AR5 Loves Steig et al 2009

Jeff Id has an excellent post on IPCC AR5 use of the highly flawed Steig et al 2009. Despite Steig’s efforts to block the publication of O’Donnell et al 2010, O2010 shows clearly that whatever is new in Steig et al 2009 is not only incorrect, but an artifact of flawed math and whatever is valid was already known.

As Jeff observes, Steig coauthor Josefino Comiso (who was very uncooperative, to say the least, in providing data underpinning Steig et al 2009) is Coordinating Lead Author of AR5 Chapter 4 on the cryosphere, where Steig et al 2009 is cited, but not O’Donnell et al.

AR5 chapter 10 cites Steig et al 2009 four times, without any citation of O’Donnell et al 2010. Jeff quotes the following from AR5:

It was concluded that the pattern of mean surface temperature trends in both West and East Antarctica are positive for 1957–2006, and this warming trend is difficult to explain without the radiative forcing associated with increasing greenhouse-gas concentrations (Steig et al., 2009).

As readers of CA and tAV are aware, Steig’s methodology smeared warming from the Antarctic Peninsula into other parts of Antarctica. Jeff observes of the IPCC draft (with considerable frustration):

That last sentence is too much for me. The positive trends are very easy to explain – bad math!!

In Chapter 5 (Paleoclimate), Steig et al cited as authority for polar amplification. In this case, they have a placeholder noting O’Donnell et al 2010 as follows:

[Reference needed: O’Donnell et al., ?]).

O’Donnell et al 2010 was the only reference among the 1000 or so citations for this chapter where the authors had not bothered locating volume citation information. It was the only reference returning the term “null” as follows:

O’Donnell, R., N. Lewis, S. McIntyre, and J. Condon: Improved methods for PCA-based reconstructions: case study using the Steig et al. 2009 Antarctic temperature reconstruction. Journal of Climate, 0, null.

One doesn’t get the impression that the IPCC authors had read, let alone assimilated, the findings.

Check out Jeff’s post.


  1. Glacierman
    Posted Dec 13, 2011 at 10:52 AM | Permalink

    Blatant, in your face, gaming of the system for all to see. This underscores how important their efforts to keep certain papers from being published were to the team.

  2. kim
    Posted Dec 13, 2011 at 11:05 AM | Permalink

    Between Scylla and Charybdis we have a blind and heavy hand on the tiller.

  3. Craig Loehle
    Posted Dec 13, 2011 at 11:44 AM | Permalink

    I just want Santa to put in my stocking whatever it is these guys have that they are never wrong. /sarc in case you can’t tell.

  4. Arthur Dent
    Posted Dec 13, 2011 at 11:53 AM | Permalink

    This is just disgraceful have these people no shame. The authors produce a ZOD not having read one of the key papers.

  5. Steeptown
    Posted Dec 13, 2011 at 12:23 PM | Permalink

    Conflict of interest anyone?

  6. Jean S
    Posted Dec 13, 2011 at 1:14 PM | Permalink

    The same chapter (Ch10) loves Santer also! It has an “interesting” way of citing McKitrick et al (2010). From the first read, it appears that you fully support findings of Santer et al (2008) and if there are any small differences, these are resolved by Santer and coauthors (2011a) (in preparation) and Santer and coauthors (2011b) (in preparation)! 😉

    • andy
      Posted Dec 13, 2011 at 2:00 PM | Permalink

      McKitrick might be busy now reviewing the responses…

  7. Hu McCulloch
    Posted Dec 13, 2011 at 1:19 PM | Permalink

    As Jeff observes, Steig coauthor Josefino Comiso (who was very uncooperative, to say the least, in providing data underpinning Steig et al 2009) is Coordinating Lead Author of AR5 Chapter 4 on the cryosphere, where Steig et al 2009 is cited, but not O’Donnell et al.

    Although Comiso did not release the data that went into the AVHRR matrix, Steig to his credit did (after some prodding) release the AVHRR matrix itself, which was sufficient to show the defects of the paper.

    • Layman Lurker
      Posted Dec 13, 2011 at 8:05 PM | Permalink

      The AVHRR matrix might have been sufficient to deconstruct S09, but a discussion of Comiso’s AVHRR cloud masking process and algorithm remains unarchived IIRC. I would argue that this process – which involves the foundation data for S09 – should have itself been submitted and peer reviewed prior to even considering the S09 submission. This is a significant issue which has received very little attention.

      • Phil
        Posted Dec 15, 2011 at 10:26 PM | Permalink

        From: Comiso, J. C. Variability and trends in Antarctic surface temperatures from in situ and satellite infrared measurements. J. Clim. 13, 1674–1696 (2000).

        The trend analysis on station data yielded on the average 0.012 ± 0.008°C yr-1 and -0.008 ± 0.025°C yr-1 for the 45- and 20-yr record, respectively. The latter reasonably agrees with the trend of -0.042 ± 0.067°C yr-1 inferred from the satellite 20-yr record. The 20-yr record length is shown to be about the minimum length required for a meaningful trend analysis study. However, interannual fluctuations of the temperatures are large and the 95% confidence level for the satellite trends ranges from -0.177 to 0.094°C yr-1 for the Antarctic ice sheet. Nevertheless, the observed cooling is intriguing, especially since it is compatible with the observed trend in the sea ice cover. (emphasis added)

        What changed for Steig 2009?

        From the Steig 2009 SI:

        In Comiso (ref. 8), cloud masking was done by a combination of channel differencing and daily differencing, based on the change in observed radiances due to the movement of clouds. Here, we use an additional masking technique in which daily data that differ from the climatological mean by more than some threshold value are assumed to be cloud contaminated and are removed. We used a threshold value of 10°C, which produces the best validation statistics in the reconstruction procedure.(emphasis added)

        The “climatological mean” used for this purpose has not been released and it is unknown how it was calculated. Also unknown is the basis for assuming that certain values are “cloud contaminated.” Thus a slight cooling was turned into a slight warming. Personally, I doubt that any of this data supports a conclusion that there has been statistically significant warming or cooling in Antarctica (except perhaps for the peninsula.) There just does not appear to be enough hard data.

  8. Tom Gray
    Posted Dec 13, 2011 at 2:00 PM | Permalink

    After Durban with its non-agreement and Canada’s formal announcement of Kyoto withdrawal, do you really think that the IPCC has any credibility left of more directly any ability to affect public policy?

    Why would AR5 matter more than any other academic survey paper?

  9. KnR
    Posted Dec 13, 2011 at 2:15 PM | Permalink

    AR5 is just following the pattern laid down , poor science treated as its the ‘word of god ‘ becasue its supports ‘the cause ‘ and comes from the right type of scientists and very mind its actual contents .

    • KnR
      Posted Dec 13, 2011 at 2:52 PM | Permalink

      and comes from the right type of scientists and never mind its actual contents .

  10. DEEBEE
    Posted Dec 13, 2011 at 7:04 PM | Permalink

    The teenager goes juvenile. Infantile here we commme

  11. Manfred
    Posted Dec 13, 2011 at 8:35 PM | Permalink

    NASA Code of Conduct

    Click to access standards_of_conduct.pdf

    NASA Ethics/Standards of Conduct

    NASA Ombudsoffice

3 Trackbacks

  1. […] S points out in a comment on the Steig thread that our findings were completely misrepresented by IPCC chapter 10 (also the […]

  2. […] AR5 Loves Steig et al 2009 […]

  3. […] grande surprise, le 5e rapport du GIEC (en préparation), a inclus la version de Steig et al. et complètement ignoré le travail […]

%d bloggers like this: