## More O.B. Confidential

Osborn and Briffa site chronologies differed from Esper site chronologies for 4 sites. Site chronologies can differ depending on the standardization method used; in order to analyze the effect, one needs to see the measurement data. Hundreds of measurement data sets have been archived at WDCP. The really weird thing is that the Hockey Team likes to use proprietary data, rather than archived data – look at mainstay Briffa studies: Yamal, Tornetrask update, Taimyr, Athabaska (Luckman version): unarchived. Since Science policy requires data, I’ve persisted in trying to get the actual measurement data used by Briffa so that I could understand what accounts for the site differences with Esper. Here is the most request (and we’re at 38 emails back and forth and counting:

3. In 4 cases, the Osborn site chronology differs from the Esper site chronology, although in the other cases the versions are identical. In some cases, the date ranges do not match. I do not believe that it is possible to replicate the Osborn version from the Esper measurement data in these 4 cases and surmise that Osborn used a different measurement data set. I therefore request measurement data used by Osborn for the following sites: Polar Urals, Tornetrask, Taymir and Athabaska.

This email did not contain a request for confidentiality. The voice is that of Sciencemag, not O&B.

Esper et al. was not the source for the four series in question, as stated in the SOM of the paper (see paragraphs c and d). The Athabasca series was replaced with a series from Luckman and Wilson. The other three series contain some non-identical tree-ring series derived from the same sites; thus the series they used can not be reproduced using the Esper et al. data; there are fewer tree cores in the Esper et al. data. The source for these three series is Briffa (2000). Osborn and Briffa did not not use raw tree-core measurements, only chronologies that had previously been assembled by others, and these have been deposited. You may want to contact those original authors or those publications if you require their raw data.

Isn’t this cute. Osborn and Briffa are taking the position that the Science 2006 article did not directly use the measurement data, but only the chronologies calculated from the measurement data. Thus, they are taking the legalistic position that the measurement data itself is not subject to Science’s data policy – a point which Science seems to be endorsing. Given that the chronologies were calculated in Briffa {2000), I guess I should try to locate this Briffa fellow. I wonder if he’s got anything to do with the Briffa of Osborn and Briffa.

Give me a break. Science should take the position that they are the big dogs. If Osborn and Briffa want to rely on non-arms-length results published in a journal which does not adhere to Science’s data archiving policies, they should not be allowed to hide the wienie. Science should tell Osborn and Briffa to produce the measurement data.

I’ve not given up on persuading Science of this. However, in the mean time, I’ve written to Osborn and requested that they voluntarily provide the measurement data, but no answer so far.

It is so ridiculous – Briffa 2000 was published 7 years ago. The data from these 4 sites (Yamal, Taimyr, Tornetrask update, Athabaska- Luckman version) are being used over and over again in Hockey Team studies and it is still unarchived.

What if Briffa, as a lead author in IPCC 4AR, hypothetically featured results from Osborn and Briffa 2006? Wouldn’t it be objectionable if Briffa, under such circumstances, didn’t readily and enthusiastically respond to requests for supporting data?

1. fFreddy
Posted May 10, 2006 at 2:13 AM | Permalink

Sorry, getting lost again.
Where was Briffa 2000 published ?
Does the Science policy explicitly say that you can either reveal all your data OR rely on some other magazine with less a explicit data policy ? Doesn’t the “reveal all data” requirement flow through to cited sources ?

2. John A
Posted May 10, 2006 at 4:24 AM | Permalink

So is Sciencemag basically asking Steve to ask Briffa to ask himself for the underlying data or did I miss a step?

3. Gary
Posted May 10, 2006 at 6:51 AM | Permalink

All this non-compliance with data archiving/disclosure policy is getting confusing to someone not deeply familiar with the details. Any possibility of putting a basic facts summary in table form (study name, authors, journal, what isn’t archived, date first requested, excuse for non-compliance, etc.)? For balance and respectfulness, include those who have complied, graciously and grudgingly both.

4. JerryB
Posted May 10, 2006 at 7:49 AM | Permalink

Such a table might be nice, but it could take quite a bit of work, it would be difficult to make it complete, or even know how complete it would be at any point in time, and maintaining it could be a nuisance.

5. Dave Dardinger
Posted May 10, 2006 at 7:52 AM | Permalink

Reading between the lines, it would seem that a lot of what we have here is an attempt by the Team to see exactly what they can get away with combined with a desire to kick the can down the road. Combined with the “we’ve moved on, why don’t you?” meme, they hope to provide a shell (hopefully not a CaCO2 shell given the worry about future undersaturation and all that), within which they can maintain the story to the public that the science is agreed upon and people like Steve M are merely nipping at their heels.

I’m sure the recent demonstration by Steve of just how quickly he can demolish new claims, provided he has the actual data used will have done nothing to make the Hockey Team desire to be more transparent.

6. Gary
Posted May 10, 2006 at 8:59 AM | Permalink

RE #4
True, so keep it as simple as possible. Annual reports for companies have a summary page of the auditor’s detailed report. At least this would quantify the non-compliance/non-disclosure problem. For example, 80% of publications failing to meet journal policies is much more important than 5% – and deserving of stronger criticism.

7. John A
Posted May 10, 2006 at 9:08 AM | Permalink

Re: #5

Dave, I think you’ll find that carbonate shells are $CaCO_3$

Yes. I’m being pedantic.

8. jae
Posted May 10, 2006 at 9:41 AM | Permalink

I wonder if some of these guys will not release data, simply because they don’t know where it is. They have all these graduate students and technicians helping, and things probably get very confused. They probably can’t even reproduce their OWN “reconstructions.”

9. John A
Posted May 10, 2006 at 10:02 AM | Permalink

Re #8

That’s not an excuse. It’s just sad.

Why are we then attaching so much importance to their work if they behave like that?

10. Dave Dardinger
Posted May 10, 2006 at 10:17 AM | Permalink

re#7

Actually I caught myself typing CaC02, backspaced to change it to 3 and then typed 2 again and failed to notice. Bad finger training! I also catch myself typing Stever instead of Steve a lot, but I think I’ve always caught that.

11. per
Posted May 10, 2006 at 6:05 PM | Permalink

I am not sure I have this right.
O&B described their methods, and gave a citation. However, this citation is wrong, and another paper (B,2000) is the actual source of the data. Is the original data from B(2000) archived ?

I can see that Science has a strong vested interest in not finding too many errors in the work it publishes in its pages. But if it is arguing that all is well when the original data reference was wrong, and the real reference doesn’t have the information required, it seems to me that a good dose of daylight on this dusty process is required !
yours
per

12. Pat Frank
Posted May 10, 2006 at 7:46 PM | Permalink

#11, Per, if Science has a strong vested interest in not finding errors in its pages, then the editorial staff should be replaced in its entirety. Science ought to have a strong vested interest in finding every single error in its pages, publicly letting everyone know about them, and then expunging or retracting them. Perhaps the most important thing any scientist should know (apart from integrity in data) is that science is not about image.

13. Steve McIntyre
Posted May 10, 2006 at 9:34 PM | Permalink

#11. There are a number of errors in Osborn and Briffa 2006 which would be frustrating to anyone trying to replicate them.

In data that I’ve looked at so far, there are a number of incorrectly cited data sets. The position of Osborn and Briffa, seemingly endorsed so far by Science, is that the incorrect identifications don’t “matter” to the final answer and therefore no Corrigendum is required.

A site “chronology” is type of standardized average taken from measurement data, which is highly sensitive to the form of standardization. Four chronologies used in the study were calculated in Briffa 2000. The underlying data for Briffa 2000 was never archived, even those these series are used in many studies (not just Osborn and Briffa 2000).

14. Jack Lacton
Posted May 11, 2006 at 3:30 AM | Permalink

I notice that none of the noisy crowd from the ironically named Realclimate (& Friends) rarely make comment on the threads about non-disclosure of data.

They’re happy to get all hot and emotional and argue that the science is right (and using the industry funding inanity to discredit people on the way through) while having NEVER seen any of those data that provide the proof of their position.

“I’m right.”
“Show me the data.”
“No.”
“No?”
“No.”
“Have you seen it yourself?”
“I don’t need to. I’m right.”
“Ooooooookay then…”