As a mild break from Lewandowsky’s fake data and false results, I am going to revisit IPCC 1990 Figure 7, which I discussed in several Climate Audit posts from 2005-2008 – a topic that was raised at Lewandowsky’s blog by conspiracy theorist John Mashey, who, rather than confronting the problems of Lewandowsky’s use of fake data, recently went into paroxysms of ecstasy at the discovery of an incorrect citation in an early Climate Audit post. An incorrect citation in a Climate Audit post – it doesn’t get much better than that for Mashey. Mashey feverishly extrapolated a simple incorrect reference to belief in a flat world.
Normally, I’d just ignore this sort of deranged commentary, but the Climategate emails contained interesting context on IPCC 1990 Figure 7.1 that I’d noticed but not previously commented on. The emails also place the discussion in Jones et al 2009 in an interesting context. Today’s discussion will not be complete: there are interesting points in the Climategate emails about the understanding of the IPCC 1990 graphic that I’ll try to return to on another occasion.
It is true that a Climate Audit post in early 2005 contained an incorrect reference. I try to be careful, but do not claim to be infallible. Recognition of fallibility is one of the reasons for replication and audits. In this case, I had incorrectly referenced a graphic from IPCC 1990 to IPCC 1995. The misunderstanding on my part was very brief. By June 2005 (here for example), I had tracked down the correct reference and used this correct reference in all subsequent posts. Unfortunately I forgot to back and correct the reference in the earlier post at the time. Now that the matter has been drawn to my attention, I have added an update at the earlier post. Mashey’s frenzied excitement at the discovery of this incorrect reference seems, however, more than a little over the top.
Mashey noted that Ross McKitrick had picked up the correct reference by July 2005, but failed to mention that I had already picked up the correct reference in June 2005 and that I had used the correct reference in many subsequent posts. By neglecting all posts and discussions from June 2005 on, Mashey built the incorrect reference in a Climate Audit post into a huge tangle of conspiracy.
I’ll pick up today with a 2007 post that was critical of the reification of the IPCC 1990 graphic in Martin Durkin’s Swindle. I thought that there was an interesting story to be told about the transition from the IPCC 1990 graphic to the IPCC TAR adulation of the Stick, but it was a different one than the one that Durkin had told. In my post, I took the position that the most important criticism of the IPCC 1990 graphic (that it was only representative of one location) applied even more strongly to the MBH Stick (whose MWP-modern comparison also boiled down to one locality, Graybill’s bristlecones, proxies that were not necessarily an improvement.)
William Connolley responded to my post in a post entitled “TGGWS / IPCC ’90 fig 7.1.c“. Connolley (as far too often) misrepresented what I’d actually said, accusing me of “uncritically” accepting the IPCC 1990 graphic. Connolley added the further curious criticism – that I’d accepted the 1990 graphic “uncritically” “despite its lack of good source”, adding that “if McI didn’t like the IPCC ’90 fig, he would be ripping into it as sourceless.” Connolley cited a Wikipedia article (that he had written) on the lack of provenance of this figure.
At the time, the lack of precise provenance of IPCC Figure 7.1 hadn’t been on my radar. If pressed, I would have presumed that it was derived from Hubert Lamb and not the stuff of mystery.
In 2008, I took a look at the provenance of the graphic. I located a graphic in Lamb 1965 (Figure 30) that convincingly seemed to be the origin of the IPCC panel. The figure below is taken from that post and shows the IPCC 1990 graphic (digitized) plotted against the data from Lamb 1965, comparing this to the top panel of Lamb 1965 Figure 3 – showing that the Lamb 1965 figure is unmistakably connected to the IPCC graphic.
Top – Comparison of digitized IPCC 1990 to re-plotted data from Lamb 1965; bottom – Lamb 1965 Figure 3 top panel.
In one of the comments, reader Brent made an inspired find that placed the iconography in a new light (one that remains unexplored). The graphic that we had been looking at was the bottom panel of IPCC 1990 Figure 7.1, which was actually a three-panel triptych, in which the top two panels were schematic temperature histories over the past million years and the past 20,000 years. Brent observed that an almost identical three-panel triptych occurred in the 1986 edition of a 1977 book by Crispin Tickell, who attributed it merely to the “British Antarctic Survey”. [Update Sep 30 11 pm: a reader has consulted Tickell 1977 and states that the triptych at Tickell's website for his book is not in Tickell 1977; the website says that Tickell revised the book in 1986; the 2nd edition needs to be checked. I've accordingly amended references to Tickell (1977) in this post to a generic reference to "Tickell" to reflect this.] The two three-panel diagrams are shown below. I don’t believe that anyone can reasonable doubt that the IPCC 1990 Figure 7.1 is derived from the same iconography as the Tickell triptych.
It’s not quite the same in detail. The top panel is most similar. The middle panel is the most different: more smoothed in the IPCC version. [Richard Drake observes in comments that the IPCC middle panel only covers 10,000 years as well.] The closing portion of the bottom panel has been changed to remove the downturn in the Tickell version – as discussed in my 2008 post, it looks like a point has been added.
Left: from Crispin Tickell website; right: IPCC 1990 Figure 7.1/
For what it’s worth, I recently noticed a graphic in a UK geography textbook that is either derived from the Tickell book or, more likely, from a common “Ur-document” that remains unknown.
Climategate Reaction to My 2008 Post
My 2008 post caused considerable consternation behind the scenes among Climategate correspondents. It turns out that they had been trying to identify the provenance of the IPCC 1990 graphic since early 2007 (see 2007 Climategate 2 emails) and were planning to publish an article including a discussion of the topic of my post: the provenance of IPCC Figure 7.1. According to 2007 Climategate emails (which I’ll review on another occasion), they had concluded that the IPCC 1990 graphic had been taken from a 1989 UK Department of the Environment pamphlet and had been inserted into the IPCC 1990 report at the very last minute. They had independently traced the 1000-year panel back to Lamb graphics, later also citing Lamb 1965 Figure 30 in Jones et al 2009 (and a similar graphic in Lamb 1982.)
My 2008 post, also tracing the provenance back to Lamb 1965, caused considerable concern among the Team that I had “scooped” them. Within hours of my original post, Gavin Schmidt wrote to Mann and Rahmstorf (CG2-3079) that I had “almost” worked out the source of the figure and that they were being “scooped”:
On another subject, McIntyre has worked out where IPCC 1990 fig 7.2 has
came from (almost). We are being scooped!
Further in this post, I will examine the exegesis of Jones et al 2009 with a view to identifying what was “almost” about my post. (It seems to me that my 2008 post contained an exact link to Lamb 1965 and that the Jones et al 2009 linkage to Lamb 1982 is either immaterial or, more probably, incorrect. It’s hard to see the basis of “almost”.)
Mann immediately forwarded Schmidt’s email to Jones, and in true Lewandowsky conspiracy style, presumed that information on their plans to reveal the provenance of the IPCC 1990 graphic had been leaked to me and that my motive for the blog post was to scoop them:
Please see the below–any update on the Wengen paper? It appears that McIntyre is trying to scoop us, must have somehow learned that we’ve tracked this down. It would be nice for the paper to be officially ‘accepted’ before he figures the story out,
Jones immediately conveyed his worry to Briffa and Osborn, worrying that we were “close” to finding out what the IPCC figure was based on. Jones also noted the unknown-to-him Tickell reference, speculating (incorrectly) that it was after 1989.
CA are getting close to finding what the IPCC figure from 1990 is based upon. They haven’t found the original source, nor any of the CRU pubs that show Lamb is wrong anyway. ..
#50 and the link to Crispin Tickell’s web page is interesting – back to BAS pub. If you have time can you follow this one up. I think CA have the dates wrong and this should be after 1989.
In their final article, they make no mention of the Tickell triptych despite its unmistakable iconographic precedence. Jones’ surmise that we had the Tickell date wrong was incorrect.
Jones et al 2009
Jones et al 2009 set out the Team’s exegesis of the provenance of IPCC Figure 7.1. Needless to say, they did not mention the independent Climate Audit discussion that had traced IPCC 1990 Figure 7.1c back to Lamb 1965 or that had identified the iconographic precedent in Tickell.
Jones et al 2009 wrote:
So where did the schematic diagram come from and who drew it? It can be traced back to a UK Department of the Environment publication entitled Global climate change published in 1989 (UKDoE, 1989), but no source for the record was given.
Jones et al 2009 did not give any further bibliographic details on the UK Department of Environment publication. According to a 2007 Climategate email, Jones got a copy of the pamphlet from David Warrilow at DEFRA; I recently wrote to Warrilow asking for the same document, but thus far have not received an acknowledgement. It will be interesting to see this graphic though I presume that it is pretty much identical to the IPCC triptych.
Jones then reported that the figure could be traced back to Lamb 1982 Figure 30 (noting that this figure was similar to the Lamb 1965 Figure 3 that had been reported at Climate Audit). Jones justified the derivation from Lamb 1982 because of its “vertical resolution”, while noting that there was no relevant difference in the underlying data between the 1965 and 1982 versions:
Using various published diagrams from the 1970s and 1980s, the source can be isolated to a series used by H.H. Lamb, representative of central England, last published (as figure 30 on p. 84) by Lamb (1982). Figure 7 shows the IPCC diagram with the Lamb curve superimposed – clearly they are the same curve. The ‘Central England’ curve also appeared in Lamb (1965: figure 3 and 1977: figure 13.4), on both occasions shown as an ‘annual’ curve together with the extreme seasons: winter (December to February) and high summer (July and August).
The IPCC diagram comes from the 1982 publication as the vertical resolution of the annual plot is greater. The data behind the 1977 version are given in table app. V.3 in Lamb (1977), but these are essentially the same as previously given in Lamb (1965). All three versions of the plot have error ranges (which are clearest in the 1982 version and indicate the range of apparent uncertainty of derived versions). The 1982 version dispenses with the three possible curves evident in Lamb (1965, 1977) and instead uses a version which accounts for the ‘probable under-reporting of mild winters in Medieval times’ and increased summer temperatures to meet ‘certain botanical considerations’. …
Here is a side-by-side comparison of the two figures. It is evident that Lamb 1965 Figure 3 top panel and Lamb 1982 Figure 30 top panel are plots of the same curve, though the 1982 graphic is simplified from the earlier graphic:
Left: From Lamb 1965; right – from Lamb 1982.
Back to the Climategate Emails
In the Climategate emails, Jones, Mann and Schmidt told one another that I had “almost” discovered the provenance of IPCC 1990 Figure 7.1, but seemed to think that I had not quite pinned it down. So what, if anything, had the Team discovered about the provenance of IPCC 1990 Figure 7.1 that had not already been published at Climate Audit?
I didn’t know about the intermediation of the 1989 UK Department of the Environment pamphlet, but that was a relative minor point in the exegesis of Jones et al 2009. Not mentioned in Jones et al 2009 was Ammann’s observation (CG2-4039) that the UK DOE pamphlet had “‘expertly’ extended” the version in Lamb 1982 (and the identical Lamb 1965):
just for comparison, here is the superposition of Lamb’s central England
(what ever season that might be) on the UK Dep. of Environment report
graph. This appears very much to be the same data, note all the little
bumps and wiggles are just the same, and then in the tails its
Other than identification of the DOE pamphlet in the lineage of the iconography, the only difference was their attribution of the IPCC graphic to Lamb 1982 rather than Lamb 1965, relying only on the vertical scale. However, the vertical scale of the medieval panel of the Tickell triptych corresponded to the vertical scale of the IPCC 1990 figure diagram – firmly refuting the Jones et al 2009 theory that the Lamb 1982 diagram figured into the iconographic history of IPCC 1990 Figure 7.1.
To the extent that they differentiated their exegesis from the Climate Audit exegesis by placing Lamb 1982 in the iconographic lineage of IPCC 1990 Figure 7.1 – a rather minor distinction on which to refuse at least an acknowledgement of Climate Audit – their iconographic history was a retrogression from the Climate Audit derivation.
Their failure to refer to or discuss the Tickell triptych and/or look for its predecessors was also an important retrogression from the Climate Audit discussion. There’s not a doubt in my mind that any iconographic scholar would unhesitatingly place IPCC 1990 Figure 7.1 in the iconographic tradition of the Tickell figure. It is virtually impossible that Tickell himself originated or even modified the Tickell triptych. There is almost certainly an “Ur-graphic” preceding the Tickell graphic, which Tickell presumably obtained (as he said) from the British Antarctic Survey. It is entirely possible, perhaps even probable, that the IPCC 1990 figure derived from that Ur-graphic without intermediation by Tickell 1977. (I’ve written to Tickell asking for the provenance of the triptych diagram, but have not yet received an acknowledgement.)
By the way, the origin of the other two panels is somewhat interesting. The top panel is almost certainly derived from a foraminifera O18 series from Emiliani that caused great interest in the 1970s. This series appears in different versions in contemporary literature; I’ve not tracked down the precise version used in the triptych. I don’t know where the middle panel comes from: from the labeling, it looks like a British source.
In hindsight, Jones et al were clearly aware of Climate Audit’s prior discussion arriving at almost exactly the same conclusion. If one of their pals had independently published a blog post identical to the Climate Audit post, I’m sure that they would have acknowledged it. But since it was Climate Audit, they didn’t.
Connolley’s role in this is curious. Connolley’s blog avatar Stoat is (aptly enough) a weasel, an avatar that Connolley regularly lives up to. Connolley was well aware that the incorrect reference in the early CA post did not imply any ongoing “confusion” on my part as to the provenance of the IPCC 1990 graphic. As noted above, he had even done a post on IPCC 1990 Figure 7.1 linking to a Climate Audit post.
But rather than pointing this out to Mashey at Lewandowsky’s blog, Connolley immediately materialized at Climate Audit, so excited that he was almost out of breath, and offered to provide me a re-education camp on the differences between IPCC 1990 and IPCC 1995 if I were “confused” on the point:
No. The pic you’re showing is from IPCC ’90. http://en.wikipedia.org/wiki/MWP_and_LIA_in_IPCC_reports has the details, if you’re confused.
Needless to say, Connolley was well aware that Climate Audit had discussed IPCC 1990 Figure 7 on multiple occasions, that the incorrect reference was limited to an early Climate Audit post and a contemporary presentation and that the correct citation had been made in all discussions subsequent to June 2005. But rather than pointing this out to Mashey, Connolley, lived up to his weasel avatar.
In reviewing this material, I also noticed interesting 2007 Climategate emails that contain new information on how the IPCC 1990 graphic was perceived in the early 1990s, leading up to the notorious “get rid of the MWP” comment, that I’ll try to review in another post.