Global Update

I’ve finished processing the southern hemisphere GHCN data, and computed the temperature according to the simple procedure for the entire globe.

The results will probably be no surprise, since several others have replicated my results already, with the same outcome. Here’s the temperature history (annual averages) for pre-cutoff stations only, compared to that for post-cutoff stations only:

And here’s the difference between the two. Note that the post-cutoff stations (the ones NOT omitted) show LESS warming than the pre-cutoff stations (the ones that dropped out):

As stated before, contrary to the claims of D’Aleo and Watts, station dropout did NOT introduce a warming trend. If anything, it introduced a cooling trend.


27 responses to “Global Update

  1. Will Watts wail with wondrous woe ?

    Or will he ignore this and move onto his next piece of dodgy deception ?

    I wonder…

    • I don’t. Oh, wait, was that a rhetorical question?

      I used to respect Watts. I used to think he was someone who was working to improve the science. Then I learned enough climate science and statistics, processed the data myself enough times, to realize that he was either a) ignorant of the science and data or b) he was lying.

      Given Watts’ response (or lack thereof) to the Menne paper, Tamino’s analysis, plus a whole slew of bad arguments and wrong science that Watts’ acolytes have made at my blog over the last few years, the preponderance of evidence at this point is that option b) is the correct one.

  2. Watts’ response: “But, but, he’s not using his real name! He didn’t post his code so he’s not doing science! (walks away triumphantly).”

    • Although I generally believe that professionalism in science is when we publicly stand by our name and credibility, we can be fairly certain that this will be Anthony Watts’ response. Refuse to rebut the substance of the post – go directly to ad hominem.

  3. In case you missed it, Watt’s source, E. M. Smith, has already started backpedaling at Lucia’s blog:

    ” I think there is clearly a pattern of bias in the data against high and cold places. I’ve demonstrated that. I take no position on motivation (malice, stupidity, ?) as I can’t see inside folks minds. I SPECULATE that this would have a warming bias on the data products, especially of things like GIStemp where anomalies are calculated “Basket A to different Basket B” as I’ve run a benchmark that shows such change can leak through to the anomaly maps. I’m WORKING ON proof that it does, or does not”

    So Smith discovered that some high latitude areas are undersampled and only “speculated” about the warming bias! How convenient!

    And now back to Watts & D’Aleo version:

    “It can be shown that they systematically and purposefully, country by country, removed higher-latitude, higher-altitude and rural locations, all of which had a tendency to be cooler.”

    “The thermometers were marched towards the tropics, the sea, and airports near bigger cities. These data were then used to determine the global average temperature and to initialize climate models. Interestingly, the very same stations that have been deleted from the world climate network were retained for computing the average-temperature base periods, further increasing the bias towards overstatement of warming by NOAA.”

    These guys are dishonest, stupid, or both. And better educated septics didn’t have the balls to tell Anthony that he was flatly wrong.

    • These data were then used to determine the global average temperature and to initialize climate models.

      Clearly they have no clue how climate models are spun up. Not surprised.

  4. And in the meantime climate skeptic Roy Spencer keeps running into record global temps: Feb 2010 2nd hottest in UAH satellite measurements

  5. Tamino,

    Well done! My results using the CAM method look quite similar:

  6. “These guys are dishonest, stupid, or both. And better educated septics didn’t have the balls to tell Anthony that he was flatly wrong.”

    That’s despite Willis Eschenbach’s tirade at Judith Curry in which he claims “climate scientists need to police their own backyard”

  7. Wattsgate? :p

    [Response: Wattergate.]


  8. Kees van der Leun,

    keep a copy of that chart… last month it showed Jan.10 as 0.723C anom, but that spike seems to have gone. in the picture. What I downloaded last month:

    2009 12 0,280 0,318 0,242 0,503 30 0,259 0,305 0,214 0,2 365
    2010 1 0,723 0,318 0,242 0,503 30 0,259 0,305 0,214 0,2 365

    What his file shows presently

    2009 12 0.288 0.329 0.246 0.510 31. 0.260 0.306 0.214 0.201 365.
    2010 1 0.724 0.841 0.607 0.757 31. 0.296 0.339 0.252 0.269 365.

    No February value yet.

    • hmm, new version loosing a whole 0.1C and some wordspiel, the venerable Wattus being reffed as the person to point out the peculiarity?

      • Roy Spencer says that their adjustments to the raw data have changed to reduce seasonal variability (apparently the RSS set had substantially less). I don’t know of any physical reason for this change in treatment of the data but it doesn’t change the overall UAH trend. You have to look at the newest post (not just the “Latest Temperature” page to find comparisons between the old/new versions and his explanation.

      • Does the seasonal adjustment address Tamino’s point from a while back?

      • Yes, and so UAH 5.3 creeps further up on what RSS has been showing since v 3.2. Latter published

        Jan. 0.640+
        Feb 0.588+

        RSS now shows:

        Jan 0.630
        Feb 0.610

        Attribution of why the change was made, well GFB, if that would go to someone in the It’s True and verified science camp.

  9. I loved it when Spencer and Christy thank AW for bringing this to their attention…

    I copied the comments from WUWT where I pointed out this error to Anthony Watts, and he “discovered” this trend. I linked to Deep Climate.

    The comments included a comment from Deep Climate giving proper credit for discovering the seasonal UAH trend.

    I can’t put it all in a comment, but if you pop me an email, I will send you a text file containing the comments from last July. It is incredibly amusing, yet discouraging reading. AW went on and on about the evils of adjusting reported data later on.

  10. He does seem to keep correcting his data, and January did go down in the latest round. January and February still very high, though.
    For the record: Below is the table I find on his homepage right now, including February data.

    2009 1 0.213 0.418 0.009 -0.119
    2009 2 0.220 0.557 -0.117 -0.091
    2009 3 0.174 0.335 0.013 -0.198
    2009 4 0.135 0.290 -0.020 -0.013
    2009 5 0.102 0.109 0.094 -0.112
    2009 6 0.022 -0.039 0.084 0.074
    2009 7 0.414 0.188 0.640 0.479
    2009 8 0.245 0.243 0.247 0.426
    2009 9 0.502 0.571 0.433 0.596
    2009 10 0.353 0.295 0.410 0.374
    2009 11 0.504 0.443 0.565 0.482
    2009 12 0.262 0.331 0.190 0.482
    2010 1 0.630 0.809 0.451 0.677
    2010 2 0.613 0.720 0.506 0.789

    • Ray Ladbury

      I don’t think we need to attribute nefarious motives to the corrections. Satellite data are VERY difficult to interpret and require lots of corrections. For one thing, the orbit is changing continually for any satellite in low-Earth orbit (LEO).

      This is one reason why I tend to believe the terrestrial record more–that and the longer baseline.

      • Deepclimate explored some of the strange seasonal anomalies in the UAH product a few months ago. It looks like UAH is correcting that problem since Jan/Feb were showing strangely high anomalies every year.

        Here’s Deepclimate’s second post on the subject.

      • I got interested in the subject because of a couple of posts that Tamino had done previously on the severe annual cycle in UAH (credit where credit is due).

        I may have been the first to show that the recent annual cycle led to a correspondingly large divergence in trend slopes by month (which showed about 0.12C/decade difference between February peak trend and May trough). For me and others, I think that made the problem more understandable in practical terms.

        I had a reasonably civil correspondence with John Christy about it (after Watts brought it to his attention). Christy had just written about the annual cycle last July (in the UAH readme) and acknowledged it needed to be fixed.

        I’d echo the sentiment that inter-satellite calibration and other adjustments are extremely difficult, and I’d still consider the surface record more reliable and less uncertain than the satellite-derived tropospheric record, whether UAH or RSS.

  11. A call out is just a little “The Wire” *grins*

    Posted a link to the GHCN post on the Guardian as they had a comment on double standards. Thought this nicely dovetails into that comment piece, hope the mods dont delete it.

  12. Sorry OT but didn’t know where to put this one.

    Looking at the current volcanic activity in Iceland I was drawn to my bookshelf in order to get reacquainted with the literature on the Laki (Skaftar Fires) eruption in 1783. I’m glad, because what I found suggests that there is some potential for an interesting summer [and winter] ahead.

    I quote from Francis & Oppenheimer (2004) Volcanoes

    An abnormal temperature decline began in Autumn 1783, hitting rock bottom between December and February 1784, when th elowest ever winter average temperature for this region [Eastern US] was recorded ; 4.8C below the 225 year mean…The bitter winter followed an uncommonly hot summer for various parts of the northern hemisphere. For instance, July 1783 is the warmest on record (at 18.8C) in central England except for 1983 (the average July monthly temp for all years 1659-2001 in this record is 15.9C) [NB. not sure that this still stands given the book’s publication date]. What is more, the heat wave across Europe coincided with fumigation of the lower atmosphere by Laki’s aerosol. The subsequent picture is less clear. Thordarson and Self found that the European summers of 1784, 1785 and 1786 were colder than normal, but tree-ring chronologies record a warm summer in 1784.

    Needless to say, I’m watching the migration of the current eruption fissure quite intently.

    From a different perspective; reading of the health and environmental impacts of the 1783 Laki eruption also makes for some sobering reflection

  13. NASA say that eruptions need to be close to the equator and also emit huge quantities of SO2 in order for them to have a significant impact on climate.

    Laki was unusual in that it wasn’t a huge quantity of SO2, it was a VAST quantity of SO2 over many months. One (unreliable) source puts it at “equivalent to a Mount Pinatubo-1991 eruption every three days” – for 8 months!

    If Eyjafjallajökull was emitting anything remotely close to this, then I hope we would have heard about it already. That sort of quantity is just plain toxic over a very wide area.

  14. Tamino, it’s been a while since you posted this, but did you compare raw v adjustments for global, too – as you did with NH temps? What was the result?

    [Response: I did. Just as for the NH, for the globe the adjusted data show less warming than the raw, especially recently. I’ll be sending the paper describing the whole thing to a few readers for comment soon. I’ll also post here when it’s ready for release.]

  15. Would it be over the line to suggest WTFgate?

    Er, yes, it probably would.

    Carry on.