pm | Reply

      MLO reports November up by 1.61 ppmv, on the high end of the October to November change series since 1958… when fossil fuel burning is quite a bit less this year than last year… Trend 388 ppmv. SOI for November negative.

    • Riccardo // December 13, 2009 at 4:17 pm | Reply

      just to better understand what mspelto explained, here is the last paragraph of Hurrell et al. paper:

      “Finally, it could nonetheless be true that both the tropospheric and stratospheric circulation trends over the last half of the twentieth Century have been tied to increased GHG concentrations, even though the source for our attribution work has been AGCM simulations using fixed atmospheric chemistry. This is because the history of prescribed SSTs likely contains an anthropogenic component, including the observed warming of Indian Ocean surface waters (e.g., Knutson et al. 1999). That the close match to the observed warm pool SST time series since 1950 by the historical CSM simulation (Fig. 13) cannot be attributed to natural variability in the coupled model is demonstrated in Hoerling et al. (2004), where we also highlight similar behavior in historical simulations with other coupled models. Thus, the results presented herein do not discount the existence of an indirect greenhouse effect through the dynamical response to the tropical ocean changes of the last half-century.”

    • Layman // December 13, 2009 at 4:46 pm | Reply

      Following up on my question above about the graphs presented at the top of the page (GISS, RSS, UAH), the graphs show temperature anomolies – this means temperature difference from some reference value? Do I understand that correctly? If so, what is (are) the reference value(s) used? As always, my apologies if I’m not getting the terms or questions right. Any and all insights always appreciated.

    • dhogaza // December 13, 2009 at 6:12 pm | Reply

      this means temperature difference from some reference value? Do I understand that correctly? If so, what is (are) the reference value(s) used?

      Yes, the UAH and RSS reconstructions use the base period explained above, I forget offhand what base periods GISTEMP and CRU use (they differ).

      I forget because they’re unimportant, choice of zero point doesn’t affect the trend.

    • RedLogix // December 13, 2009 at 7:20 pm | Reply

      I spend much of my life as an automation engineer looking at noisy real-time trend plot data from all sorts of processes. After doing this for 30 years or more my ‘eyeball’ science is as useful to me as is Tamino’s (far more sophisticated) statistical analysis. Personally I eyeball all these temperature graphs going back a century or more… and reality of AGW is bleedingly obvious.

      That many other ordinary folk like me are being deceived into not seeing this simple fact is the result of a very sophisiticated and professional campaign of misdirection.

      These sorts of propaganda campaigns have nothing to do with truth. Have you not noticed that pointing up their basic errors (some so laughable even I can see them unaided), is never acknowledged or admitted? That it never prompts them to re-think anything? Have you ever noticed that they literally never produce any new data or research of their own, that all they do is pick and carp over the work of others?

      All this is because the truth is irrelevant to them. Acknowledging error would imply intellectual integrity, and that would not serve their purpose. Original research would imply an honest endeavour to work with real information, and that does not serve their purpose either.

      Their sole purpose is to deceive, and their behaviour is proof of this. When the history of this epsode in human life is written, the root sources of the controversy will be traced to a small handful of professional agititators who have knowingly and malicously concocted lies in order to serve the interests of their masters. My contempt for these miserable traitors to our future is complete.

      The only good thing about the deniers and their entirely manufactured fear, uncertainty and doubt campaign, is that I’ve been compelled out of curiosity to learn the underlying science and maths. For this Tamino has been my first stop for some years now, and as a long time lurker I want to thank you for the fantastic effort you have made and how grateful I am for it.

    • Hank Roberts // December 13, 2009 at 7:47 pm | Reply

      Layman, an example of picking a baseline:

      The ‘zero’ line there is the “1978-2000 mean”

    • Chad // December 13, 2009 at 8:16 pm | Reply

      GISS uses the climatological mean of 1951-1980, HadCRUT 1961-1990, NCDC 1901-1000, RSS and UAH 1979-1998.

    • David B. Benson // December 13, 2009 at 8:38 pm | Reply

      Two bos global temperature response — A linear two bos model for global temperature response to forcings suggests extending the observations in
      to determine the best fitting value of R in

      T(t) = (R/ln(2))*f(t-d)

      where T(t) is the temperature at year year minus the temperature in 1880 CE, f(t) = ln(CO2(t)/CO2(1880-d)) and d is the expected lag from the two box model, about 16 years.

