In a recent post, I reported on the diagram in Jones 1998 (Science), which pushed hide-the-decline a year earlier than my previous inventory. (The Briffa bodge, an earlier technique, dates back to 1992 and Jones 1998 is a sort-of transition from the Briffa bodge to truncation as hide-the-decline technology.)
I’ve had a few requests for a fresh inventory of hide-the-decline incidents, updating the discussion of Oxburgh panelist Kerry Emanuel’s false claim to the US Congress that hide-the-decline had been limited to a “single lapse of judgement” in a “non peer-reviewed publication” (the WMO diagram).
However, rather than this being a “single lapse of judgement”, to my knowledge, there is NOT A SINGLE graphic in “peer reviewed literature” that shows the Briffa decline in a spaghetti graph comparison of temperature reconstructions.
I’ve done a quick inventory below (and other examples will come to mind) and re-examined the handling of the Briffa reconstruction in the spaghetti graph in each article. In 22 of the 28 diagrams listed below, the Briffa reconstruction has been truncated to hide-the-decline (following the practice of IPCC AR3 where Mann had been Lead Author.) As an alternative to showing the decline, Mann, in 1999, proposed that IPCC simply not show the Briffa reconstruction. This practice has been followed in 6 of the 28 listed below, including the influential 2006 NAS report and 2009 EPA Endangerment Finding (which used the diagram from the NAS report.) But remarkably, not a single one contains a graphic comparing the actual Briffa reconstruction to other reconstructions.
| Article | Figure |
| Jones 1998 (Science) | Figure captioned “Getting Warmer?” |
| Jones et al 1999 (Rev Geophys) | Figure 6 |
| Briffa and Osborn 1999 (Science) | Figure 1 |
| IPCC FOD | Figure 2.25 |
| WMO-1999 | Cover |
| IPCC AR3 SOD | Figure 2.21 |
| Crowley and Lowery 2000 | Figure 1 omits |
| Briffa et al 2001 (JGR) | Plate 3 |
| IPCC AR3 | Figure 2.21 |
| Jones et al 2001 (Science) | Figure 2A |
| Briffa and Osborn 2002 (Science) | Figure captioned “Records of Past Climate” |
| Esper et al 2002 (Science) | Figure 2 omits |
| Bradley et al 2003 (Springer) | Figure 6.5, 6.6 |
| Mann et al 2003 (EOS) | Figure 1 |
| Briffa et al 2004 (Glob Plan Chg) | Figure 8 |
| Cook et al 2004 (QRS) | Figure 1 |
| Esper et al 2004 (EOS) | Figure 1 omits |
| Esper et al 2005 (Clim Dyn) | Figure 1 omits |
| Esper et al 2005 (GRL) | omits |
| Juckes et al 2006 (CPD) | Figure 1 |
| Hegerl et al 2007 (Nature) | Figure 1 |
| Hegerl et al 2007 (J Clim) | Figure 5b |
| IPCC AR4 2007 | Figure 6.10b |
| NAS Panel 2006 | omits |
| D’Arrigo et al 2007 (Gl Plan Chg) | Figure 3 omits |
| Mann et al 2008 (PNAS) | truncated input |
| Mann et al 2008 (PNAS) | Figure 3 |
| Kaufman et al 2009 (Science) | Figure 3G omits |
| EPA Endangerment Finding 2009 (Science) | omits |
| IPCC AR5 2013 | Figure 5.7 not shown |
Within the literature, there are several articles disclosing the decline, notably by Briffa himself e.g. Briffa et al 1998 (Nature 291), Briffa et al 1998 (Nature 393), Briffa 2000 (QSR) – indeed, this was how I originally noticed that the decline had been hidden in the IPCC AR3 diagram in 2005 long before Climategate – but none of these articles shows a comparison of the Briffa reconstruction to the other (Mann, Jones) reconstructions. The spaghetti graphcs of Briffa et al 2001 and Briffa et al 2004 (Glob Plan Chg) both hide-the-decline, but there are other figures that show the decline (but these other figures do not compare to other reconstructions.) Mann et al 2008 took matters to a different plane entirely: in addition to hide-the-decline in its spaghetti graph, Mann et al 2008 replaced the modern portion of Briffa MXD data with “infilled data”.
Even technical articles on the “divergence” problem do not contain diagrams showing the Briffa et a 2001 reconstruction including decline as against other reconstructions (e.g. D’Arrigo et al 2007 (Glob Plan Chg)).
In summary, far from hide-the-decline being the “single lapse of judgement” claimed by Emanuel and the Oxburgh panel, the opposite is the case: nowhere in peer reviewed academic literature can one find a diagram showing the decline in the Briffa MXD reconstruction compared to other reconstructions.
Climate Audit 
56 Comments
A single but rather lengthy lapse of judgment?
Ten years of sleeping around turns into ‘a lapse of judgment’ when your wife calls her lawyer.
Have we reached the ‘But honey…’ stage with climate science?
The negotiating phase of grief. Initially it’s denial (so skeptics were joyfully mocked), then anger (so skeptics were loudly maligned). Finally, after negotiation (as skepticism is righteously appropriated), it’s depression (as skeptics are accepted as having been accidentally right), then acceptance (as skepticism was only faulted for being imperfect).
A bit off topic and I apologize but here is a big data server for open source R computations…..For all those who like that here on CA.
http://www.toolsjournal.com/integrations-articles/item/3299-teradata-powers-open-source-r-users-with-high-speed-processing-additional-analytics-and-more
But only a momentary lapse of reason…
What was the year where the first ‘peer-reviewed’ publication came out declaring a justification for the ultra-special present-day scenario where dendrochronology is less robust but yet really strong for all previous eras (unsubstantiated conjecture or not)? Did it just take one such paper to instantly justify the exclusion?
Maybe Kerry will have it that he meant it was a “single lapse of judgment” made multiple times.
Serial misinformation (™ Climate Science), maybe we can get Sir Paul to protest (Nurse, not McCarthy)? Of course it’s a mildly complicated matter, but I don’t see how anyone who believes in the scientific method and disclosure can condone this type of malfeasance.
This is worth a citable write up (so it in fact could be cited).
remarkable but not surprising
(given the low standards in this field)
appalling yet evidently business-as-usual
and Michael Mann, with malice aforethought, calls Judith Curry an “anti-science” renegade who is a “serial misinformer”
chutzpah does not even begin to describe the arrogance and explosive self-inflation of that man
Geoff, that would be McCartney I think?
The Sir Paul reference, in case anyone missed it, is http://www.theguardian.com/science/2014/sep/04/sir-paul-nurse-criticises-figures-distort-scientific-evidence
Yes of course, McCartney!
So when can we see M. Manns spurious paper, drawing, doodles, etc., and citations and referances to it, ripped out from the annals of science?
Not only did Mann infill his Briffa data in 2008 with temperature data, by cutting off the data and replacing it with temperature data, he then compared the data to the temperature record ostensibly for removing series which didn’t match temp. If I recall Briffa’s data 2.0 passed “correlation screening” 100% but would need to look it up. The percentage of passed series was a foundational argument in the 08 paper. The MV methods of course perform the same operation as screening but do it behind the scenes.
In light of a comment by Steve on a different thread where he stated (to RickA) that he never stated that Mann grafted temperature data on to proxies, I am confused as to what infilling data is as compared to grafting. Could you or someone else explain the difference to me?
Also, did Mann clearly identify the temperature data that he was infilling?
JD
Steve: the commenters with whom I was taking issue were making trivial and incorrect comments about standard spaghetti graphs, not MXD series in Mann et al 2008. When people make incorrect comments, it offers opportunities for Mann to point to a graph in which instrumental and proxy reconstruction are separately labeled as supposed proof of good behaviour.
Steve,
Perhaps you could post this information permanently in your sidebar Pages section, a kind of ‘History of the Decline’ and update it from time to as new ‘evidence’ comes to light.
Thanks,
W^3
Nick Stokes “the rest of the world has moved on”.
No you haven’t.
You would have stopped defending Mann and his merry band of clowns.
You would have used all your skills to defend the scietific method.
Steve, I’m probably being very dim, but D’Arrigo et al GPC 2008 “On the ‘Divergence Problem’ in Northern Forests: A review of the tree-ring evidence and possible causes” does seem to include Briffa in its Figure 3. What am I missing?
Steve: the Briffa MXD reconstruction (Briffa et al 2001) is the one in controversy. It’s from a very large population and is not shown. Briffa also developed a reconstruction using only a few sites: Yamal, Tornetrask and Taimyr ring widths (Briffa 2000) – that’s the one that’s shown here. It’s increase in the 20th century came mostly from Yamal – thus the interest in the validity of this chronology.
Steve’s commentary D’Arrigo: Making Cherry Pie on 7th March 2006 may also be worth taking in.
Steve: the Briffa quote on the decline is worth repeating over and over until the people in the field rise above cargo cult explanations:
As you say of an earlier D’Arrigo paper in that thread:
Thanks again for your meticulous care from those days to this.
i.e., since we can’t explain the decline we will assume that it does not represent anything inconvenient to our beliefs.
In particular, we will assume that there is never any reason to re-consider our assumptions, methods, and data; we will assume that science is something we simply make up to suit our convenience.
Science could always “bring the proxies up to date” and try to determine what kind of recent anthropogenic forcing(s) might explain the decline.
Maybe they’d also find whatever anthropogenic forcing(s) might be causing the pause.
The PDF has
I guess that makes it ‘D’Arrigo et al 2007 (Gl Plan Chg)’ as Steve has it above?
Thanks Steve, very helpful.
Re: Steve McIntyre (Sep 11 11:57), that plot makes the point beautifully.
Briffa had many different variations of his 2001 MXD reconstruction, eventually settling on a version using principal components that differed notably from versions that used more direct averaging. The differences arose mainly in the early portion. If there are material differences between a PCR version and direct averages, I presume that something is getting flipped in the PC process, but didn’t parse his results to try to figure it out.
an aside (but it goes to the issue of whether paleo scientists are making accurate comparisons and analyses related to the “let’s move on” theme allegedly justifying the neglect of divergence issues):
is it still the case that numerous proxies have not been updated for the periods of the past 2-3 decades?
Because to the extent that this is still the case, it would seem to indicate that practitioners in this field are not making a serious, credible effort to “move on” with more comprehensive, well-balanced studies.
Steve, I’m unclear on the grey shaded area “period of divergence” in light of the line for Briffa MXD (2001) — doesn’t it depict the divergence starting around 1960 or a bit earlier?
Steve: D’Arrigo characterized divergence as a problem commencing in the 1980s, though divergence in the Briffa MXD series obviously began much earlier.
“an aside (but it goes to the issue of whether paleo scientists are making accurate comparisons and analyses related to the “let’s move on” theme allegedly justifying the neglect of divergence issues):
is it still the case that numerous proxies have not been updated for the periods of the past 2-3 decades?”
I take “let’s move on” more as “let’s change the subject”. There have been calls by many interested parties, including SteveM, to update the proxies. If nothing else, if the trees and proxy locations used in the original temperature reconstructions were re-sampled, we would have some out-of-sample data to analyze. There appears no motivation in the climate science community to update the proxies. The reason given is often the expense to do it. SteveM and Pete of Colorado did some resampling of tree rings on their own time that SteveM termed as I recall, a Starbucks expedition. That exercise questioned the expense argument.
To be fair I think one climate scientist and dendrochronologist named, Rob Wilson , was talking about updating TRW proxies but I have not read any of the publications and analyses of that work – if there are any. There is a Mann connection for Rob Wilson in the link below. Well dated volcanic cooling is used by dendros to show the efficacy of the tree ring response even if the response is not quantitative with cooling and requires a large and abrupt temperature change to become a marker. Mann, as he is wrought to do, wrote a paper apparently hand waving the issue away by suggesting that the volcanic cooling creates years with no growth rings and thus the cooling signal is lost. Evidently dendros like Rob Wilson take great pride in accurately dating TRWs (an attribute of TRWs that are rather unique to that proxy) and Mann’s paper was an affront to those abilities.
Here’s the Briffa reconstruction placed against the D’Arrigo graphic in a longer 1400-2000 context. To place temperature scales on the reconstruction requires location and scale parameters, each of which is accompanied by assumptions and uncertainties. In a very quick inspection of the comparison below, it seems to me that the scale of the Briffa MXD reconstruction has been set smaller than the scaling of the DArrigo RCS reconstruction in the 19th century increase phase.
The problem with Briffa MXD divergence is that late 20th century values go to essentially Little Ice Age values, where as D’Arrigo RCS divergence is a lesser effect in the last 20 years. Recall that the Darrigo RCS series used the Yamal RCS reconstruction (though they said Polar Urals and refused to issue a corrigendum). Briffa’s most recent Yamal is much less HS. The long series in the DArrigo network are mostly the same as Briffa (2000) and its uptick is, as I recall, produced in large measure by the incorporation of shorter series, rather than an uptick from the long series (otehr than Yamal).
Steve McIntyre (12:16 PM):
It remains amazing that no peer-reviewed paper 1998-2013 dared to include a graph of this clarity – and relative simplicity – about something this important.
“It’s increase in the 20th century came mostly from Yamal”
And from one tree in Yamal.
Yes, the One Tree to Rule Them All…
… And In Their Ignorance Binde Them.
Mann(2008) Figure 2 shows compilations of reconstruction proxies with divergence during the modern warming period for dendro and non dendro proxies. If the paper were written in the true scientific spirit the main topic of the paper would have been: Given the general divergence problem, and even in the face of a post facto selection process, how can we rationalize the proxy responses as reliable thermometers for past climate?
http://www.pnas.org/content/early/2008/09/02/0805721105.full.pdf+html
The paper briefly and begrudgingly admits to a general problem with divergence, but without addressing the problem head on.
“Interestingly, although the elimination of all tree-ring data from the proxy dataset yields a substantially smaller divergence bias, it does not eliminate the problem altogether (Fig. 2B). This latter finding suggests that the divergence problem is not limited purely to tree-ring data, but instead may extend to other proxy records.”
If one is willing to wade through the individual proxy series divergence can be seen in many proxy responses to temperature. The Briffa MXD series is certainly not a one-time thing in the world of proxy divergences. Divergence is something to be expected from reconstructions where proxies that are truly not reliable thermometers are selected after comparing the response to the instrumental record and running out of proxies – as chance would predict – that turn upward in the modern warming period.
Attaching the instrumental record to a reconstruction merely takes attention away from the divergence and allows the implication that the proxy responses are on par with thermometer responses (at least in the pre-divergence time period) without having to explicitly state such nonsense.
Perhaps you could clarify a couple of points that might make the issue clearer to those of who don’t follow the details of the issue are carefully as you have.
Briffa’s reconstruction in question is based solely on tree rings / a dendro proxy. Questions:
1. Briffa’s reconstruction (or reconstructions plural) is important because it draws on a large body (“network”) of tree ring data. Yes?
(Network in this context means a collection of data from different locations, specie of tree, altitudes, etc. Correct?)
2. Are there other similar studies on the scale of Briffa’s (that is, based on a large network)?
Is D’Arrigo 2007 on a similar scale?
3. How many of these other temperature reconstructions are based on dendro (tree ring) proxies?
4. I’m interested here in how many studies other than from Briffa show a similar the tree ring proxy divergence. In other words, is Briffa’s graph representative of the work of others in the field?
4. By “spaghetti graph” you mean a graph that show many different reconstructions on the same graph?
5. The importance of spaghetti graphs (as contrasted with D’Arrigo 2007) is that they give an overview of the entire field of temperature reconstruction of the past using a variety of proxies.
Re graph Briffa v. D’Arrigo,
Is it physically plausible that this dendro methodology can correctly recreate temperature differences of a few tenths of a degree from several hundred years ago?
We are having problems today, doing that regionally with modern sensors designed to measure temperatures, such as thermometers.
Maybe the discussion should not be concerned with whether the decline has a mechanistic cause, but whether it is merely noise and error.
That way, tongue in cheek, there is a probability of 0.5 that post any date, the trend would be conveniently up or inconveniently down.
Geoff:
I think many here, including our host, share your doubts. It is, of course, possible that there are metrics from living and formerly living things or from the vestiges of past weather events that contain a climate signal. Tree rings,sediments and ice cores may well do so. However, as Carl Sagan noted, “Extraordinary claims require extraordinary evidence”. The claim of precision in measuring past temperatures from previous centuries strikes me as indeed extraordinary. Alas, the evidence seems far from extraordinary, but rather limited, opportunistic, selective, contradictory and otherwise highly debatable.
“Extraordinary claims require extraordinary evidence”.
Last time I read that quote, it was attributed to Lance Armstrong…
Responding to Nick Stoke, Sept 11 at 5:22
Here is a link to a new report from Nature Communications that 1) admits to the
divergence about 1960, 2) presents a hypothesis regarding the main cause of the
divergence and then 3) presents data to try to back up their case. They try to
show that volcanic eruptions emit large quantities of SO2 that migrate up to the
north pole and cause a two-year lag in cooling due to a reduction in light in
certain areas of the arctic. They go back and show that declines in proxies can
be correlated mainly to volcanic eruptions, and that areas of the arctic that
were not exposed to volcanic ash show little signs of divergence. This seems somewhat
like science to me. Perhaps, if one could date volcanic eruptions in the past, then
proxies in the areas of the eruptions could be compensated. What appears to be a
decrease in temperature in some of the proxies they think is really mainly due to
a decrease in light in those areas.
http://www.nature.com/ncomms/2014/140507/ncomms4836/full/ncomms4836.html
The way the Marcott 2013 hockey stick avoided retraction via a few weasel worlds within the paper about a lack of statistical significance of the blade versus how Mann promoted it as a super hockey stick that validated his life’s work is analogous to how the lack of contemporary validity was hidden in plain site as a “well known divergence problem” that had been already considered and widely known, blah blah blah. Try that in genetics! Or cancer research. Or dietary claims. Or particle physics. Well, they did try that in dietary claims, resulting in the diabetes-promoting upside down government sponsored diabetes sugar bomb Food Pyramid, supported by the Michael Mann of his day, Ancel Keys, promoting a former single bullet theory in which cholesterol and saturated fat took the place of carbon dioxide, to over explain via big lever effect, an actually complex stasis feedback-dominated system, falsely.
worlds = words
That’s profound.
Steve
I can not remember you ever expressing the same tone as you have here.
“You’d think that someone in this wretched field would know such things. Buncha incompetents”.
It’s about time!! How on earth you have managed for so many years to be both polite and gracious to these imbeciles i’ll never know but I sure as hell admire you for it.
CG2 email 4387
Just a trick of the trade? What is this Kelly paper?
Steve: Odd juxtaposition of words. Looks to me like the paper is Kelly, Leuschner, Briffa and Harris, The climatic interpretation of pan-European signature years in oak ring-width series http://hol.sagepub.com/content/12/6/689.short
Steyn detects what might be a weakening in Mann’s resolve:
He further says that that will not impact his countersuit.
http://www.steynonline.com/6565/the-lonesomest-mann-in-town
Better discussed on a Mann-Steyn thread like here? But note Steyn’s statement on 14th August that
A nice surprise, then, and presumably a significant one.
Who with clean hands could Mann get on his side?
Are there any papers that try to look at different proxies and based on common traits argue that the decline is recent only?
MikeN,
I asked similar and/or related questions on this thread before but received no answers. A couple of questions are begging to be answered: Why the attention on the paper(s) by this one author (Briffa)? Are Briffa reconstructions considered authoritative in the field? I’ve followed climateaudit.com since long before climategate and I don’t know (or I forget) the answer.
If climateaudit.org is getting funding by “big denier” special interests then evidently they aren’t getting their monies worth.
Just making sure Steve got to read this. The Twitter Chainsaw Massacre. http://thepointman.wordpress.com/2014/09/19/the-twitter-chainsaw-massacre/
Nick Stokes writes:
“… there was a divergence; they didn’t believe that the results reflected anything climatic, and they didn’t know what caused it”
I’m most definitely not an expert in this field, but the dendrochronology argument seems to run something like this:
1. Tree ring sizes in some trees seemed to approximately track the thermometer temperature record up to about 1980. In other trees this was not the case.
2. We have no way to predict beforehand which trees will be good “treemometers” and which ones won’t.
3. The trees we studied which seemed to be good “treemometers” up to around 1980 seemed to lose that ability in later years. We have no idea why.
4. We have postulated that this was due to some antropogenic-caused anomaly and can be ignored, but have no evidence to support this. Therefore it could just as easily be due to, say, bored and whimsical Martians corrupting the data just to mess with us. We simply don’t know.
5. For the trees that seemed to be good “treemometers” for some or all of the era of thermometer records, we have assumed that they also correctly tracked temperature in the preceding centuries. We have no idea if this assumption is correct, and have no way of validating it.
6. It is easy to think of dozens of factors beside temperature that could affect tree growth either negatively or positively. We have no way of knowing the influence of any of these factors on any historical tree records.
7. Notwithstanding any of the preceding, we believe that certain trees are warning of an impending climate crisis.
Without trying to be in any way flippant, I assume I’m missing something obvious. It can’t really be that bad can it?
I’m afraid that it is. This is the very issue that prompted me to look at this field in detail. I’ve told this story before. When I originally read Briffa’s cargo cult on divergence, I couldn’t believe that this was accepted as valid scientific procedure. Because the Mann hockey stick – also based on tree rings – was the leading article in the field, I looked at it first. But it was Briffa’s absurd “explanation” of the divergence problem that prompted me to question the specialists.
When reviewing the replies at CA I often “hit the breaks” when seeing the pink background accompanying Steve’s comments. I don’t think I have to explain, thank you Steve, you are a profound example to the science community. It is my hope that your appreciated wicked acumen and superb example amongst such a small group of knowledgeable individuals will continue to grow.
4 Trackbacks
[…] For further details, Climate Audit has a post Inventory Hide the Decline. […]
[…] Auditor, Steve McIntyre provides extra context to the ‘trial of the Century': Inventory of Hide-the-Decline […]
[…] https://climateaudit.org/2014/09/10/inventory-of-hide-the-decline/ […]
[…] is also an industry where if you are incorrect in your predictions your career is short and if you bodge the data you jail term might be […]