Coffin, meet nail.

For those who are not mathematically inclined and did not entirely follow the discussion about Eric’s reconstruction in the previous post, well, a picture is worth a thousand words.

This is what happens to Eric’s reconstruction when you:

Top row:  Add the designated trends to the Peninsula stations

Second row:  Remove the designated trends from the Peninsula stations

Third row:  Treat Byrd as a single station, and add the designated trends to Byrd and Russkaya

Fourth row:  Treat Byrd as a single station, and remove the designated trends to Byrd and Russkaya


Please note how Eric’s reconstruction responds quite well to changes in the Peninsula . . . except it teleconnects them to the Ross Ice Shelf and the south pole.

Please also note how Eric’s reconstruction does not respond at all to changes in the only two West Antarctic land stations they used:  Russkaya and Byrd (the response is even less if only the manned Byrd station is used).

Anything that he “got right” . . . as I said before . . . was by accident.


I am quite tired of people who are willing to spend tens of pages during a review making claims without bothering to check.  I am quite tired of people who are willing to spend a couple of hours writing posts about how they “got it right” and I “got it wrong” without bothering to check.

Eric . . . feel free to confirm this for yourself.  I assume you have your own code handy.


  1. mrsean2k
    Posted Feb 8, 2011 at 12:56 PM | Permalink

    Typo I think, “Third Row” and “Fourth Row” descriptive text are identical in the intro text but graphic annotation looks correct – feel free to zap

  2. Posted Feb 8, 2011 at 12:57 PM | Permalink

    It would be cool to see the same with the O10 reconstructions. So many versions isn’t necessary.

    • Posted Feb 8, 2011 at 1:14 PM | Permalink

      Yep, but it takes longer to do those. Fear not . . . there will be an update. 😉

  3. JWS
    Posted Feb 8, 2011 at 12:59 PM | Permalink

    Should the fourth row be “remove” rather than “add”?

    Nice, btw.

  4. Paul Penrose
    Posted Feb 8, 2011 at 1:01 PM | Permalink

    Thanks for all the hard work Ryan. The pictures really do make it quite clear what the effects of your sensitivity tests are on the original reconstruction.

  5. Posted Feb 8, 2011 at 1:06 PM | Permalink

    Ryan, Could you put a couple of dots on one of the maps to locate the Byrd and Russkaya stations? Perhaps Peninsula stations as well?

  6. Sean
    Posted Feb 8, 2011 at 1:10 PM | Permalink

    Be sure to create ASCII versions of the images so that Eric can read your research in his comments rather than ignore a site he never visits.

    • Jason
      Posted Feb 9, 2011 at 12:55 AM | Permalink

      OK that made me LOL.

  7. Sean Houlihane
    Posted Feb 8, 2011 at 1:20 PM | Permalink

    So… any guesses about how they would chose to spin this if they ‘read’ it. My vote is for Bizarre (using the last thread for a translation).

    • Dean
      Posted Feb 9, 2011 at 8:44 AM | Permalink

      As Jeff C posted above, they’ll just say that it’s not worthy of discussion even on the slowest of days.

      Every time I see this type of behavior I am reminded of the movie “Titanic”, when Rose says “Now they will retreat into a cloud of smoke and congratulate each other on being masters of the universe.”

  8. Posted Feb 8, 2011 at 1:25 PM | Permalink

    It is amazing that Eric Steig still pretends not to understand the basic point shown in your first row.
    In his RC post he is still warbling about “I think that they are right to have retained more EOF patterns than we did, though the main impact of this is only in capturing the strong Peninsula warming” as he did in his original paper.
    Obviously the main impact is not that retaining more PCs ‘captures’ the peninsula warming, but that it prevents the bogus smearing of the warming over the rest of Antarctica in his paper.
    Eric still seems to be in denial about this.
    It is now almost two years since I pointed it out to him on their blog.
    In fact he was probably aware of this when he wrote the paper.

    • Layman Lurker
      Posted Feb 8, 2011 at 3:00 PM | Permalink

      Right you are Paul. You reasoned the obvious flaw even in the earliest of days. I brought it up with him again a few months later on the “Overfitting” thread. Gee, do you think maybe it occured to Steig to take the 2 minutes to tweak the code and run sensitivities to check for methodological artifacts in the reconstruction? This was being suggested to him directly. It was plastered all over blog posts and discussion threads and he could have done it in a coffee break.

      Aside: After I clicked on your link I wanted to go through the comments again. This was the thread with the infamous “Matlab Class” inline snark from Eric at Jeff Id. I did not even get through the first page of comments but it was just one fascinating irony after another. Note the polite posts from Ryan O confirming the lack of impact from the “Harry” glitch. The “overfitting” post has the infamous “…someone called Ryan O…” quote from Steig (in the fine print at the bottom of the post). I think there is a likely Shakespeare quote or something that would be apropos.

    • Pat Frank
      Posted Feb 8, 2011 at 6:18 PM | Permalink

      Layman, “…do you think maybe it occured to Steig to take the 2 minutes to tweak the code and run sensitivities to check for methodological artifacts…

      Eric’s possible undeclared and unpublished sensitivity tests may have ended up in a folder called “Back to 5 PCs Censored.”

  9. tetris
    Posted Feb 8, 2011 at 2:05 PM | Permalink

    Unfortunately, all we are likely going to get from the RC Team is a couple of “elevator musak” articles on the RC site so as to wash this out of their system.
    Meanwhile, thx to you and your colleagues for doing the due diligence. And -our host’s blog policy to the contrary, which I respect – don’t ever feel bad for the short fuse and unwillingness to suffer fools, and saying so [I have a couple of t-shirts that way.. 🙂 ] .

  10. Craig Loehle
    Posted Feb 8, 2011 at 2:33 PM | Permalink

    One hypothesis is a complete lack of spatial vizualization ability on Steig’s part. This smearing problem (false correlation that is) has been so evident ever since the Chladni patterns posts way back when (and is well known, as Steve pointed out then). But it seems that the Team are not bothered by a-causal spatial connections as long as their favored tool gives it to them. Hansen also likes long distance romances, with his arctic data being influenced by land data at a huge distance (vs leaving blank areas on his maps, horrors!).

  11. GerryC
    Posted Feb 8, 2011 at 2:37 PM | Permalink


    Angry people never win a public debate!

    You’ve won the arguement (and won it with your first post)
    I would chill out and relax a bit, its still possible to lose overall unless you stay calm and focussed.
    Don’t let them provoke you more, which is the logical next move from their standpoint and don’t let them manipulate your anger with further provocations.

    • Duster
      Posted Feb 9, 2011 at 11:29 AM | Permalink

      Ah, but angry people can nail down the lid on a coffin. The debate was “won” with the original O’Donnel et al. paper. Steig however seems to be serious about trying to “spin” things so that climatologists will look less like statistical and mathematical incompetents. Even Steig should have realized that the “warm spot” in his images should not wander around the continent. That very strongly suggests that the S09 results are an artifact of the mathematics and method rather than empirical (‘real world’ for ‘post-normal’ scientists) continent.

      • Duster
        Posted Feb 9, 2011 at 11:37 AM | Permalink

        Just ran across the link to Delingpole’s column below and see that he uses the same metaphor in the title.

  12. Steve Fitzpatrick
    Posted Feb 8, 2011 at 2:37 PM | Permalink

    Ryan O,
    The beauty of the Steig(09) method is that it doesn’t rely at all on inconvenient data to get the desired…er… expected result.

    Gavin’s reply to your comment at RC just shows how little (that is, approaching zero) he understands about the effects of the number of retained PC’s on all inverse problems. Just astounding.

  13. Peter Whale
    Posted Feb 8, 2011 at 2:45 PM | Permalink

    Well done Ryan,the team have been tamed.

  14. Jeff C.
    Posted Feb 8, 2011 at 2:48 PM | Permalink

    Is there a TKO in climate science? This is getting hard to watch. The Chladni patterns and Steve’s distance correlation plots of the recons were done back in February 2009. It has been known for two years that their “innovative” statistical methods were yielding spurious correlations. Dr. Steig must be a smart guy, why in the world would he use such an easily disproven example?

    This paper had a big red bull’s eye plastered on it from day one. The Nature cover, the science by press release, the upending of the hot peninsula/cold continent paradigm, and the Mannian math ensured it would be scrutinized to a level never before seen. Yet they pressed forward as if it was the 1990’s and no one would or could check their claims. Unbelievable hubris.

    • AusieDan
      Posted Feb 8, 2011 at 8:40 PM | Permalink

      Jeff C.
      You asked QUOTE why in the world would he use such an easily disproven example? UNQUOTE

      The answer is simple GroupThink, (refer book by Irving Janis). There are the classic symptoms – Illusion of invulnerability – Closed Mindedness – Defective Decision Making.

      Group Think means mental blindeness which leads victims to stumble into disaster. Remember President Kennedy and the Bay of Pigs episode.

      Victoms of GroupThink are very vunerable.

      • Posted Feb 8, 2011 at 9:56 PM | Permalink

        You guys have apparently won and so perhaps you should direct your superior skills to oil depletion and specifically to the epic tome I would guess that it contains enough ideas for 12 PhD theses for you to try to debunk.

        • golf charley
          Posted Feb 9, 2011 at 8:40 AM | Permalink

          “You guys won” is a bit tribal. How about “attempts to mislead the scientific process lost”

        • JEM
          Posted Feb 9, 2011 at 1:26 PM | Permalink

          I wish that were so, but the Mann/Steig flavor of science won’t have truly lost until their fellow climate-science practitioners truly believe they’re putting their professional reputations at risk by following in their footsteps.

    • Posted Feb 8, 2011 at 8:57 PM | Permalink

      Jeff C who is not me.

      I’ll remind readers that Jeff produced a reasonably accurate reconstruction simply by gridding temp stations before slamming through the regEM meatgrinder.

      He also get’s credit for finding k = 7 first. Ryan and Nic simply proved it later on.

      • Steve Fitzpatrick
        Posted Feb 8, 2011 at 10:21 PM | Permalink

        Without doubt he was insightful in finding k=7. But really Jeff, I looked at Eric’s RC post on “overfitting” last year, and I could see that k=3 was way too few. My eyeball guess from one of Eric’s graphs was 5 or 6, and I know VERY little about choosing the correct number of PC’s for inverse problems outside of a few ‘rules of thumb’. I just don’t understand why Steig et al didn’t do the same kinds of tests as Ryan O describes above. Are they so very dumb as to not ever test the obvious stuff? I just don’t understand.

      • Jeff C.
        Posted Feb 8, 2011 at 10:48 PM | Permalink

        Someone asked Gavin at RC about that gridded reconstruction, the first Steig09 alternate way back in February 2009. His response was (from memory), “not worthy of consideration on even the slowest of days”. Good times.

        • Craig Loehle
          Posted Feb 9, 2011 at 9:36 AM | Permalink

          “not worthy of consideration on even the slowest of days” Translation: didn’t understand what you did.

        • Jeff C.
          Posted Feb 9, 2011 at 1:33 PM | Permalink

          Perhaps, but I’m sure Gavin is smarter than that. Their premise was that RegEM was so good at correlation mining, it didn’t matter that 40% of the data came from 5% of the area (the peninsula and islands). S09 states:

          “Unlike simple distance-weighting or similar calculations, application of RegEM takes into account temporal changes in the spatial covariance pattern, which depend on the relative importance of differing influences on Antarctic temperature at a given time.”

          The gridded reconstruction demonstrated that if the measured temperature data were supplied to RegEM in proportion to the area it represents, the trend was cut in half. If RegEM was doing what S09 claimed, it should have made little difference.

          Steve shortly followed with correlation vs. distance plots that showed that the measured station data had a predictable distance correlation. Close stations had good correlation, and a well-behaved decay in correlation was present as station distances increased. The distance correlation completely broke down in the AWS and satellite reconstructions. It was becoming clear Dr. Steig’s results were rife with spurious correlations.

          The Chladni patterns were the death knell. S09 justified their use of only the first three PCs by claiming the spatial patterns of the loadings resembled known Antarctic atmospheric patterns. Steve demonstrated that using random data on a disc shaped like Antarctica gave comparable patterns. Simply devastating. No wonder they wanted any discussion of the Chladni patterns removed from the new paper.

          Jeff Id then produced countless reconstructions using every conceivable methodology. The same results kept coming up; 0.07 deg C/decade. The only method that yielded the 0.012 deg C/decade was that used by Dr. Steig.

          Within a month of its release, it was clear the paper was broken. Of course the easy part is pointing out what is wrong; the hard part is fixing it and proving the fix is correct. The work done by Ryan, Nic, Jeff, and Steve is masterful; I suspect it will be cited in other fields when facing data scarcity issues. Not so much in climatology.

  15. P. Solar
    Posted Feb 8, 2011 at 3:12 PM | Permalink

    Hi and thanks. The last post was more explicit and had a lot more detail but was hard work to get the essentials out of .

    This has it in a nutshell. Spreading heat where it shouldn’t and not registering where it should.

    A good addition may be the “theirs and ours” graph from your post here when you first got the paper accepted.

    Having shown here that their work does not correctly follow temperatures, that graph summarised in an equally succinct way the bottom line, take home message about antarctic warming that results from the two papers.

    Excellent work by all concerned.

    • bernie
      Posted Feb 8, 2011 at 3:28 PM | Permalink

      I agree. The contrast will make the whole argument much, much clearer. Currently I had to struggle to get the point.

  16. stephen richards
    Posted Feb 8, 2011 at 3:19 PM | Permalink

    Good, intelligent well founded science. Well done Ryan O

  17. PJB
    Posted Feb 8, 2011 at 3:20 PM | Permalink

    Has anyone coined this “Peer-gate” yet?

    • Jeff C.
      Posted Feb 8, 2011 at 3:23 PM | Permalink

      Delipole has dubbed it realclimategate, not bad.

    • Frank K.
      Posted Feb 8, 2011 at 9:17 PM | Permalink

      Posted Feb 8, 2011 at 3:20 PM | Permalink | Reply

      “Has anyone coined this “Peer-gate” yet?”

      I prefer Steigate

  18. Posted Feb 8, 2011 at 3:26 PM | Permalink

    Bishop Hill (Andrew Montford) gives a better explanation (by means of the four sets of maps) of the flaws in Steig et al’s methodology than Ryan does, it has to be said. The Bish is a whizz at presenting stuff clearly for non-scientists. Steig and Mann could learn something from it, too, I dare say!

    • Posted Feb 8, 2011 at 3:43 PM | Permalink

      Yes . . . Bish’s explanation is much better than the above.

  19. NicL
    Posted Feb 8, 2011 at 3:41 PM | Permalink

    As first pointed out elsewhere by Jeff Id, the map in figure 1 of the post at:

    shows the locations of the two weather stations in West Antartica that Steig used in the S09 reconstruction, Byrd and Russkaya. Byrd is the upper red dot on the map, Russkaya is the lower red dot.

  20. Posted Feb 8, 2011 at 4:07 PM | Permalink

    Ryan —

    This is what happens to Eric’s reconstruction when you:

    Top row: Add the designated trends to the Peninsula stations

    Second row: Remove the designated trends from the Peninsula stations

    If everything is linear, as I thought it was, the effects should be opposite one another. However, in the top row WA gets warmer while S. Pole stays the same, while in the second row WA stays the same while S. Pole gets cooler.

    Is there a nonlinearity in here somewhere?

    • Posted Feb 8, 2011 at 4:26 PM | Permalink

      The infilled portion of the PCs are, indeed, linear combinations of the station values. However, a separate set of regression coefficients is computed for each unique pattern of actual/missing values. So in the data set used by Steig, there are 100 or so unique patterns of missingness, and, hence, 100 different sets of regression coefficients calculated for each iteration.

      Because not all of the stations are temporally complete in the 1982 – 2006 calibration period, this allows the PCs to affect the prediction of the stations . . . and, as a result, affect their own prediction. Because all of these influences compound with each iteration, the results can sometimes be rather counter-intuitive.

      So while the end result is that the predicted values are, indeed, linear combinations of the predictors, there are lots of different combinations used, all of which interact with each other.

    • NicL
      Posted Feb 8, 2011 at 5:27 PM | Permalink

      As I posted about at TAV back in 2009, RegEM is quite sensitive to trends. When one adds a trend to several stations, correlations between those stations can be expected to increase, but correlations between each of those stations and all other, unaltered trend, stations will be differentially affected depending, inter alia, on the difference between the existing trends at the stations with added trends and at the unaltered trend stations.

      Since the stations with unaltered trends will have different trends from those of the altered trend stations, the correlations between the altered and unaltered trend stations will change in a way that is not only complex but asymmetrical between trends being added and subtracted. So, yes, there can be a non-linearity in the response of the reconstruction to significant perturbations in the data.

      The asymmtrical effects of positive and negative trend changes on correlations between stations have a non-linear, difficult to estimate and sometimes counter-intuitive effect on the final reconstructions, for the reasons explained by Ryan.

      In my experience, using Truncated Total Least Squares to infill missing data (as Steig did for the S09 reconstruction) compounds these sorts of problem. Ridge regression, which we used, seems generally to be a more stable method.

      • Posted Feb 8, 2011 at 8:00 PM | Permalink

        Thanks, Nic — I guess this is just another inexplicable quirk of RegEM (which I still don’t trust).

      • Craig Loehle
        Posted Feb 9, 2011 at 9:38 AM | Permalink

        This just makes me distrust infilling. When the data has this many gaps, it is a real leap of faith to think any type of infilling is not creating artifacts.

        • Posted Feb 9, 2011 at 10:48 AM | Permalink

          I have to agree. Even though I know nothing of the math involved, infilling is guessing. Maybe it’s educated guessing, but it’s guessing all the same. one could guess that my average speed over a certain distance was 60mph, but if you only have 20% of the data, you could be missing some serious slow spots, and fast spots, and those matter, especially when talking about climate and natural variability.

    • Posted Feb 8, 2011 at 9:01 PM | Permalink

      RegEM does produce a linear combination of inputs, but in the steig example if one input at a particular month cut out, the solution for the coefficients changes. If 10, 14 or whatever months specific stations contain data for each month, the multipliers are the same. One drops out, the multipliers change.

  21. Jon P
    Posted Feb 8, 2011 at 5:00 PM | Permalink

    More bore hole filler for the dishonorable over at RC. They got there whole reguler crew commenting and supporting Eric w/o a single question.

    Jon P says:
    8 Feb 2011 at 4:53 PM

    It is very obvious that Mapleleaf (and others here) did not get past the title of the various posts. If they did there would have been some questions asked. I find the “Coffin, meet nail” post even better than Ryan’s first. You guys always say you want to only talk science, well there it is plain and simple and a point that continues to go over your collective heads.

    • Carrick
      Posted Feb 8, 2011 at 5:42 PM | Permalink

      Jon P:

      They got there whole reguler crew commenting and supporting Eric w/o a single question.

      That appears to be in part because they aren’t allowing any questions to be posted. (Not even to the borehole.)

      “Science” as usual, for them.

      • TerryS
        Posted Feb 8, 2011 at 5:58 PM | Permalink

        So RC now has 4 levels of moderation?

        1. Comments that appear unedited
        2. Comments that appear edited.
        3. Comments that end up in the borehole
        4. Comments that vanish

        • Robinson
          Posted Feb 8, 2011 at 6:07 PM | Permalink

          5. Comments that were never made in the first place because, you know, what’s the point?

        • Steve Fitzpatrick
          Posted Feb 8, 2011 at 7:56 PM | Permalink

          A moderation policy that can be most generously described as “evil and stupid”. Sorry Steve McIntyre; I do not mean to offend, but I am very unhappy that Eric can use moderation to avoid straight forward and reasonable questions. Please correct me if I am wrong, but it seems to me you (and indeed, most bloggers) don’t do that sort of thing.

  22. Michael Jankowski
    Posted Feb 8, 2011 at 5:22 PM | Permalink

    RC folks are the true “deniers.”

  23. Salamano
    Posted Feb 8, 2011 at 5:28 PM | Permalink

    From the looks of it, they’re relying on an argumentation that says, “if only Ryan used nicer words, his points would be considered”.

    I think they’re bringing the whole idea of coming from the ‘holier-than-thou’ conference just last week and dropping this bomb…they find it off-putting in a hypocritical kind of way.

    • Stu
      Posted Feb 8, 2011 at 7:40 PM | Permalink

      This is a reversal of the old argument that although Steve Mc and Anthony Watts were ‘polite’, that counted for nothing in a scientific context (although the scientific ‘debating’ that immediately followed that line of reasoning would mostly turn out to be nothing more than a self given permission to name call).

      So suddenly, politeness matters?

  24. dearieme
    Posted Feb 8, 2011 at 5:34 PM | Permalink

    It bears repeating to non-scientists: one feature of Climate Science is that its practitioners are, on the whole, a bunch of duds. Many scientists may be dull dogs in many ways, but they are usually intelligently competent at their own trades. Not so for your average Climate Scientist; and, I fear, as their incompetence has been revealed, they more and more have resorted to crookedness.

  25. Kenneth Fritsch
    Posted Feb 8, 2011 at 6:11 PM | Permalink

    RyanO, your sensitivity tests show graphically what Steig tended to avoid in his defense of S(09), i.e. not talking about the lesser warming of the Peninsula (as in we agree that the West Antarctica is warming and the results S(09) and O(10) are in essential agreement), but even more importantly to the discussion at hand, of where the Peninsula warming went.

    I find it ironic that, in my view, Steig’s major compliment to your work was the amount and kinds of sensitivity testing that you did and now that testing that you did and are doing and Steig failed to do and is failing to do now is further separating the results of S(09) and O(10). I truly appreciate what you are doing here.

  26. golf charley
    Posted Feb 8, 2011 at 6:33 PM | Permalink

    Ryan, Jeff et al

    I congratulate you all, not only on your maths and science, but also on your patience and endurance.

    Thank you

    No doubt the seals, whales, penguins and plankton living off Antartica, will be relieved to know that polar bear numbers in the area are to remain constant, in the absence of significant change!

    • Jason
      Posted Feb 10, 2011 at 8:02 AM | Permalink

      Constant as in zero, you mean?

  27. BobN
    Posted Feb 8, 2011 at 7:31 PM | Permalink

    Ryan – This is the clearest demonstration of the issues with the S09 methodology, though I have to admit I didn’t quite get until I read Bishop Hill’s summary (consider adding his language as an update to help new readers understand). The fact that the calcualted trend in West Antarctica increases more than the trend in the pennisula when you only add an artifical trend to the penninsula stations is clear and convincing evidence of the smearing effect that you have been describing.

  28. Ed Waage
    Posted Feb 8, 2011 at 7:35 PM | Permalink

    Wow, the visuals tell a compelling story.

    Perhaps it is time for Steig et al to write Corrigendum II, the sequel, and this time acknowledge those who pointed out the problems.

  29. Posted Feb 8, 2011 at 7:59 PM | Permalink

    Professor Steig will likely forever regret the day he stepped into a Mannhole.

  30. Pat Frank
    Posted Feb 8, 2011 at 8:56 PM | Permalink

    I’m beginning to wonder whether, right from the start, Team scientists were advised by Fenton Communications (FC) to never, ever admit error.

    That would explain the policy applied consistently across all the Team individuals, virtually right from the start.

    We know Team scientists are strongly associated with FC. They must have had joint strategy sessions. Never admitting error is the first and greatest political commandment. It seems certain to be the advice of professional and thoroughly politicized spin-meisters.

    All scientists must admit their errors. I’ve done, and know others who have done. But when has any Team member ever admitted an error? That trait has been uniform among Team members and uniformly applied.

    Conscious dissemblance and equivocation to avoid admitting error is totally anti-science and professionally derelict.

    I suspect the hand of FC in that. If it’s an applied policy, then each Team member will have had to have made a personal decision to stonewall science.

    • mark t
      Posted Feb 8, 2011 at 10:45 PM | Permalink

      Depends upon whether you’re a betting man or not, Pat. Not one of them has ever admitted error even when caught red-handed lying about it.


      • Pat Frank
        Posted Feb 9, 2011 at 12:46 AM | Permalink

        Agreed Mark. I was thinking of just that when it butted up against Fenton Communications in my mind, and the suspicion of deliberate policy advice suddenly dawned.

        I’ll be back around to JeffId’s pretty soon, by the way. Sorry to have left you and a few others as the voices of reason out there.

        • stan
          Posted Feb 9, 2011 at 11:10 AM | Permalink

          If I were a member of the team, I don’t think I’d take PR advice from Fenton. I suspect that the folks at Fenton are focused on trying to advance the portfolios of their clients. If a scientist or two has to be used as cannon fodder, the shredding of a few careers is but a small price to pay and a sacrifice Fenton would willingly make. There are lots of scientists out there.

    • S. Geiger
      Posted Feb 8, 2011 at 11:37 PM | Permalink

      For a brief time, as the smoke from climategate settled RC took on some critical posts…and although not responding to all of them they did let most through (Gavin seemed a bit stunned….he’s returned to his former self since).

      • JEM
        Posted Feb 9, 2011 at 2:24 AM | Permalink

        Which, of course, might be indicative of initial confusion and loss of direction, followed by a little spine-stiffening induced by some outside force.

      • stephen richards
        Posted Feb 9, 2011 at 11:20 AM | Permalink

        You cannot know how much was allowed through and how much was binned. What the doesn’t see the heart cannot grieve.

    • Geoff Sherrington
      Posted Feb 9, 2011 at 3:48 AM | Permalink

      Pat Frank,

      Here is some of the advice from Futerra, from documents I have dated July 2006. These come from “EVIDENCE BASE REVIEW OF PUBLIC ATTITUDES”.

      2. Forget the climate change detractors.
      Those who deny climate change science are irritating, but unimportant. The argument is not about if we should deal with climate change, but how we should deal with climate change.

      5. Climate change must be ‘front of mind’ before persuasion works.
      Currently, telling the public to take notice of climate change is as successful as selling tampons to men. People don’t realize (or remember) that climate change relates to them.

      (Comment: What, allegedly, did Prince Charles say to Camilla in that 1993 phone call? There are numerous published discourses about it, such as at In this case, the best advice would have been to say nothing – before the event.)


