I’m a bit late in covering this due to other issues of interest taking precedence, but I want to make sure this is widely known. We’ve covered the issue of the Gergis et al paleoclimatology paper this past summer, as well as noted the retraction, but the real action as usual, is behind the scenes in the emails, emails now made available via FOI thanks to Michael Kottek who posted this on Climate Audit to announce the emails were now available:
Posted Oct 28, 2012 at 7:04 AM | PermalinkThe results of my FOI request to the University of Melbourne can be seen here:
I requested all correspondence between the authors and the journal regarding the paper. The referees reports were exempted as were documents relating to the resubmitted paper.
I also requested correspondence between the authors after the paper was accepted. Once again emails relating to the resubmitted paper were exempted, and personal material redacted.
I note that emails regarding the paper that were received by one author and not forwarded to the others would not have been covered by my request.
Despite the embarrassment of the withdrawn paper, the University is to be commended for their no nonsense approach to this request. As an alumunus, I am pleased that the response is far more sensible than the approach taken by the UEA and UVa.
That’s true, because there appears to be no holding back of any important correspondence. Here’s some select portions.
Karoly’s first technical response (June 7 Melbourne) to Neukom’s confession was a surprisingly strong endorsement of criticism of non-detrended correlation, going as far as to even agree with [McIntyre] by name:
”If the selection is done on the proxies without detrending ie the full proxy records over the 20th century, then records with strong trends will be selected and that will effectively force a hockey stick result. Then Stephen Mcintyre criticism is valid”.
And then there’s this from Gergis
[…]
”Over recent days we have been in discussion with colleagues here in Australia and internationally about the use of detrended or non detrended data for proxy selection as both methods are published in the literature .
People have argued that detrending proxy records when reconstructing temperature is in fact undesirable (see two papers attached provided courtesy of Professor Michael Mann) .
While anthropogenic trends may inflate correlation coefficients, this can be dealt with by allowing for autocorrelation when assessing significance. If any linear trends ARE removed when validating individual proxies, then the validation exercise will essentially only confirm the ability of the proxies to reconstruct interannual variations. However, in an exercise of this nature we are also intrinsically interested in reconstructing longer-term trends. It therefore appears to be preferable to retain trends in the data, so that we are also assessing the ability of the proxies to reconstruct this information.”
And this admission from co-author Phipps:
Based on the various emails circulated over the past few days, it appears that we will not have a viable millennial-scale reconstruction if we pursue the detrended approach. I therefore feel that we should use the raw data to validate the proxies…
My preference is therefore for David’s Option 2, with Option 1 as my second choice. I dislike Option 3 as it will not leave us with a viable reconstruction. I also dislike Option 4 as it strikes me as essentially starting again from scratch – which seems unnecessary given how far this work has already progressed, and also seems out of proportion to what is only a matter of fixing a technical issue.
Mann, in correspondence with the authors Gergis and Karoly, in his typical style tried to sell a collection different workarounds for the problems they brought on themselves, and in the end, his advice was rejected, the JC editors told the authors the paper was not viable, and the authors were forced to withdraw the paper. Full stop.
Journal of Climate editor Chiang wrote:
After consulting with the Chief Editor, I have decided to rescind acceptance of the paper- you’ll receive an official email from J Climate to this effect as soon as we figure out how it should be properly done. I believe the EOR has already been taken down.
Also, since it appears that you will have to redo the entire analysis (and which may result in different conclusions), I will also be requesting that you withdraw the paper from consideration. Again, you’ll hear officially from J CLimate in due course. I invite you to resubmit once the necessary analyses and changes to the manuscript have been made.
I hope this will be acceptable to you. I regret the situation, but thank you for bringing it to my prompt attention.
The end result of the AMS putting the authors on notice, plus the admissions in the emails, rather puts this early defamatory remark from Dr. Mann in perspective:
Well I’m afraid Mclntyre has probably already leaked this anyway. I probably don’t have to tell you this, but don’t trust him to behave ethically or honestly here, and assume that anything you tell him will be cherry-picked in a way that maximally discredits the study and will be leaked as suits his purposes.
Read the whole episode here at Climate Audit.
It should be noted that commenter Jean S. contributed the first valid critique, which then later grew into this full retraction. Kudos to him too.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
One might just as well “calibrate” the height of people with temperature, and conclude that temperature was colder in the past because people were shorter.
phi:
Your post at October 31, 2012 at 6:01 am says
I DO know that and I DID substantiate – with evidence – one of the reasons I know that with absolute certainty.
I suggest you read a post before making such untrue and unsubstantiated claims about it in future. In case you have forgotten, my post you failed to read but misrepresented is at October 31, 2012 at 5:33 am.
And had you read it then you would not have made your later post at October 31, 2012 at 7:50 am which refers to “calibration”.
Richard
I guess we can look forward the shameless sycophants and defenders of Mann and his ike of Climate scientists to attempt to defend this latest misbhevaiour from climate scientists. Joel Shore and Phil should soon be cropping up here with their defense of the indefensible.
richardscourtney,
Your post October 31, 2012 at 5:33 am. is not a demonstration of the uselessness of dendro but of the real trap that represents screening.
My answer to wobble refers to another issue related to calibration and not screening.
Steve McIntyre says:
October 31, 2012 at 8:02 am
HaroldW writes:
“I think it’s a disservice to refer to this as “McIntyre’s triumph over Gergis, Karoly, and Mann.”
Nor do I regard this as a “triumph” over them. When I read the cavilling of Karoly and Gergis, I feel a sense of disappointment rather than vindication. I would prefer that the headline be more neutral.
I agree with Steve and Harold. My reaction was “Ouch!” when I read the headline. …this is after following the entire story as it unfolded.
It may be a triumph of the scientific method. Perhaps Persistence of McIntyre encourages reluctant acceptance of Scientific method. Subtitle: Jean S. shows that thoughtful careful analysis will always win the day.
Just sayin’
“The growth of plants is limited by the availability of the least available nutrient.” Liebig’s Law.
It has always bothered me that so many scientists seem to be so limited in their knowledge outside of their chosen field. To know the truth of Liebig’s law one needs to do no more than attempt to grow a garden. Tree ring growth depends primarily on the availability of water and oxygen access to the root system. Temperature per se, within a fairly wide but varying range from species to species, has no great impact.
ferd berple says:
One might just as well “calibrate” the height of people with temperature, and conclude that temperature was colder in the past because people were shorter.
~~~~
Fully agree. May I also suggest that we select the Dutch sample? It has a better trend.
My statement to you was directly related to your comment to richardscourtney. I’m not sure why you are suddenly trying to claim that it’s unrelated to the paper of Gergis.
Btw, feel free to share the studies that show “quite remarkable results.” I’m very open to looking at them objectively – seriously.
Harold W: I think it’s a disservice to refer to this as “McIntyre’s triumph over Gergis, Karoly, and Mann.”
What makes it a victory is the history of Mann’s attempts to portray McIntyre as a dishonest, incompetent, crank — instead of responding in thoughtful ways to the merits of McIntyre’s actual criticism. And the history of the attempts to stifle getting this information into the public domain. This is a victory. And it may be a warning to UVA and PennState that they are protecting the correspondence of a scientist whose view of his own records is not to be trusted.
It is also a breakthrough (I think, maybe it’s happened before) that the peer-reviewers took McIntyre’s statistical critique seriously, and that the record of their doing so has been published. In the McShane and Wyner dialogue published in the Annals of Applied Statistics McIntyre and McKittrick contributed a comment, but here there is the additional measure of respect that a paper was rejected and withdrawn.
I am only one reader of this stuff, but it seems to me that the word “victory” is appropriate to this outcome.
Steve McIntyre: When I read the cavilling of Karoly and Gergis, I feel a sense of disappointment rather than vindication. I would prefer that the headline be more neutral.
I empathize and I suspect that most other readers to as well. But I think that for a lot of us viewing the struggle, this is a triumph for science, led in this case by you and your technically expert readers.
wobble,
The critics of Gergis paper relate to the method of selection of proxies and not to the divergence. Regarding richardscourtney, he demonstrated, in my opinion correctly, the weaknesses of screening but not the fact that “treemometry cannot work”.
Regarding studies with MXD, see eg Briffa 1998, Briffa 2001, Cook et al. 2002, Esper 2010, Grudd 2008.
The issue of the divergence is effective but no one has yet proved that it was a divergence of dendro and not a divergence of thermometers.
phi:
Your post at October 31, 2012 at 9:00 am says in total
My post at October 31, 2012 at 5:33 am IS a demonstration of one of the reasons for “the uselessness of dendro” unless, of course, you explain how it is possible to escape what you call “the real trap that represents screening” (i.e. the impossibility of defining a calibration set).
Care to give that explanation a try?
hint – you won’t be able to copy an answer from SkS, RC, etc. because they don’t discuss the real problems of thermometry but, instead, make excuses for ‘straw man’ arguments as a method to pretend thermometry is not nonsense.
Richard
phi says:
The issue of the divergence is effective but no one has yet proved that it was a divergence of dendro and not a divergence of thermometers.
I opine: You keep using the word “divergence”. I don’t think it means what you think it means.
phi:
At October 31, 2012 at 10:40 am yo say to wobble,
The tree rings are calibrated against “the thermometers”.
If “the thermometers” are wrong then the calibration of the treemometers is wrong so the results of treemometry are wrong.
Treemometry has as much relationship to science as palmistry but is more misleading.
Richard
richardscourtney,
It is easy to avoid the trap of screening. Just don’t do screening. Studies based only on MXD don’t do screening.
coalsoffire,
And what it means ?
In the same comment thread, is reported a quote from Gavin Schmidt to Gergis, after Gergis was complaining about the Climate Audit community requests for all the data, including the rejected proxies. Gavin recommended providing all the data, and also cautioned: “While there is no chance whatsoever that they [“Steve McIntyre et al”] will examine your work and find no faults.” Sounds like grudging respect and painfully-gained wisdom on Gavin’s part.
I don’t know, this sounds hopeful. Perhaps the conflict between the scientist’s wallets and their ethics is slowly being resolved. They have perceived ClimateAudit and others to be a major threat to their billion-dollar taxpayer-purchased CAGW rice bowls and public accolades, and have reacted primitively.
I don’t know, I’m an engineer, I want people to find faults in my work. When engineers make mistakes, people die. When scientists make mistakes, it just means we pound money down a rat-hole until somebody wises or fesses up.
coalsoffire,
Would you say that one diverges always from something? You obviously right and I should have clarified. I mean of course a divergence from actual regional temperatures.
richardscourtney,
“The tree rings are calibrated against “the thermometers”.
If “the thermometers” are wrong then the calibration of the treemometers is wrong so the results of treemometry are wrong.”
Indeed it’s a valid objection. But you can avoid this problem by calibrating on detrended data.
People are complaining about treemometers. That’s why the tree-ring proxies are selected from tree populations that seem to be limited by temperature, e.g. the northern larch forests of Siberia. On the Taimyr peninsula for example, the southern part is wooded, the northern is treeless. What’s the difference? Temperature, most likely, is the factor limiting the growth of trees. It’s a reasonable hypothesis. Both the northern and southern parts get the same amount of precipitation, and have the same soil chemistry.
One could say that the limiting nutrient is still water because the wooded areas hold the snow (and more of it) longer into the spring, leading to a longer-duration water abundance. And that may be so, but what then causes the tree-line? Temperature is still the answer.
It’s a reasonable hypothesis, but as a theory, it still needs lots of work.
The next big question to answer, do tree-line regional temperature variations correlate well with global temperatures trends? And that is not clear at all, perhaps because regional temperature trends are used to infer global trends (cart before horse), and because trees at the edge of starvation, so sensitive to temperature, the regional signal may mask the global.
Briffa, Mann, and many others are trying to figure this out, but their maths are suspect, and they perceive valid scientific criticism as threats to their CAGW rice bowls, and have behaved in decidedly non-scientific ways.
I see no reason not to show real temps from real thermometers on the same chart as the tree ring proxies. However there has to be enough of both to allow the reader to see what ther heck is going on. As faras I see, see if the two curves/lines agree for a resonable length of time (like 25 years). I think there is divergence in the ancient past as clear as there is deviation in the present.
In my experience with numerical analysis, the only way to properly test a model against raw data is to do so directly. *Any* manipulations of the data from the original raw data introduces new ‘artifacts’ – effectively systematic errors that are a direct consequence of incomplete analysis if the raw data.
The most honest treatment of the data uses the model to calculate the value of a dependent variable (i.e. a temperature observation) that results from applying the model to the full set of independent variables (insolation, altitude, latitude, longitude, etc.) associated with the value of the dependent variable.The difference between the observation and the calculation of the independent variable is squared and summed with all other such squared differences. This sum (of squared deviations = SSD) tells us of the uncertainty inherent in the model.
By recalculating the SSD as the original independent variables are varied systematically (using curve-fitting algorithms such as the Simplex) to determine the values of these independent variables (and their uncertainties) that produces the “best” fit (smallest SSD) to the observed data. This requires running the model hundreds or thousands of times (something they probably could do) and will let you know exactly which variables in your model are most important, least important, and correlated (either positively or negatively) to other variables.
This would possibly provide infromation that contradicts the narrative, and if there is a competent numerical analysts on staff anywhere, this has probably already been brought to their attention.
In lieu of ‘shooting the messenger,’ you can always lock him out.
Crispin gets around, eh? ☺
phi:
Your post at October 31, 2012 at 11:33 am makes no sense. It says
Say what!!!???
In what way does or can using detrended data overcome the thermometers being wrong?
And you still have not addressed the problem (which you called the “trap”) of it being a physical impossibility to calibrate treemometers.
Richard
richardscourtney,
“In what way does or can using detrended data overcome the thermometers being wrong?”
Because thermometers are affected by drift roughly constant since the early twentieth century, detrending and calibrating on this period you get a suitable relationship between actual temperatures and MXD (high frequencies are not affected). Then you check with other proxies that you have correctly analyzed the situation.
See this sample : http://img38.imageshack.us/img38/1905/atsas.png
“And you still have not addressed the problem (which you called the “trap”)…”
See my post October 31, 2012 at 11:11 am.
phi:
I am starting to think you are being deliberately obtuse.
Your post at October 31, 2012 at 12:20 pm says;
Firstly, if you think the divergence problem is thermometer calibration drift then you are deluded. But if that were true then, as I said, that means the thermometers are wrong so the temperature data is wrong because the trees are calibrated to the thermometer data. Detrending the calibration set does not overcome this.
You are claiming a proxy indication of temperature is more accurate than thermometer data which is used to calibrate it!
And I read your post at October 31, 2012 at 11:11 am. It says
The method REQUIRES “screening”, i.e. selection. It is part of calibration.
Either you select for trees which indicate the gross temperature trend or you select for trees that indicate the variations in detrended data. In either case, the selection assumes that correlation indicates causation: it doesn’t.
The growth of trees is dependent on the limiting factor for their growth, and this is rarely temperature. Moisture, sunlight and nutrient availability are usually much more important than temperature, and they can all vary throughout the life of a tree (e.g. most people know about – what my son calls – ursine defecation in an arboreal environment).
Assuming you are not being deliberately obtuse, then I strongly suggest you do a lot of reading before again posting on this subject to WUWT. Most ‘regulars’ here know sufficient about the subject to assess the worth of your comments.
Richard