Neukom and the Steig Over/Under

Earlier this year, I reported on the refusal of Raphael Neukom, an associate of IPCC confidentiality advocate and WG1 Co-Chair Thomas Stocker at the University of Bern, to archive data used in a then recent multiproxy study, Neukom et al 2011 (Clim Dyn). In his refusal letter, Neukom stated that

Most of the non-publicly available records were provided to us for use within the PAGES LOTRED-SA initiative only

Neukom’s website lists a set of multiproxy data, many of the series said to be “available on request for LOTRED-SA contributors”.

Despite Neukom’s “excuse” that the data had been provided to him for use “within the PAGES LOTRED-SA initiative only“, Neukom et al 2011 has been prominently used in the IPCC AR5 draft assessment report. The Neukom reconstruction is used to compare the performance of models to paleoclimate reconstruction in South America in FOD Figure 5.9 as shown below:

Figure 1. Neukom et al 2011 south South American temperature reconstruction (heavy black). Colored lines show results from various climate models.

Unless you are a climate scientist, you would probably not describe the paleoclimate reconstruction as cohering particularly well with the various models, but that’s a story for another day.

In the running text of the draft AR5, Neukom et al is cited on several occasions, including as authority for the observation that SH summer temperatures in the Medieval Warm Period (you know, the one that is supposedly regionally restricted to Greenland and a few counties in England) were “mostly warmer than the 20th century climatology”:

Progress has been made in the SH since AR4, where new tree ring records from the Andes, northern and southern Patagonia, Tierra del Fuego, New Zealand and Tasmania (Boninsegna et al., 2009; Cook et al., 2006; Villalba et al., 2009), ice cores, lake sediments and documentary evidence from southern South America (Neukom et al., 2011; Prieto and García Herrera, 2009; Tierney et al., 2010a; Vimeux et al., 2009; von Gunten et al., 2009) and terrestrial and shallow marine geological records from eastern Antarctica (Verleyen et al., 2011) allow a better understanding of past temperature variations (Neukom et al., 2011). A multi-proxy reconstruction for southern South American (Neukom et al., 2011) finds austral summer temperatures between 900 CE and 1350 CE that are mostly warmer than the 20th century climatology (though associated with large uncertainties), with a sharp transition after 1350 CE to colder conditions that last until approximately 1700 CE.

Citation of Neukom et al 2011 by IPCC clearly takes its use outside the realm of LOTRED-SA associates, but Neukom has thus far taken no steps to ensure that proxy data of Neukom et al 2011 (now used by IPCC) is available to anyone outside his circle of cronies.

Neukom (together with Joelle Gergis) has also published a recent survey of Southern Hemisphere proxies Neukom and Gergis 2011, one of the objectives of which was to report on the “availability” of SH proxies. Unfortunately, with the sort of blindness all too familiar from Climategate, Neukom and Gergis conflate availability of data to cronies with public availability – a situation that I am presently testing with a long-unarchived Eric Steig data set.

Neukom and Gergis 2011 is one of a number of recent multiproxy articles appearing more or less on the eve of AR5, as they themselves note:

Given the importance of global circulation features like ENSO, IOD, SAM and the IPO, a concerted effort is now underway to consolidate existing high-resolution palaeoclimate records from these regions in time for the IPCC fifth assessment report (AR5) (Gergis et al., 2011; Neukom et al., 2010, 2011).

They provide a list of 174 proxies (expanded from the list on Neukom’s website) that supposedly meet the following criteria:

– extend prior to 1900
– are calendar dated or have at least 70 age estimates in the 20th century
– extend beyond 1970 to allow sufficient overlap with instrumental records
– are accessible through public data bases or upon request from the original authors

Series said to be available “upon request from the original owners” are, for the most part, the same series that had previously been described as only “available on request for LOTRED-SA contributors”. Neukom and Gergis report that there were only 14 records extending back to 1000 AD that met the above criteria. Somewhat surprisingly, although IPCC had claimed “progress since AR4” in the development of SH proxy data, only one of the 14 series appears to have been developed since AR4 (with most of the series being developed prior to AR3.)

Many of these records are listed by Neukom and Gergis as only “available upon request”. The most intriguing such example is Eric Steig’s Siple Dome dD (and d18O) series – see excerpt below – which Neukom obtained as a “personal communications”.

The Siple Dome core was drilled in 1993-4: the data was publicly funded. Even by Lonnie Thompson standards, this is a long time for the data to remain both unpublished and unarchived, particularly given the scarcity of long SH proxies. One cannot help but think that the data set would have been promptly published if it had HS shape and, ergo, my prediction is that, if and when, the data ever is made “available”, it will not have a Hockey Stick shape.

As a test of Neukom and Gergis’ statement that the data sets are “available upon request”, I submitted a request for this data to Eric Steig (cc Neukom) a couple of days ago, thus far, without acknowledgement from either.

Dear Dr Steig,
Neukom and Gergis 2011 stated that the datasets listed in their tables were accessible “upon request from the original authors”. Among their series are Siple Dome delD (1000-1993) and d18O (1654-1994) series said to have been sent to Neukom as a “pers comm.” In 2010. Could you please provide me with a copy of the series that you provided to Dr Neukom.

Thank you for your attention,
Stephen McIntyre

My early line on the over/under for getting the data is the publication date of AR5.

Six more of the 14 long series were obtained by Neukom and Gergis from crony contact (rather than public archives) including: Law Dome (Curran); ocean sediment series 106KL (Rein); four South American tree ring series: Central cluster 3A/CAN Composite 20; Central Cluster 3C/CAN Composite 25 and Central Cluster 3B/CAN Composite 23 (all said at Neukom’s website were said to be “unpublished” and to be only “available on request for LOTRED-SA contributors” and Lenca was said to be “Not public”.) The Aculeo temperature reconstruction from pigments is only available in a smoothed version of the regression product – the underlying data has not been archived. In the case of the South American tree ring series, there are public archives of measurement data that appear to overlap the versions used here. The existence of different versions makes the need for careful archiving of versions as used all the more important.

Despite IPCC’s puff about “progress”, a more objective assessment of actual progress in the field would be to inform readers from other fields that there has actually been negligible progress since AR4 in the development of new long proxies.

Tree rings: the Lenca series ends in 1987; the three long Central Cluster/CAN Composite series are Fitzroya cupressoides (FICU) series that end between 1990 and 1995. One of these series is almost certainly a version of the Rio Alerce FICU series used in the AR3 reconstructions Jones et al 1998 and Mann et al 1998-99. At most, the Neukom and Gergis versions appear to be re-processing of data collected in the early 1990s.

Ice cores: the “14” series of Neukom and Gergis appear to include five ice core series, none of which are new since AR4 (and again, most are older than that.) Two series from the 1993 Siple Dome core appear to be included (Na- Mayewski et al 2004; and dD – Steig and White, unpublished and unarchived); a Law Dome series (presumably related to the series illustrated in Jones and Mann 2004); a series from Berkner Island (Mulvaney et al 2002); and Quelccaya, Peru (Thompson 1983, updated 2003). The Law Dome series was definitely available for AR4. As an AR4 reviewer, I had asked that it be included in the illustration of actual SH proxies. IPCC correspondence about the decision not to show this proxy (which has an elevated MWP) is in the Climategate dossier. CRU’s Tim Osborn (also Lead Author in AR5), knowing that the proxy showed an elevated MWP (and that I knew this and that I knew that they knew that I knew…), proposed that they cover themselves by mentioning the proxy in the running text but not show it in the diagram, a device readily agreed to by IPCC authors reluctant to show a SH series with an elevated MWP.

Coral: the discontinuous Palmyra series of Cobb et al (2003). See 2006 CA discussion here.

Sediments: ocean sediment 106KL was already available in Rein (2004), not Rein (2007) as shown by Neukom and Gergis. The Aculeo pigment reflectance series of von Gunten et al 2009 is new.

As I presently understand the matter, only one of the 14 data sets is “new” since AR4: the Aculeo pigment reflectance series. And while the development of novel proxies is a good idea, until the pigment reflectance methodology is replicated in other lakes, little weight should be placed on this result (particularly without the underlying data being available for analysis.)

In my opinion, a candid assessment of progress in the field by IPCC would clearly state that (1) Neukom and Gergis’ list shows that there has been essentially no development since AR4 of SH proxies that permit comparison of the medieval and modern periods; and (2) that much (far too much) of the limited data still remains unarchived. Unfortunately, WG1 Co-Chair Stocker seems more concerned with ensuring confidentiality of IPCC drafts and internal correspondence.


  1. oMan
    Posted Jan 19, 2012 at 11:41 AM | Permalink

    I just wanted to say how impressed I am with the work you have done, and that you continue to do. It is a standing example of how to do things right: the systematic approach; the thoughtful design of the inquries; the attention to every detail; the careful archiving and exposition of what you seek and why; and what you’ve been told or given (or, too often, not given); above all, the calm and professional manner. You’re creating a significant body of high-quality evidence that, sooner or later, will be cited in the indictments (whether legal or merely political and cultural) against people who have trampled on scientific integrity and the public trust. Thanks.

    • TAC
      Posted Jan 19, 2012 at 12:31 PM | Permalink

      I share oMan’s view. Please add my thanks to his.

    • William Larson
      Posted Jan 19, 2012 at 12:32 PM | Permalink

      oMan: I second what you say, and will add that Mr. McIntyre’s writing tends to be eminently readable and often spiced with humor/wit. You get a lot of bang for your buck here, no question.

  2. Pat Frank
    Posted Jan 19, 2012 at 1:35 PM | Permalink

    This comment,

    CRU’s Tim Osborn (also Lead Author in AR5), knowing that the proxy showed an elevated MWP (and that I knew this and that I knew that they knew that I knew…), proposed that they cover themselves by mentioning the proxy in the running text but not show it in the diagram, a device readily agreed to by IPCC authors reluctant to show a SH series with an elevated MWP. (my bold)”

    … is as damning a piece of evidence of the mendacity virtually universal among AGW-promoting scientists as any we’re likely to see.

    1153233036.txt is the Climategate email. Tim Osborn makes the suggestion to leave out the Law Dome series because of its SH MWP, and Jonathan Overpeck agrees this is a good resolution.

    Tim Osborn rationalizes his dereliction by suggesting the Law Dome series is thought by some to show “atmospheric circulation changes rather than temperature changes.” And then, showing the opportunism in the “community,” here we go with Neukom and Gergis using Law Dome in a temperature study. Did Tim Osborn file a protest?

    And the AGU, AMS, and NAS? All unshaken by the pervasive evidence of f{snip] staring them in the face. Recently joined by the NCSE.

    • Craig Loehle
      Posted Jan 19, 2012 at 4:22 PM | Permalink

      You know, it is funny, but if I submit a paper and neglect to cite studies that get a different result from mine, the reviewers remind me of it. One is supposed to cite ALL relevant work, not just the work you like. If you can’t explain the contradictions (and there are always some) you must at least mention them as requiring further study. At least IRL.

    • Posted Jan 20, 2012 at 9:03 PM | Permalink

      From Pat’s linked email:

      (3) Dahl-Jensen showed the temperatures obtained
      by inverting the borehole temperature profiles.
      This has a colder MWP relative to the recent
      period, which shows strong recent warming.

      Simply amazing.

  3. Jit
    Posted Jan 19, 2012 at 2:35 PM | Permalink

    Do any of the models have a better fit than anomaly = 0 ?

  4. tomdesabla
    Posted Jan 19, 2012 at 2:56 PM | Permalink

    Yeah, Steve’s ok, for an oil shill denialist ; )

    No, seriously, I have liberal pro-agw friends who I have pointed to this site, but I don’t think a single one has actually showed up yet.

    You may be able to lead a person to Steve McIntyre, but you can’t make him think.

    • William Larson
      Posted Jan 19, 2012 at 3:11 PM | Permalink

      tomdesabla: OK, here I go wandering snippably off-topic again, but your last line reminds me of that great comment from Dorothy Parker about horticulture: “You can lead a horticulture, but you can’t make her think.”

  5. alex verlinden
    Posted Jan 19, 2012 at 2:56 PM | Permalink

    if it wouldn’t be so tragic (for Science, for financial losses, for making the wrong decisions for e.g. our energy future, etc. …) this would be soooo funny !

    these are absolutely great pieces of investigating and writing, and I second oMan’s reply too …

  6. Craig Loehle
    Posted Jan 19, 2012 at 3:03 PM | Permalink

    I note that in Fig 1 the red line is on a steady upward trajectory of 1.7 deg C/1000 yrs without any help from us. They clearly aligned the models to the 20th Century to all look good then.
    This failure to understand the importance of audit/verification is so baffling. Do they invest in companies whose books are secret? I sure would not.
    What’s the over/under on the confidence intervals? I’m betting they did them wrong.

    • Braddles
      Posted Jan 19, 2012 at 3:25 PM | Permalink

      Since these models run forwards in time, they should be aligned at the starting point, not the end. Aligning at the end gives a false impression that everything is in agreement for the 20th Century.

      • Steve McIntyre
        Posted Jan 19, 2012 at 3:40 PM | Permalink

        Re: Braddles (Jan 19 15:25),

        Yep, good point.

        BTW, I recall reading a recent blog post( which I forgot to mark) which claimed consistency between models and paleoclimate reconstructions. If anyone else has noticed such a statement recently, could you place a link in the comments here. I’ll check AR5 on the topic as well – as this is a topic that they are ramping up.

        Neukom would seem to stand against that proposition.

        • WB
          Posted Jan 19, 2012 at 5:34 PM | Permalink

          Steve is it the quote in this Bishop hill post?

          Also in your opening, I think you mean last year.

          Love your work.

        • Steve McIntyre
          Posted Jan 19, 2012 at 5:51 PM | Permalink

          Yes, that’s exactly the quote that I was looking for – original is here

          To ensure the accuracy of the models at projecting future climate trends, the models are often run backwards in time to “retrodict” past climate changes, and then compared with paleoclimate observations. The models through this process have become remarkably accurate and give the climate research community confidence that the future projections are robust.

          I think that one can reasonably disagree with the contention that the models have, in this case, yielded a “remarkably accurate” representation of the paleoclimate reconstruction. (Not that much weight can be placed on the paleoclimate reconstruction.)

        • Neil Fisher
          Posted Jan 19, 2012 at 8:54 PM | Permalink

          As I commented at BH, that depends on your definition of remarkable – “remarkably accurate” would be commonly assumed to mean “highly” or “very” accurate, but it could be argued that “remarkably accurate” and “remarkably inaccurate” amount to the same thing. Certainly I would imagine that many people, being shown the details of said accuracy, would indeed suggest that it is “remarkable”, but by the same token I can’t imagine many suggesting it is “remarkably accurate”.

          This is “team work” don’t forget…

        • Posted Jan 20, 2012 at 7:04 AM | Permalink

          Steve McIntyre (Jan 19 17:51):

          Not that much weight can be placed on the paleoclimate reconstruction.

          Pity. I was getting remarkably excited.

        • Tom Gray
          Posted Jan 20, 2012 at 11:42 AM | Permalink

          And if the reconstructions turns out to be inaccurate with say missing a MWP, what does this say about the models.

          Two edge swords cut both ways

        • Craig Loehle
          Posted Jan 20, 2012 at 12:45 PM | Permalink

          Yeah, so much for the paleo-recons “don’t matter”–just a pea-thimble trick. If you compare climate models to unverifiable proxies which don’t agree with each other, what do you get? (that’s a rhetorical question…)

        • Tom Gray
          Posted Jan 20, 2012 at 2:33 PM | Permalink

          This verification is not double blind. it is not even blind. The models and the reconstructions have evolved with knowledge of each other. Certain reconstructions have been accepted and others have been rejected. It is not surprising that results that have evolved together by the same community are similar.

          The comparison presented may be a statement in climate physics or it may be a statement in the sociology of science. The comparison may be good because both sets of results are made by people who know each other. This is not a blind experiment in any way.

        • Posted Jan 21, 2012 at 11:19 PM | Permalink

          At an AGU 2010 gene wahl asked a very interesting question from the floor. ( dont ask why I rememember shit like this)

          A GCM study was done of the Iberian ( as I recall) area. They had proxies and paleo runs of the GCM. In theory its a really cool approach ( using the GCM to construct a prior) In this particular case their was one proxy at odds with the GCM. Wahl was there, malcom hughes. (john
          mashey too)

          Wahl: “so which is right? the GCM or the proxy?”
          warmed my heart

        • Jeff Norman
          Posted Jan 22, 2012 at 5:49 AM | Permalink

          The word “retrodict” does not show up in my dictionary.

          Retro is “a prefix occurring in loanwords from Latin meaning “backward””.

          There is no “dict” but “diction” means “style of speaking or writing””.

          So retrodiction could mean speaking out your asterisk.

        • Mark F
          Posted Jan 22, 2012 at 11:44 AM | Permalink

          Retrodict was easy to find using public dictionaries. Your polyps must have been blocking your view.

        • theduke
          Posted Jan 22, 2012 at 1:35 PM | Permalink

          Retrodiction: It’s not in my dictionary either. It is in the online dictionaries: “the explanation or interpretation of past actions or events inferred from the laws that are assumed to have governed them. . .”