    • Mike B
      Posted Feb 9, 2011 at 12:43 PM | Permalink

      Pat, you pose an interesting question, and I think the Fenton folks are heavily involved in a lot of what goes on at RC. On the other hand, I think it quite likely that many of the crew at RC don’t need any advice for them to never, ever admit error.

  31. Tom C
    Posted Feb 8, 2011 at 10:03 PM | Permalink

    The similarities of this fiasco to the MBH – MM saga are numerous. In both cases the Team used a PC technique to fish out some elusive temperature signal from sparse data. In both cases they mis-used the technique, either willfully or due to incompetence. In both cases it took careful, exhaustive work to trace the errors and point them out. You can bet that in both cases they will fight tooth and nail, for years on end if necessary, to avoid admitting the errors. I think Pat Frank is right regarding the strategy.

  32. DM OF WA
    Posted Feb 9, 2011 at 12:38 AM | Permalink

    Thank you very much for your well-argued and
    excellent post. It’s the kind of article that
    makes Climate Audit my favourite climate science
    Web site.

    However, can I respectfully suggest you refrain
    from the vitriol? I know it must be difficult
    to remain restrained in the face of attacks.
    Nevertheless, rising to the bait and pouring
    vitriol on your opponents only diminishes you,
    Climate Audit and sceptical thinking in general.


    • tetris
      Posted Feb 9, 2011 at 3:04 AM | Permalink

      Noble thought and very much in keeping with our host’s blog policies.
      Problem is, the Team has a track record of using their “scientific” concoctions for clear political purposes: few commented on the fact at the time, but it so happens that the original Steig paper was [conveniently] published a few days before Al Gore was scheduled to testify on AGW in Congress, which allowed the latter to ever-so-convenniently state that he now had proof positive that Antarctica was warming too. The last “non-warming” hold-out finally subdued…
      Ryan, Steve, Jeff and Nick tried to meet science with science and were received with Team politics and vitriol, both part of the same arsenal..

      • Posted Feb 9, 2011 at 10:57 AM | Permalink

        Indeed. The Team have a habit of simply sticking their fingers in their virtual ears and chanting “LA LA LA I CAN’T HEAR YOU!”, by way of their blog moderation.

    • dougieh
      Posted Feb 9, 2011 at 7:10 PM | Permalink

      i agree (to an extent) with your comment re – “pouring
      vitriol on your opponents only diminishes you,
      Climate Audit and sceptical thinking in general.”

      however, there comes a time when it becomes apparent that reasonable debate (plus evidence of all your worst fears(gate)) will not lead to resolution of differing points of view as you hoped/expected.

      Steve has always expected science or the governing bodies to eventually sort out its own mess on the climate debate, but, I think even he is growing weary of the stupidity of the run-arounds including his attempts & now Ryan et al.

  33. ThinkingScientist
    Posted Feb 9, 2011 at 4:33 AM | Permalink

    Would it be possible for RyanO to also present the results as difference maps relative to the original S09 map? I think this would help to highlight the effects. Also, on each map a symbol marking the location of the data points modified to test the sensitivity in each case would also be very helpful, rather than just describing the positions modified as “peninsula” etc

    These additional displays (together with the original displays) would make it much clearer and easier to communicate these results to a lay audience.

  34. ianash
    Posted Feb 9, 2011 at 6:14 AM | Permalink

    Ryan O’Donnell, Eric has completely destroyed the points you make here. Why not grow a pair and discuss it at Realclimate?

    Real simple…stop abusing Eric and start debating. Be a grown up and do it. My guess is you wont but maybe you have a little more zazz than the usual denialist trade that post here.

    • Gord Richens
      Posted Feb 9, 2011 at 7:48 AM | Permalink

      Post a comment at RC with a link to this page. See if it sticks.

    • Frank K.
      Posted Feb 9, 2011 at 8:21 AM | Permalink

      Eric Steig is the one in denial…he truly has gone off the deep end. Again, I hope he has the support he needs to get through this rough patch in his life…

    • mpaul
      Posted Feb 9, 2011 at 10:32 AM | Permalink

      Better yet, have Steig debate it here where the moderation policies will ensure that both sides will be allowed to post without moderator manipulation.

      • Posted Feb 9, 2011 at 10:41 AM | Permalink

        Better yet, find a neutral corner of the web and debate it there. I’m sure one can find an unbiased moderator somewhere.

    • Hu McCulloch
      Posted Feb 9, 2011 at 1:26 PM | Permalink

      Posted Feb 9, 2011 at 6:14 AM

      Ryan O’Donnell, Eric has completely destroyed the points you make here. Why not grow a pair and discuss it at Realclimate?

      Ryan sensibly refrains from posting at RC because of their censorship practices. Likewise, Steig may not want to dignify CA by posting here (a stupid reason, but his choice), or may suspect that CA might be as underhanded as RC (an unwarranted reason, in my view, but at least a logical one).

      However, this kind of blog exchange is already taking place, just not on one blog: Eric posted his informal blog reply to O10 on RC, and now Ryan has replied with 2 posts here on CA.

      Obviously, Ryan has noticed the RC post, but team Steig might (somehow) not happen to have noticed what Ryan writes in reply here. Ryan should therefore politely advise Steig and all his co-authors of his posts by e-mail, and indicate here that he has done so, leaving it to them to reply here or on RC or by e-mail or not at all as they choose. Meanwhile, anyone who is watching can see whether or not there was a reponse.

      According to you, ianash, Steig has “completely demolished” the points Ryan has made here. You could be helpful instead of just demonstrative if you would be specific with quotes from Steig’s posts or comments that you find particularly persuasive. Ryan (or Nic or Jeff or Steve) could then either respond briefly in a comment or at length in a new post or not at all and anyone watching could see their response or lack thereof.

      Ryan, for his part, has made the point here that artificially changing the peninsula data changes the W. Ant. (or S. Pole) trend using the S09 method but not the peninsula trend, while changing the W. Ant. data has no effect on anything. That sounds pretty damning for S09. How has Steig (or you) reponded to that?

    • Michael R
      Posted Feb 9, 2011 at 6:36 PM | Permalink

      Unfortunately to debate or argue a topic requires that both sides get equal time. RC Comments are moderated in a way that anything remotely making the people at RC look like they were wrong are deleted or now moved to the borehole.

      The last comment in the borehole is a perfect example of how a reasonable question that is both valid a topical to the issue at hand is removed simply because they do not like it.

      In fact the moderation policy there, not to mention the sarcastic and sometimes petulant responses at RC spured my interest in investigating the science behind climate change. In that respect I should thank them. Otherwise I cannot understand how they do not see that their actions are singlehandedly the best (worst?) way to lose trust in their skills and any work produced.

      In fact, I have a hard time even making myself read a paper with any o ftheir names atributed with any kind of fairness and open mmindedness (though I try) and that was because of RC.

  35. Lazlo
    Posted Feb 9, 2011 at 6:42 AM | Permalink

    Publish his ridiculous review. You can never have too much information..

  36. Posted Feb 9, 2011 at 7:51 AM | Permalink

    why do you take O’Donnell at his word? And now he’s my word: His allegations have no basis in fact. Now you have my word against his. Now try thinking,-

    So Eric did not review the paper?

    Perhaps he was one of them who gave a favorable review.

    I think Eric Steig should simply come out and reveal the identities of the reviewers.

    • TomRude
      Posted Feb 9, 2011 at 11:29 AM | Permalink

      Perhaps as guessed by Jeff Id, reviewer A was a Team effort… and thus despite his email acknowledging he was reviewer A, Steig’s contortion might have some truth to it? Sounds like Furlong…

      • Posted Feb 9, 2011 at 11:44 AM | Permalink

        If Reviewer A was a team effort, then Eric should come out with it and clear his own name. Why should he take the bullet for a collective poisoning carried out in his name?

        • glacierman
          Posted Feb 9, 2011 at 11:58 AM | Permalink

          I am pretty sure that that is against the reviewer rules, or code of conduct. They are not supposed to involve anyone else unless a specific expert is needed and in those cases it is required to be divulged. There are many ethical issues at play here.

        • tetris
          Posted Feb 9, 2011 at 2:46 PM | Permalink

          Since the Team’s modus operandi is to steadfastly reject as a group any accusations of wrongdoing by anyone of its members, nobody takes a bullet for the collective. Organized collective denial serves as shield.

        • Jason
          Posted Feb 9, 2011 at 4:38 PM | Permalink

          There are limits to this.

          I haven’t seen any recent posts by Gavin attempting to support Eric on the merits of the math in Steig et al.

  37. golf charley
    Posted Feb 9, 2011 at 8:36 AM | Permalink

    Shub, yes I noticed that remark at RC. The magazines editorial staff will know the truth

  38. PJB
    Posted Feb 9, 2011 at 9:00 AM | Permalink

    Why not have both parties submit their papers, methods and code to a statistics journal for peer-review by statisticians and see which passes?

  39. AJ Abrams
    Posted Feb 9, 2011 at 10:26 AM | Permalink

    Steve, Ryan or Jeff

    What is the issues of K_gnd that Eric thinks wasn’t addressed? I have looked through the remarks there and here and can’t seem to grasp what straw he is grasping at.

    Thanks in advance

    • Steve McIntyre
      Posted Feb 9, 2011 at 12:22 PM | Permalink

      Re: AJ Abrams (Feb 9 10:26),
      look at the Review Comments from Reviewer A at

      • AJ Abrams
        Posted Feb 9, 2011 at 12:37 PM | Permalink

        Steve, I did which is why I’m confused by his latest statement. It seemed to me that they (Ryan, you etc) did address it.

        • Mark T
          Posted Feb 9, 2011 at 1:00 PM | Permalink

          I think the word for that is disingenuous.


      • AJ Abrams
        Posted Feb 9, 2011 at 1:48 PM | Permalink

        I tried reading it again. I think it helped because I wasn’t concentrating so much on trying to get the math, but instead was looking at what was being done overall. My opinion was that it was a moving target manipulation.
        I wanted to believe on the first reading that Steig really didn’t understand why K_gnd=7 was appropriate and therefor was arguing from a position of ignorance, but in the reread, it really is apparent that Ryan’s first explanation was easy to understand. The second and third reviews then each changed the nuance of his argument and it reads as deliberate. Again this is opinion, but his email to Ryan and Jeff support this opinion. To say that his concerns over K_gnd weren’t addressed really was disingenuous and manipulative. I could understand if he wanted to readdress a certain point – say his 3rd review accusation of ignoring K_gnd in favor of K_sat, but that isn’t what is being said, instead he’s trying to reframe the argument – yet again.

        I’d say really good job on the responses by the way. You kept the pea in front of you and seemed not to fall for misdirection issues. You addressed them without letting them become “the” argument.

        My end take is thus though – Steig is going to be able to keep the masses questioning your results regardless of how clear said results are in the same way that has been done before; staw men arguments. They tend to give people the impression that the argument is about something it is in fact, not about.

        • Mike B
          Posted Feb 9, 2011 at 3:28 PM | Permalink

          It’s obvious to me why Steig wasn’t satisfied with O10’s choice of k_gnd: it didn’t give the result Steig wanted, i.e. significant warming in West Antarctica.

    • Posted Feb 11, 2011 at 9:05 AM | Permalink

      The paper was about addressing Kgnd, every reply was about addressing Kgnd, Eric just didn’t like/want the answer it gave. Iridge takes the decision out of our hands yet gets the same answer as the methods we originally had used to chose Kgnd. Now Eric knows full well that all the methods got the same answer, yet he is still hollering on about Kgnd. Certainly our choice can’t be as bad as he prefers if we get the same answer right?

      If Kgnd was almost the point of the paper, yet he’s not happy. Kind of makes its own point doesn’t it.

      The SI and first submission are on line (linked in the previous post) I think we all like that one better than the final – minus a few grammatical corrections. The SI has pictures/descriptions of Chladni patterns and side by side of the methods with area weighted included.

  40. Salamano
    Posted Feb 9, 2011 at 11:36 AM | Permalink

    So where do things stand now?

    R.O. Has presented all sorts of work and effort and posted it here, and in addition has divulged his perspective on the inner-workings on how the peer-review process was less-than-spectacular in this case.

    E.S. Disagrees with R.O.’s perspective of ‘how it all went down’ but prefers to not contribute or respond to the specifics that R.O. efforted in the last two posts. Perhaps (a) because he already did a bunch of it in his reported 88-page review and would prefer to keep that stuff confidential, or (b) because the only way to respond favorably for ‘his side’ is when armed with new information or new results that can be published specifically to address his concerns without acknowledging that they were either salient or strong (or his) in the first place.

    I suppose I can understand E.S. unwillingness to comment on the ‘peer-review’ part of it, because it has traditionally been confidential (and by not talking about it no one can really know the whole story– which the uncertainty might be preferable to him). However, using his disagreement regarding the peer-review of RO10 as a way to indicate justification for not responding to these posts is off-the-mark.

    Either he already has responded in a relevant and significant way that has not been mentioned (and feels he doesn’t need to do it again); or he wants to spend his time on something else (which has not been stated plainly without piousness); or he wants to avoid it until a well-constructed new effort can be published to step on whatever is appearing in a blog rather than a publication– so they can promote ‘further confirmation’ without acknowledging any discrepancy ever existed or was ever ‘helpful’.

  41. Salamano
    Posted Feb 9, 2011 at 1:42 PM | Permalink

    “[I’ll have a post later today addressing these points. Forgive me for being annoyed with you for assuming it might actually be true that I am a deceptive duplicitous idiot.–eric]”

    I guess we’ll finally see what semantics or interpretations E.S. or R.O. is operating with (or hiding behind).

    For the record…I haven’t seen an ‘idiot’ label levied anywhere just yet.

    • glacierman
      Posted Feb 9, 2011 at 1:53 PM | Permalink

      I predict the “well our methods may have had issues, but we got it right” argument. A long-time favorite of the team. Everything else will basically be avoided/ignored. Look for a lot of bringing up other data/studies that show something “consistent” with Steig09 as evidence that his study is correct. There will be alot of pointing to some other study or post that the followers will reference whenever this issue gets brought up along with the meme “Steig09 has been vindicated by X….this is just another denier attack on science…” (maybe it is an attack on post-normal science).

      Let’s see, how many times have we seen that done?

    • Salamano
      Posted Feb 9, 2011 at 4:48 PM | Permalink here it is…

      Looks like he was upset that his original criticism was never addressed (K_gnd=7)…and felt that since he didn’t see the 4th iteration of the paper he could rightly say he never saw certain things.

      He also wants to qualify that his behavior was not duplicitous, and requests a retraction to that effect.

      Apparently he uses the example as a reason why anyone he names should not be an equitable participant in the climate science literature…mostly because he sees them as working harder to prove people wrong, rather than to identify what is right.

  42. Stacey
    Posted Feb 9, 2011 at 2:48 PM | Permalink

    Ryan O and Jeff Id you are in very good company, Professor Freeman Dyson, a physics giant is being beaten up by the pygmies at UnReal Climate one of which is Eric Steig.

    You saw it here first:-
    [Response: Thanks. My mistake — will correct.–eric]

    Shame he didn’t do the same for you guys?

  43. movielib
    Posted Feb 9, 2011 at 5:03 PM | Permalink

    Eric Steig has replied to Ryan:

    There seems to be a lot of arm waving about O’Donnell being wrong about… well, everything.

    There is what I’d call a personal attack against “O’Donnell, Condon, and McIntyre,” comparing them unfavorably with such “legitimate, honest commenters” as “Susan Solomon or J. Michael Wallace, or, for that matter, Gavin Schmidt or Mike Mann or myself [i.e. Steig].” You see, he thinks people like O’Donnell and McIntyre are not legitimate honest commenters. The compulsory word “deniers” is also thrown in.

    Steig claims O’Donnell is going to “retract [his] allegations” against Steig. It’s very vague and I sure don’t know what he’s talking about.

    He says he was a reviewer for the first three drafts of the O’Donnell et al. paper but not for the “markedly different” fourth draft so he hadn’t seen it before publication.

    Curiously, Steig does not address the point that is the subject of this thread.

    • Salamano
      Posted Feb 9, 2011 at 5:13 PM | Permalink

      How does saying that if only a paper was “Fundamentally Reworked”, it would then be a valuable contribution…How does that constitute something positive or encouraging?

      I’m sure a few ex-girlfriends would line up and talk about how if only their men were ‘fundamentally’ redone…

      I think it’s a good thing that all the text of the various reviews are available in this case…so that we can see if there are any clever exerpting behavior going on.

      • Posted Feb 9, 2011 at 7:15 PM | Permalink

        Yes…an odd juxtaposition of text strings from Steig:

        “The fact of the matter is that my reviews of O’Donnell’s paper were on balance quite positive.” versus “I emphasize that I think that a fundamentally reworked version of this manuscript could potentially provide a useful scientific contribution”. Steig places these lines conveniently almost next to one another in his RealClimate piece.

        (I wonder if Gavin has given up peer reviewing for the team?)

        Funnily enough Steig doesn’t mention in this piece the inappropriateness of reviewing (endlessly) a paper drawing attention (like this post itself) to the inadequacies of Steig’s own work.