      The ln(2) bit is so R corresponds to the temperature increase to a doubling of CO2 concentration. The result is only roughly comparable to the TCR values from different GCMs, but I predict is about 2 K.

      I’m not set up to do the calculations but now that we have 130 years of GISTEMP (close enough) the value of R might be of some general interest.

    • Nick Barnes // December 13, 2009 at 8:54 pm | Reply

      Layman: essentially local mean temperatures are calculated monthly (e.g. “how warm was Seattle on average in November 2009?”), and then compared to local mean temperatures for that calendar month over some baseline (e.g. “how warm was Seattle in November on average from 1979-1998?”) to calculate an anomaly (“how much warmer than usual was Seattle in November 2009″). Then these local monthly anomalies are combined into zonal, hemispheric, and global monthly anomalies (“how much warmer than usual was the world in November 2009″), and these are also then averaged over a year to calculate annual anomalies (“how much warmer than usual was the world in 2009″).

      The order in which one chooses to do the averaging makes no difference to the result, and the choice of baseline period just adds or subtracts a constant to the answer. Different global surface temperature anomaly datasets mainly differ on the source data (some include sea-surface temperatures, some have different sets of surface stations, etc) and on the method of geographic averaging (e.g. GISTEMP defines an 8000-cell grid, and surface stations contribute to a grid series if they are within a certain distance, with weights depending on their distance).

      I hope this helps.

    • Gavin's Pussycat // December 13, 2009 at 10:06 pm | Reply

      you made my day.

    • Hank Roberts // December 13, 2009 at 10:40 pm | Reply

      > NCDC 1901-1000


      It’s good to cite sources when giving information — that way people can look this stuff up for themselves, rather than just trusting some guy on a blog.

      NCDC’s explanation is at:

      ———–excerpt follows————
      7. How is the average global temperature anomaly time-series calculated?
      [see original at the link for details] … Global-average anomalies are calculated on a monthly and annual time scale. Average temperature anomalies are also available for land and ocean surfaces separately, and the Northern and Southern Hemispheres separately. The global and hemispheric anomalies are provided with respect to the period 1901-2000, the 20th century average.

    • John Mashey // December 14, 2009 at 5:16 am | Reply


      tamino is not only an expert, BUT he also demonstrates skills that not all experts have.

      a) Some technically-expert folks jump into detailed esoterica at the drop of a hat, and even more, seem to want to wallow around in there, and lose track of what one is really trying to do.
      Sometimes, that’s a purposeful misdirection tactic: use a lot of formulae to impress people. (I wouldn’t want to name names).
      Sometimes, it’s just the way people think if they are natural complexifiers. (I’ve had to fire a few like that.)

      A rather good statistician where I used to work named John Tukey used to whack this stuff hard when he saw it, and our organization was quite serious about using good statistics,so people listened.

      b) You should follow this blog because tamino is just the opposite of such complexifiers. He can give lucid discussions at various levels of sophistication, and explain each quite well and appropriately for the level. Even better, he tends to use the *simplest* techniques that work well enough, rather than complexifying. I think Tukey would approve, were he still with us.

      c) Hence, if you want to learn, this is a good blog to follow, because
      - some things may be over your head
      - some may be just right to learn something new.
      - Finally, even if you already know about something, it is well worth watching how an {expert+good communicator} presents it.

      • guthrie // December 14, 2009 at 9:24 pm | Reply

        John Tukey is regarded as a rather good statistician? He was mentioned in some of the stats I’ve been doing recently, I kind of assumed he was pretty good. I guess it makes me feel a bit better about hanging out here.

      • Maarten Nieber // December 15, 2009 at 8:07 pm | Reply

        Hi John,

        thanks, I will keep that in mind. Right now, I am just curious to learn about what the discussion topics are within the climate debate (I suspect that the regular media are not the best information source for this), but if I want to get a more technical explanation of some issue, I will look for his posts.

        Best regards

        • John Mashey // December 17, 2009 at 5:22 am

          Depending on where you are, first build a coherent knowledge base, like by reading:

          1) General: David Archer, “The Long Thaw”
          or for more detail, with a bit more math & physics:

          2) David Archer, “Global Warming: Understanding the Forecast”. (textbook for undergraduate science majors).

          3) Spencer Weart, The Discovery of Global Warming: Revised and Expanded Edition”, or online at AIP.

          If you already know all that, then try IPCC AR4 WG I Technical Summary, and if that’s not enough, go the full report, and read at least the first chapter of the WG II report.

          About the most confusing thing possible is to plunge into the blogosphere without first building a solid base of understanding. You can easily be pulled into a strange alternate universe where CO2 isn’t really a greenhouse gas, conservation of energy isn’t a law of physics, and the statistics of noisy time series are suspended. Some people never escape.

    • Kevin McKinney // December 14, 2009 at 5:26 am | Reply

      Second that, RedLogix.

    • cce // December 14, 2009 at 7:16 am | Reply


      Sorry I missed your comment. The Vinnikov link works for me, but the Sherwood link is dead. Here is a new link:

      In any case, these are the cites:

      Temperature trends at the surface and in the troposphere
      Konstantin Y. Vinnikov,1 Norman C. Grody,2 Alan Robock,3 Ronald J. Stouffer,4, Philip D. Jones,5 and Mitchell D. Goldberg2
      Received 20 June 2005; revised 12 October 2005; accepted 7 November 2005; published 11 February 2006.

      Robust tropospheric warming revealed by iteratively homogenized radiosonde data
      Steven C. Sherwood, Cathryn L. Meyer and Robert J. Allen Yale University, New Haven, CT, USA (, Holly A. Titchner Met Office Hadley Centre, Exeter, UK

    • cce // December 14, 2009 at 7:51 am | Reply

      Here’s a newer paper on radiosonde homoginization. They take some model data, trash them like the radiosonde data have been trashed, and then see if their homoginization algorithms can recover the original trends.

    • Sekerob // December 14, 2009 at 10:41 am | Reply

      I trust that NCDC and others are fully weighting in the Hemispheric ratios of land versus sea for some seem to just add the NH/SH halves together and divvy this by 2. E.g. without naming a supplier of land temps.

      2008 NH 0.768 anomaly at ~ 2/3rd
      2008 SH 0.303 anomaly at ~ 1/3rd
      2008 GL 0.535 anomaly = plain average

      Seems wrong to me, but I’ve probably overlooked something or there being a logic why that is done… Could speculate on this here, but won’t.

      PS: No references for you Hank to check personally ;>)

    • Hank Roberts // December 15, 2009 at 1:12 am | Reply

      > Layman // December 14, 2009 at 3:44 am
      > The link is great, thanks. I was just about to
      > ask question number 6!

      And at 3:44 am! That does make me smile.

      But — always, remember, I’m an amateur, I take a few seconds, look for some examples, but I rarely make the effort to find the _best_ information. I just try to show how easy it is to begin looking for stuff.

      Once you find what might be a good answer, always,when you post something based on that, point to your source. It might be the best one to rely on; or not.

      Someone told me back in the 300-baud acoustic coupler days: asking questions on the Internet is like walking into an immense, dark library and shouting out your question, and getting several different answers back. Then your work begins.
      Shorter version: “Post what you think, and await correction.”

    • Maureen Vilar // December 15, 2009 at 4:37 am | Reply

      I’m not a scientist but am a moderator on the ClimatePrediction (CPDN) forum:
      At the moment we have a new member who knows that the oil film on the ocean surface decreases evaporation. He is convinced that this will lead to decreased cloud cover and thereby an increase in net global heat loss, ultimately leading to global cooling.

      He does not appear to have considered whether this effect, which seems to me plausible to some extent, is large enough to more than offset the warming effect of the increasing atmospheric concentration of greenhouse gases.

      I can find no studies of the effect of the marine oil film on global warming. Does anyone know so we can respond in more detail to this member?

      • Kevin McKinney // December 15, 2009 at 1:18 pm | Reply

        I don’t have that information, and don’t know where to look, unfortunately.

        However, I’d note that your member should consider that “cloud cover” is hopelessly vague. The effects of cloud are not simple–they can be either warming or cooling, depending on altitude and (IIRC) droplet density and location, and I believe there’s a lot of work still underway to clarify this. He’s stepping into a really thorny problem, it seems to me.

        Another question that could be asked is, what do data on humidity levels tell us? He is assuming atmospheric water vapor content as a driver of cloud cover, is that correct? If so, that could be an independent check on the notion.

    • Ray Ladbury // December 15, 2009 at 10:59 am | Reply

      Sceptical Guy, I have to admit that I find it a bit odd that you would choose to listen to Watts and Monckton over the 90% of climate scientists who admit the reality of anthropogenic climate change.

      One thing I would counsel, though, and this in all seriousness. When you look at a “scientific” argument, ask yourself whether it is looking at all, or at least the preponderance, of the data. One common characteristic I have seen in all anti-science is that it seizes on a single result or analysis or feature that looks odd and then tries to argue that the scientists (you know, the guys who’ve studied the field for 30 years or more) have got it wrong. Whether it’s a “missing transitional fossil” or a “missing hot spot”, they try to get you to focus on only that aspect while ignoring the astounding completeness of the fossil record or the astounding success of the consensus model of Earth’s climate. Another thing you will notice is that the target audience of such tripe is usually non-specialist.

      Science is a human endeavor. The process is not always pretty. However, in the end, it delivers truth more reliably than any other human endeavor. That is worthy of consideration.

      • Sceptical Guy // December 16, 2009 at 12:21 am | Reply

        Thanks Ray. That’s sage advice – and I agree with you (except that I wouldn’t compare the “astounding completeness of the fossil record” to the “astounding success of the consensus model of Earth’s climate” _just_ yet – but maybe soon ;). When I started this journey, many people would tell me that “the science is settled”, but very few could tell me what that science actually was – meaning they didn’t know. With hindsight I admit I probably started from the wrong point of, “Well, what’s the OTHER side of the story?” and got a bit carried away with that. I also admit to being seduced by Watts’ project – seemed so simple – “Hey, has anyone checked how good the thermometers are?”

        Like I said, I do always try to look for rebuttal of every article I read (from BOTH sides), but probably not as thoroughly or as diligently as I should. But I am getting better, and I’m also getting better at spotting and filtering out the bunk in some websites I come across.

        This is also why I asked my last question. May I ask your opinion of the work of the Pielke’s? Especially Sr.? I guess I want to ask, “Are there any scientists on the sceptical ’side’ (for want of a better term) that you think ARE worth listening to?”
        E.g. – Do Dr Spencer’s views on intelligent design (I’m a non-religious evolutionist BTW) totally invalidate all his climate work? Does Axel-Morner’s belief in water divining make ALL his sea-level work bunk?

        I totally agree with you about science. I see it as a perfect practice, practiced by imperfect beings, eventually leading them towards truth.


        • Maureen Vilar // December 16, 2009 at 9:21 am

          Sceptical Guy, I don’t think a person’s thought processes in one area can a priori be considered indicative of their thought processes overall.

          Isaac Newton wrote a treatise on alchemy and tried to turn lead into gold. That doesn’t invalidate his other work. We’re all part of the society and culture we’ve lived in – probably not very scientific even now.

          If we tried to apply the scientific thought processes of our area of expertise to every aspect of life we might become obsessive and dysfunctional. We do need to think about different things in different ways.

    • Ray Ladbury // December 15, 2009 at 1:53 pm | Reply

      Maureen, I would consider several things. First, what is the percent solid angle of the oil slick. Second, is there a reasonable-sized chunk of ocean upwind. Third, clouds can warm as well as cool. Fourth, an oil slick has a finite lifetime and breaks up on a timescale of months.

      Personally, I think that if we get to the point where enough of the ocean is covered with oil that it makes a difference in solar radiation reaching the ground, that will be the least of our problems.

    • Hank Roberts // December 15, 2009 at 4:21 pm | Reply

      Maureeen, that might be a serious question, if so Dr. Le Quere’s work on biological feedbacks would be a good place to start. We keep finding far more life and far more complicated life in places we thought were just simple substances; the thin film at the top of the ocean is full of newly discovered living organisms. Some hits here:

      But ‘oil’ might come from a guy with a book about world wars’ military marine activity causing climate change, who pops up under many guises. Try Google with some phrases from the person’s posting and see if they’re often used.

    • David B. Benson // December 15, 2009 at 8:05 pm | Reply

      Maureen Vilar // December 15, 2009 at 4:37 am — The group out of Rome, Italy, which tracks global percipitation has found no cchange in 25+ years of data; another effort finds a slight increase in 50+ years of data. Both are consistent with a nearly constant global relative humidity. As the temperature of the air increases, its absolute humidity must then be increasing. This implies that evaporation must be very slightly increasing to maintain the relative humidity at close to constant.

      Therefore I conclude that the oil slick problems, while acute locally when a spill occurs, have no observable effect on global evaporation rates.

    • Maureen Vilar // December 15, 2009 at 10:11 pm | Reply

      Thanks to everyone for the useful ideas and information.

      Athough we all know there have been local changes in precipitation, eg probably more in Antarctica but less in SE Australia, I did think that if there were important global changes in measured or predicted precipitation we’d all have heard about them.

      The posting member has indeed written a book ( which, judging by voluminous postings on their forum-cum-website, pursues the same thesis: the oil film problem is the fault of the oil companies (not caused by war).

      The graphics of the climate models I run at home don’t predict future global drying apart from the occasional result that’s an outlier among the ensemble, though regional changes show quite frequently in model types whose results allow this display.

      I think the scenario where the oceans heat up because the oil film doesn’t allow ocean heat to transfer to the atmosphere while the land, largely devoid of clouds, cools down is one person’s nightmare. It’s certainly a distraction from the usual sceptical arguments. Let’s hope it’s a complete red herring.

    • David B. Benson // December 16, 2009 at 1:47 am | Reply

      Sceptical Guy // December 16, 2009 at 12:21 am — Roy Spencer is a meteorologist; when he just did that he made a definitve, even prize-winning contribution (study of MJO).

      That said, hiw more recent attempts regarding climate response to CO2 are deeply flawed, even sometimes blundering; there is a series of thread here on Open Mind about one such blunder. In particular, it seems he refuses to consider the evidence offered by the paleodata; ice core proxies. So he has published recent papers which are wrong on the face of it and AFAIK nobody bothers to attempt peer-reviewed commentaries, in contradistinction to Schwartz’s erroneous paper(s).

    • Ray Ladbury // December 16, 2009 at 2:21 am | Reply

      Sceptical Guy,
      I think you are starting to ask the right questions. Certainly, some science is settled–we understand thermal radiation quite well and the mechanism of greenhouse warming. All of that, we’ve understood for over a century.

      That CO2’s effect is not saturated we also know, and have done so for ~60 years. And I think you will find that the important aspects of the theory of Earth’s climate are well understood. These, unfortunately are precisely the aspects that point toward anthropogenic climate change.

      Since you ask my opinion on the Pielkes, I would say that they are certainly a cut above the average denialist–they at least admit there is a greenhouse effect. I’ve been disappointed with their tendency to fixate on one fact or analysis while ignoring mountains of evidence that contradict their position. I don’t think they have a deep understanding of the physics of climate, and I think they have occasionally been too quick to see dishonesty where there is none.

      Spencer’s and Christy’s religious views do not detract from their ability to do science. However, when I hear someone claim that Intelligent Design is a scientific theory, it does make me wonder how well they understand science. ID is not and can never be scientific–you can even prove this with information theory.

      Perhaps most disappointing, I’ve seen all these dissenting scientists overstate their case when speaking to or writing for laymen (e.g. in the editorial pages of the Wall Street Urinal). That I have a problem with.

      Ultimately, I am sorry for Pielke et al. Their rejection of those aspects of climate science that imply anthropogenic causation place them at a severe disadvantage when it comes to publishing. They’re trying to understand climate with one cerebral hemisphere tied behind their back.

    • Ray Ladbury // December 16, 2009 at 2:40 pm | Reply

      Maureen Vilar,
      While I would in general agree that one can be an expert in one’s own field and an absolute crank outside of it, I think the case of ID is sufficiently grave to cast doubt on one’s ability to do real science. Intelligent design has unlimited explanatory power–an omnipotent deity can by definition do anything–but it has absolutely no predictive power. It is the predictive power one is trying to maximize in science.

    • John Mashey // December 17, 2009 at 5:59 am | Reply

      Recall that Spencer & Christy wrote a chapter for the Jastrow, Nierenberg, Seitz 1990 book, i.e., GMI-George C. Marshall Institute, arguing strongly for use of satellites (yes, OK), and about the problems with ground stations (uhh), and noting there was no evidence of warming from satellites 1979-1988. But, curiously, even with their own data (which turned out to have errors), they didn’t show a regression line. Admittedly, 10 years is too short. All in all, this was not a crazy article, although working this early with GMI makes one nervous.

      Based on my usual Catalog of Reasons, this looks like (at that point):

      TEC1: Long anchor (take a very early position, and not change it in face of data).

      TEC4: Intra-field fights: a few observational folks have not been keen on climate models.

      Later on, there may have been other reasons.
      Spencer has become a GMI Director, and both Christy & Spencer are Heartland “global warming experts.” (and certainly, they know a lot more than most on that list … but it’s not a list most scientists would wish to be on.)

    • Sceptical Guy // December 17, 2009 at 9:45 am | Reply

      Thanks Ray, David, & Maureen. I appreciate your honesty, candour, and tact ;). I admit I’m not very familiar with the details of ID (besides what I’ve read in a few SciAm editorials), and frankly, I’d like to stay that way. But I can see how a true creationist (who believes the earth is < 10000 years old) would have trouble studying ice-cores! That's partly why I asked.

      I admit to liking Pielke's stance more and more as it just seems to fit with how I already perceived things to work. And I admit to being disturbed by the current global fixation on CO2, as this helped give rise to the recent ethanol craze, which I'm sure you would agree created more problems than it solved. Of course, liking one thing and being disturbed by another is an emotional response – not a scientific one. The forcings and feedbacks don't care how I feel about them – they are what they are. Again, that's why I asked your opinion.

      I would suggest Ray that factions on both sides could be accused of overstating their case in certain media. Mr Gore and Greenpeace for example. And in this respect I actually think we have been quite poorly served by our main-stream media which tends to focus on scandal and sound-bites. That just drives people like me to the blogosphere – where we find all sorts of weird and wonderful ideas! However Gore is an ex-politician, whereas Pielke et al are scientists – they should know better.


      • Kevin McKinney // December 17, 2009 at 1:05 pm | Reply

        Just seizing on one point you mention–the ethanol issue.

        Don’t confuse this thought with a thorough analysis of the question, but my perception of the US ethanol legislation is that it was primarily driven by 1) desire for increased energy independence, itself partly driven by concern about rising fuel costs; and 2) it was a great opportunity to bring in another farm subsidy scheme.

        It seemed to me at the time that though ethanol could, if done right, reduce emissions, there was no attempt to try to make systematic changes needed to ensure that outcome. Thus, it seemed to me that emissions mitigation was an issue that was hijacked as a talking point by folks pushing other agendas. (Certainly I was making the argument at the time that ethanol shouldn’t be viewed as an emissions fix.)

        It is worth noting that Brazil’s sugar-cane-based ethanol economy is working beautifully at present, and avoiding large amounts of CO2 emissions. Details on Wikipedia:

        It’s a fascinating read. The sophistication of the system is remarkable, embracing development of numerous varieties of sugar cane (including genetic modification efforts), deployment of co-generation using the “bagasse” from the cane, sophisticated flex-fuel vehicle technology, and the development of highly-efficient industrial processes.

        On the emissions front, the consensus seems to be that for the life-cycle, the system does reduce emissions considerably. One report even concluded that Brazilian ethanol could end up being carbon-negative in some cases–startling if true. (I haven’t tracked down the analysis that led to that claim.)

        And it’s worth noting that Brazilian ethanol is the object of protectionist efforts from the US and others. (Going back to the ag subsidy theme.)

    • Ray Ladbury // December 17, 2009 at 1:12 pm | Reply

      Sceptical Guy says: “And I admit to being disturbed by the current global fixation on CO2, as this helped give rise to the recent ethanol craze, which I’m sure you would agree created more problems than it solved. ”

      No. The ethanol craze in the US has been fueled by domestic politics. It is a chance for legislators to dole out billions to well heeled agro-corporate donors. It actually uses more energy that it produces!

      Now in Brazil, alcohol produced by sugar cane actually has weaned the country off of imported petroleum–as well as providing a much needed stimulus in the economically disadvantaged Nordeste. This would not work in the US here sugar is also heavily subsidized.

      As far as Gore, yes he is a politician (and not a very good one, IMHO), and yes he does push the science beyond the envelope in some cases. On the other hand, his basic message is scientifically sound, and even his extreme statements rest (tenuously) on some scientific justification.

      The Pielkes on the third hand are on shaky ground scientifically, as there is no way land use is going to significantly modify estimates of CO2 sensitivity. It just won’t happen. There’s too much independent evidence.

      • Sceptical Guy // December 19, 2009 at 10:40 am | Reply

        @ Kevin & Ray
        Fair enuff – I retract my statement on ethanol. I was of course referring to corn-based ethanol, and my opinion was based mainly on: this article (,9171,1725975,00.html), which links corn-based ethanol demand to increased deforestation; this World Bank report (, which links corn-based ethanol demand to increased poverty; and this UN FAO report ( which links corn-based ethanol demand to higher food prices and increased worldwide hunger.

        I admit to not keeping up with the state-of-the-play in biofuels much, apart from the “Grassoline” article in the July SciAm (which I read with great interest), and the recent trials of aviation biofuel (from jatropha I think) by Boeing and Air New Zealand. I completely missed this article (,8599,1735644,00.html), which describes the Brazilian sugar-cane ethanol industry that both of you are referring to.

        I try to avoid commenting on US internal policies – ‘cos I’m not American, and I think I would find it a tad rude in reverse. In my own country in 2008 we nearly wound up with a govt mandated biofuel level which was sold to us MAINLY as a “green” initiative, and as “doing our bit for climate change” (my country is a Kyoto signatory). This was around the time of the food supply ‘crisis’, “2nd generation” biofuels were still a fair ways off, and local production (from beef tallow and whey curds) would not meet the mandated levels. So this meant probably importing ethanol, and probably US corn-based ethanol. Then we had a change of government and the new administration adopted a ‘wait-and-see’ approach.

        Don’t get me wrong – I’m all for alternative energy, I’m all for reducing our dependence on oil (a finite resource), reducing pollution, and recycling (I would LOVE it if my lawn clippings could help power my car ;). I think my position on biofuels is actually similar to yours Kevin. I think they have a great potential in mankind’s future, PROVIDED THAT their introduction is properly managed, and they do not result in increased deforestation, unaffordable food prices, and other environmental problems such as over-use of nitrogen. An example being Brazil’s use of sugar-cane ethanol.

        But then I am nobody, this is just my opinion, and it’s worth what you’re paying for it – i.e. nothing. Thanks heaps for the real-world example of how biofuels can and _should_ be done. Much appreciated.

    • Scott A. Mandia // December 17, 2009 at 7:01 pm | Reply

      @ Sceptical Guy:

      In their research articles, scientists tend to highlight uncertainties because it is these uncertainties that require further research.

      Furthermore, none of these scientists wishes to be wrong in print so the human tendency is for these folks to downplay certainty.

      To the non-scientist, this published work must look like the scientists do not know very much.

      That is why it was EXTRAORDINARY that the IPCC 2007 conclusion was that the recent warming is very likely due to human activities.

      Some claim the IPCC was “alarmist” but the latest data shows the IPCC was too conservative in some of its key projections.

      Just look at Synthesis Report from Copenhagen and Copenhagen Diagnosis (linked below):

    • Berry // January 2, 2010 at 8:04 pm | Reply

      These days, sceptics proudly present Lubos Motl exercise on UAH data, stated “No statistically significant warming since 1995″

      As an expert, what is your opinion about this?

    • Hank Roberts // January 2, 2010 at 9:06 pm | Reply

      Berry, here’s how to look up prior mentions.
      Search in Google using this string: lubos motl

    • Ray Ladbury // January 2, 2010 at 9:30 pm | Reply

      Barry, FWIW, Tamino mostly agrees. See here

      Of course, given the noise in the system, this is not really all that significant. Thirty or more years is climate. Less than that is short-term variation.
      And we do have a pretty good feel for short-term variation and what causes it. When you account for some of those factors, the results change quite a bit.
      and here:

      Motl, probably knows this. He probably even knows that it’s not reasonable to pick only a single temperature database. However, he tends to disdain reality unless it has 10 or more dimensions.

    • deech56 // January 2, 2010 at 9:46 pm | Reply

      Pick NASA-GISS (1995-2008) and warming is significant. Pick a more reasonable starting year like 1975 and warming is highly significant.

    Leave a Comment