Dec 22 2011
The Third Nail In The CRU/IPCC AGW Coffin
Update At End
Major Update: Well, well – it seems Tim Osborn admits to the artificial adjustments used by CRU (and shown at the end of this post) in a Dec 2006 email to coworker Thomas Klienen (email #4005):
Unfortunately we haven’t yet published the details of how the gridding and calibration were done. Also we have applied a completely artificial adjustment to the data after 1960, so they look closer to observed temperatures than the tree-ring data actually were — don’t rely on the match after 1960 to tell you how skilfull they really are!
Looks like Osborne just confirmed my analysis of Osborn-Briffa for me! – end update
Since the Climategate 2 emails have come out I have been able to gain a lot more context on the workings (or should I say shenanigans) of the Hockey Team and their efforts to hide the fact the science was rapidly pointing away from their claims of historic recent warming. Not only was the CRU temperature record completely unverifiable and unmaintained (see here), the premise behind tying today’s temperature back 1500 years via tree rings was also a fool’s errand and completely unfounded (see here).
These two revelations alone are enough to deem the AGW science D.O.A, since the CRU gridded data sets cannot be replicated even by CRU (let alone independent analysis) and the foundational principle that supposedly connects tree rings (and other biological proxies) to local modern temperatures cannot be applied. Mathematically the story is a mess.
In these posts and others I have laid out how the Hockey Team responded when faced with the realization that the science was not confirming their ideology. Instead of rejecting their conclusions and following the data, they attempted to cover the problems up, obscuring the problems from the science community. And in this post I show how bad it really got.
The epitome of this disinformation was to hide the decline seen between the modern (and most accurate) temperature record and tree rings – the so called “decline” or “divergence”. Steve McIntyre has done a great service in exposing this spin campaign, showing how graphs of tree ring data aligned with temperature data would look under full disclosure of the results:
Click to enlarge and note how the orange line drops down from around 1940-1960. When this happens it means – as far as tree rings are concerned – the climate today is the same as in the Medieval Warm Period. At least the local ecosystem is the same.
So how did the Hockey Team deal with this most inconvenient of truths? They truncated the data and slapped the modern temperature changes (not values!) from the CRU temperature record at the end of the plot (the red line). This is the same CRU gridded data that has not been verified and cannot be reproduced.
This is all known because of the infamous Phil Jones’ email talking about how he used Micheal Mann’s Nature trick to hide the decline (eamil #3451 from November 1999):
Dear Ray, Mike and Malcolm,
Once Tim’s got a diagram here we’ll send that either later today or first thing tomorrow.
I’ve just completed Mike’s Nature trick of adding in the real temps to each series for the last 20 years (ie from 1981 onwards) amd from 1961 for Keith’s to hide the decline. Mike’s series got the annual land and marine values while the other two got April-Sept for NH land N of 20N. The latter two are real for 1999, while the estimate for 1999 for NH combined is +0.44C wrt 61-90. The Global estimate for 1999 with data through Oct is +0.35C cf. 0.57 for 1998.
Thanks for the comments, Ray.
A trick indeed! Many times the Hockey Team has stated there are many independent studies confirming the Hockey Stick. Well I dove back into the Climategate 1 files and discovered a folder call “Harris-Tree”, with some interesting IDL code in it. Most intriguing is the fact many of the files (“recon_xxx.pro”) are nearly identical in structure. These are:
- recon_esper.pro (Feb 2002)
- recon_jones.pro (Apr 1999)
- recon_mann.pr (Oct 2000)
- recon_overpeck.pro (Mar 2002)
- recon_tornyamataim.pro (Apr 1999)
- recon1.pro (Jan 1999)
- recon2.pro (Aug 1999)
When I first looked at these files I initially though someone was just doing a batch run of series – until I noticed the file dates (in parenthesis). These dates are the ‘modified/last updated’ dates in the files, so they actually span 3 years, not just 6-7 supposedly independent studies. In addition, there is another infamous file in the batch called briffa_sep98_e.pro dated Sep 1998. I assume this is part of MBH 1998 study. This file has the stunning admission that the data has been fudged to make the modern period look warmer:
;
; PLOTS ‘ALL’ REGION MXD timeseries from age banded and from hugershoff
; standardised datasets.