          In other words, reverse soothsaying or predicting the past with knowledge limited by the parameters of the present.

        • HAS
          Posted Jan 22, 2012 at 2:02 PM | Permalink

          This definition is what my Sorter Oxford says, noting [ORIGIN from retro- + diction, after prediction.]

          While one can question how well the retrodiction works, and therefore the quality of the assumption over the laws that governed the past actions, for this very reason prediction/retrodiction is an honourable part of science.

        • Punksta
          Posted Jan 23, 2012 at 1:48 AM | Permalink

          The word “retrodict” does not show up in my dictionary.

          Prefer “hindcast” ?

        • theduke
          Posted Jan 23, 2012 at 11:10 AM | Permalink

          “Hindcast” does not show up in my dictionary either. Of course I insist on using my old Webster’s New World of the American Language (college edition) circa 1963. I wore out the first volume and bought a used one in good condition to replace it. I rely on the internet for newly coined terms.

          Yes, a lot of words have been invented since then, some for the sake of expedience. As it relates to science, are the results of “hindcasting” or “retrodiction” falsifiable? Or even worth attempting to falsify?

          (Note: I am a layman askings questions for which I do not have an answer.)

      • Posted Jan 20, 2012 at 11:09 AM | Permalink

        Interesting, let’s try with some reconstructions first:

        • RomanM
          Posted Jan 20, 2012 at 11:27 AM | Permalink

          Nice work, UC!

          The animation graphs are a good way of showing the chosen “anomaly period affects the look of the graph.

        • Posted Jan 23, 2012 at 5:55 AM | Permalink

          Neukom et al:

          In order to minimize such reductions of
          variance back in time, the temperatures reconstructed by
          PCR were rescaled to the mean and standard deviation of
          the predictand in the calibration period (e.g. Cook et al.
          2004). CPS avoids this loss of variability by simply scaling
          a composite of the predictor data against the predictand in
          the calibration period (e.g. Esper et al. 2005; Jones et al 1998).

          All resulting reconstructions then have equal standard deviation over the anomaly period? I need to try that next. Should I scale the smoothed values or the originals? Which way is more accurate?

        • RomanM
          Posted Jan 23, 2012 at 9:20 AM | Permalink

          The scientifically more reasonable approach would be to scale an unsmoothed series to unsmoothed calibration data, particularly if one is interested in properly calculated standard errors. However, when comparing several series to each other, I think it could depend on what smoothing method is used and what final product is desired. For some methods there may be no difference.

          In this case, it might be interesting to see the result both ways.

        • Posted Jan 23, 2012 at 11:57 AM | Permalink

          UC — The calibration period and the anomaly period are different critters. CCE, ICE, and CPS all give the same mean as the instrumental data over the calibration period. Usually the instrumental data is given as an anomaly relative to its mean during a recent 30-year period, so it will have zero mean during this period, but then even though the reconstruction is to be interpreted relative throughout relative to this reference period, it will not have zero mean during the reference period. Changing the anomaly period in this sense simply moves all curves up or down by an equal amount.

          Usually one would not use a short pre-calibration anomaly period, since there is no instrumental data there, and the proxies are noisy. It sometimes does make sense to use the full reconstruction period as the reference period for a reconstruction, as in Loehle and McCulloch (2008).

          I haven’t looked at Neukom’s paper, but it sounds like the “loss of variance” he refers to is that from “ICE” — regressing the instrumental temperature on the proxy as if the proxy caused global temperature. CPS (matching variances during the calibration period) increases the standard deviation of the reconstruction during the calibration period (and therefore elsewhere, relative to the calibration mean) by a factor of 1/R, while CCE increases it by a factor of 1/R^2. Since only CCE (regressing the proxy on temperature and inverting) is consistent in the usual case that all the error is in the proxy, CPS still has inadequate volatility. (If there is equal noise in the instrumental data and the proxy, TLS would make sense, and CPS is likely very similar to TLS despite its lack of a theoretical foundation.)

          But the equality of variance observed in CPS should be in the calibration period, and about the calibration mean, not in the anomaly reference period or about the anomaly mean.

        • RomanM
          Posted Jan 23, 2012 at 12:21 PM | Permalink

          My interpretation of UC’s question was in the context of his doing a similar animation as he had done earlier by varying the anomalising period to see how the various time series matched up except this time by rescaling the series to equal variance for the anomalised period.

          In the common anomaly exercise, the reconstructions moved up and down by different amounts (sometimes in opposite directions). In the common rescaling period there could be similar effects as well reflecting possible different time-related characteristics of the recons.

        • Posted Jan 23, 2012 at 3:34 PM | Permalink

          I understand from the Neukom quote that rescaling of PCR result is an additional step that improves the reconstruction. Then changing the anomaly / calibration period shouldn’t change the relations between reconstruction much. We don’t know the past stds, but we can assume some arbitrary value, say 0.1 (smoothed) :

          For CCE we have inconsistency statistic, but with variance matching it seems a bit difficult to derive similar

        • Posted Jan 31, 2012 at 9:53 AM | Permalink


          UC — The calibration period and the anomaly period are different critters. CCE, ICE, and CPS all give the same mean as the instrumental data over the calibration period. Usually the instrumental data is given as an anomaly relative to its mean during a recent 30-year period, so it will have zero mean during this period, but then even though the reconstruction is to be interpreted relative throughout relative to this reference period, it will not have zero mean during the reference period. Changing the anomaly period in this sense simply moves all curves up or down by an equal amount.

          Got it, good point. I was a bit distracted by 0183.txt

          we used the ’61-90 base period for the absolute anomaly scale, but
          we aligned the series based on an earlier (’31-60) interval of the
          instrumental record, which pre-dates (largely) the recent decline in the
          Briffa et al series

          The 30 year alignment is a bit short, but the video is for illustrative purposes only 😉

          ps. Take a look at the AR4 Fig 6.10 c and compare with 6.13.d. It seems that reconstructions are shown as most accurate (more ‘scoring’) during 1880-1960 in the former and 1450-1900 in the latter. Is it due to different centering period?

        • Steve McIntyre
          Posted Jan 20, 2012 at 12:15 PM | Permalink

          Nice graphic – what reconstructions are in it?

        • Posted Jan 20, 2012 at 2:51 PM | Permalink

          First set I found, from Juckes folder, let me check..

        • Posted Jan 21, 2012 at 4:40 PM | Permalink

          They are from Juckes “cited reconstructions” csv-file,


          lowpass.m smoothed with f=1/40, and “minimum roughness” boundary constraints (my favorite, see )

        • S. Geiger
          Posted Jan 21, 2012 at 11:09 PM | Permalink

          I don’t get the white noise vid. The lable keeps saying unprecidented even when the points plotted clearly are not (?)

        • Posted Jan 22, 2012 at 9:27 AM | Permalink

          Thanks, UC. However, like S. Geiger, I don’t see why there are so many “unprecedented” cooling and warming points.

        • Posted Jan 22, 2012 at 9:42 AM | Permalink

          I get it now — Each “unprecedented” endpoint is the highest or lowest on the newly resmoothed curve. With Mannian “minimum roughness” (which as Steve has shown amounts to endpegging), the endpoint has much higher variance than the internal smoothed points.

        • Posted Jan 22, 2012 at 12:44 PM | Permalink

          Yes, the filter is not time-invariant, so the ‘old record’ is smoothed away and we have a new record almost every time a new sample arrives.

        • Scott Brim
          Posted Jan 22, 2012 at 1:15 PM | Permalink

          Re: UC (Jan 22 12:44),

          Hu, you say, “Yes, the filter is not time-invariant, so the ‘old record’ is smoothed away and we have a new record almost every time a new sample arrives.”

          May I inquire as to what kinds of things might happen to the record if a new sample arrives which had been collected in the form of a tree ring proxy temperature value taken after about the year 1960?

        • Tom Gray
          Posted Jan 22, 2012 at 3:29 PM | Permalink

          If this were an engineering report, the properties of the smoothing and other procedures would be noted. As in : All charts will sue the Smoothing method of Smith and Jones (1962). Smith and Jones (1962) would be well known and available in text books or something specific would be supplied with the report. Readers of the report would be provided with enough information to give it a proper assessment.

        • Posted Jan 23, 2012 at 12:08 PM | Permalink

          I see GISSTEMP is not currently endpegging its 5-year running mean, and simply gives no smoothed values for the last 2 years. See I recall a prior discussion here that prior to 2009, when temperatures were rising, GISS used Mannian double flip endpegging which gave a dramatic conclusion, but then in 2009 when the index took a nose dive, they suddenly saw the error of their ways and started truncating the smoother instead.

          Does anyone recall whether GISS temporarily reverted to endpegging in 2010, when the series ended on a high value?

        • Posted Feb 7, 2012 at 9:25 AM | Permalink

          Can’t recall GISS doing such tricks, but

          Rahmstorf changed the filter on the fly:

          Met Office had some problems:

          In March 2008, some diagrams were placed on this web site which showed smoothed annual series that included data for 2008. The annual value for 2008 was based on the only two months of data – January and February – that were available at the time. January and February 2008 were cooler than recent months, leading to a marked downturn towards the end of the smoothed series (Figure 2, orange line) that caused much discussion.

        • Posted Jan 22, 2012 at 1:45 PM | Permalink

          Are the smooth lines in Neukom’s graph similarly end-pegged?

        • Posted Jan 20, 2012 at 1:18 PM | Permalink

          Needs a soundtrack. How about

        • Posted Jan 20, 2012 at 1:51 PM | Permalink

          I’ve just tried exactly that. Superb.

        • Posted Jan 20, 2012 at 3:02 PM | Permalink

          This is one option,

          “Sounds of Trees”

        • srp
          Posted Jan 20, 2012 at 9:52 PM | Permalink

          That animation actually made me a little seasick. It’s no wonder I can’t read in the car.

        • DeNihilist
          Posted Jan 20, 2012 at 10:20 PM | Permalink

          Thanx UC, you made a metallica fan outta me finally!

        • Posted Feb 15, 2012 at 6:56 AM | Permalink

          Hu ( ) ,

          Changing the anomaly period in this sense simply moves all curves up or down by an equal amount.

          I think AR4 Fig 6.13. is an exemption. If all the curves are moved by equal amount, then the coloring should stay the same. Here’s 6.10.c:


          and 6.13.d:


          It is interesting that the reconstruction ensemble gets more accurate for 1500-1899 in the latter. I don’t have the raw data, has only ‘as plotted’-version, so I cannot YouTube this.

  7. Daniel
    Posted Jan 19, 2012 at 3:16 PM | Permalink

    This post is again very informative and shows more obvious defaults in IPCC’s AR5 process, and a further confirmation of IPCC’s carelessness in implementing 2010 IAR report’s recommendations. I have the impression that your posture, your post drafting is changing somehow ; I mean, you go to the point more directly than you used to previously ; you seem to have reached a point where too much being too much, you tend to show that all these IPCC contributors do not deserve the level of formal manners you were used to maintain previously.

    Am I right ? Is this the next step after your turning down being a AR5 report reviewer ?

  8. Tony Mach
    Posted Jan 19, 2012 at 3:25 PM | Permalink

    As a slightly OT remark: I find the choice of colors in that spaghetti-graph by Neukom et al interesting. Lots of warm yellow-red-brown colors? The blue and green tends to be dark. SRSLY? I wonder what the perception would be if these lines instead were in more clearly distinguishable colors like bright pink, bright red, bright blue and so on.

  9. acob
    Posted Jan 19, 2012 at 3:28 PM | Permalink

    “Unless you are a climate scientist, you would probably not describe the paleoclimate reconstruction as cohering particularly well with the various models,…”
    The immense restraint it must take to write such a deadpan description – but I can’t stop chuckling about it.

  10. Steve McIntyre
    Posted Jan 19, 2012 at 3:33 PM | Permalink

    For the record, in contrast to my experience with Stocker’s Neukom, I recently requested data from Fredrk Ljungqvist for his reconstruction and received it promptly and cordially. I plan to discuss this soon.

    • William Larson
      Posted Jan 20, 2012 at 12:53 AM | Permalink

      “…received it promptly and cordially.” The irony here is that this feels like a breath of fresh air, whereas it ought to be the humdrum everyday norm.

      • Arkh
        Posted Jan 20, 2012 at 8:41 AM | Permalink

        What should be “humdrum everyday norm” is not to have to ask for the data but have it available with the paper or somewhere mentionned in the paper.

        • Posted Jan 20, 2012 at 9:40 AM | Permalink


        • William Larson
          Posted Jan 20, 2012 at 12:59 PM | Permalink

          You bet.

        • Punksta
          Posted Jan 23, 2012 at 1:55 AM | Permalink

          What should be “humdrum everyday norm” is not to have to ask for the data but have it available with the paper or somewhere mentionned in the paper.

          The primary blame here lies with journals.
          What can be done to imbue them with the spirit of real science ?

  11. EdeF
    Posted Jan 19, 2012 at 4:13 PM | Permalink

    Do we have instrument temperature data for South America from about 1920 to present, and how does it
    compare with the proxy re-constructions?

    • EdeF
      Posted Jan 22, 2012 at 3:08 AM | Permalink

      Here is a plot of South American Instrumental temperature data from CRU. Notice that it does not start warming again until about 1910, then has the
      similar pattern found in N. American, that is an
      increase in temperature until 1940, then a drop until the 70s, then further warming until recently. This differs from the proxies above, which show constant
      warming from about 1850 on, and barely a blip at 1940.

      Which one is correct, the proxie or the instruments?

  12. Kenneth Fritsch
    Posted Jan 19, 2012 at 4:15 PM | Permalink

    “What’s the over/under on the confidence intervals? I’m betting they did them wrong.”

    Not sure which way you think the CIs should go, but I see them as sufficiently wide for the reconstruction that it can be claimed that the model results are all withing the range of the CIs of the reconstruction except for the orange line result and part of the gray line result.

    I think the CI strategy might be applied here as it was with the Tropical surface to troposphere warming comparing climate models to observations. You could, of course, argue that we should use the SEs for the model results, but in the case of the IPCC the argument more likely will be ike that of a defense attorney.

    • Steve McIntyre
      Posted Jan 19, 2012 at 5:53 PM | Permalink

      I’d be amazed if the “confidence intervals” illustrated here have any basis in statistics known to mankind.

      • Phillip Bratby
        Posted Jan 20, 2012 at 10:26 AM | Permalink

        But probably known to Mann’s-kind.

      • Posted Jan 20, 2012 at 10:33 AM | Permalink

        But definitely known to Mannkind. 😉

  13. David L. Hagen
    Posted Jan 19, 2012 at 4:20 PM | Permalink

    IPCC policy allows for unpublished data but indicates that it be archived in hard copy and its location be made cited!

    4.3.3 Preparation of Draft Report
    Preparation of the first draft of a Report should be undertaken by Coordinating Lead Authors and Lead Authors. Experts who wish to contribute material for consideration in the first draft should submit it directly to the Lead Authors. Contributions should be supported as far as possible with references from the peer-reviewed and internationally available literature, and with copies of any unpublished material cited. Clear indications of how to access the latter should be included in the contributions. For material available in electronic format only, a hard copy should be archived and the location where such material may be accessed should be cited.

    Click to access ipcc-principles-appendix-a-final.pdf

    IPCC further allows for unpublished non-peer reviewed literature.

    . . .and selected non-peer review literature according to Annex 2 and IPCC Supporting Material (see section 6). Material which is not published but which is available to experts and reviewers may be included provided that its inclusion is fully justified in the context of the IPCC assessment process (see Annex 2).

    Disparate views should be included!

    Lead Authors should clearly identify disparate views for which there is significant scientific or technical support, together with the relevant arguments.

    Appendix A to the Principles Governing IPCC Work
    Adopted at the Fifteenth Session (San Jose, 15-18 April 1999) amended. . .Thirty-Fourth Session (Kampala, 18-19 November 2011)

  14. KnR
    Posted Jan 19, 2012 at 4:57 PM | Permalink

    Once again poor practice that would not be accepted coming form an undergraduate, is consider OK from those claiming to represent the ‘best’ of the science in the area . Is it really to much to expect these academics to meet the standards they would demand of their own students?

  15. AJ
    Posted Jan 19, 2012 at 9:01 PM | Permalink

    Steve… have you decided that you can best be an AR5 reviewer by working outside “the process”? That is, to cite and critique the drafts here at CA. If so, you have my vote!

  16. ausiedan
    Posted Jan 19, 2012 at 9:22 PM | Permalink

    It seems to me that the document reviewed here has not been replicated by other similar studies and cannot therefore be regarded as suitable for refernece by any scientific publication.
    I therfore confidently predict that it will not be referenced in AR5.

    Whoops – my silly – I forgot that AR5 will NOT and is NOT INTENDED to be a serious scientific document.
    So all is well.

  17. Alexej Buergin
    Posted Jan 20, 2012 at 5:13 AM | Permalink

    Maybe the problem is not bad scientific practice. Maybe these people learned how to treat data in Swiss banks, and now they do just as the banks do.

  18. Posted Jan 20, 2012 at 6:13 AM | Permalink

    Hi all,

    I had a good interview with Dr. Neukom some time ago. Here is the link:

    • David Silva
      Posted Jan 20, 2012 at 9:35 AM | Permalink

      Hmmm…I did click the link, but unfortunately I don’t speak German. Any English translation done on this?

      • Matt Skaggs
        Posted Jan 20, 2012 at 11:06 AM | Permalink

        Some translated excerpts from the interview:

        “Mit Hilfe dreier verschiedener statistischer Verfahren”
        “with help from three different statistical approaches”

        “Temperaturen in Südamerika wie praktisch überall auf der Welt in den letzten 150 Jahren stark angestiegen sind”
        Like nearly everywhere temps in SA have climbed steeply in the last 150 years.

        “Die heutigen Temperaturen scheinen in Südamerika jedoch (noch) nicht wärmer zu sein als in früheren warmen Perioden während der letzten 1100 Jahre, was in Kontrast steht zu den meisten Resultaten aus der Nordhemisphäre.”
        Modern temps are not higher than in earlier periods within the last 1100 years, unlike the northern hemisphere.

        The host then asks if SA is “terra incognita” for temp proxies.

        Neukom responds that SA is not “terra incognita” but that many long records have uncertain dates. The data is very limited in certain regions. Also, there is little data before 1500, which is clearly stated in the paper.

        The host asks about the MWP.

        “Insbesondere was die Amplitude, d.h. das Ausmaß der Warmphase im Vergleich zur darauf folgenden kühlen Phase und den heutigen Temperaturen angeht möchte ich meine Hand nicht ins Feuer legen.”
        There was clearly warming between 1150 and 1350. However, there is considerable uncertainty about the magnitude of the warming with respect to both later cooling and the modern period. “I would prefer to not “stick my hand in the fire” on that question.”
        Neukom then goes on to say that the peak of the MWP varied across SA. We need to understand the forcings on a finer scale, and that will help with future reconstructions.

  19. Ftzr
    Posted Jan 20, 2012 at 10:47 AM | Permalink

    I love the mystery – ‘unarchived data’ – what a challenge!

    Let’s see: Google “Steig White Siple Dome”

    First page shows Siple Dome Isotope Highlights at

    That leads directly to the NSIDC site:

    which, it appears has a ton of high resolution isotope data, that appears to have been archived some time ago.

    … and I thought there would be a story here. Oops.

    Steve: While the data sent by Steig and White to Neukom is undoubtedly related to this data, none of the datasets on these pages appears to me to be candidates for what was sent. First, the data sets at this link are all depth vs isotope. The UNH core gives an idea on accumulation rates, but for use in paleoclimate analyses the conversion from depth to age has to be done and isn’t done in these data sets. Most of the cores listed here as “high-resolution” are about 30 meters deep; this probably doesn’t cover more than about 150 years. Other cores are 100 meters deep – probably going back only to 1250 or so.

    Also note that some isotope data in these lists is available only to WAISCORE PIs.

    • Posted Jan 20, 2012 at 12:27 PM | Permalink

      Chuckle !!!
      Nice try !!

      Good followup Steve!!

    • Ftzr
      Posted Jan 20, 2012 at 1:20 PM | Permalink

      Depth Age scale for Siple-A is described here:

      and the latest version is available here:

      You appear to not have noticed that this is a much better way to archive data. Since age models can be adjusted over time (new dates, new techniques, new 14C calibrations etc.), fixing the date to the samples in the archive is not robust. Indeed, you might want to investigate why the GISP2 cores at NOAA are still on their original time scale in the archive, rather than on the much better age scale that was developed for NGRIP.

      But if you prefer to invent problems to bash scientists with, go right ahead. I’m sure there is something else you don’t understand that can be used against them.

      Steve: archiving of the age-depth relationship used in a result is an integral part of an adequate archive – something that I’ve encouraged over and over. WHy on earth would you suggest that I am against details?? I’ve endorsed proper and detailed archives. Indeed, I’ve disapproved of the failure to archive age-depth relationship in, for example, Thompson’s Quelccaya data set that also is in the Neukom data.

      Thank you for drawing my attention to the SIDADS age-scale for the A-core. This is helpful, but is not a substitute for Steig archiving the data set that he provided.

      The data for Core A that is presently available was described as “low resolution”. The webpage for Core A ( states “Note: White’s data are currently available to WAISCORES PIs only.” I presume that the data said to be available ony to WAISCORE PIs is what was provided to Neukom.

      In any event, I can think of no good reason why the version provided by Steig/White to Neukom should not be archived.

      Nor do I understand why you wouldn’t simply agree that Neukom should have ensured that he was able to archive the supporting data for his calculations.

      • Ftzr
        Posted Jan 20, 2012 at 4:47 PM | Permalink

        I love the way you have completely imagined some conversation that Neukom and Steig might have had. Maybe Neukom never even talked to Steig. Maybe he got it from White. Or perhaps he spoke to Mr Google and no-one else. Perhaps White simply pointed him to the NSIDC archive (mentioning that the core index pages were out-of-date since the isotope data are indeed available) and Neukom put together his own annual series (core a seems to be roughly bi-annual, core b is sub-annual). I mean, how can one trust the isotope measurement unless you know exactly how Neukom came across this info?

        So clearly any data archive has to include all emails that anyone even remotely connected has ever sent to anybody – just in case Steve McIntyre wants to forensically put together some history on data that he isn’t the least bit interested in, but that happens to be connected to someone he wants to harass, just because.

        We should try and get that inserted into the NSF regulations or something.

        To go back to being just a little bit serious, the White and Steig data are archived, with an age model, and anyone who actually cared about what they showed can easily examine them (as I just have). So stop playing games.

        As for neukom’s archive, that should all be online. But your rush to find fault with Steig is just grandstanding.

        • Posted Jan 20, 2012 at 7:27 PM | Permalink

          you are not fooling anyone. Read the post. Steve sent a request to Steig already. now start to agonise about your grand-children

    • Squanto McButterpants
      Posted Jan 20, 2012 at 6:44 PM | Permalink

      This link says White’s data is only available to WAISCORE PIs:

  20. Posted Jan 20, 2012 at 12:32 PM | Permalink

    “Progress has been made in the SH since AR4, where new tree ring records……”
    Zoinks !!
    More tree rings.

  21. Crusty the Clown
    Posted Jan 20, 2012 at 3:57 PM | Permalink

    I’m just a clown, so don’t listen to me, but those pesky editors at the “International Journal of Clownatology” insist that if the correct URL for the supporting data is not in the paper at the time of publication, then the data is not considered archived. If only other Int’l Jour. of C.s would adopt similarly rigorous requirements for the publication of articles!

  22. Eric Steig
    Posted Jan 20, 2012 at 7:12 PM | Permalink


    As one of your more adept readers discovered, the relevant data were archived many years ago — including both depth and age.

    While the format is not as convenient as it could be — and here I would agree with you that we could (and we generally do) do better — the data have hardly been ‘refused’ to anyone. There was no stipulation put on the data; and the old ‘available to PIs only’ listed next to White is simply something that wasn’t removed. It would be normal to have in place while the preliminary data are still being produced, to help avoid incomplete or erroneous data sets from propagating. No one had asked for these data, and evidently the data center folks didn’t go back and check on whether they could be accessed. I will alert them to this.

    Meanwhile, I’ve gone ahead and put the data in more convenient form and you can go to my website ( to get them.

    Email me (I will read it) and let me know if you have any problems getting the data.

    Eric Steig

    Steve: thanks for this undertaking. The link doesn’t work at present, but I presume that this will be tidied.

    In addition to doing your part as you’ve done, I think that you should contact Neukom and urge him to ensure that the other data as used in his study is properly archived and accessible. This sort of urging from you would presumably have more effect than mine and help eliminate criticism on issues that should not be at issue.

    • TerryMN
      Posted Jan 20, 2012 at 10:01 PM | Permalink


      There may be a permissions problem or something else going on with the URL – the /sipledome URL gets a not found, and the only two directories visible from data directory are itase and taylor. Thanks.

      • Brandon Shollenberger
        Posted Jan 20, 2012 at 11:47 PM | Permalink

        I get the same thing.

    • Neil Fisher
      Posted Jan 20, 2012 at 10:13 PM | Permalink

      Thanks and good to see your participation – most welcome!

    • WB
      Posted Jan 21, 2012 at 12:50 AM | Permalink

      Hi Eric Steig, good to see you comment and within just a few days of Steve McIntyre asking for stuff but I get the ‘not found’ error from your link. I guess that makes you one of Steve’s less adept readers 😉 (just kidding).

      Quick question – if, as you say, the data in question in Steve’s post used by Neukon was archived years ago how can it be ‘new’ for the purposes of IPCC AR5’s assertion also noted in Steve’s post? To paraphrase Steve, unless you are a climate scientist, you would probably not describe old archived data as new.

