A few days ago, WUWT pointed out that the American Meteorological Society webpage showed that the Gergis et al paper had been officially “withdrawn”. However, readers should know better than to presume that this would have any effect on IPCC use of the reconstruction.
The withdrawal of the Gergis article hasn’t had the slightest impact on IPCC usage of the Gergis reconstruction, which continues to be used in the recently released AR5 Second Order Draft, thanks to academic check kiting reminiscent of Ammann and Wahl. Tim Osborn of CRU is a Lead Author of the AR5 chapter (as he was in AR4) and would be familiar with the technique from AR4.
Although David Karoly had denied that Gergis et al had been withdrawn, the AMS finally admitted this at their website here.
However, the Gergis reconstruction continues to be used in the IPCC Second Order Draft (released for review in early October). The figure below shows its use in the First Order Draft on the left and its use in the Second Order Draft on the right – the two reconstructions are identical up to smoothing.
However, the attribution has changed. The First Order Draft attributed the reconstruction to Gergis et al (then submitted to Journal of Climate, later accepted and then withdrawn. See CA posts here). The Second Order Draft attributes the Gergis reconstruction to PAGES 2K Consortium (submitted to Science).
I presume that the PAGES 2K Consortium has done a little academic check kiting a la Wahl and Ammann i.e. that they have cited the Gergis et al reconstruction even though the article has been withdrawn. The IPCC then cited the article that kited the check (a kitation? :))
It would be interesting to see precisely how the PAGES 2K Consortium (of which Gergis was a member) stickhandled their citation of the withdrawn Gergis et al article. I doubt that they were entirely candid with Sciencemag on the matter.
The PAGES 2K consortium article should be online to Second Order Draft reviewers at the WG1 website, but I haven’t seen it because I haven’t signed up as a reviewer since I am not prepared to agree to the secrecy demands instituted by Thomas Stocker at Phil Jones’ urging. But for any IPCC reviewers reading this post, take a look. See if this most recent check-kiting makes you proud to be part of the organization.
Climate Audit 
62 Comments
It is worse than that….. There are 4 scientists who are the “Steering Committee and Data Managers” for the “Australia 2K” section of the PAGES project. 3 of these 4 people are……
Gergis, Phipps, Neukom
http://www.pages-igbp.org/index.php/workinggroups/aus2k
Yes, they basically ARE the Australia section of the PAGES 2K project.
That is “Australasia” region for “Aus 2K”
although the Gergis/Karoly team has not itself been too consistent or rigorous about when they are describing Australia and when they are talking about a wide region “Australasia”…. and as I recall their paper set definite long/lat boundaries for their Australasia region but then went outside that box without explanation or justification in using some proxies….
Kitation – the ‘bustitution’ of the paleo climate mafia. Looks wrong, sounds awkward, but works perfectly. đ
Are they allowed to tell us just how much pride they feel?
The camel’s back must be getting tired.
Wow. “Retract and use anyway” was not even on my list of possible outcomes, so I did not see that coming. Now I want to see what the curve would look like using only the proxies that actually passed significance testing
At least for myself, I could use some more calendar attributions. What day was Gergis paper retracted? I went to the WUWT link, it shows a third party tweet announcing withdrawal October 18, which I assume is after the release of the “Second Order Draft” in “early October.” Maybe I am out of the loop on this one news-wise, but speaking for out-of-the-loopsters, I could use a little more chronology.
Please feel free to withdraw my words here if they are not helpful.
Steve: the precise dates are obscure. Karoly was a withdrawal-denier before he was a withdrawal-admitter.
The paper was ‘put on hold’ on 8th June 2012. Whether it had been withdrawn was left ambiguous, as Steve implies. On 9th September a commenter on a Kiwi blog, wat dabney, checked a URL provided by another poster, for the Gergis paper, and reported the exact wording quoted by Steve above. The WUWT report originated with Jean S here on CA on 18th, who found these words.
Make of the semantics what you will but some time between 8th June and 9th September ‘put on hold’ became ‘withdrawn’. The IPCC had no doubt by then hatched its escape plan for the dodgy reconstruction, by using another paper quoting the Gergis, in whatever strange transitional state it was deemed to be in.
No government should listen to this organization until it changes its ways.
the data link at ncdc has a note from 13 June speaking of resubmission. If it was resubmitted until end of July then it’s use in the IPCC is OK. I’d think the reference to the consortium paper is mainly meant to have one source for all/most regional reconstructions.
PS: the innuendo of the this blog-post is once more … annoying.
Steve: elsewhere it’s been noted that they didn’t make their July 31 deadline. While you may not like the implications, we’ve seen this before with Wahl and Ammann.
When you speak of innuendo, what are you referring to specifically?
Do you find it more or less annoying and destructive than the citation of a paper known to be faulty for some time and ultimately withdrawn?
Re: meteo (Oct 23 03:41),
no, it was not resubmitted by the end of July but likely sometime between 5th and 8th of September, see here.
meteo,
you seemed to have known this already on August 2nd:
Did you have inside information, or was it just a wild guess?
If it was resubmitted until end of July then itâs use in the IPCC is OK.
Ahhh, the refreshing world of climate science! Where a paper showing major methodological flaws that has been hastily withdrawn and resubmitted is “OK”, provided it is (a) submitted before the entirely arbitrary date, and (b) provides the “correct” answer.
In the rest of science the silly-billies are far more concerned about whether the science might be actually correct and verifiable.
What can you say , further indications of how much you can place on IPCC’s work .
And there to stupid to understand or to arrogant to care , why this looks awful to the public .
“And there to stupid” means what?
I presume that the PAGES reconstruction in the second figure is the thick black line with the gray uncertainty region? I don’t think that it’s a smoothed version of the original Gergis et al. curve, although it’s definitely missing pretty much all high-frequency information. For example, the prominent peak c.1310 of Gergis et al. is entirely missing; that is a low period in the PAGES curve. I suspect it’s a new reconstruction from the revised submission.
IPCC SOD says of this:
The comment about proxy counts certainly *sounds* like the original Gergis et al. reconstruction, which has 4 or more proxies only after 1428. Could be smoothing, I suppose, or perhaps a differnt method of reconstruction.
The new curve has significantly wider error bars than the original, but they still seem too narrow for the pre-1400 era when there are only 3 (often 2) proxies. In the original the computed error bars did not include a term for sampling error. Perhpas they added such a term (which made the error intervals wider) but it doesn’t vary much with the number of proxies. That might account for the fact that the error interval even in the 20th century isn’t a whole lot less than that of 1000 years ago.
“IPCC SOD”
Highly appropriate acronym/mnemonic. đ
One assumes that minor modifications were made in order for the authors to claim that the work is different to that which was withdrawn. This way they can attempt to protect themselves against charges of poor ethics (something many of us take extremely seriously in scientific research).
Chutzpah:
Murdering your parents then throwing yourself at the mercy of the court because you are an orphan.
Klimate Zience:
Chutzpah.
Steve, can we get a new category, Kooky Klimate Kitations? Or will the acronym drive the SS crowd Kraaaaazy?
Citation laundering?
Depends on what you mean by withdrawn.
Still waiting on the MSM to issue corrections and the funding to be repayed!
Crickets…
Climate science papers are never retracted, They just get reconfirmed by later papers.
This reminds me of a paper I recently read by Lewandowsky and others (including John Cook of Skeptical Science fame). I wouldn’t recommend anyone bother reading it in its entirety, but there are some relevant aspects. I think the best quote is:
Sound familiar? How about:
Where was the “healthy sense of skepticism” for Lewandowsky’s results which we now know to be basically meaningless?
Steve, you are becoming a master of social satire as it relates to climate shenanigans. The powers that be on the warmist side seem intent on providing you with an unending source of comedic material for your posts.
Thanks for keeping your blog accessible and enjoyable for those of us who don’t always understand the science and related data.
Steve: Is there any reason to believe that the new Figure in the SOD is not an integral part of the paper submitted to Science? Of course, those peer-reviewing the SOD have no idea what this submission to Sciencemag actually says, nor can they be sure it will survive peer review. The Paleoclimate chapter of AR5 will say whatever the Lead Authors want it to say.
BTW, TIm Osborne was a Contributing Author to WGI Chapter 6 of AR4. Briffa was a lead author. http://www.ipcc-wg1.unibe.ch/publications/wg1-ar4/ar4-wg1-chapter6.pdf
Steve: well, the CLimategate correspondence shows that Osborn was involved in chapter 6 strategy,
I guess this has something do with this little workshop:
http://www.pages-igbp.org/calendar/upcoming/1-pages-sponsored/231-2kwritingws
A possible scenario:
-âPAGES 2k consortiumâ uses (in the May draft) the original graph from Gergis et al (with the paper status “in press”)
-Gergis et al is “put on hold” (beginning of June), which still allows âPAGES 2k consortiumâ to cite the original manuscript when the final submission (to Science) is done (before July 20)
-after âPAGES 2k consortiumâ submission, Gergis et al is officially withdrawn
-Gergis et al is resubmitted (in September?)
-when the âPAGES 2k consortiumâ paper comes back from the review (around this time or later I would guess), the citation is changed to the new Gergis et al submission
Vola, the magic is done. Easy, isn’t it? đ
It’s also easy to find PAGES 2k newsletters on the web, for example newsletter 4, May 2011, that talks of a consortium paper that ‘was invited by the IPCC’ for AR5. The group leaders for the Australian part of the consortium are J Gergis and C. Turney.
Sorry, to find these, search for ‘circular’ not ‘newsletter’.
Somehow that’s appropriate give Jean’s comment above.
Re: Paul Matthews (Oct 23 05:16),
Chris Turney is in the news as leader of the expedition of activists and tourists presently waiting for a rescue in Antarctica.
I hope nobody is going to make silly remarks about this, as FoxNews.com did yesterday:
CAGW omnipresent and mysterious the world over. Call those rescue helicopters.
I thought I recognized that name:
http://www.realclearpolitics.com/articles/2013/12/31/global_warming_researcher_gets_stuck_in_ice_121091.html
It’s official:
Media manager for the Australasian Antarctic Expedition and the ARC Centre of Excellence for Climate System Science, University of New South Wales. Who wouldn’t want a job like that.
BTW PAGES 2K links to Climate Audit’s online version of Esper’s Polar Urals chronology (otherwise unavailable). But rather than using http://www.climateaudit.info, they use the IP address (actually the former IP address of http://www.climateaudit.org, changed in 2009).
He Who Must Not Be Named becomes the The Domain Name That Cannot Be Spelled Out?
er, sorry that replaces steve’s image with green (the same green used to highlight the header of both posts) – that’s in hierarchical mode using ca-assistant in firefox. certainly not my intention.
At least you can’t complain it’s not archived, lol
I hope the Wayback machine has a copy for future historians?
It made me check, and I found this:
http://www.pages-igbp.org/workinggroups/2k-network
“To facilitate the distribution and assessment of climate data and to encourage contributions by data generators, all 2k reconstructions and associated proxy records will be permanently archived on internet-accessible data archives, such as the World Data Center for Paleoclimatology at NOAA (http://www.ncdc.noaa.gov/paleo/paleo.html).”
I guess you fall under the “such as” clause?
This publication list for Phipps says it was updated Oct. 18, 2012 and it lists a “submitted” article for Gergis et al (2012), submitted to Journal of Climate:
http://web.science.unsw.edu.au/~sjphipps/publications.html
It would be good to know if this is a new submission or just an old listing, but it would be strange if he had not updated this item considering what has occurred with it….
Re: Skiphil (Oct 23 11:07),
relevant items from the Phipps publication list and works “in progress” which I linked above (it will be interesting to determine whether there is indeed overlap between the J. Climate and the Science submissions, journals usually frown severely and penalize/retract/reject any overlap of submissions if they are pending at same time!!):
[34] Gergis, J., R. Neukom, A. J. E. Gallant, S. J. Phipps, D. J. Karoly and PAGES Aus2k Project Members, Evidence of unusual late 20th century warming from an Australasian temperature reconstruction spanning the last millennium, Journal of Climate, submitted.
[30] PAGES 2k Consortium (Ahmed, M., K. J. Anchukaitis, A. Asrat, H. P. Borgaonkar, B. M. Buckley, U. BĂŒntgen, B. M. Chase, D. A. Christie, E. R. Cook, M. A. J. Curran, H. F. Diaz, J. Esper, Z.-X. Fan, N. P. Gaire, J. Gergis, J. F. GonzĂĄlez-Rouco, H. Goosse, S. W. Grab, R. Graham, N. Graham, M. Grosjean, S. T. HanhijĂ€rvi, D. S. Kaufman, T. Kiefer, K. Kimura, A. A. Korhola, P. J. Krusic, A. Lara, A.-M. LĂ©zine, F. C. Ljungqvist, A. M. Lorrey, J. Luterbacher, V. Masson-Delmotte, D. McCarroll, J. R. McConnell, M. S. Morales, A. D. Moy, R. Mulvaney, I. A. Mundo, T. Nakatsuka, D. J. Nash, R. Neukom, S. E. Nicholson, H. Oerter, J. G. Palmer, S. J. Phipps, M. R. Prieto, A. Rivera, M. Sano, M. Severi, T. M. Shanahan, X. Shao, F. Shi, M. Sigl, J. E. Smerdon, O. N. Solomina, E. J. Steig, M. Thamban, V. Trouet, C. S. M. Turney, M. Umer, T. van Ommen, D. Verschuren, A. E. Viau, R. Villalba, B. M. Vinther, L. von Gunten, S. Wagner, E. R. Wahl, H. Wanner, J. P. Werner, J. W. C. White, K. Yasue and E. Zorita), Continental-scale temperature variability over the Common Era, Science, in review.
“Update October 2012: The manuscript âEvidence of unusual late 20th century warming from an Australasian temperature reconstruction spanning the last millenniumâ by Joelle Gergis, Raphael Neukom, Ailie Gallant, Steven Phipps and David Karoly has been re-submitted to the Journal of Climate and is being reviewed again.”
http://theconversation.edu.au/post-1950s-warming-in-region-unmatched-in-1-000-years-7081
Skiphil:
The very day Jean S, then WUWT, highlighted the changed status of Gergis Mark I to ‘withdrawn’.
Spooky.
For IPCC climate scientists “withdrawal” for them amounts to merely a “coitus interruptus” in their ongoing work on the world’s body politic. Decidedly snip-able.
From the introduction of the withdrawn Gergis paper we have the following excerpt attempting to explain the motivation for the temperature reconstruction presented in the paper and its scope:
“In response to the lack of continental-scale climate reconstructions in the IPCC AR4, in 2009 the International GeosphereâBiosphere Programmeâs (IGBP) Past Global Changes (PAGES) initiative developed the Regional 2k Network, a set of working groups to collect and process the best available proxy data to develop climate reconstructions in eight regions of the world (http://www.pages-igbp.org/workinggroups/2k-network; Newman et al., 2009). The Australasia
(Aus2k) working group is examining the IndoâPacific region consisting of the landmasses of Australia, New Zealand, the Indonesian archipelago and the neighbouring islands of the Pacific Ocean. This paper is the Aus2k working groupâs regional consolidation of temperature proxies to provide a âbest estimateâ of Australasian temperature variations over the past 1000 years.”
I think the motivation described above obviously puts the Gergis paper author’s hastiness to get the proper message out for AR5 in perspective. There were obvious “official” deadlines to meet and, of course, a result to be obtained that would not allow the paper to be ignored by the AR5 reviewers.
As for the scope of the paper, it is rather evident that some major maneuvering was in order. First of all the largest single land mass in the Australasia region as defined in the Gergis paper, the mainland of Australia, is without proxy representation in the reconstruction. There are 2 coral proxies located off the Australia mainland coast that are in the reconstruction.
The 12 TRW proxies are clustered together closely in New Zealand (9) and Tasmania(3).
The two ice core proxies are from the Antarctica and coincide with the same geographic location of Vostok. These proxies are located well outside the defined Australasia region and are 2 of 7 proxies used in Gergis that are outside the defined Australasia region.
Of the 13 coral proxies, 8 are in the defined Australasia region and 5 are outside. Most of the coral proxies represent a rather modern time period from 1750-2000. The only long period proxy, Palmyra (1149-1998), is a discontinuous proxy that was pieced together and has a high leverage on the modern warming period. Palmyra and two other coral proxies used in the Gergis reconstruction, Rarotonga and Rarotonga 3R, are far out of the defined Australasia geographic box.
As I recall, the proxies from outside the Gergis paper defined Australasia region had high leverage on the modern warming period and would suppose that once the defined geographic box was allowed to be violated other alternative proxies were available. At this point it should be remembered that the Gergis authors unabashedly preselected proxies. They thought they were using a high frequency correlation when in fact it was a low frequency one for selections. Given that that error caused the paper to be withdrawn I wonder what that portends for the paper selecting proxies outside the define area in the resubmitted paper.
Serious question: what if any overall effect on the paper would occur if the “defined area” was recast to include the locations of the current out-of-box proxies ?
I mean that if the area was redefined to include all current proxy locations, would this irreparably damage other sections of the paper ? In my experience, when authority looks like losing an argument, it simply changes the rules
I said this last week over at WUWT:
” Michael J Alexander says:
October 18, 2012 at 8:57 am
But it will still be referenced in the next IPCC report. The other three mentioned here on WUWT strengthening the (already known) reality of the MWP, however, will not.”
I love it when I’m right.
It’s an interesting (and new) definition of ‘Australasia’. The way the term is commonly used and understood in the relevant part of the world is Australia and New Zealand, or Australia, New Zealand and New Guinea (and their islands).
Certain climate scientists are now redefining geography in addition to the several other branches of learning they have stamped with their unique approach to semantics.
johanna,
It’s not just the geography definition that’s a problem. The bigger problem is to find a set or sets of temperatute/time relations for each proxy site. It is of course subjective whether one requires 100 years of observations from a set of instruments adjacent to the tree whose rings are examined; or whether a whole world temperature can be used for the calibration period; or something in between.
The greatest weakness by far that I see with the science aspects of this paper are lack of suitable temperature records. Ed Cook has admitted this for some of his Huon Pine Tasmania reconstruction. There is a reconstruction in the pipeline from the Baw Baw region east of Melbourne, where the most relevant station was started in Sept 1991. Likewise, at Mt Read (Huon Pines) there’s a newish station started Sept 1996.
The satellite period since 1979 is under a cloud because of the divergence problem with some proxy methods. The ground temperature data prior to that have been homogenised to a degree that causes loss of confidence in their suitability. Sea surface temperatures underwent a jump with Argo, but some of the sites used here might be too shallow for Argo and hence unreliable. One could write a paper about the fantasy land of associating distant temperatures with diverse proxies. One site used in the Gergis paper is described at responsive to rainfall, but nontheless a temperature signal can be extracted. This paper is for true believers with eyes firmly closed.
Where is the relevance of peer review?
“The greatest weakness by far that I see with the science aspects of this paper are lack of suitable temperature records.”
The greatest weakness I see is the authors inability to rwealize the statistical consequences of their selection process. If you have looked at lots of proxy series you know that these series show meanderings which can end with an upwardess, downwardess or flatess trend. If one is allowed to willy nilly select based on the recent time trends it is rather evident what the end result will be.
Of greater interest to me is that given this selection latitude (pun intended) just how difficult it apparently remains to find the “correct” ending trends within the bounds of the Gergis paper defined Australasia boundaries. I went back and checked the ending trends of the 7 proxies that Gergis used from outside the defined area and 5 of them show exaggerated ending trends and greater than what is seen within the bounded Australasia region. The 2 Antarctica proxies of O18 and accumulation are more subtle as to the ending warming trend and particularly with regards to temperature proxied peaks going back in time.
Thanks, Geoff
That is the exact point to my question above. Redefining the geographical boundary box to include all proxy locations irreparably weakens their relevance through improbably large distances … but I expect that in the end-game this will not be regarded as important
These documents need to include a link to future declarations of retraction. Example:
1.0 Applicable documents
1.1 The following document or documents form a part of this document to the extent indicated.
1.1.1 Declaration of Retraction http://hostofthisdocument/thisdocument-id-retracted (Free access)
That document will say “This document has not been retracted” or the contents will include the statement of retraction and other information deemed useful by signatory parties. No more mystery about document validity.
Because this will impose honesty on all it will never happen.
I asked for a copy of the PAGES preprint “Continental-scale temperature variability over the Common Era” and got the following reply:
Will it make the July 31 deadline?
Did you register as an IPCC SOD reviewer? If you’re a SOD reviewer, you’re entitled to see the article.
It is also of concern that some studies were still at the work-in-progress stage when the paper was peer reviewed. One might speculate that the objective had been settled before review and the belated insertion of certain numbers might be an inducement to junior authors to reinforce what was already submitted. It’s work done the wrong way; it’s not classical science.
Have a heart, fellas. I love Steve’s work, but Bishop Hill has a role to fulfill in the community too, and how is he going to do that without another “Jesus Paper” to explain lucidly to the average thickie like myself?
“IPCC –Keeping Sceptics Gleefully Busy Since 1988”
Here is another example of a problem that could be caused by the urgency of meeting a cut-oo date for inclusion in AR5. The paper was peer reviewed before this storm broke on 5 June 2012, so it is strange to read here that –
How can a person conducting a peer review do so when the material is incomplete, including to the stage where a conclusion cannot be reached?
This is another severe blow against the sanctity of peer review.
JoC has updated the original manuscript page by adding a sentence:
I find it strange that a page for a withdrawn manuscript is informing about a re-submission under review. I wonder if this is an indication that there is some type of technical loophole which allows them to keep the original submission date for the re-submission in order to meet the 5AR requirements.
Kind of divisional patent application?
4 Trackbacks
[…] Steve McIntyre noted […]
[…] Steve McIntyre: Ă terdragandet av Gergils artikel har inte haft den minsta betydelse för IPCCs anvĂ€ndning av Gergis rekonstruktion, vilken fortfarande anvĂ€nds i den nyligen publicerade Second Order Draft av AR5 ⊠Tim Osborn frĂ„n CRU Ă€r huvudförfattare av kapitlet i AR5 (liksom han var i AR4) och torde vara bekant med den teknik som anvĂ€ndes i AR4. […]
[…] IPCC Check Kites Gergis « Climate Audit A few days ago, WUWT pointed out that the American Meteorological Society webpage showed that the Gergis et al paper had been officially âwithdrawnâ. However, readers should know better than to presume that this would … IPCC Check Kites Gergis « Climate Audit […]
[…] unknown and unheard of prior to December 24 (except, perhaps to Steve McIntyre who recently recalled that Paul Matthews had observed that Turney and Joelle Gergis were the “leaders” of the […]