Survivorship Bias

bender has sent in the following interesting graphic showing how drought-caused dieback could bias a tree ring chronology and interfere with interpretation of past records, emphasizing the continued need to keep U-shaped temperature response functions firmly in one’s mind when considering these series.

[image being re-sized]

(2012 note) For reference, here are some related postings discussing upside-down quadratic response and potential evidence of this in recent northern Canadian and Alaskan tree ring series (an explanation for the Divergence Problem not properly canvassed by the NAS panel).

Upside Down Quadratic
Twisted Tree Heartrot Hill
Wilmking in Alaska
Positive and Negative Responders
bender on Gaspé

Other potential biases to tree ring chronologies neglected by the NAS panel (and discussed here from time to time) are altitude changes, "modern sample bias", standardization bias, pith offset bias.

What is the evidence against warmer MWP?

Lee has criticized me for not fully canvassing the supposedly manifold lines of evidence marshalled by the NAS panel against a warmer MWP. So I’ve done a little exercise to summarize the evidence AGAINST the MWP being warmer than mid-20th century, disaggregating what I believe to be the salient information from the spaghetti studies. The information is familiar, but it’s arranged below a little differently than I normally arrange it.

In my opinion, there’s a fair bit of evidence the other way – some examples mentioned here from time to time include: Cuffey’s Greenland reconstruction/Dahl-Jensen’s Greenland boreholes; Naurzbaev in Siberia, Millar et al in California, Law Dome dO18 isotopes in Antarctica, Pollissar on glaciers in the Venezuelan Alps, higher treelines in Scandinavia, Lamb’s European evidence; the Polar Urals update, … But that’s not what we’re here for today.

So let’s turn the question around – what is the evidence AGAINST a warmer MWP?

(1) bristlecone and foxtail ring widths (especially those collected by Graybill in the 1980s) are wider in modern times than in medieval times. (OK, the NAS panel has discounted this, but it’s obviously been used over and over as evidence against a warmer MWP in the spaghetti studies.)
(2) ring widths at Yamal, adjusted for age, are wider than in modern times than in medieval times;
(3) the percentage of coldwater diatoms offshore Oman is higher in the 20th century than in MWP;
(4) dO18 levels in some of Thompson’s tropical ice cores and in the overall average is higher than modern levels;
(5) combinations of the above 4 proxies under a weighted average with small numbers of other mostly nondescript proxies show mid-20th century indices slightly higher than the highest corresponding index in the MWP (the spaghetti graphs);
(6) supposedly some evidence from Antarctica according to the NAS Panel, but they did not provide any evidence and I don’t know what it is;
(7) 5000-year organics from Quelccaya (Thompson 2006, cited by NAS panel)

I realize that my arrangement of 1-5 is based on a POV, but have I left anything out from the NAS panel? I haven’t posted anything on (7) yet, but intend to do so. I’ve written extensively on items 1-5; I’ve got a note up on point 6 inviting a response, but nobody has so far volunteered a guess on what the NAS panel, those sturdy engineers and bridge designers, had in mind.

If the parties were in civil litigation or other proceeding to be decided on balance of probabilities, one would weigh the evidence from treelines, crops etc, on the one hand against the evidence listed in points (1-7) or any others to be added to this list. The case for the Team is far from overwhelming expressed like this.

Sciencemag on House E&C Hearing

Richard Kerr of Science has reported on the House Energy and Commerce Committee hearings. Having lived through the hearings, it’s interesting to see how they get characterized. For example, Kerr says:

He [North] said he doesn’t disagree with Wegman’s main finding that a single year or a single decade cannot be shown to be the warmest of the millennium. But that’s only part of the story, he added. Finding flaws “doesn’t mean Mann et al.’s claims are wrong,” he told Barton. The recent warming may well be unprecedented, he noted, and therefore more likely to be human-induced. The claims “are just not convincing by themselves,” he said. “We bring in other evidence.” The additional data include a half-dozen other reconstructions of temperatures during the past millennium. None is convincing on its own, North testified, but “our reservations should not undermine the fact that the climate is warming and will continue to warm under human influence.”

One thing that I’ve noticed in climate science is how seldom people use the exact words of their opposite party, preferring to re-describe their "main finding" in alternative words. Nine times out of ten, they mis-describe the finding – or, if the describer is on the Team, 100% of the time. How can anyone say that Wegman’s "main finding is that a single year or a single decade cannot be shown to be the warmest decade of the millennium". Not only is this not Wegman’s "main finding", it’s not even a finding of his at all. He didn’t say that the enterprise is impossible; he did say that the Mannian methods that he studied (which were not all of Mannian methods) were flawed in the way that we described them to be flawed and that the flawed method had the properties that we described. He did not purport to say that conclusions about warmest year or warmest decade could not be reached by some other method, although he raised doubts about whether methodologies used by the Team in other reconstructions are likely to provide such a method.

I’ve said over and over how frustrated I am that the due diligence of the NAS panel was so negligible and slight and that they relied on mere literature review for so much of their study. It’s ludicrous for them to say that bristlecones should be "avoided" in temperature reconstructions and then to "bring in other evidence" – a "half-dozen other reconstructions" that use bristlecones – without testing for the impact of bristlecones on these reconstructions. I’ll do the testing of the impact of bristlecones on the other reconstructions, but the NAS panel should have done it themselves.

Their report is most usable on specifics, where they make many useful comments, so their poor performance in one area does not make the report unusable in total and, to date, I’ve tried to focus mainly on the positive aspects of the report.

But when I see how Science describes North’s testimony, one realizes that the failure to assess the impact of bristlecones is not just silly, it’s negligent. Let’s say that the NAS study had been done by an engineering firm – large or small, Bechtel or your local P.Eng. I don’t think that you could find an engineer who could say on the one hand that you can’t use bristlecones ("strip-bark") and then, on the other hand, produce drawings using bristlecones. They’re trained not to do stuff like that.

My guess is that, under the circumstances described here – that bristlecones should be "avoided" in temperature reconstructions, it would be professional misconduct for an engineer to produce drawings which used bristlecones. If an engineer said that certain materials should not be used in a bridge and then produced drawings for a bridge that used these low-quality materials, it would punishable if it were the subject of a professional misconduct complaint. Why should a NAS panel conduct itself with lower standards than engineers?

Some Links to Pielke Jr

Roger Pielke has a few threads recently dealing with the House Energy and Commerce Committee hearings – today there’s an op ed by von Storch and Zorita. A few days ago, he posted his own report and then posted up my response letter, both of which have attracted a few comments.

I agree with both VZ and Pielke Jr a lot of the time, perhaps even most of the time (except for my beef with VZ on how they applied the Mannian PC method to a tame network). In their Op Ed, VZ didn’t mention that Barton appears to have already decided to request an independent review of climate models by a non-climatological panel of NAS. I think that anyone who is concerned about human impact on climate should view this very positively.

Data Request to NAS

I’ve set up a new category at right entitled "Archiving" which cross-classifies many of the posts on data archiving and requests. A review from this time last year is here . Since then, I’ve had correspondence with Moberg and Nature, which has resulted in winkling out the other Moberg series (which I’ve yet to process); extensive correspondence with Science about Esper et al 2002 and Osborn and Briffa 2006, which has been fairly productive, but still not quite complete. I haven’t posted up or discussed all my correspondence by any means. It’s frustrating to look back and see exactly how much time has been consumed in quasi-litigation over data.

The NAS panel relied heavily on articles using unarchived data and I plan to send a letter to Ralph Cicerone, President of NAS, asking him to directly request unarchived data from the various authors. Here is a draft. If anyone can think of things that I’ve forgotten to mention or has other suggestions, I’d welcome them. I want to send the email out in a day or two while matters are still fresh. At that point, I’ll substitute any changes in the version posted here. Continue reading

Splicing in Rutherford et al 2005

Post Washington, I was browsing through the turgid prose of Rutherford et al 2005 – which actually uses the original MBH98 data set (PC series and all) for nearly half their analysis, they also consider the mystery Briffa et al 2001 sites, which they also do not reveal. The SI says – contact Tim Osborn, which I’ve done with no success. I might as well be talking to a wall – a stone wall.

Mann has used this article to support the claim that he can "get" a HS without PC analyses. Of course, the reason for using PC methods in the first place was to achieve geographical balance – otherwise, people might have been worried that the reconstruction was being overwhelmed by southwest U.S. tree ring data and merely be a local effect. Abandoning PC methods without an alternative summarization allows the bristlecones to dominate the reconstruction through the back door.

Rutherford et al 2005 shows a reconstruction without PCs, which was presented to the House Energy and Commerce Committee. I thought that it would be interesting to compare this to the other reconstructions and, in doing so, noticed something interesting. BTW Jean S has written me offline showing a number of very bizarre and unsupportable aspects of Rutherford et al 2005, which we will pursue, but he’s on holiday and offline for 4 weeks. Continue reading

Trip Report

Back from Washington. I think that I’ve had enough Washington for a while. It takes me a lot of time to prepare; I don’t begrudge it and it’s useful to try to put what you’re thinking about into short statements, but it still takes time and it’s tiring. It’s a nuisance that Mann couldn’t get a babysitter last week and this had to be done twice. Continue reading

Whitfield subCommittee II: Mann under fire

Today’s hearing in Congress will be sound-only (for some reason), although it might be visually on CSPAN3 (check your listings)

The speakers will be:

Michael Mann
Edward Wegman
John Christy
Ralph Cicerone
Stephen McIntyre
Jay Gulledge

As soon as the link is available, I’ll post it up.

Update: Link is here and video feed from CSPAN3 is here

bender on Gaspé

New poster bender has written some very thoughtful posts, including some comments on Gaspé which I I’d like to recover from deep in a political thread. The growth pulse in Gaspé cedars seemed very improbable to me as a temperature proxy; bender has some detailed knowledge on the topic. Continue reading

Survivor Season 8: the Hockey Team – the Mann Overboard Episode

Well, back from Washington. I’m not very good at describing reactions and impressions and touchy-feely stuff like that; I’m more comfortable describing what’s different with the 53rd and 84th series out of 112, but I’ll try. Continue reading