    • Geoff Sherrington
      Posted Feb 10, 2011 at 7:30 PM | Permalink

      This was snipped at ReacClimate in the past 24 hours. The earlier post referred to the Bishop Hill pictures, which Eric sid he has trouble folloing in between the ad homs.:

      40Geoff Sherrington says:
      9 Feb 2011 at 10:17 PM
      Where is the response to the Science?
      Dr Steig, the pictures that are telling a simple scientific story are the results of scientific investigation, not ad homs.
      It should not be beyond your wit to say that within the constraints of the discussion, images a,b,c, are correct; d,e,f might be and h,i,j are not.
      It you are going to write a paper for peer review, you must have decided approximately what to write; and that it is reasonable to assume that the answers to the above are known to you. Why not come out with your intended conclusions now, rather than having us wait for another 80 pages of review?
      That’s how a scientist would handle a challenge.

  44. jo
    Posted Feb 9, 2011 at 6:22 PM | Permalink

    i’m on tenterhooks waiting to hear what ryan O is gonna retract/apologise for….
    (not that i would hold it against a man for making a retraction where necessary)

  45. Steve Reynolds
    Posted Feb 9, 2011 at 7:02 PM | Permalink

    Interesting and very good (IMO) comment that surprisingly got through at RC:

    Ged says:
    9 Feb 2011 at 5:37 PM
    As a published scientist in another field, I am discouraged by what I see here.

    Why is there being given any attention to personal attacks? This should be ignored as irrelevant! The only thing that should be talked about here is the actual science.

    O’Donnall gave a scientific rebuttle to your previous post in his, which has not been addressed here, and which -should be- above all the primary focus of any forward discourse.

    For, you see, science is based only on data and evidence, not personal reputations. The data will stand for itself, and any who attack you personally will have their own words heaped on their heads if you simply ignore it and deal only with the scientific matters at hand. The science is the only proof you need!

    And yet, I see no science here. Their most serious allegations are those showing the Steig et al (09) model’s algorithm/methods to be seriously flawed in its infilling response to data changes in the stations used. A very serious matter indeed for all reconstruction and modeling.

    We scientists should be swayed only by data and results, and I would like to see the truth of the matter, scientifically, illuminated by yourself so we can produce the most accurate representation of what is occurring on our planet. The personal allegations against yourself, which as a reader on this from the start I believe have been overly hyped and given too much importance by both sides, will burn themselves out before actual information. Otherwise, this will devolve into high school “He said she saids”. You must rise above it, and set an example in so doing so.

    In the end, all choices are yours, if you heed or care about my words or not. I, and other people and scientists interested in the actual information and data, long greatly for a scientific, not personal, rebuttle by yourself to O’Donnall’s points.

    [Response: I don’t disagree. I wrote a scientific commentary already, in which I pointed out some problems with their arguments, and also acknowledged that they had some good scientific points. This is the way it ought to work, and indeed many people wrote in to thank me for the substance and style. Unfortunately, O’Donnell’s reponse was to call me ‘duplicitous’. I could have ignored it, but since many people — who ought to know better — were not ignoring it, but seemed to be believing it, I felt the need to correct the misconceptions. Get it? –eric]

  46. two moon
    Posted Feb 9, 2011 at 7:24 PM | Permalink

    Steig is now up on RC claiming that Ryan has retracted. Hmmm.

    • TomRude
      Posted Feb 9, 2011 at 9:09 PM | Permalink

      Steig’s post and the subsequent changes in title at CA elicit two comments:
      1) Steig writes: “With respect to O’Donnell’s lengthy discussion of the technical aspects of the difference between our papers, I’m not complaining.” That’s the meat.

      2) The change of title at CA and words from Ryan’s post need more than a small footnote. Obviously there is always more than meet the eyes here but the changes show that in one, perhaps narrow, instance Steig got a point, or did he? May I suggest that the best way to disprove the fluff in Steig’s new post is in fact to isolate that very point and own it if a mistake was made in clear, unequivocal ways?

    • Salamano
      Posted Feb 9, 2011 at 9:43 PM | Permalink

      I think that all the ‘snark’ and ‘vitriol’ and whatever else that is being bandied about perhaps should just be ignored, unless/until it results in actual material harm.

      It’s pretty clear that both camps have wavered into that end of the pool occasionally, and simply redressing titles of posts won’t really placate anyway. It’s a red herring. Seriously… Stieg is replying to O’Donnell’s post using the title of “O’Donnellgate” and declaring justification for doing so.

      Hell…the Israelis and Palestinians do whatever they want, and can point to whatever they feel like to justify their behavior, going back hundreds (thousands) of years if they have to. At some point, justification for taunts should matter about as much as the taunts themselves…nothing.

  47. Posted Feb 9, 2011 at 8:51 PM | Permalink

    Anyone with any sense ethics or honor would have signed their review of a paper critical of theirs. Eric’s conduct should be investigated if the peer review process is to regain any credibility.

    • Salamano
      Posted Feb 9, 2011 at 9:39 PM | Permalink

      There seems to be a suggestion in the comments section at RC that J.Climate investigate R.O. for publishing the reviews and the run-up as if that is the greatest evil precipitated here.

      IMO, it sort-of makes the J.Climate look better, because it followed a protocol that still resulted in the journal being published…

      …inviting the chief author of the rebuttal piece to an unsigned review notwithstanding.

      • Posted Feb 9, 2011 at 9:55 PM | Permalink

        The JoC did fine in my opinion too. Eric, I am pretty sure had the option of signing his reviews. It was not the JoC’s place to require him to. Anonymity should be for the benefit of reviewers not involved directly in the paper. In any case, had Eric openly submitted his review, none of this side show would have happened.

      • Davos
        Posted Feb 9, 2011 at 10:55 PM | Permalink

        I’ve been trying to get this posted at RC, but it keeps locking me out…perhaps a different screen name will do…

        E.S. says,

        “…But in any case all these criticisms are already in the paper, some are right, some are wrong. A response will be forthcoming in the peer reviewed literature. So you can stop asking this question. You’ll get an answer in due coure.–eric]”

        I’m wondering how is it that some authors can ‘guarantee’ that new works will automatically achieve publication within peer-reviewed literature, and others can never be so bold, or have to fight so hard for their words to achieve that result..? It doesn’t seem to me that all manuscripts and authors approach the peer-review process on equal neutral ground.

        😉 I wonder if this future editor will also find it important that O’Donnell be a reviewer of this forthcoming response.

  48. ianash
    Posted Feb 9, 2011 at 10:05 PM | Permalink

    From Steig

    “Some months ago, O’Donnell cordially (though quite inappropriately) asked me if I was one of the reviewers, and also promised not to reveal it publicly if I didn’t want him to. I told him I was, but that I would prefer this not be public since the ‘opportunity for abuse’ was simply too great. Talk about prescience!”

    • Posted Feb 10, 2011 at 3:54 AM | Permalink

      Re: ianash (Feb 9 22:05), So where is the evidence that he ‘promised’?

      • Posted Feb 10, 2011 at 4:13 AM | Permalink

        Re: PaulM (Feb 10 03:54),
        Here is what Ryan says at Lucia’s:
        “I did not explicitly tell Eric that I would keep the information about him being a reviewer confidential. However, he did request that I do so, and I fully intended to do so.”

    • DEEBEE
      Posted Feb 10, 2011 at 5:46 AM | Permalink

      And now we are going to get into another steriledebate about whether Ryan is a bad scientist because he promised or Eric laid a trap, or was wrong because he should not have told in the first place etc. etc.

    • tetris
      Posted Feb 10, 2011 at 12:44 PM | Permalink


      It reads like this: “I know you are in the process of distorting available information, but because of the “rules of the game” I undertake not to make this public. However, now you and others are using this against me and are using my undertaking of silence actively to hang me. So I in turn, am going to expose you.

      As I have been arguing all along, from an ethical point of view, please tell me and all the others here what is wrong with that.

      From my perspective, because I am considerably less polite than Ryan, I would not only have broken the silence earlier, but also have done whatever I could to silence the bastard who double crossed me, any which way and once and for all.

      Always keep in mind, Ryan is the polite and nice guy in this story.

  49. Jeff C.
    Posted Feb 9, 2011 at 10:19 PM | Permalink

    Dr. Steig and the rest of the RC crew are public figures as a result of their advocacy and “science by press release” mode of operation. If you want to play at this level, one has to expect criticism. Nothing Ryan wrote remotely resembles libel, certainly not in the US.

    You guys want to have it both ways. You call press conferences, send talking points to the sympathetic journalists, and publicly lobby lawmakers. When anyone dares to point it out, you start the poor scientist toiling in obscurity routine. Toiling for six-figure salaries, by the way.

    Obscure scientists don’t get media lessons and press releases written by left-wing PR firms.

    • Frank K.
      Posted Feb 10, 2011 at 8:42 AM | Permalink

      The “team” at RealClimate have clearly lost this battle. Steig should have done a mea culpa, but apparently he’s digging in. He’s clearly lost it (mentally and emotionally), and apparently needs to sort out some issues in his life. I wish him well.

    • Jeff C.
      Posted Feb 10, 2011 at 11:20 AM | Permalink

      My comment was in reply to a ridiculous accusation of libel from one of the RC minions. Looks like it got snipped making my comment an orphan. Serves me right for arguing with a troll.

      • Luis Dias
        Posted Feb 11, 2011 at 10:26 AM | Permalink

        Jeff, cut the losses and just forget about it. They are not worth to even try.

        This only proves Curtis’ idea that the internet is not linking people, but balkanizing them into self-serving and echo-chambered groups. People then like PZ Myers, Phil Plait, etc., who are in completely different fields will just skim these posts and declare that “once again, the denialists…just like the homeopaths / creationists”, etc. Knowing this, RC declares that CA, tAV, Lucia’s, etc, are all “denialist scumbag places” and should be ignored, and they hook, line and sinker.

        This is the why of the commotion. We all know RC’s importance on creating impressions in a very wide and big group of people out there, who fervently believe in everything these scientists spew, and see this as just a part of the “science wars”. Many good people do not like to be called “denialists”, “creationists”, “crooks”, etc., for being skeptical at the proclamations of these “real” scientists.

  50. Posted Feb 10, 2011 at 4:40 AM | Permalink

    I hope everyone is aware that the response on RC – O’donnellgate, is not aimed at Climate Audit, or the blogs..

    It is aimed at the media and politicians. It is a mastercraft of PR, as I imagine anyone NOT upto speed with the intricate detail will read that and believe the RealClimate version of events…

    Yet, the good news, that the science has been improved, Antartica is not warming as much as was originally thought, and that the front cover of Nature that got media attention around the world, was a little alarmist with hindsight and further scientific endeavour.

    This is all forgotten and goes unnoticed by the public at large.

  51. DEEBEE
    Posted Feb 10, 2011 at 5:36 AM | Permalink

    Anything that he “got right” . . . as I said before . . . was by accident
    Ryan, not criticizing you, but to me it clearly shows that the “team” practitioners of climate science have made it a soft science. Perhaps it was inevitable since the involvement of politicians was so premature.

  52. stephen richards
    Posted Feb 10, 2011 at 7:59 AM | Permalink

    Steve Reynolds
    Great comment but look at the response. At the end Steig still could not conceal his arrogance.

    I felt the need to correct the misconceptions. Get it? –eric

    We have argued until blue in the face with people from and supporters of RC and for many years. Their responses have always been either ad homs or distorted or manipulated english and data.

    Steve well done, you hit the nail squarely but I doubt it penetrated.

  53. Dominic
    Posted Feb 10, 2011 at 8:34 AM | Permalink

    I ask this question innocently and perhaps naively.

    Where are the GOOD climate scientists ?

    By GOOD I mean scientists who adhere to the scientific method, who use methods which are suitable, who do not try to hold up critical papers, who admit mistakes and accept that mistakes are part of the scientific process and who enjoy criticism because it allows them to hone their arguments and theories and do the necessary experiments until they are rock sold.

    I suppose what I am wondering is whether the RC scientists who we are always running into here are the climate science community or is that just what they claim. Perhaps there are other GOOD scientists out there who are too busy working and publishing but keep their heads down because their results are too weak or inconclusive to allow them to make exaggerated claims.

    Because if the best climate scientists are the likes of Gavin Schmidt, Hansen, Mann and Steig then the state of climate science is totally rotten.

    • Steve Reynolds
      Posted Feb 10, 2011 at 9:10 AM | Permalink

      “Where are the GOOD climate scientists ?”

      Reviewers B, C, and D might be. Unfortunately, they don’t get the publicity the team does.

    • Posted Feb 10, 2011 at 9:49 AM | Permalink

      Re: Dominic (Feb 10 08:34), Yes, there are some good climate scientists, busily working away on observational data and computer models and writing papers. But you don’t see them. You won’t find them at Realclimate. You won’t see them on TV or read about them in the newspapers. You won’t see their papers on the front cover of Nature. You won’t hear about them winning multi-million dollar research grants. That’s because they don’t exaggerate global warming or the significance of their own research.

      There is much criticism by the skeptic community of the peer review process. There is very little criticism of the process of winning research grants or getting published in “top” (?) journals like Nature, where in order to succeed it is necessary to overhype your results and overstate the importance of your work. IMHO this is a more serious problem than peer review.

  54. TanGeng
    Posted Feb 10, 2011 at 12:30 PM | Permalink


    Could you do something similar with your methodological improvements? It’d be interesting to see a better response to an artificial signal.

  55. Posted Feb 10, 2011 at 2:01 PM | Permalink

    Dear Ryan, thanks for your good work – and pretty clear words about what has been done. You’re almost certainly right.

    Still, there are people on both sides who have prejudices about the Antarctic climate. I don’t see any robust enough reason that would make me certain about the past Antarctic temperature changes. The previous data I have gone through indicated a slight cooling; other data could show something else, including warming.

    In some sense, it would be good to filter comments and papers only meant to amplify a pre-existing assumption. Unfortunately, this set of stuff that should be filtered away probably includes Steig’s paper.

    Some years ago, I verified that his multi-megabyte files were compressible to a few kilobytes: they just contained some tensor products of eigenvectors in two spaces. It wasn’t really a serious work to publish an overwhelming amount of data which were really trivial. It looked like a work by a student who is learning linear algebra and still has big problems with it; or a person who deliberately produces fake data and tries to present them as real ones to make the point.

    So I wouldn’t read any paper by Steig in too much detail because I don’t consider him a competent scientist. It’s just a waste of time. If one wants to have a reliable evaluation of some available data, one obviously needs better people than Steig. While you’re almost surely right about your critical remarks on Steig, I am not sure whether your own version of the scientific story is 100% accurate because I haven’t verified it myself.

    Still, I can’t hide that these arguments about Steig are much ado about nothing. I don’t think that decent scientists do or should waste their time with some semi-professional confusion that a sub-par scientist of Steig’s caliber writes in some papers or on Real Climate. It’s just a waste of time.

    Antarctica could be evaluated honestly and I think it’s damn obvious what are the right procedures to deduce the most accurate possible trends etc. from the available data. Still, even if this were done, I don’t think it’s too important. Some people view the Antarctica as another battle place. I don’t. I don’t give a damn about the trend in Antarctica in the last 50 years. I think it was probably slightly cooling or nearly constant but if it were strongly cooling or warming, it won’t impact my opinions about the big questions in any significant way.

    • Posted Feb 10, 2011 at 8:03 PM | Permalink

      Agree with all of this Lubos. The only wider importance of the story for me is the screwing around with peer review, revealed much faster than the notorious Climategate examples. Why do these guy bother? It beats me.

      • Posted Feb 11, 2011 at 6:26 AM | Permalink

        Right, Richard, the peer review hasn’t worked well and has been abused, too. In an ideal world – and, in many cases, in the real world – the peer review works and improves the quality of published texts.

        However, when more accurate tests may show that it doesn’t work, it just doesn’t work. Worshiping this method doesn’t help science. Unfortunately, exactly whenever the peer review doesn’t work, it’s being most intensely used as an argument to silence real analyses of the issues.

  56. Alex
    Posted Feb 10, 2011 at 2:46 PM | Permalink

    I asked Eric to please comment on this post at RC since either this is wrong or his method is wrong. This comment never showed up, not even in the borehole or what they call it and I know I posted it since I accidentally posted it twice an got an error message the second time telling me that I double posted. I guess he is not interested in if his method is broken or not.

  57. Paul_in_CT
    Posted Feb 10, 2011 at 7:36 PM | Permalink

    I have to say, as a longtime fan of this blog, that I am embarrassed by Steig’s apparently sincere and reasonable response, and I would ordinarily have expected a “counter response” to have been posted by now, but none has been posted. Which makes me think there is no there there, and that Steig is speaking truth.

    A bad week! But maybe even worse than just bad?


    • mark t
      Posted Feb 10, 2011 at 8:47 PM | Permalink

      That does not make any sense… how are you embarrassed by Steig’s response, and for that matter, what has he said that is sincere?


    • Salamano
      Posted Feb 10, 2011 at 9:24 PM | Permalink

      It could be possible (likely) that they BOTH are sincere though.

      Steig referring to anything negative done on his part as ‘in O’Donnell’s head’ obviously doesn’t ascribe him that benefit, but neither does a charge of overt duplicity.

      Just like it took Steig a little while to formulate a response, so too will it probably take R.O. to do the same.

      Personally, I don’t think Steig is being ‘accurate’ when he says that a ‘fundamental rework of a paper is necessary for it to be viable’ constitutes acceptance or encouragement…But I also see that all of R.O.’s model renditions seem to indicate warming of the arctic under any settings and trends. [either this supports the main point of what Steig wants to say, or it shows that nothing is understandable from the data as manipulated] … If the latter is the case, than Steig may have a point as well when he talks about the difference between scientists who try to figure out what the data is saying, vs. others who try to figure out what it doesn’t say (and I do recognize both types should theoretically exist).

      I think it’s only a matter of time before a new publication comes out where the author takes the data down there, and manipulates it differently, but comes to the same conclusions regarding Antarctic warming. (As R.O. indicated, Steig may well still be right about Antarctic warming, but, so far, only by chance). As long as y’all are prepared for that possible reality…or perhaps there should be more PhD ventures to Antarctica setting up more boreholes? There really is no time to increase the resolution of on-the-ground data stations, because no one’s going to want to wait 30-years to get a trend.

      I wonder if R.O. will get to be a reviewer of that work, and/or if it will lead to so many concerned rewrites and re-reviews as it did in this case.

    • PedroS
      Posted Feb 11, 2011 at 8:21 AM | Permalink

      As a long time fan of this blog, I am embarrassed by the way this scientific argument descended into a flame war and to allegations of duplicity (now withdrawn). A 24-hour cooling-off period and personal communication to Steig would have been much better for all concerned. What did Ryan O get from his angry tirade? Nothing: Eric Steig was blown off and therefore did not address any of the questions; he turned from a sympathetic adversary to an embittered one; CA got egg on its face due to the vitriol; RC and their regulars now have another very strong reason not to address enquiring “skeptics”.

      I always considered CA took the moral high ground. Now I don’t.
      Pedro S.

      • PedroS
        Posted Feb 11, 2011 at 8:26 AM | Permalink

        In my previous post,
        “I always considered CA took the moral high ground” should be

        “I always considered CA HELD the moral high ground”

      • bernie
        Posted Feb 11, 2011 at 8:40 AM | Permalink

        If you have not read Steig’s three reviews of the O’Donnell et al paper and, equally importantly, the Editors letters to O’Donnell et al based on these reviews, I suggest that you do. Your point that O’Donnell should have written privately to Steig first is, in an ideal world, one that I would urge. However, Steig’s piece on the O’Donnell article made that option somewhat difficult especially when coupled with RC’s approach to moderation.

        Finally, I would urge you to make the same point at RC – though it appears that the comments have been prematurely, preemptively and surprisingly closed.

        • PedroS
          Posted Feb 11, 2011 at 9:42 AM | Permalink

          bernie: I understand your points, but I do not think that the dreadful RC approach to comments is an excuse: Eric Steig’s email is public, and can be Googled easily.
          I did not make my comments in RC because I know that any points critical of Steig will be buried, and any criticism of Ryan O by CA-followers ther would be trumpeted as an RC public relations victory.

  58. Pete M
    Posted Feb 10, 2011 at 11:05 PM | Permalink

    What I find interesting is that people on all sides of the climate change debate are unwilling to admit any of their research findings are invalid. Given the findings of “Why Most Published Research Findings Are False” ( and the statistical nature of climate change research (eg paleclimatology) it would not surprise me that 50%+ of climate change studies are invalid. Perhaps climate change scientist feel the figure is closer to 0% but I’m sure scientists in every field feel this way about their own field.

  59. AntonyIndia
    Posted Feb 10, 2011 at 11:15 PM | Permalink

    Eric Steig, WA, February 11th, 2011, 2:03 am


    This is at first glance reasonably fair commentary, but I don’t think you are right that this ‘raises questions’ about peer review. Or if it does, it only raises the same questions that McIntyre has raised, but in reverse. He said, for example here [Link fixed by ACR: ], that the journal “Climatic Change” ‘betrayed him’ when he wasn’t sent a final version of W&A (2006): “CC made a fundamental breach of its reviewer agreement with me by failing to send the revision to me before accepting it.” I’m unaware of his having complained to Journal of Climate for cutting me out of the review process. (In point of fact, I think that this was just fine — editors have to make decisions and not all reviewers or authors are going to like it. But O’Donnell and company are totally hypocritical in now claiming a problem with the review process when the shoe is on the other foot). The fact is that the review process works well, as testified to the fact that both my paper AND O’Donnell’s paper — both of which are substantial and important contributions to the advancement of science — were published. It really is that simple. Continually ‘raising questions’ about the process, as you have done here, but never answering the question is helpful to anyone.”

    • mark t
      Posted Feb 11, 2011 at 1:11 AM | Permalink

      He really doesn’t get what they did to his “contribution.”


  60. mct
    Posted Feb 11, 2011 at 2:26 AM | Permalink

    after a merre 123 posts (not incuding those boreholed or othgerwise ignored, it is – surprise! – all over…

    {Comments are off. Case closed. Thanks for your support (most of you). Comments on the “West Antarctica is still warming” post are still open. Science only please.–eric]

  61. toby
    Posted Feb 11, 2011 at 2:44 AM | Permalink

    There’s nothing worse than the stink of self-righteous indignation.

  62. Jean Demesure
    Posted Feb 11, 2011 at 6:13 AM | Permalink

    “I hope everyone is aware that the response on RC – O’donnellgate, is not aimed at Climate Audit, or the blogs..
    It is aimed at the media and politicians”

    @Barry Woods
    Exactly, that’s why RC is set up and how it works. The aim is not to address a scientific controversy but to have the last word, so the AGW followers can always claim : see, you denier have been “thoroughly debunked”.
    The operating mode is so predictable, be it for the Hockey Stick, Climategate or Warming Antarctica.

  63. TonyG
    Posted Feb 11, 2011 at 9:20 AM | Permalink

    I asked Eric on RC (Comment by TonyG — 9 Feb 2011 @ 9:29 PM) for his comment on these very telling graphs, very well explained by Bishop Hill–and received a quick to-the-point reply.

    Ryan, would you now please post a reply to Eric’s arguments? Which parts do you agree with, and which ones do you disagree? The logic of these arguments is sometimes becoming too confusing for a nonspecialist to untangle–and form an informed opinion… Thanks!

    ——-Here’s my RC post–and Eric’s reply———

    Can we please go back to science? I am extremely interested in reading here a statistics, not psychology, -based response by Eric to Ryan O’Donnell’s latest post on CA which discusses in suggestive graphics the way Eric’s reconstruction responds to changes in the Peninsula… Please provide your technical rebuttal of O’Donnell’s criticism; that’s the only way this matter should be argued.

    [Response: One of the things that those plots show is that there is still warming in West Antarctica even if the trends on the Peninsula were zero. So one of O’Donnell’s main claims is shown by O’Donnell to be wrong. Of course, it is certainly true that the Peninsula warming — to the extent it is correlated with the West Antarctic warming — has an influence, as indeed it should. The stronger effect of our having using just 3 PCs is that Peninsula cooling is damped, which is the opposite effect. Of course, some of both happens, but in all the tests I’ve done, the latter is a large effect.–eric]

  64. RandomM
    Posted Feb 16, 2011 at 9:41 PM | Permalink

    I’m always curious as to backgrounds of people in debates on the publication level and I’m trying to find CV’s for all of those involved in the two papers but can’t seem to find a site that will give me Ryan O’Donnell’s CV, alma mater, publications, unless he is the researcher at Carnegie Mellon, which I was told elsewhere was not the case.

    • Posted Feb 16, 2011 at 9:45 PM | Permalink

      Given that I don’t know what you mean by “CV”, you may safely assume that I am not a researcher by trade.

      • HaroldW
        Posted Feb 16, 2011 at 11:07 PM | Permalink

        CV = curriculum vitae. In a word, it’s a résumé. An academic typically lists all publications, making it longer than the résumés with which you (Ryan) would likely be more familiar.

        An example:

11 Trackbacks

  1. […] by Ryan O’Donnell via Climate Audit […]

  2. By Coffin, meet nail. « Bee Auditor on Feb 9, 2011 at 9:30 PM

    […] Source: […]

  3. […] I've sent out a variety of queries related to hist post and will write a new piece soon. He added this one for good […]

  4. […] At CA, this commenter sums it up pretty well: movielib Posted Feb 9, 2011 at 5:03 PM | Permalink […]

  5. […] I've sent out a variety of queries related to hist post and will write a new piece soon. He added this one for good […]

  6. By Top Posts — on Feb 10, 2011 at 7:12 PM

    […] Coffin, meet nail. For those who are not mathematically inclined and did not entirely follow the discussion about Eric’s reconstruction in […] […]

  7. […] O’Donnell hat eine großartige illustrierte Darstellung der Merkwürdigkeiten veröffentlicht, wie Eric Steig die Trends in der Antarktis erzeugt. Wenn Sie […]

  8. […] R code, as reviewer of their paper, a fact which was had remained unknown up-to that point. The ensuing conflagration is now […]

  9. […] Hey all,There’s a feud going on pertaining to this post on RC by these two by climate audit: […]

  10. By Behind the SKS Curtain « Climate Audit on Nov 20, 2013 at 4:16 PM

    […] […]

  11. […] ………in  blog articles titled Steig’s Trick and  Coffin, meet nail. […]

%d bloggers like this: