More Evasion by Thompson

I’ve been trying since 2003 to get detailed sample information from Lonnie Thompson on his tropical ice cores, some drilled 20 years ago. I reported on my most recent effort on Apr 19, 2007 under PNAS policies here.

Thompson has once more obfuscated a journal by falsely telling them that everything already is archived (without providing links), which the journal has duly retailed to me. I think that the journal editor should have been able to tell that Thompson was unresponsive, but I’ve written back one more time, providing a trail by which the journal can validate for itself that Thompson’s answer was false and unresponsive.

Here is an excerpt from my initial request in which I asked (one more time) for a sample-by-sample archive of isotope and chemistry information for all cores:

Thompson et al 2006 describe results from ice cores drilled at Dunde, Guliya, Dasuopu, Puruogangri, Quelccaya, Huascaran and Sajama. For each core, several thousand samples were taken and analyses on a sample-by-sample basis made for isotopes, chemistry and other indicators. The information for each core constitutes a large data set within the meaning of your policies. There is an excellent public repository for ice core data at the World Data Center for Paleoclimatology, which satisfies your definition of a public repository. Under your policies, Thompson et al had an obligation to archive this data as a condition of publication, but this appears to have been overlooked. Although Thompson et al provided a highly abbreviated summary of isotope information as Supplementary Information, the Supplementary Information is incomplete and not compliant with journal policies.


I request that you ensure that Thompson et al comply with your data policy by forthwith archiving the large datasets used in the PNAS article for each individual ice core (Dunde, Dasuopu, Guliya, Puruoganri, Quelccaya, Sajama, Huascaran) and for the entire suite of isotopes and chemistry. In addition, because the discrepancies may result from changing algorithms for dating the ice cores, I further request that the dating procedure for each core be made available under your Unique Materials policy.

I received the following response today

Thank you for your messages and your interest in PNAS. I apologize for the delay in getting back to you, but I wanted to speak with Dr. Thompson about this request personally and he was out of the office for quite some time. I was able to reach him via phone the other day, however, and can now address your query. According to Dr. Thompson, the data you seek have all been deposited in the archive you specifically mentioned as well as being mirrored on his own website. Let me know if you have any further questions.


Michael Baden-Campbell
Senior Editorial Coordinator, PNAS

Here’s what is presently archived at WDCP (and even this incomplete archive is a result of other attempts by me) for Dunde. You will readily see that it is not sample-by-sample – it’s decadally averaged according to one of a number of inconsistent dating schemes; it does not have a complete chemistry report, it’s only dO18; and it doesn’t cover the entire core, it only goes back to what they date as AD1000.

And yet Thompson has the cheek to say – everything’s already archived. In fairness to the journal, I can understand why an editorial coordinator would presume that what Thompson says is true. Unfortunately, it isn’t true. I’ve been through this dance with Thompson before. He said the same thing to NSF and that attempt went nowhere. In the hope that PNAS will do a better job and actually investigate Thompson’s false claim that he has provided a data archive compliant with PNAS policies, I sent the following letter, re-iterating that the supposed complete data archive did not exist, providing a trail by which PNAS could ascertain this for themselves. Maybe PNAS will actually enforce their own policies.

Dear Mr Campbell,

Unfortunately, the following response from Dr Thompson is simply false: “According to Dr. Thompson, the data you seek have all been deposited in the archive you specifically mentioned as well as being mirrored on his own website”

I am perfectly aware of the highly incomplete summary information archived at WDCP and at Dr Thompson’s website. Indeed, I used this information to plot the attached figure. You can readily verify for yourself that Dr Thompson’s answer is false.

My request was as follows: “Thompson et al 2006 describe results from ice cores drilled at Dunde, Guliya, Dasuopu, Puruogangri, Quelccaya, Huascaran and Sajama. For each core, several thousand samples were taken and analyses on a sample-by-sample basis made for isotopes, chemistry and other indicators. The information for each core constitutes a large data set within the meaning of your policies.”

In a responsive data archive, you could identify the sample number, top, bottom, isotope, chemistry and other indicators. Since several thousand samples were taken for each core, there would be several thousand lines in the archive. If there was more than one core for a site, each core would require a separate data file.

In the case of (say) the Dunde ice core, the only information archived by Thompson at WDCP is here:

This only covers isotope information for part of the core and this is not an a sample-by-sample basis but has been aggregated into decadal averages. The same for other sites.

I re-iterate my request that PNAS ensure that Thompson comply with PNAS policies on these data sets.

Regards, Steve McIntyre


  1. Reference
    Posted May 10, 2007 at 10:53 AM | Permalink

    The Climate Reconstructions section of the World Data Center for Paleoclimatology displays the above graphic – although tiny, it’s possible to see the total abolition of both LIA and MWP periods – no source is given but the name of the graphic is “mann.gif”

  2. JerryB
    Posted May 10, 2007 at 11:22 AM | Permalink

    One minor datum: while communicating with scientists about
    data, treat the word data as the plural of datum, not as
    a singular. 🙂

  3. Steve Sadlov
    Posted May 10, 2007 at 11:30 AM | Permalink

    RE: #1 – mann.gif …. Hahahaha! They must be naive. Or masochistic.

  4. MarkW
    Posted May 10, 2007 at 3:16 PM | Permalink


    I’d put my money on arrogant.

  5. fFreddy
    Posted May 10, 2007 at 4:32 PM | Permalink

    Re #1

    no source is given …

    The Alternate text for that gif is “Mann et al. 1999 temperature plot”

  6. Doug
    Posted May 10, 2007 at 10:27 PM | Permalink

    Re: #2

    After working with seismic data for 25 years, I know that one should say “these data”, not “this data”. I also know that is the fastest way to show that you don’t know what you are talking about.

  7. Phil_B
    Posted May 10, 2007 at 11:12 PM | Permalink

    Most writing style guides say data can be used as a collective noun and hence used in the singular, e.g. ‘the data shows …’.

    Personally, I find using data as a plural irks me, but then my background is in IT, where using data as a plural would probably get you laughed at.

  8. Willis Eschenbach
    Posted May 11, 2007 at 12:26 AM | Permalink

    I find very little data to support the conclusion that the way one uses “data” is any indication of … well … anything …


    (And as an aside, how may people would say “I find very few data to support …”, which is the plural form … )

  9. Jeff Norman
    Posted May 11, 2007 at 5:11 AM | Permalink

    How about “a data point”? Is that okay?

  10. bender
    Posted May 11, 2007 at 5:14 AM | Permalink

    ‘Data’ is plural. Like it or not, that’s the culture, if not the rule. As in: the data of MBH98 ARE most peculiar. ‘Datum’ is singular.

    Failure to conform to this conjugation is an indicator to a specialist that you are an outsider. Of course, most administrators are outsiders too, so use of non-conformist language would probably not affect the chances of a data request being fulfilled.

  11. bender
    Posted May 11, 2007 at 5:15 AM | Permalink

    ‘Data point’ is a compound noun. Singular.

  12. Posted May 11, 2007 at 5:31 AM | Permalink

    Re: Data vs datum

    English being my second language (while Danish a distant third), I tend to be more of a stickler for rules. For example, it kills me that people from the NYC area tend to “wait on line for coffee”.

    Over the first week of my Stats class, I emphasize that the word ‘data’ is plural. However, the way people use this word has changed. Sometimes people (even those of us with Ph.D.’s) just use it as a shorthand for ‘data set’.

    When I write formally, I am careful adhere to ‘correct’ usage. If I am posting on a blog or just having an informal conversation, no one really cares if you say “the data shows this, that and the other”.

    I care more about, for example, whether people can distinguish among observational units versus variables versus a particular observation for a variable in a data set versus statistics calculated from those observations. Someone else’s correct or incorrect use of the word ‘data’ does not convey an y information to me regarding their knowledge or expertise in topics of substance.


  13. Rich
    Posted May 11, 2007 at 5:53 AM | Permalink

    The Greek word “data” is plural. “The English word “data”, derived from the Greek word, is now singular especially in IT contexts. Similarly, the plural of the French word, “gateau” is “gateaux” but the plural of the English word “gateau” is “gateaus” (see any country tea-room menu). Conclusions drawn about people’s education or, worse, quality as individuals, from minor variations of usage in English is snobbery; there’s a lot of it about though.

    My only authority for these pronouncements is that I’m English, speaking English as my mother tongue. Like anyone’s going to listen to me!

    (That was fun. Let’s get back to climate)

  14. mikep
    Posted May 11, 2007 at 6:17 AM | Permalink

    The “Greek” word data is the plural of the Latin word datum – a given….But what about some substance here and less pedantry.

  15. Dave Dardinger
    Posted May 11, 2007 at 8:26 AM | Permalink

    Ok, I’m going to go to the experts. From the 1985 Dictionary of Contemporary Usage, William & Mary Morris. Their panel was asked whether they’d use “The data is…”?

    In writing Yes 49%. No: 51%
    In casual speech Yes: 65%. No: 35%.

    I think the remarks of Isaac Asimov is the most telling:

    “What’s the use of saying, ‘The data are’ when to say it will cause everyone who hears it to consider you illiterate. ‘Data’ is plural in Latin, singular in English.”

    And I’m sure in 22 years data has become even more singular. And don’t start in on “more unique” please!

  16. Stan Palmer
    Posted May 11, 2007 at 8:45 AM | Permalink

    Data is used as a mass noun n English. The concepts of singular and plural do not apply to it

    Wikipedia explains this:

  17. Ian Blanchard
    Posted May 11, 2007 at 10:05 AM | Permalink

    Both my Doctorate supervisor and current boss are linguistic pedants, and both insist on ‘data’ being used in the plural only. Not sure if there is a difference in this between English English and US English (Two nations divided by a common language).

    In common useage, ‘data is’, in science ‘datum is…data are’

  18. John Nicklin
    Posted May 11, 2007 at 10:28 AM | Permalink

    Ok, so you say tomato and I say tomato. Are we trying to prove that Thompson evaded archiving a datum or that he evaded archiving his data?

    It was a nice discussion on the etymology of data however.

  19. bender
    Posted May 11, 2007 at 5:01 PM | Permalink

    The point is: if you want your data requests to be taken more seriously, you’re best off using whatever conventions the people in control of the data use. It’s not about rules, it’s about cultural conventions. And not amongst IT folks, but amongst climatologists. When in Rome …

  20. Curt
    Posted May 11, 2007 at 5:15 PM | Permalink

    Does anyone say, “The agenda for the meeting are…”? It’s the same issue, although I don’t think “agendum” ever really caught on in English.

  21. bender
    Posted May 11, 2007 at 5:33 PM | Permalink

    It’s not about what’s right or what’s consistent with some other arbitrary conventions. It’s about what the prevailing culture is. Get over it.

    Here’s a test for you to try. Try publishing a climatology paper with “the data suggests …” in it, and see how far you get. I know what the result will be. You will be told to change it to “the data suggest …”.

    You are appealing to the wrong authority.

  22. Roger Bell
    Posted May 11, 2007 at 7:55 PM | Permalink

    The Senior Editorial Coordinator of PNAS is too far down the food chain. Look to see
    which funding agency Thompson thanks in his papers.
    If I were you I then would write to the following people complaining that Thompson is not meeting his obligations to science by not archiving his data:
    The President of Ohio State University;
    The President of the National Academy of Sciences;
    The Head of NSF – assuming Thompson is funded by them;
    Anyone at NASA who is funding Thompson;
    The President of the Hational Science Board – NSF’s Board of Directors.
    Roger Bell

  23. Steve McIntyre
    Posted May 11, 2007 at 8:40 PM | Permalink

    #22. The complaint was to PNAS not the editorial assistant. I’ve already tried : Science, NSF and the President of NAS, all of whom refused to intervene.

  24. James Lane
    Posted May 12, 2007 at 3:03 AM | Permalink

    I agree with Bender and others that the audience is relevant. In my field we use the word “data” all the time. I always use “data” as plural, and it really grates when I hear someone in my industry use it singular. One of the great things about the English language is its precision, and data is not equivalent to datum. If I hear a professional use the word, I want to know whether s/he means an individual statistic or a series (or whatever). At the same time, it doesn’t really bother me to hear data as singular on TV or in social conversation.

    In written English “data” should always be plural IMO. That said, I would expect that in 50 years the singular form will be standard. I’m not sure the word even entered common usage before the 1960s, but don’t quote me on that.

  25. mikep
    Posted May 12, 2007 at 3:19 AM | Permalink

    RE 20, Agenda is “things to be discussed”, so properly describes what we call an agenda. Agendum is presumably one of the items on the agenda.

  26. Steve McIntyre
    Posted May 12, 2007 at 5:06 AM | Permalink

    Enough parsing please!

  27. KevinUK
    Posted May 12, 2007 at 8:24 AM | Permalink

    So Steve,

    Just what does Lonnie Thompson have to hide then? He must surely have something to hide (a lot IMO) otherwise why else would he be so obstructive? Do you suspect that he has more to hide than Michael Mann did for example? Ross and yourself have done a great job in revealing that that particular Emperor (the hockey-stick) had no clothes so I suspect LT’s primary reason for being obstructive and so avoiding proper application of the scientific method by independent ‘true’ scientists like yourself is that as in the case of the hockey-stick (and recently in the case of the SST ‘bucket adjustments’ and Phil Jones UHI effect none existent but most definitely required adjustments) that LT’s work has little if any substance to back his conclusions but instead is clearly politically motivated and funded.


  28. Ron Cram
    Posted May 12, 2007 at 8:44 AM | Permalink


    Why not complain to the Wall Street Journal (same reporter who reported Mann’s data withholding) or to Congressman Barton? It seems to me that got results last time!

  29. Al
    Posted May 12, 2007 at 11:58 AM | Permalink

    If you read through here: The Nation You’ll find this quote:

    As Richard Kerr, Science’s man on global warming, remarked, “Climate modelers have been ‘cheating’ for so long it’s almost become respectable.”

    It might be worthwhile to write either Richard Kerr or the author of the article in The Nation. The Thompson data has appeared in Science – by inclusion if by no other method. And having a respected journal asking questions might throw a little more weight behind the requests for clarification.

  30. dave
    Posted May 12, 2007 at 2:15 PM | Permalink

    You can access articles on the WSJ site for free using a netpass from

    All the brokers in our office use that.

  31. Captain Kirk
    Posted May 13, 2007 at 9:00 AM | Permalink

    Data are an android.
    Data is an android.

  32. Posted May 30, 2007 at 7:58 AM | Permalink

    It looks like data hiding pays off. This from

    COLUMBUS, Ohio, May 29 /PRNewswire-USNewswire/ — Lonnie Thompson, the Ohio State University glaciologist who has probably spent more time at high altitudes than any other person, was named today to receive the National Medal of Science for his work providing explicit evidence of global climate change.

    The award, arguably the highest honor the United States bestows on an American scientist, caps nearly three decades of research by Thompson in some of the world’s most remote regions. Thompson is one of eight researchers who will receive the award later this summer during a formal ceremony at the White House.

  33. Posted May 30, 2007 at 8:38 AM | Permalink


  34. MarkW
    Posted May 30, 2007 at 2:12 PM | Permalink

    A bunch of policians, rewarding their pet scientist who has reported what they want to hear.

    Big whoop.

    And Jimmy Carter really has done something to deserve his Nobel Peace Prize.

  35. D. Patterson
    Posted Jan 7, 2008 at 3:27 AM | Permalink

    Hold that expedition!
    Before Dr. Lonnie Thompson departs for New Guinea to put some more glacial ice samples beyond the reach of other researchers, may we ask him to please give full disclosure of the previously undisclosed sample data? After all, if Dr. Thompson should meet with an unfortunate accident on the expedition to sample New Guinea’s glaciers, the world may be forever deprived of the original data and Dr. Thompson’s familiarity with his data collection methods from the previous decades of research in addition to the loss of the forthcoming glacial data from New Guinea. Indeed, perhaps it would be better that Dr. Thompson send someone else whose unfortunate loss would not so jeapordize so many years of undisclosed data and research methodology? On the other hand, if Dr. Thompson would fully disclose his data and methodologies, then it would not be necessary to…Hold that expedition!

  36. bender
    Posted Dec 9, 2009 at 11:10 AM | Permalink

    Maybe Thompson would be willing to co-operate now?

4 Trackbacks

  1. […] (CA 12/3/06), “Dunde: Will the real Slim Shady please stand up?” (CA 4/12/07), “More Evasion by Thompson”. Possibly related posts: (automatically generated)Irreproducible Results in PNASGore Scientific […]

  2. By Cicerone Then and Now « Climate Audit on Feb 4, 2010 at 5:53 PM

    […] In 2007, I tried to obtain Lonnie Thompson’s data following his publication in Proceedings of the National Academy of Sciences see CA here. Despite clear policies that on their face require the archiving of large data sets, PNAS refused to require Thompson to provide a comprehensive archive of his sample data. The complicity of PNAS in Thompson’s evasion is reported at CA here. […]

  3. […] I’ve been trying for several years to obtain Thompson’s sample data so that these inconsistent results can be reconciled without any success. Last year, Thompson published yet another inconsistent version of his ice core data in the Proceedings of the National Academy of Sciences. PNAS has policies that explicitly require authors to archive data sets – see here. I reported earlier this year on letters to NAS asking them to require Thompson to comply with their data policies. My initial inquiry got nowhere as noted here. […]

  4. […] core information for Dasupo, Dunde, and Gulaya, is detailed (inter alia) here, here, here, here, here, here, and […]

%d bloggers like this: