the Air Vent

Because the world needs another opinion

Global Temperature Deconstruction – McKitrick

Posted by Jeff Id on August 3, 2010

Watts Up With That has run an article on an important paper by Ross McKitrick covering the major issues of the global temperature record.  The article collects and describes the data and problems as we have discussed so many times here.

While we have covered many of the points here and on other climate blogs, I’m guessing that most who are familiar with the land and ocean quality problems will still learn from this article – I learned some new stuff.   Ross meticulously explains the major datasets, non-independence, urbanization, collection problems, and presents what I believe are the only reasonable conclusions.

One example of the conclusions in the report which I cannot emphasize enough is that when so many of us reproduce the same results from GHCN temperature data, the conclusion is not that “climatology is correct”, just that the methods of averaging the same data don’t cause serious problems.

The similarities across all methods imply that given the input data, the subsequent processing steps do not make much difference to the global average.

Nothing else can be concluded from Zeke, Nick, tAV or others results.  How simple and relaxing is that.

In total, the paper is an honest and fair critique of data quality – nothing more or less. Not skeptic or believer, it’s the kind of thing which makes engineers all warm and fuzzy inside.  Of course there has been plenty of resistance to anything critical of temp records in the eco-RC crowd.  The last section of the report describes some of that reaction, but the data is the data.  We have to accept it here, AGW eco-advocates need to learn the same.  The strongest point in the article is the section on comparison of trends vs economic output.  Ross has published several articles on the topic which summed simply, are a PROOF of measurable urban heat island warming in the record.  This last section will be confusing in that many don’t understand (because they haven’t studied) the math, sometimes I wish I didn’t.

My conclusions don’t change,  before we spend the big bucks, it’s time for some serious thermometer QC.  The argument for elimination of fossil fuel industry by the IPCC is based on the ‘huge’ expense of adaptation, it is of course, the weakest part of the IPCC case.  If warming is even slightly less than argued, the whole industrial suicide ‘solution’ falls apart.  Maybe it’s time that eco’s considered the poor of the world and practiced a little of that precautionary principle.

Anyway, Anthony Watts writes that it is a must read.  If you spend any time seriously considering AGW science, he’s right.

Link to the PDF is here.

#Mann PRoxy Data and Info
if (method==”offline”) {load(“d:/climate/data/mann08/”)
source(“d:/climate/scripts/mann.2008/instrumental.txt”)} else {

18 Responses to “Global Temperature Deconstruction – McKitrick”

  1. kim said

    This is the shoe we’ve all been waiting to drop.

  2. josh said

    I think the “Chimney Brush” graphic of GHCN adjustments should replace the Hockey Stick as the icon of the AGW movement.

  3. GregO said


    “before we spend the big bucks, it’s time for some serious thermometer QC.” Amen brother.

    I’m skimming through McKitrick’s paper now (darn day job!) and more than one thing sticks out but check out “Figure 1-10 Changes (“delta”) in the global average temperature resulting from GHCN adjustments…” and I quote from the paper page 15:

    “There are two notable features of the graph. The first is that the adjustments are mainly negative prior to 1940 and positive up to about 1990, effectively “cooling” the early part of the record and “warming” the later record…” Further; “The second and more obvious feature is the chimney-brush appearance of the graph, indicating a massive increase in the volatility of the adjustments after 1990. The instability dwarfs the size of the century-scale global warming signal, with fluctuations routinely going +/- 0.5 degrees C from one year to the next…” (!!!)

    How can anyone take CAGW seriously when it is clearly shown the data has been adjusted to show the past cooler; the present warmer; and since 1990 the volatility adjustments themselves are noisy enough to dwarf the alleged warming trend value.

    How can this CAGW scam go on?

  4. DaveJR said

    I’m sure I read in one of the threads on Lucia’s that Zeke et al used GHCN data before adjustments. Can anyone familiar with the datasets, and who used what, clear this up?

  5. Jeff Id said

    #4 Yes, most of us used the pre-adjustment data.

  6. Jeff Id said

    There are things done to the pre-adjustment data also. I don’t have time to look up the references though, sorry.

  7. steven Mosher said

    Well, if you see Zeke and My article on this, that was our claim. The methods are NOT the issue.

    Wish some people who whine (like I used to) about Giss1200km would get the exact point.

    the mthod differences are small. INTERESTING but small.

    Ross stuff on the station stats is misleading, however.

    provance stuff is all good. sample bias stuff needs work

  8. Jeff, All,
    it would make very little difference to your analysis if you had used the adjusted data – I think – what I mean is the the adjustments seem to be so finely balanced that the there is little change overall. At least that is what I have been seeing – and it has intrigued me since December, and I have been trying to find ways to quantify it. It is not just the adjustment, but the spatial and temporal balance of the adjustment so that averaged over the globe the differences are very small ~0.1 deg of trend.

    Look at the nightlights radiance adjustment implemented in January.
    It caused very little overall difference, but I had a look at how it affected the data by lattitude and there were reasonable differences in the latitude bands, but they cancelled each other out at global level. Dig down to country level, even grid square location and individual station level and there are very major changes. This is something we all need to look into more.

    Peter O’Neill is having a serious crack at it:

  9. Steve Fitzpatrick said

    Dr. McKitrick has done a good job pointing out the problems with the historical surface data.

    The satellite lower troposphere data is far more robust in coverage and data treatment, and shows a trend about 75% as large as the surface data since 1979. While this does nothing to address the problems with the pre-1979 surface data, it does provide additional evidence that the methodologies used on the post-1979 surface temperature data are indeed very suspect. The difference between the satellite trend and the calculated surface trend is a reasonable estimate of the maximum net influence of the problems described by Dr. McKitrick over the post 1979 period; on the order of +0.15C bias (or 0.05C per decade).

    On the other hand, the satellite data say that there has been real warming since 1979, even if it is only ~75% as large as GISS, Hadley, etc. calculate; lower troposphere warming since 1979 (~0.45C increase) can’t be simply dismissed as an artifact of inadequate data treatment or biases in the surface data.

  10. stan said

    Steve Mosher,

    We are constantly bombarded with appeals to authority. Since climate scientists overwelmingly agree we are supposed to accept our fate …..

    Their credibility is absolutely crucial to this argument. So adjustments which they have messed up are a big deal, even if the impact isn’t huge. I don’t want brain surgery from a doctor who can’t cut properly, even if the botched cut I know about is on my skin and easily stitched.

    This becomes even more significant when all the other incompetence is brought into the picture. Sloppiness abounds (Mann, Jones, IPCC, Briffa, Rahmstorf, Steig, Karl,…. the hits just keep on coming).

  11. kim said

    It’s a house of cards, structurally unsound, but erected against the force of reality by the pull of power and money. Steve had to huff and puff to break the first weak bond, but now light zephyrs detach the crumblilng remains.

  12. kim said

    In other words, it’s an illusion generated by the darker dimensions of post normal science.

  13. Rob R said

    Good stuff from Ross McKitrick. But I suspect there is more to come. This relates to the great dying of thermometers in GHCNv2, part of which I suspect is far from accidental and does not relate to dimunition of the number of real thermometers being monitored by the various national meteorological and climatological organisations.

    For instance in GHCNv2 there are about 6 (or so) onland (rather than offshore islands) temperature records for the South Island of New Zealand, only about 4 of which have records through to 2010. However, within the NZ national climate database (cliflo) managed by NIWA (National Institute of Water and Atmospheric Science) I have found about 55 stations reporting monthly during 2010. Similarly going back in time over several decades the number of stations reporting monthly data in any one year max’s out at about 90.

    So in this single small part of the globe there is a profusion of surface temperature data that does not make it into GHCNv2, and never did in the past either. It is most unlikely that the South island of NZ is unique in this. I know this to be true because the same situation also applies for the North island of NZ.

    I have had a cursory look at the temperature adjustements applied to these individual records to produce GHCNv2Adj and my immediate impression is that whatever process is used it is throwing up results that bear little relationship to reality.

    The fact that GHCNv2 finds only 4 currently reporting records for the South Island is, quite frankly, a very bad joke. Its not much more than 5% of what is readily available from the cliflo database. And its not as if they can’t get the information. It’s available for free. What is more the head guy at NIWA (Dr David Wratt) has been one of the IPCC “team” for years and co-publishes climater research with a whole bunch of those guys. All they need to do is ask for the data and NIWA will likely scramble all available resources to provide it.

    I suspect the numbers of stations reporting to the GHCHv2 database has been emptying out partly because it suits certain scientists for this to be happening. But then I also seem to be becoming steadily more paranoid with respect to my views on the “climate scientist” establishment. Am I justified?

  14. Brian H said

    Rob R;
    Yes, the same situation applies throughout Canada, especially Arctic Canada. Considering that grid size is a huge issue with the veridicality of any model constructed from a sample set, trimming the available sources by 90% or so is not only counter-productive, but mathematical and statistical insanity. Or blatant misrepresentation. Or both.

  15. PhilJourdan said

    I read the paper after following a link on WUWT. As I explained over there, while I do have a background in statistics, I have not used it in quite some time. However the paper is easy to read and understand, probably even for someone with no statistical background! As Jeff Id says, a must read to understand where the data is coming from and what it truly represents!

  16. Andrew said

    Careful examining of the surface record is a very good idea, but land/station data is just the beginning of that (sure to be interesting, on going) story. There has been relatively little interest in the sea surface records and the problems and adequacy or lack thereof of the “solutions”. I hear the Hadley team are going to get to a major adjustment of some of the early and even recent sea surface data soon-but it’s not clear to me that anybody has shown how to do such adjustments correctly.

    Very troubling if you ask me.

    At any rate, McKitrick’s stuff is very good as usual. He has put the economic correlation findings through an enormous gauntlet of statistical tests, and anything that survives that is pretty much as robust as it gets IMAO.

  17. The version of Ross’ paper linked here had some issues. Ross has produced a new version, for reasons mentioned at WUWT. Version 3 is here.

  18. The version of Ross’ paper linked here had issues.Version 3 is here.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: