UPDATE3: professor Rob Wilson leaves some scathing comments about the Mann paper. See below.
UPDATE2: There’s been some additional discussion on the dendro listserver, and it seems quite clear now that the scientists in the dendrochronology field don’t think much of Dr. Mann’s effort – and it appears there is a rift now between former co-authors. See the must read below. I’ll make this a sticky for about a day, and new posts will appear below this one. – Anthony
==============================================================
People send me stuff.
In case you don’t know, ITRDBFOR is an electronic forum (a listserver) subscribed to by most of the world’s dendrochronologists. What is most interesting is that Hughes and Briffa are co-authors of the response to Mann.
—– Original Message —–
From: Rob Wilson
To: ITRDBFOR@LISTSERV.ARIZONA.EDU
Sent: Sunday, 25 November, 2012 20:43
Subject: [ITRDBFOR] Comment to Mann et al. (2012) at Nature Geoscience
Dear Forum,
In February of this year, Mike Mann and colleagues published a paper in Nature Geoscience entitled, “Underestimation of volcanic cooling in tree-ring based reconstructions of hemispheric temperatures”. Their main conclusion was that a tree-ring based Northern Hemisphere (NH) reconstruction of D’Arrigo et al. (2006) failed to corroborate volcanically forced cold years that were simulated in modelling results (e.g. 1258, 1816 etc). Their main hypothesis was that there was a temporary cessation of tree growth (i.e. missing rings for all trees) at some sites near the temperature limit for growth.
This implies Dendrochronology’s inability to detect missing rings results in an underestimation of reconstructed cold years when different regional chronologies are averaged to derive a large scale NH composite.
We scrutinized this study and wrote a response to Nature Geoscience. We are pleased to announce that our comment, along with a reply by Mann et al., was finally published on Nov. 25, 2012 (http://www.nature.com/ngeo/journal/vaop/ncurrent/index.html) – 8 months after submission.
Our comment focuses on several factors that challenge the Mann et al. (2012) hypothesis of missing tree rings. We highlight problems in Mann et al.’s implementation of the tree ring model used, a lack of consideration for uncertainty in the amplitude and spatial pattern of volcanic forcing and associated climate responses, and a lack of any empirical evidence for misdating of tree-ring chronologies.
We look forward to a continued discussion on this subject.
Kevin J. Anchukaitis, Petra Breitenmoser, Keith R. Briffa, Agata Buchwal, Ulf Büntgen, Edward R. Cook, Rosanne D. D’Arrigo, Jan Esper, Michael N. Evans, David Frank, Håkan Grudd, Björn Gunnarson, Malcolm K. Hughes, Alexander V. Kirdyanov, Christian Körner, Paul J. Krusic, Brian Luckman, Thomas M. Melvin, Matthew W. Salzer, Alexander V. Shashkin, Claudia Timmreck, Eugene A. Vaganov, and Rob J.S. Wilson
—
———————————————————————–
Dr. Rob Wilson
Senior Lecturer in Physical Geography
School of Geography & Geosciences
University of St Andrews
St Andrews. FIFE
KY16 9AL
Scotland. U.K.
http://earthsci.st-andrews.ac.uk/profile_rjsw.aspx
“…..I have wondered about trees. They are sensitive to light, to moisture, to wind, to pressure. Sensitivity implies sensation. Might a man feel into the soul of a tree for these sensations? If a tree were capable of awareness, this faculty might prove useful. ”
“The Miracle Workers” by Jack Vance
———————————————————————–
UPDATE: RomanM locates the Mann paper in comments, writing:
The original Mann article seems to be available at his web site:
http://www.meteo.psu.edu/holocene/public_html/shared/articles/MFRNatureGeosciAdvance12.pdf
==============================================================
UPDATE2: More from the listserv
From: “Malcolm Hughes” <mhughes@LTRR.ARIZONA.EDU>
To: <ITRDBFOR@LISTSERV.ARIZONA.EDU>
Sent: Monday, 26 November, 2012 16:42
Subject: Re: [ITRDBFOR] Comment to Mann et al. (2012) at Nature Geoscience
> Ron – no dendrochronologists were involved in the offending Mann et al
> 2012 paper. What Rob described was the response of a number of us to
> some of the multiple flaws in the original paper. Cheers, Malcolm
>
> Malcolm K Hughes
> Regents’ Professor of Dendrochronology
> Laboratory of Tree-Ring Research
> University of Arizona
> Tucson, AZ 85721
—– Original Message —–
From: RONALD LANNER
To: ITRDBFOR@LISTSERV.ARIZONA.EDU
Sent: Monday, 26 November, 2012 03:48
Subject: Re: [ITRDBFOR] Comment to Mann et al. (2012) at Nature Geoscience
“a temporary cessation of tree growth” resulting in no rings for all trees? Now that is a hypothesis that I am willing to bet good money has no empirical support since studies of trees began 200 years or so ago. Speculation this bald could give dendrochronologists a bad name.
=============================================================
UPDATE 3: Rob Wilson leaves this comment at Bishop Hill today, bolded section is my emphasis:
Nov 26, 2012 at 9:00 PM | 
Rob Wilson
Hi Again,
Our comment and Mann’s response to it can be accessed from this link:
http://www.st-andrews.ac.uk/~rjsw/all%20pdfs/Anchukaitisetal2012.pdf
his original paper is here:
http://www.meteo.psu.edu/holocene/public_html/shared/articles/MFRNatureGeosci12.pdf
Hmmm – what do I think of Mann’s response. Where does one start!
Well – he has provided NO evidence that there are stand (regional) wide missing rings for major volcanically forced cool years. Let’s focus on 1816 as an example – The “Year without a Summer” – where historical observations clearly show cool summer conditions (related to Tambora in 1815) throughout NE North America and Europe. Using either long instrumental records or historical indices, there is no evidence of a stand-wide missing ring in temperature sensitive tree-ring chronologies in Labrador, Scotland, Scandinavia or the Alps. Mann would probably turn around and say – well, actually, my model says that 50% of the sites would express missing rings – just not those in NE America and Europe. Sheesh!
To be less flippant, and putting aside criticisms of tree-ring series as proxies of past climate, the method of crossdating is robust and easily verifiable by different groups. I would be surprised if Mann has ever sampled a tree, looked at the resultant samples and even tried to crossdate them. He has utterly failed to understand the fundamental foundation of dendrochronology.
I undertook most of the analysis in D’Arrigo et al. (2006) and we clearly stated in the original paper that due to the paucity of sites (only 19) around the northern hemisphere, the reconstruction was most robust at time-scales greater than 20 years. Using the D’Arrigo reconstruction to look at inter-annual response to volcanically forced cool summers was a poor choice. Maximum density records, as shown in our response, would clearly be a far superior tree-ring parameter to use for such an exercise – as Briffa clearly showed in 1998. See also this paper:
http://www.st-andrews.ac.uk/~rjsw/all%20pdfs/D’Arrigoetal2009a.pdf
There is a lot more I could say, but this can all wait until next week at the AGU Fall Meeting.
One final observation is I urge you to look at Figure 1 in Mann’s original article. The instrumental record (black line) in Figure 1a (upper panel) clearly does not show strong cool temperatures in 1884 related to Krakatoa as seen in the two models. Following Mann’s hypothesis, the instrumental data must be wrong.
Time for some red wine
Rob
Mark Steyn should highlight this event where Mann’s colleagues are now complaining that Mann made claims unsupported by the evidence.
The consensus of Mann’s colleagues in the dendro field is that it’s very likely that Mann mostly made things up .
When is it ever good sense to use data that has failed to pass peer review?
When it was the implications of accepting the data, and not the correctness of the data, that prevented peer review by the orthodoxy.
Assuming some tree line trees are so temp/light limited that they fail to grow a ring after a major eruptions. Some trees which are less stressed, would still have rings even if they are a bit thinner than usual. Matching growing patterns before and after the event will rapidly show if there are missing years.
Sounds like they just need find the sampled trees and sample trees nearby, but at a different elevation. As this divergence puts in queston the reliability of the proxy thing, you would think they would be keen to settle the question with a field trip not models.
Maybe Josh could do a cartoon with Mike the Stump fighting back Pine Bark Beetles.
OK, so let me see if I’ve got it right here:
Dendro’s believe that: All else being equal, a tree ring’s width will be proportional to temperature during the growing season.
All else being equal, a tree ring’s width will be proportional to available water during the growing season.
All else being equal, a tree ring’s width will be proportional to available fertilizer during the growing season. (More nitrates will increase growth, more phosphates will increase growth, more sodium (salt) will decrease growth, etc.)
All else being equal, a tree ring’s width will be proportional to available soil and sunlight during the growing season. (A nearby fire will remove trees, increase light, and therefore increase growth of this particular tree.)
All else being equal, a tree ring’s width will be proportional to stress during the growing and passive seasons. (More stress on the tree (insects, fire damage, falling limbs, greater winds, roots pulled, nearby erosion, etc.) will reduce growth.
What then, is Mann’s “correction” for the increased CO2 yielding 20 to 27% MORE growth in trees and plants worldwide in last half of the 20th century – regardless of worldwide temperatures?
Even if you assume no other factor has ever affected “his” trees over the past 1200 years as he tries to eliminate the Medieval Warming Period and Little Ice Age, how has he corrected his tree ring widths for increased CO2?
Policy Guy asks above (November 25, 2012 at 9:37 pm) “So how does one demonstrate that a tree does not generate a ring under certain circumstances?”
That’s a good question and should be fairly easy to experimentally demonstrate. Plant a tree in a greenhouse. Control the conditions for whatever set of “certain circumstances” you wish to test. Hold the temperature arbitrarily low, for example. Plant a control tree in the next room. After a couple of years, take cores and look to see if the tree rings really stopped. Given enough time and a modest budget for greenhouse space, it should be fairly easy to find out what conditions (if any) will result in a complete shutdown of tree rings.
This, of course, begs the question of whether that research has already been done. If so, I would have expected it to be cited either in the original report or the rebuttal. If not, it seems to be a serious overstep to make the claim that the shutdown occurs without attempting to prove that the phenomenon occurs.
kadaka (KD Knoebel) says:
November 26, 2012 at 7:48 am
From Gail Combs on November 26, 2012 at 6:58 am:
The best comments on the BEST data.
Drop the parts of the URL’s from question mark to end to make the links work.
_____________________
Thanks, problem did not show when I tested links on my machine.
Mann:
“Sorry, there are some tree ring missing.”
Nobel prize committee:
“This is dynamite !
You don’t have to be a dendro.. to observe that if you do indeed suspect that there are missing rings because of volcanic cooling on your marginal limit-to-growth lattitude, simply move down south a hundred kilometres and see if rings appear. I’m am always amazed how much apriori reasoning (the kind your teenage child is forced to use in an argument because of lack of experience) that is appealed to by CAGW scientists. Get out more! If there was a Starbucks in the vicinity, mathematician Steve McIntyre could probably settle this question.
Will Mann, Noble prize winner (not) (etc.), sue?
Mughal says: @ur momisugly November 26, 2012 at 10:20 am
…And does anyone really want to encourage an environment where scientists (or anyone, including bloggers) can’t make a mistake?
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
The problem is not that Mann made a mistake it is his Attack Dog attitude, going so far as having Reporters in Canada try to dig up dirt on Steve McIntyre, getting journal editors fired for publishing the wrong paper, keeping other scientists papers from being published if they do not regurgitated the party line, keeping peer-reviewed papers out of the IPCC…. All these were revealed in the Climategate e-mails as actions of Mann or the ‘Team’.
None of those are the actions of a scientist. Mann is not a scientist but an activist sacrificing his soul for $$$$ since I am sure he and the rest of the Team do not believe the malarkey they spew. If they did they would be cheerful that warming has stopped instead of calling it a ‘travesty’
Rob Wilson has posted a reply to a commenter(s) at Bishop Hill:
Dear All,
Mann’s major flaw was to see something in his model which did not agree with “nature” and assumed that there must be something wrong with nature. Alas, if he had taken the trouble either (1) to speak to some of his dendrochronological colleagues or (2) look at some real tree-ring data to learn what “crossdating” is, he would have quickly realised that his hypothesis was wrong and would not have wasted a lot of time for many people.
In my opinion, not reading a paper just because Briffa is a co-author seems rather narrow minded. Feel free to froth at the mouth, but this just highlights how entrenched SOME of you are w.r.t. your ideas, opinions and biases.
Simon Hopkinson – such a comment is really not very helpful – but did, I guess, provide the impetus for me to write a quick comment.
Rob
http://bishophill.squarespace.com/blog/2012/11/26/lonely-old-mann.html
Nobel prize? huh…
On dendrochronology: Much as I dislike writing this, I strongly suspect that carefully selected trees/sites can be used ad a rather noisy proxy for temperature.
Now before jumping on me, read on…
If you have ever been up a mountain as far as the tree-line, it becomes obvious that the line is pretty well defined. Visit often enough and you will see that the tree-line is pretty much a constant distance (elevation) away from any “snow-line”. Its reasonable to conclude that the major factor in determining where the tree-line is, and why it exists at all, is temperature.
Stick to the West side of the mountain, and rainfall will be fairly abundant and plentiful.
There are very few beasties depositing manure around these trees, and dying at the base of them.
Few trees actually make it to maturity, so on the tree-line itself, crowding is not a big issue.
Under these circumstances, I can be convinced that average temperature is the biggest determinant of tree growth in any year, and that a core sample may well be able to be correlated to average temperature.
I would consider studies in this environment to be reasonable empirical science to validate other temperature proxy/anecdote measurements — for that locality, and during the period covered by other evidence.
It is the extreme extrapolation in time and space that I find difficult to accept. Although conditions in these locations are relatively stable, just take a look at a typical tree-line and you will often see that there are dead trees further up the slope. I have no real idea of the significance of that, but think it reasonable to suspect that it may be an indication that conditions were warmer when those trees grew. There are often dead trees extending well into the tree-line (downhill). What caused these to die? Old age, possibly, but maybe also an indication of a colder period?
To extrapolate in time, we have to assume that conditions didn’t change significantly. That is a huge assumption.
For me, dendrochronology has a place – as supporting evidence of other co-incident temperature measurements/indications. Beyond that, its wishful thinking.
Bob asked
“Speaking of the BEST dataset, can someone tell me why the thing diverges so much from all other datasets?
http://i45.tinypic.com/23tqi3m.jpg ”
—– ——
There is quite good correlation between BEST and the grandaddy instrumental record of them all, CET.
http://wattsupwiththat.com/2012/08/14/little-ice-age-thermometers-historic-variations-in-temperatures-part-3-best-confirms-extended-period-of-warming/
It does demonstrate that CET has a valuable role as a reasonable proxy for global temperatures (should such a thing exist) although it probably suffers-as does BEST-from an inadequate allowance for uhi.
tonyb
Duke C. says:
November 26, 2012 at 11:38 am
Perhaps not very helpful in getting Mann’s ox out of the mire, but that’s exactly where Mann’s ox belongs.
It simply amazes me that Mann still has a small cadre of comrades willing to put up with his lousy manners and defend his horrible “science”.
As for me, I find it immensely entertaining.
There’s a hole in my tree rings, dear Gavin, dear Gavin,
Then mend it, dear Mickey , dear Mickey, mend it.
With what shall I mend it, dear dear Gavin, dear Gavin?
With MWP data , dear Mickey , dear Mickey, with MWP data
The MVP data are too high, dear Gavin, dear Gavin
Then use LIA, dear Mickey , dear Mickey, use LIA
The LIA data are too low, dear Gavin, dear Gavin
Then squash it, dear Mickey , dear Mickey, squash it
What shall I squash it dear Gavin, dear Gavin
Hit it with a hokey stick dear Mickey , dear Mickey, a hokey stick
vukcevic said on November 26, 2012 at 10:28 am:
The original Wilcox Solar Observatory data, field strength given in 1/100 Gauss (18 is 0.18G) (10,000 Gauss = 1 Tesla):
http://wso.stanford.edu/Polar.html
Your graph includes the note on the 30-day averaging. But you have obviously not used the filtered data, which is described (with incorrect grammar) by:
A 20nhz low pass filtered values eliminate yearly geometric projection effects.
The importance of using the filtered values is shown in the North Pole readings, which has apparently flipped polarity three times this year by the unfiltered values. Filtered shows the North has flipped once. The South is still declining by both and has yet to flip.
The filtered values show the average has not yet passed through zero. You are chasing transient noise.
Latest News re court case against Mann, and more
Click here.
“….no dendrochronologists were involved in the offending Mann et al 2012 paper.”
“””””””””””””
Did the editor send this paper to any dendrochronologists for review? If the paper is so clearly flawed that so many scientists raise issues, how could it have been passed for publication?
The clearest explanation of cross-dating I have read is by dendrochronologist Harold C. Fritts in The Wisdom of the Ancients:
“Instead of simply counting, dendrochronologists use a procedure called cross-dating to establish the exact year each ring grew from the beginning of the trees life to the time it was sampled.
“The growth rate of bristlecone pine can be so slow in very dry or cold years, that some trees fail to grow so that the ring is missing at the base of the trunk where they would naturally be counted. If the rings had been counted rather than cross-dated, the tree age would be one year less for every year that the tree did not produce a complete ring.
“The procedure of cross-dating matches the patterns of wide and narrow rings in a large number of trees. A narrow ring can be found in faster growing trees during the same year that the ring is missing in slower growing ancients. When the pattern of wide and narrow rings is memorized or recorded in someway, it is easy to identify where in a stem a growth ring is missing or when mistakes or other types of problems occur. This allows dendrochronologists to identify the chronology (timing) of every ring to the exact year in which it grew.”
http://tree.ltrr.arizona.edu/~hal/tancient.pdf
“Well, Mann did have the good sense to use BEST data in his response.”
BEST isnt data, is a statistical model, same as all the global
“datasets”. It drive me crazy that people think you can take a bunch of raw data, adjust and manipulate it 7 ways to Sunday and then call the output “data”.
So the sum contribution of Mosher in this thread is bugger all, as per usual. Brilliant.
“Mann’s major flaw was to see something in his model which did not agree with “nature” and assumed that there must be something wrong with nature. ”
This is the major flaw with much of what is called “climate science”. Hopefully this episode opens the eyes of some of the dendro’s, especially guys like Briffa and Hughes who have been involved in other faulty Mann papers.
“I strongly suspect that carefully selected trees/sites can be used ad a rather noisy proxy for temperature.”
I agree, there may be useful proxies in sites that are carefully selected. To determine this the sites must be selected beforehand based on the known science (ie before taking any core samples, you decide on which sites should offer the best temperature proxies based on local conditions), not selected afterwards as Mann et al love doing. The proxy can only be validated through such specific selections, and not by wiggle matching after the fact which is always a biased exercise and introduces false matches that will only correlate for a period due to chance.
I’m sure a lot of the dendro’s know this and use such methodology. Its the “climate science” folk that need to understand correct scientific methods.
Perhaps trees also grow by reducing tree rings… LOL
joeldshore writes “For example, one time we had some OLED devices coated and I found that my optical model was quite insistent on the notion that the layer of Aluminum (Al) was about 300A thick, not 200A thick and this seemed to be a quite robust result not sensitive to other assumptions ”
But it was the very precise measured data coupled with the (presumably accurate) model in a very tightly constrained situation that allowed you to do that.
How would your model have faired if the only data you had to work with was “OLED is on” and “OLED is off”? Conversely would you have come to the same result if the model didn’t accurately model the effect of thickness?
This is the situation the GCMs are in. They dont do anything well enough to be useful when conditions differ from those under which they’ve been built and tuned.
Any gossip on the twitter vine? Don’t do it myself.