; Reads Harry’s regional timeseries and outputs the 1600-1992 portion
; with missing values set appropriately. Uses mxd, and just the
; “all band” timeseries
;****** APPLIES A VERY ARTIFICIAL CORRECTION FOR DECLINE*********
;
yrloc=[1400,findgen(19)*5.+1904]
valadj=[0.,0.,0.,0.,0.,-0.1,-0.25,-0.3,0.,-0.1,0.3,0.8,1.2,1.7,2.5,2.6,2.6,2.6,2.6,2.6]*0.75 ; fudge factor
In 1998 the team was very unsophisticated in their ‘hiding’ efforts. They simply fudge-factored it to bump up the modern temps by around 2+ °C. But afterwards things become much more ‘coordinated’ and sophisticated, because the next 4 years would see the application of ‘the trick’.
Before I go any further, let me draw attention to email #0112 from Feb 2007, which highlights a common Mann assertion:
You’re not even remotely correct in your reading of the report, first of all. The AR4 came to stronger conclusions that IPCC(2001) on the paleoclimate conclusions, finding that the recent warmth is likely anomalous in the last 1300 years, not just the last 1000 years. The AR4 SPM very much backed up the key findings of the TAR The Jones et al reconstruction which you refer to actually looks very much like ours, and the statement about more variability referred to the 3 reconstructions (Jones et al, Mann et al, Briffa et a) shown in the TAR, not just Mann et al.
Mann is acting like Jones 1999, Mann 2000, Briffa (unknown) were completely independent – but the CRU IDL code files prove otherwise. All these studies use the same CRU gridded temperature data Phil Jone conveniently lost. But moreover, they all go to great lengths to hide the decline using a common code base. Here is the header from each, in time order (note the period start [perst] and period end [peren] dates hard coded into the programs):
recon1.pro (Jan 1999)
;
; Computes regressions on full, high and low pass MEAN timeseries of MXD
; anomalies against full NH temperatures.
; THIS IS FOR THE AGE-BANDED (ALL BANDS) STUFF OF HARRY’S
;
; Specify period over which to compute the regressions (stop in 1940 to avoid
; the decline
;
perst=1881.
peren=1960.
;recon_tornyamataim.pro (Apr 1999)
;
; Computes regressions on full, high and low pass MEAN timeseries of MXD
; anomalies against full NH temperatures.
; THIS IS FOR THE AVERAGE OF TORNE-YAMAL-TAIMYR
; CALIBRATES IT AGAINST THE LAND-ONLY TEMPERATURES NORTH OF 20 N
;
; Specify period over which to compute the regressions (stop in 1940 to avoid
; the decline
;
perst=1881.
peren=1960.
;recon_jones.pro (Apr 1999)
;
; Computes regressions on full, high and low pass MEAN timeseries of MXD
; anomalies against full NH temperatures.
; THIS IS FOR THE Jones NH10 reconstruction
; CALIBRATES IT AGAINST THE LAND-ONLY TEMPERATURES NORTH OF 20 N
;
; Specify period over which to compute the regressions (stop in 1940 to avoid
; the decline
;
perst=1881.
peren=1960.
;recon2.pro (Aug 1999)
;
; Computes regressions on full, high and low pass MEAN timeseries of MXD
; anomalies against full NH temperatures.
; THIS IS FOR THE AGE-BANDED (ALL BANDS) STUFF OF HARRY’S
; CALIBRATES IT AGAINST THE LAND-ONLY TEMPERATURES NORTH OF 20 N
;
; Specify period over which to compute the regressions (stop in 1940 to avoid
; the decline
;
perst=1881.
peren=1960.recon_mann.pr (Oct 2000)
;
; Computes regressions on full, high and low pass MEAN timeseries of MXD
; anomalies against full NH temperatures.
; THIS IS FOR THE Mann et al. reconstruction
; CALIBRATES IT AGAINST THE LAND-ONLY TEMPERATURES NORTH OF 20 N
; IN FACT, I NOW HAVE AN ANNUAL LAND-ONLY NORTH OF 20N VERSION OF MANN,
; SO I CAN CALIBRATE THIS TOO – WHICH MEANS I’m ONLY ALTERING THE SEASON
;
doland=1 ; 0=use Mann NH 1=use Mann land north of 20N
;
; Specify period over which to compute the regressions (stop in 1940 to avoid
; the decline
;
perst=1881.
peren=1960.recon_esper.pro (Feb 2002)
;
; Computes regressions on full, high and low pass Esper et al. (2002) series,
; anomalies against full NH temperatures and other series.
; CALIBRATES IT AGAINST THE LAND-ONLY TEMPERATURES NORTH OF 20 N
;
; Specify period over which to compute the regressions (stop in 1960 to avoid
; the decline
;
perst=1881.
peren=1960.
Now we have to be careful here not to jump to too many conclusions, but the evidence on its face is pretty damning. We did perform some code comparisons and the differences appear to be basically changes for input files, some filtering and output files. The code of these recon files is definitely an evolutionary branch through time.
While the common header is used throughout, the programmer took time to correct it for each ‘independent’ study. In fact, by the time we get to Esper 2002 someone finally notices the ‘stop in’ year is not 1940 anymore, but set in peren=1960. In Esper this long running disconnect is finally corrected. So this tells me the header is valid and all these runs work to ‘hide the decline’ – as Phil Jones admits he did in November 1999, copying Michael Mann.
What I see is strong evidence of collusion to imply a range of independent studies are confirming MBH98. But of course they are not independent if they use the same code and same processed CRU temperature record. And all of them truncate their series to hide any exposure of how tree rings indicate modern cooling (especially compared to the Medieval Warm Period).
Where was this code used? Were its outputs used in the final papers? No one can tell for sure right now – but if someone could rerun this code, produce the graphs and compare them to the independent studies we would have a strong indication. My guess is the likelihood is very high since the code was adapted over the course of 3 years and maintained with updated comments.
BTW, These are not the only files with headers/comments claiming to hide the decline.
For example there is hovmueller_lon.pro from Feb 2007, with this header:
;
; Plots a HovMueller diagram (longitude-time) of meridionally averaged
; growing season reconstructions. Uses “corrected” MXD – but shouldn’t usually
; plot past 1960 because these will be artificially adjusted to look closer to
; the real temperatures.
;
This header is identical to a Nov 1999 version of the same program – so was the header not updated in 2007 or is it still accurate? Who knows. All the way into 2007 the team could have been using ‘corrected’ tree ring densities and truncating series to hide the decline.
Same thing with data4alps.pro dated Aug 2008, which has this interesting warning:
if doabd ne 0 then begin
openw,1,’osborn_briffa_WITHLOWFREQ.alps.dat’
endif else begin
openw,1,’osborn_briffa.dat’
endelse
printf,1,’These are gridded, calibrated estimates of mean warm-season’
printf,1,'(April-September) temperature anomalies (degC with respect to the’
printf,1,’1961-1990 mean), based on tree-ring density records.’
;
printf,1
printf,1,’IMPORTANT NOTE:’
printf,1,’The data after 1960 should not be used. The tree-ring density‘
printf,1,’records tend to show a decline after 1960 relative to the summer‘
printf,1,’temperature in many high-latitude locations. In this data set‘
printf,1,’this “decline” has been artificially removed in an ad-hoc way, and‘
printf,1,’this means that data after 1960 no longer represent tree-ring
printf,1,’density variations, but have been modified to look more like the
printf,1,’observed temperatures.’
Incredible. The true measurements have been ‘artificially removed in an ad hoc way‘ and replaced with fictional data that ‘have been modified to look more like observed temperatures‘. And this is in 2008 (if the comments are up to date).
It seems clear to me there has been rampant collusion across many studies and almost a decade of work to hide the data that would end all the claims of historic recent warming. The code released in 2009 in Climategate 1 has plenty of examples of this same kind of disclaimer in the code. Mike Mann’s claims of independence ring hollow in light of these discoveries.
Some has a lot of explaining to do.
Update: Found one more file called calibrate_nhrecon.pr from Jan 2005:
;
; Calibrates, usually via regression, various NH and quasi-NH records
; against NH or quasi-NH seasonal or annual temperatures.
;
; Specify period over which to compute the regressions (stop in 1960 to avoid
; the decline that affects tree-ring density records)
;
perst=1881.
peren=1960.
This file also has the correction for 1960 in comment field. More interestingly are the list of series than can be run through this truncation function:
- openr,2,’/cru/u2/f055/data/paleo/esper2002/esper.txt’
- openr,2,’../tree5/phil_nhrecon.dat’
- openr,2,’../tree5/mann_nhrecon1000.dat’
- openr,2,’../tree6/tornyamataim.ave’
- openr,2,’/cru/u2/f055/data/paleo/ipccar4/data/mann03_orig.dat’
- openr,2,’/cru/u2/f055/data/paleo/ipccar4/data/crowley03_orig.dat’
- openr,2,’/cru/u2/f055/data/paleo/ipccar4/data/rutherford04_orig.dat’
Interesting list of names from IPCCar4.
Their philosophy relies on two key concepts:
1. Climate over the past 1000 years was very stable.
2. It only recently became unstable.
They do this with BOTH the temperature record AND the CO2 record (take ice core data, smooth it, tack on an instrument record at the end, similar to taking tree-ring data, smoothing it, and tacking instrument data at the end).
http://chiefio.wordpress.com/2011/12/20/co2-past-and-problems/
Another key point is how they compress their hockey stick graph to hide another inconvenient truth. Notice in the diagram at the beginning of this article (Hockey Stick graph) there are actually TWO rises in temperature. There is a rise up, a little hook down, and a second rise up. Briffa’s orange line gets hidden near the top of the first rise.
Now note that the first rise is practically identical in magnitude and duration, identical in nearly every way, from the second rise that starts around 1970. A clearer picture can be seen here:
http://bobtisdale.files.wordpress.com/2011/12/figure-19.png
They don’t want to draw attention that first period of rise from about 1915 to about 1945, though, because it shoots their AGW theory down if they do. That rise in the early 20th century (and there was another in the late 19th century, too, of about the same duration) happened when there were no significant CO2 emissions. In fact, their models completely break down outside of the 30 year period from 1975 to 2005.
They have to make very sure not to mention the early 20th century rise. What caused that one? Same thing the caused the late 20th century rise: recovery from the Little Ice Age that started in the mid 1800’s. We have had three 30 year periods of warming as we recover from the Little Ice Age. We had a late 19th century warming, an early 20th century warming, and a late 29th century warming all of about the same duration. You can see part if it here:
http://www.woodfortrees.org/plot/hadcrut3gl/from:2004/plot/hadcrut3gl/from:2004/trend/plot/hadcrut3gl/from:1975/to:2004/plot/hadcrut3gl/from:1975/to:2004/trend/plot/hadcrut3gl/from:1942/to:1975/plot/hadcrut3gl/from:1942/to:1975/trend/plot/hadcrut3gl/from:1911/to:1942/plot/hadcrut3gl/from:1911/to:1942/trend/plot/hadcrut3gl/from:1879/to:1911/plot/hadcrut3gl/from:1879/to:1911/trend/plot/hadcrut3gl/from/to:1879/plot/hadcrut3gl/from/to:1879/trend
But the point is that they must suppress any idea of previous natural variability and accentuate any warming since about the end of WWII. In fact, Jones notes that “natural variability dominates before 1970” as noted in 5187.txt. Then they notice that the air is getting warmer … well yes it is … because the planet IS getting warmer because we are recovering from the Little Ice Age and things are warming up to what they had been before that period (we still aren’t all the way there, yet, either).
The observed temperatures have been diverging from the IPCC models ever since the rate of modern warming slowed in about 2000:
http://rankexploits.com/musings/wp-content/uploads/2011/11/CherriesAppliesOranges.png
And current observations are outside the lower “error bar” for the IPCC models. The trend since 1989 is not only much less than forecast, it is outside the IPCC’s calculated “confidence” region. This basically means that the models have no skill in forecasting temperatures. If you go to time periods previous to to 1975, you find the same problem, the models have absolutely no skill in forecasting temperature change even with only natural forcings:
http://bobtisdale.wordpress.com/2011/12/14/it-really-should-go-without-saying-but/
The models are wrong with both anthropogenic forcings and with them removed as Bob Tisdale has shown with several postings on the subject.
The reason is that their models are basically too flat when you remove CO2 and their climate sensitivity is too great causing too much of a rise when the CO2 forcing is added in. In other words, the models do not include whatever the natural variables are that actually drive climate to change over time.
This will get worse for AR5 because those models have EVEN HIGHER CO2 sensitivity than the AR4 models do, which have already been shown wrong.
Jones and Mann *must* suppress natural variability in order to remain the celebrity darlings of the “progressive” movement. If they were show that climate varies naturally and does not follow CO2 to the extend they want you to believe it does, they would be dropped like hot potatoes.
Dangit, my posting hit the mod bucket. Probably too may links (I think I had three).
CP,
You had more than 3!
BTW, the uniformitarian principle also assumes the proxies and their environmental responses are the same – which is nuts. As I mentioned variability on local ENVIRONMENT will introduce a different response in the proxy (wet an cold will look like dry and warm in tree rings).
Well, tree rings are only an indication of July temperatures. Even in the LIA we had years with hot summers that were followed by extremely cold weather.
Excellent piece of code mapping 🙂
Maybe pass this on to Monkton for his fraud dossier ?
troppo19,
Thanks. If you know how to pass it on, please feel free!
The other issue is that we know with absolute certainty that a 2 degree rise in temperatures will not cause any catastrophic environmental disaster because for most of the period of this interglacial, temperatures were 2 degrees higher than they are now. Every species alive today survived about 5000 years of temperatures higher than today.
We also know that any species that has existed for more than 100,000 years (virtually all of them) survived a period of 5 degrees warmer than today during the warm period before the last ice age.
The “climate catastrophe” being hyped by these people can be shown to be absolute nonsense by the simple fact that we have already experienced those levels of temperatures and everything survived just fine.
AJ,
Followed you over from CA. Some of what you are claiming is not right. This claim:
“And all of them truncate their series to hide any exposure of how tree rings indicate modern cooling”
is not right. The tree rings do not show cooling. When Hantemirov applied the “corridor” method to the same data (actually predating Briffa et al), he did not get a decline. Briffa et al tried a new method and bollixed it up. There is now good evidence that the “decline” was a statistical artifact caused by having no young trees from the late 20th century, although the exact provenance remains to be determined. Finnish researcher Mauri Timonen has quite a bit on this at his site. Layman Lurker did an interesting statistical analysis at the Air Vent as well.
This leads to your second error. The problem is not data uniformity. There is a vast literature that supports tree rings at tree line as temperature proxies, and when Esper collected a better data set (and processed it properly), he got a MWP and LIA from tree rings. Try going to the CG grepper and reading the hits on searches of “divergence” and “Esper.”
[…] The Third Nail In The CRU/IPCC AGW Coffin Like this:LikeBe the first to like this post. […]
Matt, appreciate the comments, however even if there are some theories of why the data diverged (and young trees make no sense since they show more rings, e.g. more warmth), the fact they hid the decline before having some new theories is still correct.
And as someone with a biology degree I can confirm trees can no more indicate temperature than a rock. Of all the factors, temp will be last to influence, behind water, nutrients, canopy, pestilence and length of optimum days with good environment.
AJ,
Please read this before you make any more blanket statements about treemometers or “the decline:”
http://lustiag.pp.fi/ClimateFromTreeRings_gb.htm
No one is arguing that the Team did not hide the decline. I share your view that this was wrong. But dendrochronologists that look for temperature signals are not buffoons reading tea leaves. You might also be interested in following Climategate threads on the grepper under search terms “divergence” and “Esper.” Ed Cook, Jan Esper, and to a large extent Keith Briffa actually appear as protagonists, doing battle with the nefarious Michael Mann.
“Of all the factors, temp will be last to influence, behind water, nutrients, canopy, pestilence and length of optimum days with good environment.”
What that is true in a general sense, they do tend to try to pick trees in places where temperature is the major constraint. That is trees growing in places that get no summer precipitation or a surplus of it that are near the forward edge of high latitude or high altitude tree lines.
But again, the tree rings only indicate two things: 1. June/July temperatures and possibly 2: winter snowpack as a delayed melt will delay start of growth until the soil thaws.
Matt,
I never said dendrochronologists are buffoons. What I said is they overstate how much temperature signal they can detect because they cannot go back in time and ascertain the primary drivers that act BEFORE temperature. It could be a cool summer, but lots of sun, nutrients and water would provide ample growth.
CP – tree lines are better indicators, and since they move over time there is no guarantee those selected today were in a good position years ago. I understand there can be a lot learned, but they cannot make the claims they are making.
Now, if someone wants to claim the can detect a range of local environment to +/- 5°C in rings, I could buy that.
But of course, that means there is no way to compare the past to today at the fraction of a degree level, and any comparison would be local (not global).
Matt,
Actually we agree on Briffa and Cook, etc. I think even Bradley tries to stay off the Kool-Aid. I once thought Briffa was part of the FOIA release as someone tired of being run over by Mann and Jones and their Jihad.
For those who remember the Monty Python Spam Song, a variant of it runs through my head whenever I read about the further antics of “The Team”:
picture them sitting together, singing “fraud, fraud fraud fraud, fraud fraud fraud fraud, Fraudity Fraud! wonderful Fraud! Fraudity Fraud! Wonderful Fraud!”
the really sad part is that when this finally breaks out into full public recognition of what happened (and it’s already well on its way) it is Science itself that is going to be discredited for the next 100 years. Why would anyone in the future believe what “science” has to say when this many scientists conspired to pull a fraud this big, just to pad their own departments and get bigger grants?
People still aren’t getting it, though. According to the Rio treaty and a bunch of other agreements we have signed including such policies as the “Precautionary Principle” it doesn’t actually matter if greenhouse warming is actually occurring or not or if their models are correct or not.
What we have got ourselves into at this point is a situation where all that is needed from the IPCC is an assessment that CO2 emissions COULD cause warming and that warming COULD cause detrimental environmental impact. That binds us to act according to Agenda 21 of the Rio declaration.
Cause/effect does not need to be established nor do we need to concern ourselves with such issues as “uncertainty”.
http://en.wikipedia.org/wiki/Precautionary_principle
So doesn’t matter if people prove the graphs are all made up, the models are junk, and it starts cooling for 30 years. The UN can still demand we waste billions of dollars on the problem.
CP,
The UN has demanded our for decades. They aren’t going to get it – evah!
At least it brings into focus the fact that the UN only exists because Europe, the third world, and China hoped they could use it to neutralize the US. Now it is dissolving into irrelevancy.
Let the UN die. Do nothing to resurrect it. Turn the building into housing for the homeless, a worthwhile purpose for a change.
that may sound over the top, but this is serious: Crosspatch, in order to solve the situation you point out the UN has to die. Fortunately I think this will happen over the next few years.
(political aside: for anyone who thinks this newsletter issue was “invented” to drag down Ron Paul, I saw several of those newsletters 20 years ago. That’s why I’ve always known he was a nutcase, just like I’ve always known that they would pop back up and destroy him if he ever threatened to break out of the “angry internet crank” phase of his career.)