Since Chefen has brought this figure into play, there’s much to consider about it. [Also see followup here.] For interested parties, the data is at WDCP here . I haven’t got to checking Chefen’s results yet, but wanted to table interesting results in passing. As Jean S noticed, the Corrigendum stated that an “old” version of the reconstruction was used in these calculations, but it didn’t “matter”. I’ve plotted the two versions together and, for the purposes of his correlations, the difference between the two versions probably doesn’t “matter” but it’s worth wondering why the two versions differ. Secondly, Figure 7 contains a splice of instrumental and proxy records, which presumably carry forward into the correlations. I’ll review what Mann has said in the past about such splices.
First here is the original Figure 7 shown in full with original caption. Notice the change in the line at the right of the top panel and the caption. Mann has here spliced the proxy reconstruction with instrumental data from 1980-1995.
Original Caption. Figure 7. Relationships of Northern Hemisphere mean (NH) temperature with three candidate forcings between 1610 and 1995. Panels, (top to bottom) as follows. “ÅNH’, reconstructed NH temperature series from 1610–1980, updated with instrumental data from 1981–95. “ÅSolar’, reconstructed solar irradiance. “Ålog CO2′, greenhouse gases represented by atmospheric CO2 measurements. “ÅDVI’, weighted volcanic dust veil index. Bottom panel, evolving multivariate correlation of NH series with the three forcings NH, Solar, log CO2. The time axis denotes the centre of a 200-year moving window. One-sided (positive) 90%, 95%, 99% significance levels (see text) for correlations with CO2 and solar irradiance are shown by horizontal dashed lines, while the one-sided (negative) 90% significance threshold for correlations with the DVI series is shown by the horizontal dotted line. The grey bars indicate two difference 200-year windows of data, with the long-dashed vertical lines indicating the centre of the corresponding window.
The next figure compares the “old” and “new” versions of the MBH reconstruction. The range of differences is from -.2 to 0.27. I wonder how this is dealt with in confidence interval calculations. The blue on the right shows the spliced instrumental record.
Top: black – “Old” reconstruction; red- “new” reconstruction; – blue – instrumental splice.
Some time ago, we discussed Mann’s statement at realclimate that:
No researchers in this field have ever, to our knowledge, “grafted the thermometer record onto” any reconstruction. It is somewhat disappointing to find this specious claim (which we usually find originating from industry-funded climate disinformation websites) appearing in this forum.
I had discussed in a previous post a splicing of proxy and instrumental records executed by Crowley and was able to show convincingly that Crowley had effected the type of splice that Mann said no one, to his knowledge, had ever done. I then pointed out that Mann had switched Crowley versions between Mann et al (Eos 2003) and Jones and Mann 2004, using an unspliced Crowley version in the latter. I surmised that the Eos 2003 spaghetti graph (and these spaghetti graphs are hard to read) contained a spliced Crowley version and that Mann knew this, calling the above claim into question.
Lambert examined the Eos graph and suggested that the graph shown as being from Crowley was actually a displaced version of the MBH99 reconstruction and it was simply another Mann screw-up, but did not provide evidence that Mann “knew” of the Crowley splice. Since Mann has not issued a corrigendum explaining exactly what the screwup in Eos 2003 was, it’s impossible to know for certain – this is the Hockey Team. However, I agreed with Lambert that it might have been simply yet another Mann screwup and did not show that Mann mis-spoke at realclimate.
However, here we have a splice in Mann’s own work, in MBH98 itself no less. How can someone reconcile the splice in Figure 7 and related Supplementary Information with the claim that “No researchers in this field have ever, to our knowledge, “grafted the thermometer record onto” any reconstruction “? I’m sure we’ll find out soon enough. Lambert’s pretty resourceful at explaining away his client’s DNA.
Climate Audit 
22 Comments
This is just silly. You know perfectly well that Mann was referring to presenting a splice as a reconstruction. There was some context to his comment, you know.
You go out of your way to construct some bizarre reading of Mann and then use that misreading to accuse him of lying. I suppose this stuff plays well to the cheer squad.
And I’ve pointed out several times that you have misrepresented my position in your earlier post. Your refusal to correct it is dishonest.
do you have any comments on the Ritson AR1 piece, Tim?
Tim, let’s start and see if we can agree on this. I submit that Mann has “grafted” a thermometer record onto a reconstruction in this Figure and in the related correlation calculations. This can be dealt on its merits and has nothing to do with any past exchange. How can you disagree with this?
I disagree with doing such grafting. It is a bad practice. Tim seems to think it’s only a bad practice if done secretly? But that was not the discussion.
I am assuming that the solar irradiance and DVI index graphs were created by proxy studies with the CO2 graph from ice core analysis, all without any error bars included.
Steve, pardon me for using this thread to ask a general question, but I don’t know where else to ask this on this website.
My question is, what are the alternate energy technologies which, if the Manns of the world had their way, we would now be using? I simply don’t see windmills, solar cells and hybrid cars making much of a difference. Even nuclear energy faces a shortage of fuel.
Thanks
Very amusing to see Lambert accusing someone else of dishonesty and misrepresentation. But I suppose you know very well how to “do” both of those eh Tim ?
BTW, written up your “Mann Screws it Up for the umpty umpth Time” yet ? You sure have plenty of material to go with !
#6. David Smith, I really try to stay away from discussions like that as much as possible. There are plenty of other forums for it.
My own views on economic issues are actually that, if there’s a job to be done, we’re better off getting started on it and trying to figure out those problems. Usually it takes a combination of things. But I’d like to know the job specs a lot better than I think we do right now.
From a mineral exploration point of view, uranium prices have been very low for a generation because of low demand growth and a second factor which people outside minerals businesses would miss – there have been sales from the U.S. government stockpile for a generation, constituting a large proportion of supply. If you surveyed gold exploration prospects in 1976 (and I remember this), there had been virtually no gold exploration for a generation. But in the 1980s, people figured out new exploration models and new targets and there was lots of successful exploration. I have little doubt that higher uranium prices will usher in a new pulse of uranium exploration and lots of new mines will develop. So that would be the least of our worries.
Steve, since your practice is to ignore my answer, invent a position and attribute it to me, it is unnecessary for me to answer your question. Just make something up! It’s more efficient that way.
If you were really interested in a genuine discussion with me, you wouldn’t demonstrate your bad faith so openly.
Tim, I actually try pretty hard to exactly characterize people’s positions and to quote them whenever possible. Let me ask a simple question the answer to which wil be very difficult to mischaracterize:
Do you agree, yes or no, that Mann has “grafted” a thermometer record onto a reconstruction in (a) the top panel of Figure 7; (b) the supplementary information containing the digital version of this series; (c) correlation calculations using the digital data in the Supplementary Information?
The caption to Figure 7 says:
But, Steve wrote over on the other thread about Mann’s Figure 7:
Ummmm … regression coefficients are not correlations. So, did Mann calculate regression coefficients and mistakenly call them correlations in the caption? If so, it would be unbelievably amateurish. Are you folks sure of this?
If they are actually regression coefficients, then interpretation of the results would be very different. It would mean that the magnitude of the impact of CO2 on climate had changed dramatically over time, i.e., THE PHYSICAL SYSTEM ITSELF changes dramatically over time. For instance, CO2 is estimated to have a NEGATIVE effect on temperature for quite a while, and then, in the later part of the period, it is estimated to suddenly have a large impact. (Remember, the regression coefficient is interpreted as the impact on temperature PER UNIT of CO2, not the total impact of the total amount of CO2.
Also, if the regressors are highly correlated (the collinearity problem Steve alluded to) then that could explain the instability of the regression coefficients. Collinearity is notorious for giving results like this that flip flop when there are small changes to the data. If you run univariate regressions for each right-hand-side variable, what do the plots look like?
When I see something like this, I always suspect that we haven’t correctly understood what the author did — it is hard to believe a published paper could be this amateurish.
Re my #11:
Maybe I spoke too soon. If the data is standardized, perhaps a regression coefficient CAN be equal to the correlation (at least in the univariate case). It is not so clear what it is in a multiple regression context.
The point about collinearity is probably still true, though.
Nevermind (perhaps).
Terry, what would an interaction term between Solar and GHG in a linear regression mean? I assume that 3 units of Solar and three unist of GHG would equal 9 units of effect instead of 6. Similarly, 0.5 units of Solar and 0.5 units of GHG would equal 0.25 units of effect, instead of 1. Course this is an argument for the alarmists if it were real, but couldn’t interaction produce the change in CO2 effect you describe?
Sorry but I’m struggling to understand that as a description of atmospheric chemistry. The way I interpret the CO2/temperature relationship is that the pre-industrial levels of CO2 to temperature should be regarded as a stable relationship. Therefore any consideration of the impact per unit of CO2 increase to temperature starts from a neutral position. I understand that CO2 can both reflect certain short-wave radiation and absorb certain longer-wave reflected radiation. It is difficult to see how the swing from a negative effect to a positive effect as Terry describes can occur other than in a marginal sense.
Are you really saying that in order to get the right end result, Mann didn’t care or understand the unreality described by the maths in order to reach that point?
Needless to say, if anyone made a comment on Tim Lambert’s site such as he made here then it would be censored (same happens at “real”climate). Tim knows that Steve doesn’t censor and plays on that to deny and/or obfuscate.
Needless to say, Jack Lacton has made many comments on my site without being censored. On the other hand, I’ve had many comments I’ve made here deleted. To be fair, the deletions have stopped since John A vanished, so it’s pretty obvious who was doing it.
re: #14
I’ve never heard of CO2 reflecting short-wave radiation. What frequencies and to what extent?
Jack, I have not experienced any censorship at Tim’s site though I have disagreed with him there on many occasions. Whatever Tim’s opinions may be on other subjects, he appears to be an ardent supporter of free speech and I salute him for that.
#16
Tim,
In one of my first posts to your blog I made a comment about whatever topic and then described you as something along the lines of being a modern day Lamarckian Lysenko character. I believe you censored that one, which, by the way, I didn’t have a problem with and I’ve moderated my approach subsequently. Your use of the term “many comments” is a bit of a give away, really.
In regards to this site, I’d be interested in understanding what you’ve had censored. I’m a long term reader of climateaudit (though infrequent contributor) and can’t imagine what you’d be censored on given what is posted by others. Without trying to get into the ad hominem, discredit the person approach so favoured by pro-AGW sites, I’d reasonably suggest that you’re hardly a noted expert in the area of climate science, statistics or earth sciences, which puts you in the category with some of the other contributors such as Hearndon, TCO et al who seem to have no problem with unresearched and/or biased postings.
Regards
#17 – Dave I have no joy as yet locating where I read that. It may well be that I’ve misunderstood (wouldn’t be the first time) what was being said. I was just trying to be as fair as possible to Mann as regards what Terry explained. IIRC, its a really small effect anyway but I’ll keep looking to clarify the point.
So Tim, are you going to address the point, or just continue to rope-a-dope?
You look about as classy as your hero when he dismisses criticism of his work with haughty remarks like
Now there’s a class act.
I’d have to agree with Tim Lambert here that MBH98 Figure 7 top merely plots the two series as separate lines (one solid, one dotted) on the same graph, identified as such in the caption.
There may be some trick smoothing in MBH98 fig. 5b, but that’s a different matter. I’ve asked UC about it over at http://signals.auditblogs.com/2010/02/26/the-trick-timeline/comment-page-1/#comment-208.
2 Trackbacks
[…] an interesting knock-on effect from the collapse of MBH98 Figure 7 (see here and here). We’ve spent a lot of time arguing about RE statistics versus r2 statistics. Now think about […]
[…] week, through Chefen, Jean S and myself, here here here and here , we showed that MBH98 contained questionable statistical methodology for assessing […]