    • Steve McIntyre
      Posted Jan 22, 2012 at 4:40 PM | Permalink

      I’ve now examined the data posted up at In his prior comment, Eric Steig stated:

      As one of your more adept readers discovered, the relevant data were archived many years ago — including both depth and age.

      This is not the case.

      The new archive includes the annual delD series from 1000-1993 that Neukom had cited. No delD data was in the archive that Steig said had been ‘alertly discovered”. It lists data for a number of cores, including Core A here. An age-depth scale for Core A is here.

      None of the previously available data includes delD data (i.e. the back-to-1000 date provided to Neukom.) Steig’s statement that this data had been previously available and that he has simply placed the data in more “convenient” form is not correct. The delD data cited by Neukom was not previously archived in any form – convenient or otherwise.

      Nor does the O18 data in Steig’s new archive correspond to the previously available data. The new archive includes the annual O18 data from 1654-1994 provided to Neukom. O18 data for core A had been previously archived but this data (after conversion to a date scale) is NOT the same version as that in the present Steig archive.

      Here is a plot of the “new” data (black) with the red (bottom panel) showing the previously available O18 data (under its age scale):

      It is commendable that Steig archived this data in response to the Climate Audit request. However, it is very tiresome that Steig chose to pretend incorrectly claim that the data was already available and that the present archive was simply providing it in a “more convenient” form. The proper course of action would have been for Steig to issue a small apology for not previously placing the data online, rather than pretending incorrectly claim that it had been online all the time.

      Script for graphic is:

      #siple age: core A

      plot(work$bp,work$O18,type=”l”,xlim=c(2000,-50) )
      #[1] 252



      title("Siple Dome")

      legend("topleft",fill=1:2,legend=c("New","From Old"))

      • DGH
        Posted Jan 23, 2012 at 12:36 AM | Permalink

        What causes the apparent change in the behavior of the data, referring to the smoothness, from roughly 1725 to 1825?

  23. Wayne
    Posted Jan 21, 2012 at 10:24 AM | Permalink

    My question is: who is using software that generates such horrible graphs? I know that climate science is full of roll-your-own FORTRAN code rather than using modern tools (or at least tools that have been well-maintained for decades). I don’t even know of free software that can make such an ugly graph. Or did someone work hard to make it so ugly?

    Just because something looks good doesn’t mean it’s correct, but when something looks so sloppy, I wonder at the rest of the thought process behind the data.

    (To name a couple of specifics: the codors stink, the zero line is the same color as one of the traces, and the lines making the axes and ticks are too thick and don’t even meet properly at the corners.)

  24. Posted Jan 21, 2012 at 1:57 PM | Permalink

    OK, If all documents were made open to the public for comment before publication. Just how much would get published. There would be continuous ranting by the great informed and intelligent ones (at least that’s what they imagine of themselves) hoping to forward their views as the One Truth.
    Please see WUWT for 3000+ posts trying to disprove/prove GHGs act like greenhouse gases. And there are still 3 camps The warmists that believe in GHG theory and thermodynamic laws. Those that believe that gravity acts like maxwells demon. Those that believe that flugal dust warms the planet. Those that believe that black body radiation is stopped if there is conduction occurring.
    Comments are becoming interesting with Hans Jelbring calling out Physicist Robert Brown:
    “It is good to know that you and willis are great IPCC supporters.”

    Imagine if blog review of publications really happened. There would be NO scientific advancement!

  25. Eric Steig
    Posted Jan 21, 2012 at 4:22 PM | Permalink


    my apologies, I gave you the wrong link.
    should have been simply:

    • Jeff Norman
      Posted Jan 21, 2012 at 8:41 PM | Permalink

      Which says it was last modified “20-Jan-2012 16:13” nine minutes for your post. Interesting.

      • DGH
        Posted Jan 22, 2012 at 1:08 AM | Permalink

        Did Steig suggest that the data had been archived at that location for a long period of time?

        He wrote, “Meanwhile, I’ve gone ahead and put the data in more convenient form and you can go to my website”

        • Earle Williams
          Posted Jan 23, 2012 at 3:36 PM | Permalink

          I think it being actually available online is rather convenient, don’t you?


    • igsy
      Posted Jan 22, 2012 at 3:55 PM | Permalink

      In my line of work, the regulatory standards bar is set at a level to deal with the plausible deniability problem. Whilst I am not aware of any legally enforceable institutional regulatory standards for Climate Science Research, it would be nice to have a link to the data in its initial and apparently inconvenient form, together with timestamp or equivalent original proof of posting. Are you able to oblige in this regard, Dr Steig? I would appreciate it if you would, since that would take the plausible deniability issue off the table, leaving us to deal with matters of greater importance.

      Thanks in advance.

  26. Chas
    Posted Jan 22, 2012 at 7:16 AM | Permalink

    Picking points off the graph every 25 years (and using some ‘imagination’ during the last century) I find that the mean of the models : the Proxy recon has an R2 of 0.048
    Whilst the mean of 11 cumulative normal random number series achieves, on average(over 10K tests), an R2 of 0.080 with the proxy recon.

  27. steven mosher
    Posted Jan 22, 2012 at 4:10 PM | Permalink

    Steve and Others

    Let me suggest that you take the time to watch this entire video.

    And let me spoil the punchline

    1. reducing uncertainty about sensitivity is worth trillions of dollars. that is, climate science
    and economics tells us that we can reduce potential costs of mitigation by reducing the uncertainty
    surrounding sensitivity.

    2. Work in paleo is critically important to reducing this uncertainty. In terms of bang for the buck
    dollars spent on paleo have huge leverage. Again, this is entailed by IPCC science.

    If we accept, for sake of argument, these findings, one thing becomes clear. Paleo is a good investment.
    Paleo data and analysis is critically important. Its worth trillions. As it stands, we under invest in
    Paleo relative to the ROI. Given its importance ( accepting their argument ) we could reasonably expect
    this data and the methods should be subject to the most exacting standards we have.

    Bottom line: looking at the ROI of the science its clear that we need more money for paleo. We also can suggest that those charged with this vital data and science should
    be willing to accept more scrunity. It’s mission critical data.

    • Bob Koss
      Posted Jan 22, 2012 at 4:58 PM | Permalink

      I didn’t listen to the program, but from what you say, the calculations are based on mitigation being a necessity. I’m not aware of that having been conclusively established. It seems to me the trillions saved by increasing accuracy is nothing but speculation.

      I wouldn’t object to more funds being put toward learning more about paleo in general, not just temperature, but I think taking the funds from the climate modeling allocation is the way to go. Their really isn’t any reason for having so many models when they have no objective way to declare any of them as being incorrect.

      • Posted Jan 22, 2012 at 7:34 PM | Permalink

        the calculations are not based on mitigation being a necessity.

    • Bruce
      Posted Jan 22, 2012 at 5:40 PM | Permalink

      What if Jørgen Peder Steffensen is right and “We started to observe meteorology at the coldest point in the last 10,000 years.”

      Why spend one more penny to prove that we are damn lucky the LIA came to an end and we should be really, really grateful it is a tiny bit warmer.

      Temperatures in Greenland were about 1.5 C warmer 1000 years ago than now.

      It was perhaps 2.5 C warmer 4000 years ago.

      The period around 1875, at the lowest point of the Little Ice Age, marked the coldest point in the last 10,000 years.

      Other evidence from elsewhere in the Northern Hemisphere confirms this picture.

    • Geoff Sherrington
      Posted Jan 22, 2012 at 7:09 PM | Permalink

      Maybe dendrothermometry is facing an uncertain future, one that could lead to its demise. There are two broad ways to scope error, especially bias. One is to take periods of featureless tree ring response and ascertain if temperature was also invariant within sensible limits. The other is to examine calibrations within the instrumented period – and here we meet the problem of divergence. A modern summary paper dealing with these is badly lacking, though it would be simple to construct if openness was the criterion.
      I’ve looked again at a subset, the Tasmanian data. Some is quite old, but it cannot be updated because of a lack of relevant temperature records. The main station of relevance for the Huon Pine in the Mt Read wet mountainous area has been recording on BoM public records from late 1996.
      Eastern, drier lowland sites involving other tree genera are being reported, but are not encouraging. Partly, this could be because of poor temperature records again, plus the divergence problem again.
      A point will be reached when it is realized that you can’t make a silk purse from a sow’s ear.
      Like all of us, I wish for a good proxy for temperatutes, but the more I read of dendro, the coarser the sow’s ear grows.

    • Howard
      Posted Jan 22, 2012 at 7:35 PM | Permalink


      As far as I remember, you have always stated that paleo is not important. Did this video change your mind?

      • Posted Jan 22, 2012 at 7:43 PM | Permalink

        Read more carefully.

        LGM versus the past 2K.

        when it comes to sensitivity the past 2K dont currently constrain sensitivity very well. Hence, not currently that important. LGM.. totally different beast.

        watch the vid. I’ll try to be a bit more explicit so that confusion doesnt arise.

    • Posted Jan 22, 2012 at 8:25 PM | Permalink

      not mutually exclusive

    • Geoff Sherrington
      Posted Jan 27, 2012 at 5:07 AM | Permalink

      This video delivered by Held of the Postdam Institute is one of the most blatant examples of anti-science I have had the misfortune to follow. The first eror is the unquestioned assumption that CO2 and other GHG are going to cause damage in the future through a rise of global temperatures. As usual, no new evidence is adduced. Then there are several methods suggested as to how to qauntify cost, benefit, risk, uncertainty, tipping points (do they exist in this field?) and time frame. Each of these is slyly combined to give a scenario that leaves early adoption of the precautionary principle as a top option. Then for cream on the cake, there are illustrations of totured economics that purport to show that the early adoption of massive alternative energy will give a more rapid – and positive – payback for those wise enough to follow the path. The path is to make money, for it is conceded that the proposed power changes will do bugger all to GHG effects for decades.

      Unfortunately, elsewhere, there are groups like electric power authorities with teams working 24/7 to maximise investment opportunities. These are likely to be close to optimum because the demonstration of realised profit now is a powerful motivator.

      OTOH, the pie in the sky modelling at Potsdam does not follow the same course that power authorities would follow if unhindered. Instead, we are told that the precautionary principle should be the main God to guide us, then profit in the bank will follow.

      The precautionary principle, as many examples have shown, is usually a way to disguise unfinished work especially when the outcome starts to diverge from the salivating response that was expected.

      Unfortunate incidents like the failure to find the predicted tropical tropospheric hot spot, Trenberth’s missing heat, the almost constant global temperature for the last 2 decades, the constancy of recent ocean heat content, Envirosat and Topex 2 showing ocean levels are now falling – inconveniences such as these are glossed over, and not adequately taken into the PIK economic modelling as described by Held. To him, a sensitivity analysis seems to end with a recommendation on when to invoke the precautionary principle, the sooner the better, wink, wink, because the climate sure as hell ain’t following doomsday obediently enough to scare the peasants and the longer the P.P. is delayed, the more the climate has a chance to diverge from models.

      It is shameful that educated economists are taken in by this tripe. Held does not even seem to recognise that the large scale adoption of wind and solar cannot be done without parallel expansion of spinning reserve, such as gas turbine generation. He’s one of the old school who thought that enough windmills would almost always turn enough on average to negate the need for base load. That’s the type of major equation that PIK should be examining, not pseudo science mixed with economics dressed up in a way to con bankers and insurers.

      Now, Germany is in deep economic trouble. It is a 50:50 as to whether it will go into deep recession and take a lot of Europe with it. It would be trite to say that this has happened because the German economic planners believe the PIK muddle and are implementing it. We’ve seen Angela Merkel wipe nuclear off the slate. That could be the single, dominant move that plunges Europe into recession. I say, let’s sit back and see and not offer them a cent in bailouts.

      ROTFL when you hear the dictum “Uncertainty Under learning”, an aphorism for “We reserve the right to change future figures without explanation or apology when things fail” – as they will. Stupid PIKS.

  28. Eric Steig
    Posted Jan 22, 2012 at 5:44 PM | Permalink

    Some people are never satisfied. The data I sent you are the data I would have sent to Neukom, which are the data you asked me for. You appear to be comparing apples with oranges. The data you link to is a different data set– same core, but sampled at much lower resolution.

    steve: I expressed my appreciation for your placing the data online. I object to you saying that “the relevant data were archived many years ago”, which appears to be not the case. This statement was made with the apparent intent of showing some sort of omission on my part, which was not in fact the case. All you needed to do was to place the data online (as you did), together with a small apology for your previous oversight in failing to archive it in the past. There wasn’t any need for claiming that the data had been “archived many years ago”. It wasn’t.

    • Posted Jan 22, 2012 at 6:59 PM | Permalink

      Re: Eric Steig (Jan 22 17:44), You missed the pea Steve. This is a very carefully worded reply. He didn’t say the data requested was already archived years ago, he said the “relevant” data was previously archived. Thus implying:

      1. While replying to your request and demonstrating cooperation, your annoying little request is irrelevant, but he showers you with largesse anyway showing how scientifically cooperative he is.

      2. Diverting attention away from his lack of previous archiving.

    • Posted Jan 22, 2012 at 7:59 PM | Permalink

      “would have sent”

      1. would have sent but didnt
      2. did send, but cant recall exactly

    • Neil Fisher
      Posted Jan 22, 2012 at 8:58 PM | Permalink

      Some people are never satisfied.

      Indeed, but I hardly think that is the case here. Please understand that what SM is attempting is an audit of the various papers he posts about. Surely you will have noticed the attention to detail, the apologies where they are due, and the corrections where they are required (most in plain sight using strike-out).

      I don’t believe that SM in particular or CA readers in general (there will be exceptions) expect perfection, but we do expect honesty – if have have made an honest mistake, or had intention to post data but for any reason did not, there is (or at least should be) no problem with admitting said error/lapse here. Indeed, you would gain great credibility here by doing just that (if you should need to).

      I am sure there are many others here who would join me in saying “please stay engaged” with us. While I certainly understand defensiveness given the history of animosity, please leave such baggage “at the door” – if you are offended by any comments, ignore them as they likely do not represent the majority opinion here. As a word of advise: please read SM’s requests carefully, as he is very specific about what he wants.

  29. Eric Steig
    Posted Jan 22, 2012 at 8:28 PM | Permalink

    Steve: sigh… damned if I do, damned if I don’t. Your ability to find fault, feel insult, and assume mallintent astounds me. While you appear to be correct that one of the ten Siple Dome data sets we produced more than ten years ago was not archived along with everything else more than ten years ago, this is hardly ‘not archiving’ as you characterize it.

    You write that:

    “One cannot help but think that the data set would have been promptly published if it had HS shape and, ergo, my prediction is that, if and when, the data ever is made “available”, it will not have a Hockey Stick shape.”

    This is ridiculous: with all the data available, even if there were a ‘hockey stick’ in this one record, it would have had to be compared with all the other data that were available from Siple Dome. Furthermore, your prediction is wrong: the highest values appear to be in the most recent part of the record. Sure looks like a hockey stick to me. Why on earth would anyone have tried to keep this a secret?

    One cannot help but think that you are not interested in the facts here, only in furthering your (I will admit, very effective) campaign at casting doubt on the integrity of scientists.

    • Steve McIntyre
      Posted Jan 22, 2012 at 9:35 PM | Permalink

      Re: Eric Steig (Jan 22 20:28),

      OK, there are a couple of layers of points here.

      I took issue with the assertion that the delD data set had been archived years ago. It hadn’t. You now concede that I was correct on this point. You say that your failure to archive the delD “is hardly ‘not archiving'”. I disagree. Not archiving the delD data is ‘not archiving’ the delD data.

      You objected to my statement:

      “One cannot help but think that the data set would have been promptly published if it had HS shape and, ergo, my prediction is that, if and when, the data ever is made “available”, it will not have a Hockey Stick shape.”

      A couple of comments.

      As it happens, my prediction has proved correct. The delD data from Siple Dome did not have a HS shape.

      This implied criticism originates in prior incidents involving other scientists. Several incidents in particular: Lonnie Thompson’s failure to publish the Bona-Churchill isotope results – my prediction was that Bona-Churchill would not have elevated 20th century O18 (or else it would have been published promptly.) My prediction was correct. Similarly with Linah Ababneh’s bristlecones from Sheep Mt – which contradicted the very HS-shaped Graybill bristlecones that are the active ingredient in the MBH (and several other) HS-shaped temperature reconstructions. And, of course, the substitution of Yamal and failure to update Polar Urals with additional data.

      You observed that you did archive low-resolution O18 data from core A and high-resolution data from cores and argue that this mitigates your failure to archive the delD data. I agree that this constitutes a partial mitigation, but only a partial mitigation.

      There is a shortage of annually resolved proxies reaching back to the medieval period. Although the delD record from Siple Dome has been theoretically available for some time, it hasn’t been mentioned in multiproxy studies prior to Neukom (including Mann et al 2008). Nor was it mentioned in the 2006 NAS panel review of proxies nor in the 2007 IPCC report. I think that you should simply concede that you should have archived this data rather than trying to justify its non-archiving.

      I remain of the view that the delD data would have been prominently published and archived if it had had A HS-shape. I agree that it’s unfortunate that this sort of suspicion arises. At this point, unfortunately, the only way to overcome this sort of suspicion is for authors to be very proactive in reporting “negative” results.

      If you wish to contribute to overcoming this sort of suspicion, it would have been worthwhile for you to speak out against conduct like East Anglia refusing to disclose their Yamal-Urals regional chronology or Thompson’s serial non-archiving.

      Again, I appreciate your finally archiving the delD data and do not wish this point to be lost.

    • Craig Loehle
      Posted Jan 22, 2012 at 10:41 PM | Permalink

      Eric: reminder that the topic of this post is the inability to get the data underlying the Neucom paper. To say that the restrictions listed at the site are “old” and no longer apply implies insider knowledge. Further, Neucom refused to provide them. Sure looks like unavailable data to an observer.

    • Salamano
      Posted Jan 23, 2012 at 7:08 AM | Permalink

      At the very least. I think Eric Steig should be applauded for coming over here and contributing to the story, because it has indeed advanced the discussion.

      Echo-chamber-itis has doomed many-a-blog comments section over the years, and it’s nice to see SM’s well-researched and cited commentary encouraging (or inciting) a meeting of the minds. It’s also nice to see that various perspectives are allowed their full say without being edited or sniped.

      • Steve McIntyre
        Posted Jan 23, 2012 at 9:28 AM | Permalink

        it’s too bad that Mann and Jones declared a fatwa against participation at Climate Audit by climate scientists. It would have been a good idea for them to defend themselves against criticism rather than sulk and gossip among themselves by email.

        • BillC
          Posted Jan 23, 2012 at 10:26 AM | Permalink

          Steve or Steve,

          If you or one of your readers know it will be quicker than me looking it up.

          Is a map of the proxies used in the HS available? Even better is there somewhere one can look at maps of proxies used in multiple paleo studies?


        • steven mosher
          Posted Jan 23, 2012 at 6:36 PM | Permalink


  30. Punksta
    Posted Jan 23, 2012 at 2:24 AM | Permalink

    Some background on the PAGES LOTRED-SA initiative


    Scientific goals are:

    (i) to collate, maintain and share a common state-of-the-art protected data base (for contributors only) with the available multi-proxy data sets

    (Emphasis added). Yet again, a cult of secrecy in climate science.

    and from

    … to use the Mann et al. (1998), Luterbacher et al. (2004) and Moberg et al. (2005) methodologies …..

  31. Punksta
    Posted Jan 23, 2012 at 3:07 AM | Permalink

    As regards Steve’s predictions about the chances of non-HS data ever being archived, the general point is that we need to know not only what people are telling us, but what they are deliberately not telling us.

  32. WB
    Posted Jan 23, 2012 at 4:34 AM | Permalink

    Eric Steig, SM’s “ability to find fault…astounds me.” Mate, that says so much about you, I wonder if you even realise.

  33. Wayne
    Posted Jan 23, 2012 at 9:31 AM | Permalink

    @Steve I really do think that a different tone is called for when someone is a guest here. Steig came here — unlike the vast majority of his peers — and interacted, including posting data on his website. Your tone is still accusatory, which I think is inappropriate.

    Yes, hold his feet to the fire if he says data’s in there and it’s not. Yes, point out that the authors of the paper in question refused to share data so Steig’s assertion that that’s an outdated restriction need to be followed up on. Yes, point out if the archived data is not the same as previous versions of the same data.

    BUT please pull back on the hard-hitting tone. I wouldn’t be surprised if Steig never comes back again, and that’s a loss, in my opinion. It simply strengthens The Team when people see that attempting to parlay with “the great unwashed masses” is met with disdain.

    As it is, you’ve won a battle but increased the other side’s resolve.

    • Steve McIntyre
      Posted Jan 23, 2012 at 11:52 AM | Permalink

      Re: Wayne (Jan 23 09:31),

      There’s a bit of history with Steig, who obstructed publication of O’Donnell et al 2010, and who’s made a variety of scurrilous accusations about me at realclimate. Nonetheless, considering the present exchange de novo.

      Reader Fitzr had implied that I’d been overlooked obvious data sources. I had responded that the data to which fitzr had pointed was related to the data provided to Neukom, but wasnt the same data. (BTW given the existence of different data versions, I think that it is important to confirm the version actually used, but that’s a different issue.)

      Steig’s opening comment here was:

      As one of your more adept readers discovered, the relevant data were archived many years ago — including both depth and age.

      In my opinion, Steig was being snarky here – endorsing the “adept” fitzr as compared the non-adept who had been unable to discern the data provided to Neukom in the archives pointed to by fitzr. The problem – as Steig later conceded – was that this statement was untrue. The data provided to Neukom was different. None of the archived data was delD and the O18 data was a different version. I pointed this out.

      Steig’s statement about the prior availability of the data was untrue. Re-appraising the situation in light of your comments, the most generous interpretation of Steig’s erroneous statement is that his assertion about prior availability of data made carelessly, with no attempt to check whether it was correct or not. It’s unfortunate that Steig should venture into this sort of exchange with such lack of care about facts, but I agree that this is a possible interpretation and I will amend some of my comments accordingly.

  34. geronimo
    Posted Jan 23, 2012 at 10:47 AM | Permalink

    “One cannot help but think that you are not interested in the facts here, only in furthering your (I will admit, very effective) campaign at casting doubt on the integrity of scientists.”

    Eric, the stakes are very high here and the integrity of scientists is of the utmost importance because your work is being used by environmentalists to drive a green agenda which involves swingeing taxes on energy, and will almost certainly lead to fuel poverty here in the UK. It doesn’t help that the people you associate with refer to “the cause” when describing the intent of their scientific work. Publishing papers and then refusing access to the data and methodology is about as unscientific as you can get, and needs no comment from Steve Mc for lots of us to draw our own conclusions about the integrity of such scientists. Some climate scientists appear to be scouring the records to find data that supports the “cause”, and hiding, or ignoring data that doesn’t, they don’t need help having people doubt their integrity.

    It’s good that you’ve archived the data, and it’s good that you’ve come over here to put your case. It would have been great if you could have joined in the discussion we had here on Steig et al 2009. there were lots of issues we could have addressed with you on this blog, and who knows? maybe you could have set us straight.

  35. PaulW
    Posted Jan 24, 2012 at 8:58 AM | Permalink

    Re: Raphael Neukom claim that data had been provided to him for use “within the PAGES LOTRED-SA initiative only“,

    I have been assured by a Mr Chandrasekhar of PAGES, that they have a completely open policy regarding data.

    Which presumably means Raphael Neukom is subverting his agreement with them in some way.

    Steve: why don’t you ask Mr Chandrasekhar to inquire into the matter?

  36. Kenneth Fritsch
    Posted Jan 31, 2012 at 12:28 PM | Permalink

    I missed UC’s videos the first time through – making points with points. Good stuff.

  37. Kenneth Fritsch
    Posted Jan 31, 2012 at 12:57 PM | Permalink

    “If we accept, for sake of argument, these findings, one thing becomes clear. Paleo is a good investment.
    Paleo data and analysis is critically important. Its worth trillions. As it stands, we under invest in
    Paleo relative to the ROI. Given its importance ( accepting their argument ) we could reasonably expect
    this data and the methods should be subject to the most exacting standards we have.”

    Sounds like a marketing campaign to me, Mosher. I would judge that looking more critically at reconstructions would simply begin with looking at the divergence problem in dendro and non dendro reconstructions – and without a major cost to bring the initial proxy sites up to date. A second phase would be to do reconstructions using a prior criteria, based on reasonable science, for selecting proxies and then reporting the results without any posterior eliminations. And of course this assumes the methodology limitations of doing reconstructions are well acknowledged.

    I do not judge that doing paleo better is a so much a matter of funds but rather of a lack of unbiased approaches to the analysis. I suspect that funds for paleo would go to the currently prominent scientists in the field who would in turn limit the studies to attempts to show that divergence was of an anthropogenic source.

    By the way with regards to that video you linked in the post: I believe that climate models have a better chance of getting it right eventually than economic models. Climate models can be based on physics while economic models are based on assumptions that attempt to make human actions as predictable as established physics.

3 Trackbacks

  1. […] example of why so many people are skeptical of mainstream climate  science: “Neukom and the Steig Over/Under“, Steve McIntyre, 19 January 2012 — It’s a sad […]

  2. […] McIntyre recently accused Eric Steig of suppressing ‘inconvenient’ results from an ice core record from Siple Dome (Antarctica). […]

  3. By Steig’s “Hockey Stick” « Climate Audit on May 12, 2012 at 3:57 PM

    […] complaint arose from a CA post earlier this year, in which I criticised Neukom for their failure to archive proxy data in a study […]

%d bloggers like this: