RealClimate logo

Lu: from ‘interesting but incorrect’ to just wrong

Filed under: — gavin @ 5 July 2011

Some readers might recall a story from a couple of years of ago relating polar ozone depletion to cosmic rays and the subsequent failure of predictions made using that theory. The idea came from from a Qian-B. Lu (U. Waterloo), and initially seemed interesting (at least to those of us who were not specialists). Perhaps cosmic ray induced chemistry was playing some part in releasing chlorine from CFCs as well as the more accepted idea of heterogeneous chemistry on polar stratospheric particles? Lu’s predictions for increased polar ozone loss in 2008/2009 as a function of the low solar activity (and therefore higher CR flux) did not come to pass. Worse (for this idea), new analyses demonstrated that the hypothesized CR-induced CFC loss wasn’t detectable at all.

Undaunted, Lu continued to publish his ideas, though without really dealing with the criticisms, and indeed extending his scope to the issue of climate change as well as ozone depletion. He made a new claim that since CFC concentrations correlate better with temperature change, and that implies that CO2 can’t have an impact on climate. Very odd logic indeed. Unsurprisingly, his newest contributions have ended up in less and less mainstream publications. His last paper (Lu, 2010) was in the “Journal” of Cosmology – a recent online production that has been associated with a number of ‘fringe’ ideas (to be polite).

The paper before that Lu (2010, Phys. Rep.) has now come in for a real spanking from Grooß and Müller (2011) in “Do cosmic-ray-driven electron-induced reactions impact stratospheric ozone depletion and global climate change?”. From the abstract:

Here we show that these arguments based on the CRE mechanism are inconclusive. First, correlations of satellite data of CFC-12, N2O and CH4 from ACE-FTS show no evidence of significant loss of CFC-12 as predicted by the CRE mechanism. Second, conclusions drawn about a possible CRE impact on the atmosphere, based on correlations of different observed atmospheric parameters, do not have a physical basis. Finally, predictions … based on these correlations are not reliable for either the ozone hole or global surface temperatures.

In my opinion the term ‘inconclusive’ is very polite indeed. The paper shows very clearly that there is no loss of CFCs through interactions with cosmic rays since if there was you’d see a change in the ratio of CFCs to CH4 or N2O (relatively long-lived gases) in the stratosphere. And you don’t. This was exactly the same (and completely valid) point made by the same authors in their rebuttal of Lu’s earlier paper (Müller and Grooß, 2009). However, since Lu obviously took no notice of that earlier criticism, it is impressive that Grooß and Müller took the trouble to rebut his claims even more thoroughly.

As we discussed recently, the role of providing rebuttals to bad papers in the literature is mostly thankless, but it is necessary. Hopefully for Müller and Grooß this is the last time they’ll need to.

40 Responses to “Lu: from ‘interesting but incorrect’ to just wrong”

  1. 1
    Nick Barnes says:

    That’s Müller, not Muller.

    [Response: Oops. fixed. Thanks. – gavin]

  2. 2
    Andy Park says:

    I think that this story illustrates the perils of overspecialization. Specialization in research is a necessity, but when you become so specialized that you lose sight of the big picture of your discipline – in this case it appears that Lu lost sight of both the big picture of atmospheric chemistry and basic logic (since even if CFCs did have a role in warming, logically that does not preclude a role for CO2). A lot of climate skeptics seem to make the same basic error – using the type of logic that “if the sun, then not CO2”, rather than “if the sun, then possibly the sun and CO2”.

    The big question is, of course, how can people gain extremely specialized knowledge, yet still keep the big picture in sight?

  3. 3
    Bushy says:

    That assumes that your interpretation of the big picture is correct.

    [Response: Hmmm…. Let’s think about the two options – nobody knows anything about anything in which case everyone’s opinion is just as valid as anyone else’s, or there is a scientific method which allows opinions/theories/hypotheses to be tested and for the ones that don’t work very well to be put aside allowing us to ratchet up to a better state of knowledge over time. Or it’s all a conspiracy. Your call. – gavin]

  4. 4
    Donald Oats says:

    The answer to the big question posed by at the end of the post by Andy Park [#2] is: “Read widely.”

    [The extended version is: “Read widely, but stick to the best of the best of peer review.”]

  5. 5
    Bushy says:

    Well, Gavin, you gave me three options there, not two.
    I will discount the conspiracy theory as well as the “nobody knows anything about anything thing”.
    However science and progress has shown time and time again that informed dismissal of hypotheses has not been productive in so many scenarios.
    The whole CO2, Ozone, feedback issue is so mired in uncertainty that I fail to see where you can even start to discount this chaps work without at least giving it some consideration or at least. If you are going to discount it so forcefully in public – your scientific rebuttal instead of just contempt and dismissal is required.

  6. 6
    Lauren says:

    Yes, it is very important to rebuttal bad literature since there are so many people out there making false claims. How do you distinguish what literature is “bad” though?

  7. 7
    Bushy says:

    Looked at your reply again and option 2 0f 3 is best of a bad bunch.

  8. 8
    Bushy says:

    Why do you feel that it is ok to just edit and distort the comments by your readers? The comment at 6 is not what I said except for a cherry picked few words that are useless without context. Good heavens chaps, you have to man-up or state your case. Obfuscation is going to get you absolutely nowhere. You are being observed by so many.

    [Response: I have no idea what you are talking about. If comments are edited it is noted clearly with an “[edit]” and this is only done in the event that an interesting point is obscured by something off-topic or abusive. Your comment #6 was approved without delay and was not edited by anyone. You did have a comment that got caught by the spam filter (which happens) which I have pulled out. – gavin]

  9. 9
    Steve Fish says:

    Bushy. If all three of Gavin’s choices make up a bad bunch, I would be fascinated with what you think is a good fourth option. Steve

  10. 10
    One Anonymous Bloke says:

    Lauren #5 Did you read the article? Did you notice that Lu’s hypothesis didn’t stand the test of observation? Pretty straightforward really. Once you get the hang of it you can apply this method to all sorts of things.

  11. 11
    Ed Davies says:

    In my undergraduate days one of the post-grad students I knew well plastered his office door with all sorts of weird rantings on a Sunday afternoon. We all thought he’d been taking something less than completely legal and looked forward to telling not to be such a twit when he came down. In a way he had been: in effect his sports training and intense approach to his work had resulted in some condition where he basically overdosed on his own adrenalin. Luckily for him the lab manager’s son had had a similar problem so she recognized what was going on and medical intervention was arranged by the end of Monday morning.

    Obviously it’s impossible to tell from a distance but I do wonder whether some other scientific careers aren’t blighted by similar but less acute problems. It raises all sorts of questions about how feedback should be given and if, after help, it’s possible for somebody’s career to be put back on track.

    The rather intensely polarized atmosphere surrounding climate science would, presumably, make dealing with this sort of thing even harder.

  12. 12
    Jack Mott says:

    Lauren – either you need to be a little bit smart and do some hard work to check the literature, or you need to be really smart and do a little bit of work to check the literature. The key here is work. You might have to, get a calculator out, or look things up, or think carefully about the logic behind conclusions to see if they hold.

    you might even have to go to the lab or into the field and take some observations and perform experiments!

  13. 13
    simon abingdon says:

    #2 Andy

    “A lot of climate skeptics seem to make the same basic error – using the type of logic that “if the sun, then not fairies at the bottom of the garden″, rather than “if the sun, then possibly the sun and fairies at the bottom of the garden″.”

  14. 14

    I’m repeating myself here (from posts long past), but I’d still like to see a very simple database that merely cross references papers that reference, support or rebut other papers, with the intent of being able to see not only if a paper has been published, but if it has stood the test of time.

    It’s far too easy these days (on the Internet and blogosphere, mind you, I’m sure not among real scientists who keep up with their fields) to cite a paper just because it’s been published somewhere by someone. I still run into people citing Lindzen and Choi (2009) as the final nail in the AGW coffin.

    I’d love something like google.scholar with simple refute/support checkboxes on the search, or columns/tags on the search results.

    Of course, I suppose then people would just publish as many supporting papers as they could at the Journal of Saying What I Want To Hear Said (which is probably the most prestigious and long standing journal I know of).

  15. 15

    Simon at comment 13 demonstrates nicely another denier (I refuse to use the term skeptic) approach, which is to view anything they don’t want to believe as pure magic, so magical that they view it as beyond even being worthy of dispute. The witch scene from Monty Python and the Holy Grail always comes to mind when I see such displays of ignorance scientific acumen.

  16. 16
    Edward Greisch says:

    How do we simulate cosmic rays in the lab? With a particle accelerator beam. We could do the experiment, but it seems pointless.

    It seems to me that solar particles have plenty of energy to create and destroy ozone and destroy CFCs, so what would be the difference? The energy required to do chemistry is very low compared to cosmic rays. We use high energy machines to probe things much smaller than molecules.

  17. 17
    simon abingdon says:

    #15 Bob

    Did you consider whether I might be supporting Andy’s position? Or did you not?

  18. 18
    Jim Eager says:

    Only an imbecile could equate “fairies at the bottom of the garden” with CO2 in an analogy.

  19. 19
    simon abingdon says:

    #15 Bob

    It’s evidence that convinces me, not appeals to Monty Python.

  20. 20
    David McCabe says:

    Lu is a truly special case. Does his university still put out the fawning press releases about how he is changing our understanding of atmospheric chemistry with his groundbreaking research?

    I never read Lu 2010, despite his climate claims, but Lu 2009 (in PRL) was just scary. Actually, Lauren and others, if there was more time in the world, I wish I could go through and chart the logic of the paper, to show you how illogical it is if you just strip the jargon away. If you are really interested, get the article here:

    and look at how Fig 2 is interpreted on in the paragraph beginning “One might argue…” at the end of p 3.

    A) if you plot two periodic phenomena with similar frequencies next to each other, especially for 1 – 3 cycles, you have not proved causation

    B) you cannot quantitatively diagnose non-linear systems with simple linear arguments, that just does not make sense.


    And, agreed…Rolf Müller’s work has been tireless to document that these articles are meaningless, illogical speculation.

  21. 21
    One Anonymous Bloke says:

    Simon Abingdon #19 Legions of policy makers are salivating with anticipation of your decisions, are they?
    “Carbon tax or cap and trade?”
    “Hmm, I’m not sure, we’d better check what Simon Abingdon thinks…”
    “Are you sure he has the technical expertise to assess the evidence?”
    “He must have, he comments on blogs!”

  22. 22
    Hank Roberts says:

    > It seems to me that solar particles have plenty of
    > energy … compared to cosmic rays.

    Logic leads you astray; these are not large objects with fixed diameters like hailstones or bullets, where being hit is fairly certain and damage follows.

    At the scale of atoms and molecules, most of everything is space — empty — and a higher energy particle is more likely to pass through more material interacting, while a lower energy particle meandering through the same stuff has a higher chance of an interaction.

    Search term suggested:
    “capture cross section: A measure of the probability that an incident particle or photon will be absorbed …”

  23. 23
    Eli Rabett says:

    What bothered Eli about the Physics Reports paper was that Sigrid Peyeimhoff, the editor who oked it maybe because Lu traced his idea back to a paper of hers. Peyerimhoff is no slouch but she let Lu off the leash in the last half of a review paper on dissociative electron attachment.

    There were lots of reasons for not taking Lu seriously, and Eli named a lot of them at the time, but the damage was done by how the rejectionists ran with the paper.

  24. 24

    “If you are going to discount it so forcefully in public – your scientific rebuttal instead of just contempt and dismissal is required.”

    Um, isn’t that rebuttal provided in Grooß and Müller (2011) which is linked above? Or do you think Real Climate needs to come up with more reasons that Lu’s work is incorrect?

  25. 25
    WVhybrid says:

    Regarding Sphaerica (Bob)’s request that “I’d still like to see a very simple database that merely cross references papers that reference, support or rebut other papers”, there is such a listing for debunked drivel. Try looking at:

  26. 26
    Eli Rabett says:

    Eli has done a quick read through and provides some detail, in addition to the detail from earlier on.

  27. 27


    Actually it looks like you lean toward the conspiracy side. With well crafted spin to wrap it in so that it looks innocuous. Sort of the, I’m not a conspiracy theorist, but I do run into these folks like ‘Gavin’ who are editing my comments to ‘hide the truth’ thus supporting the conspiracy theory that … and I’m not saying it’s a conspiracy, but hey look my comments are mysteriously being edited…

    This line of inferred reasoning reads like Fox News reasoning on how they would never use Nazi references on their broadcasts:

    It’s a style of inferred argument that I find quite distasteful.

  28. 28
    deconvoluter says:

    Re #24

    there is such a listing for debunked drivel.


    Anti-AGW papers debunked

    But Solomon et al’s paper is neither drivel nor an anti-AGW paper.
    I think it is a mistake to include a minority of papers classified with a little (M) meaning a misunderstood main stream paper. They should be put into a separate list to avoid confusion.

    Otherwise it won’t be long before you see the authors of such (M) papers claimed by contrarians as supporting their position.

  29. 29

    25 WVhybrid,

    Thanks, I was aware of that. I was actually thinking of something more comprehensive, and more scientific (i.e. not “blog” rebuttals… the deniers would have a field day making every real paper that contributes to the science look like it’s been “debunked”).

    And I’d think it would apply to papers regardless of what they prove. There are lots of things that science “believed” ten years ago that are now being reconsidered, and my interest applies to those cases as well as papers that are quite simply wrong the day they’re published. I’d just like a way of making sure that I’m always referencing all of the most current and valid knowledge, rather than something that has either since turned out to be less solid than once thought, or was perhaps (as in Lindzen & Choi, 2009) immediately refuted.

    I would think such a tool would be useful to grad students, and to professional scientists who are doing research or applying factors from outside of their usual area of expertise.

  30. 30
    Hank Roberts says:

    > I’d just like a way of making sure that I’m
    > always referencing all of the most current and valid knowledge

    Say you do find someone who claims to be providing this service for you.
    You’d trust them to be doing it for your benefit?

    Nobody will do this for you.
    The world changes every day.
    Any summary, review, consensus statement, or article is out of date.

    Ask a librarian.
    Learn how to use the tools.

    Yes, publish/print/post what you find citing your sources.
    But the next person along still has to check what _you_ claim.

    You may be able to trust the various different services that give you “cited by” lists for articles — but their lists differ, and the lists have to be updated by people, it’s not magic.

    Your choices: do the work, or wait patiently for the Google “Wisdom” button to be implemented.

  31. 31

    Andy #2
    The common way a researcher can cover broad areas necessary for publishing a paper with implications beyond one’s specialty is to collaborate with experts in those areas. Someone who publishes by themself is either writing something wholly-contained within their expertise or is a crank.

  32. 32
    Eli Rabett says:

    Hank cries hammock and looses the librarian avengers.

  33. 33
    WVhybrid says:

    Sphaerica (Bob) said ” I was actually thinking of something more comprehensive, and more scientific”.

    Over the years several rebuttal articles have been discussed on this blog. The archives probably contain many if not most of the peer-reviewed rebuttals you are talking about. Bob, please don’t let me stand in your way of developing the database you are talking about. Maybe you could even get some of the contributors to this site to help with it. For instance, I might help.

  34. 34
    One Anonymous Bloke says:

    Sphaerica (Bob) et al. Aren’t you describing Wikipedia or something similar? To my ‘mind’ synergy is better than invention. “Wikipedia Scholar???”

  35. 35
    WVhybrid says:

    Re: 34 One Anonymous Bloke

    Are you suggesting a Wikipedia article about a climate issue? Ah, you go first. I’ll check your spelling.

  36. 36
    richard pauli says:

    The classification of “Not Even Wrong” may apply. Nicely described by Wikipedia

    “The phrase implies that even a wrong argument would have been better than the argument proposed, because an argument can only be found wrong after meeting the criteria for a scientific hypothesis (proper assumptions, falsifiable, makes predictions). Arguments that are not even wrong do not meet these criteria.
    The phrase “not even wrong” is often used to describe pseudoscience or bad science and is considered derogatory.”

  37. 37
    Brian Dodge says:

    Bushy says: 5 Jul 2011 at 10:43 AM “…scientific rebuttal instead of just contempt and dismissal is required.”
    simon abingdon says: 5 Jul 2011 at 3:43 PM “It’s evidence that convinces me, not appeals to Monty Python.”
    Back in 2009, I downloaded monthly data from Jan 1969 through fall 2009 (ozone from and GCR from and did a scatterplot using Appleworks. I’ve put it on the net at I’ve also compared Oulu GCR to HadCRUT(from There’s not an obvious correlation between O3 and GCR; if anything, there’s a slight indication that higher GCR rates give higher temperatures, but the error bars are large; and increasing CO2 correlates to increasing temperatures, expected from the radiation physics.

  38. 38
    Brian Dodge says:

    Wups – in my comment @2:49. I left off the link for Oulu vs HadCRUT –

  39. 39
    DP says:

    A query concerning the global temperature anomoly for June 2011. The NOAA puts it at 0.58C which makes it the 5th warmest on record. NASA/GISS on the other hand puts it at just 0.5C. Which is likely correct or is there a misprint on one of them?

    [Response: The anomalies on each data set are with respect to different baselines, and so you need to adjust for that prior to comparing the numbers (reduce NOAA by 0.03). Additionally, the methods, interpolations/extrapolations are performed differently so that on a month to month basis you don’t expect them to be identical – although the correlation is pretty high (0.96 or so). – gavin]

  40. 40
    One Anonymous Bloke says:

    WVhybrid, I think the wikipedia model might be useful, yes. I don’t think a database like this would be comprehensive unless it were the work of a group (a bit like science, really), although I don’t think it would be a good idea to have editing privileges so widely distributed.
    If the idea is that junk gets debunked, you have to display the junk too.