11 July 2011

Making Stuff Up at Real Climate

Real Climate has an interesting post up today in which Gavin Schmidt reports that, according to a new analysis, errors in climate data (ocean temperatures) may lead to a reduction in the 1950-2006 global temperature trend of 17%, perhaps more.  This is a big deal scientifically and speaks to the certainty – in this case excessive -- with which climate science is often reported, especially by the IPCC.

As far as Real Climate is concerned it is interesting that they seem to think that the story here is not the revisiting of the science of global temperatures, but how they can score some points against fellow bloggers (Maybe trying to change the subject?).

Unfortunately, in my case (and according to Steve McIntyre, in his as well) they engage in some fabrication to try to score those points, incorrectly claiming that I had offered a “prediction” of how the science on this issue would evolve. When called on this, Gavin first admitted that he could be confused (he was), but when I pointed out to him exactly how he was confused, he decided to dig in his heels.

Actually, on Prometheus I and a number of commenters did what people normally do when they hear about interesting science -- we discussed, probed, questioned, hypothesized, explored. Schmidt seems upset that people engaged the subject at all. For my part, I discussed the issue of temperature adjustments in some depth (e.g., here) and offered up a few conditionals that spanned the scope of possibilities (and event had an exchange with Gavin et al. on the subject). But I offered no predictions of how the science would turn out.  As readers here know, I predict football but not science.

Taking a look back at my discussion of the temperature trend issue at Prometheus from 2008 for the first time since it was written, it actually stands up pretty well:  I asked, “Does the IPCC’s Main Conclusion Need to be Revisited?”  The answer would seem obviously to be “yes” if it is indeed the case that 17% of the global surface temperature trend that the IPCC thought it had fully accounted for was actually measurement error.  Oops.  But that sort of thing -- learning something new about something we thought we had settled -- happens in science, and it should not be a surprise or a scandal.

But that is just science.  On the apparently much more important issue of the blog wars, Gavin Schmidt has decided to let his fabrication stand and has encouraged and published the usual cheerleaders piling on, adding to the misinformation in the comments, unfortunately making this post necessary.  Richly ironic.  I do not miss sparring with those guys.


  1. Perhaps the best way to describe Real Climate accurately is that it is an Astroturf site, paid for by outside interests to promote a specific cause.
    Schmidt, as a paid astroturfer, is not going to go gently into the night.

  2. I had commented on the WW2 problems with SST long before Thompson et al 2008.

    In a CA post http://climateaudit.org/2008/05/31/lost-at-sea-the-search-party/ on the matter on May 31, 2008 I stated:

    "Let’s say that the delta between engine inlet temperatures and uninsulated buckets is ~0.3 deg C (and here we’re just momentarily adopting one of the canonical Folland numbers as this particular number surely deserves to be cross-examined). Insulated buckets would presumably be intermediate. Kent and Kaplan 2006 suggest a number of 0.12-0.18 deg C. So for a first rough approximation to check our bearings on this – let’s suppose that it’s halfway in between. Maybe it’s closer to engine inlets, maybe it’s closer to uninsulated buckets. We’re not trying to express viewpoints on such conundrums here – we’re merely examining what assumptions are latent in the temperature estimates."

    To clarify the post of a day or two earlier which Schmidt had already sniped at:

    "In doing an “audit”, if an auditor identified a situation where the assumptions did not warrant the conclusions (e.g. the IPCC hypothesis of a sudden and permanent changeover to engine inlets at Pearl Harbour), he would not substitute his own assumptions; he would reject the assumptions and throw the problem back at the authors. I’ve been pretty consistent about this in the proxy areas, avoiding the temptation to posit what really happened. I don’t want to go beyond this policy here either.

    Now I did a graphic showing the impact of abandoning the Pearl Harbour hypothesis under certain other assumptions. Please construe any such calculation as equivalent to a sensitivity calculation to show the effect of (say) excluding bristlecones, to illustrate the impact of certain assumptions, but not advocacy of a specific alternative. My language at the time may not have been explicit on this, but it’s the way that I do things. There are any number of intricacies in the interpretation of SST buckets. What are the effects of hull sensors? Drifting buoys? Etc. etc. I have no information on such matters at present. Could hull sensors and drifting buoys offset changeover from insulated buckets to engine inlets? Could be. I’m not opining on this. I’d like to see proper expositions of these populations and biases and one of the good outcomes of Thompson et al is that it will almost certainly achieve this goal."

  3. "if it is indeed the case that 17% of the global surface temperature trend that the IPCC thought it had fully accounted"
    Roger, you weren't so iffy in that 2008 post. Or speaking of 17%. You said:
    "But we know now that the trend since 1950 included a spurious factor due to observational discontinuities, which reduces the entire trend to 0.06 [from 0.11]"
    We now know...
    And your bolded propositions for the IPCC, leading up to "D. The IPCC statement needs to be fundamentally recast" rest on that new "knowledge".

  4. The title of your 5/29/2008 Prometheus post posed a question. In the post itself you gave four possible answers:


    A. The entire trend of 0.06 per decade since 1950 should now be attributed to greenhouse gases (the balance of 0.06 per decade)

    B. Only >0.03 per decade can be attributed to greenhouse gases (the "most" from the original statement)

    C. The proposed adjustment is wildly off (I’d welcome other suggestions for an adjustment)

    D. The IPCC statement needs to be fundamentally recast


    By your post here, you seem to be saying the correct answer was *D* - I'm not a climate scientist, but wasn't the correct answer *C*?

  5. -3, 4-Nick, Kevin-

    Thanks for your comments ... IPCC said that at least 50% of the trend was due to GHGs, leaving at most 50% to be explained. If 17% is explained by error then that is 1/3 (min) of the residual trend.

    We can of course agree to disagree, but that seems fairly important with respect to a finding that was expressed with 90%+ certainty. Wildly off? Not nearly.

  6. I'm still not a climate scientist, but the IPCC AR4 covered the years 1956 - 2006. So, rather than 17% explained by error shouldn't that be 11%?

    And if the answer isn't *C*, then you believe the IPCC statement needs to be fundamentally recast?

  7. It's worth noting that 1956 starts at the bottom of a La Nina trough and 2006 ends at an El Nino peak. I'm not a climate scientist either but it looks like a bit of a cherry picking might be involved in the start point. Half a century is an arbitrary choice, and a start date covering a year in with similar ENSO conditions would make more sense. Given that the variability axcross the range of ENSO can cause temperature variations of an amplitude comparable to the entire change in global average temperature during the C20th. e.g. from the end of 2006 to the start of 2008.

  8. "they engage in some fabrication to try to score those points"

    One could well accuse you of the same sin Roger. Not your finest hour. IMO you should admit your mistakes and leave it at that. humble pie doesn't taste as bad as you think.

  9. -8-Marlowe

    Thanks, as you are a reader of this blog you can point to mistakes I've made and corrected ;-). It goes with the territory of course.

    In this case what mistake are you referring to? Please provide a quote from my own writings, thanks!

  10. From Urs in 2008:


    The 90% confidence relates to the assertion that "most of the trend is related to GHG's", not to the amount of the trend, and it does not mean that the anthropogenic part is at least 90%. O.k.?

    Even if the trend would be reduced by 50% (which will not be the case, as you could see in your calculation, it's likely less than 20%), this assertion is still true, because the influence of other known factors on the trend is very low (volcanoes negative, sun likely less than 15%). The GHG influence would still be more than 70%, which corresponds to "most"."

    And yet no retraction from you but rather more defensiveness about you not having said %90...

    a commenter at RC makes a similar point as Urs:

    "What is shockingly ill-advised to me is that the Pielke and McIntyre projections both required, in order to fit with their hoped for story line, that the adjustments not only affect the period from 1945 to 1960, but also extend beyond that into the late 90s, in order to level the more recent temperature increases so as to both make the rate appear less dramatic and the amount of recent, CO2 forced warming less of a concern.

    And yet Thomspon et al 2008 explicitly states (emphasis mine):

    The adjustments immediately after 1945 are expected to be as large as those made to the pre-war data (Fig. 4), and smaller adjustments are likely to be required in SSTs through at least the mid-1960s, by which time the observing fleet was relatively diverse and less susceptible to changes in the data supply from a single country of origin.


    The adjustments are unlikely to significantly affect estimates of century-long trends in global-mean temperatures, as the data before ,1940 and after the mid-1960s are not expected to require further corrections for changes from uninsulated bucket to engine room intake measurements."

    As you know I agree with much of what you say, but it's instances like these that make wonder why your the most misrepresented/misunderstood person in the blogosphere...

  11. -10-Marlowe

    I have no idea what you a talking about, sorry. What is it that I wrote that you think is in error? Please cite my words specifically. I am happy to consider your comment, Thanks!

  12. At climateaudit Steve McIntyre runs the numbers and finds:

    "in a discussion of HadSST, even at realclimate, it wouldn’t have been out of place for Schmidt to actually mention the decrease in 1950-2006 trend from HadSST2 to HadSST3: 29.5%."

    which is exactly one of the scenarios that I looked at based on McIntyre's hypothetical ... Ouch.

  13. Brilliant post, here and at CA. I miss the sparring too...

    Popcorn time!

  14. "in a discussion of HadSST, even at realclimate, it wouldn’t have been out of place for Schmidt to actually mention the decrease in 1950-2006 trend from HadSST2 to HadSST3: 29.5%.

    which is exactly one of the scenarios that I looked at based on McIntyre's hypothetical ... Ouch."

    In your 2008 article, you said:

    "The graph below shows a first guess at the effects of the Real Climate adjustments (based on a decreasing adjustment from 1950-60) based on the graphic in The Independent."

    And then you put a graph based on the independent graph, a graph which was how ocean temp (HadSST) effects global temp (HadCRT).

    Now let's turn back to the original RC post:

    "For reference, the 1950-2006 trend changes from 0.11±0.02 to 0.09±0.02 ºC/decade – a 17% drop in line with what was inferred from the Independent graphic."

    Factoring sea ice changes, your 30% might double the the real number (and that gets smaller working back to 1945). You do know that, right? You know McIntyre's 30% is HadSST, right? Your graphs and the Independent graph and the RC graph are based on HadCRU. This is a very important distinction that if overlooked will give people bad information. We don't want that.

  15. -14-Grypo

    I looked at 3 different hypothetical scenarios for how temperature trends might change -- low one was 15%, high one was 50%. I explicitly asked Gavin (and anyone else) to offer up values they'd like added to the exercise. It wasn't rocket science -- it wasn't even climate science;-) -- it was just an open exploration of the issue.

    Given that the value that Gavin likes now -- 17% -- falls within the spread I looked at I remain baffled as to what the issue is here.


  16. "Given that the value that Gavin likes now -- 17% -- falls within the spread I looked at I remain baffled as to what the issue is here."

    The issue is that neither the 30% or the 50% projections has any basis in physical reality. Read what James Annan and other were trying to tell you 3 years ago. And then you used those invalid boundary parameters to question basic conclusions of the science community. So, of course, when it turns out that the invalid projections were wrong, it is going to be pointed out. A simple admission that these were wrong would suffice, I'm sure. Why is that so hard. I think egos are getting in the way. It's not like several people didn't tell this would happen 3 years ago.

    If the IPCC makes a projection based on several scenarios, all but one having no basis in reality, I'm sure people would like that pointed out as to prevent further error and speculation that is surely to lead people in wrong conclusions. This isn't about Roger, or Gavin, or Steve.

  17. -16-Grypo

    I encourage you to go back and read those posts, the 15% scenario resulted precisely from my interactions with Gavin, Urs, James etc. Of course at the time they disavowed any commitment, but please don't suggest that I ignored their input;-)

    More importantly you seem to gloss over the entire point of the sensitivity analysis exercise, which is to explore the sensitivity of conclusions to assumptions, not to predict a single "right" conclusion. In this case the need to revisit the IPCC conclusion is insensitive to whether the community accepts a surface temperature change of 50%, 30% or 15% -- the implications are robust to such uncertainties.

    I have published peer reviewed papers using such sensitivity analyses to explore the robustness of conclusions, see e.g., this paper on the implications of different assumptions in future hurricane behavior on damage (in which I knowingly select scenarios that might be considered physically implausible, that ensures I've bounded the scientific debate):

    Of course, the scientific community is still to address this issue formally, and a 17% reduction on a blog at Real Climate may or may not be what is ultimately concluded. But from my perspective, I don't really care if it is 17% or 50% or 10%, as the implications for how science was presented in AR4 are essentially the same.

    As far as the efforts by Gavin to score silly points in a blogdebate, that speaks for itself.


  18. Roger:
    Your point is clear. What is really odd is the length of time between the raising of the issue and corrections to a key data series. Gavin is trying, but largely failing, to score cheap points. It is a shame given that a 17% change in a key trend is both scientifically and practically an important change.

  19. Grypo said...

    If the IPCC makes a projection based on several scenarios, all but one having no basis in reality, I'm sure people would like that pointed out as to prevent further error and speculation that is surely to lead people in wrong conclusions.

    Yeah, because the IPCC has been really good about correcting previous errors in the past! No defensiveness at all about Himalayan Glaciers or Amazonian rainforests or sea level rises.