Skip to content

AdjustmentGate–South America

April 3, 2012

By Paul Homewood



                                   Before                                                                                      After

First it was the Arctic, then Australia, now it’s "Invent The Incline” in South America. It seems GHCN have been up to their tricks again.


However, as I am accused of not showing adjustments that produce a cooling trend, I can report that I have actually found a couple; Jujuy in Argentina and this one in the town of Arica, Chile! Take a good look at it. These adjustments seem to be like Sheffield buses – you wait ages for one to come along, then have to wait even longer for the next one! (There is also the odd case of San Luis where GHCN have adjusted the 1940 temperature up, but also adjusted the 1939 down, effectively cancelling out – temperatures have also been adjusted downwards from 1945 to 1973).




Don’t get too excited though, because it appears GISS have forgotten to make their homogeneity adjustment for UHI for them, despite the fact they have populations of 124000 and 88000 respectively.(More on this later).

As usual, I have gone back to 1940 to compare the unadjusted and adjusted temperatures. (Temperatures for 2010 are the same in both versions). Appendix A shows the results for all GHCN stations with records back to 1940. Out of 37 stations, 14 have past temperatures adjusted down, 20 remain unaltered and only 3 are adjusted up, giving, as a crude average, an artificial warming trend of 0.27C. Add this to the adjustments in the Arctic and Australia of 0.70C and 0.30C respectively and a pattern seems to be building up.

Finally, I have listed populations and indicated which are airport sites. As you can see, there are only four rural sites, i.e. less than 10,000, and two of these are based at airports. Do these stations show a different trend to the rest and have the urban sites been adequately adjusted for their UHI effect? We’ll take a closer look at this tomorrow.


                                                                                     APPENDIX A

Station Population
x 1000
Airport? Original Mean
GHCN Adjusted
Mean Temp
Salta 260 Y 16.98 15.88 -1.10
Santiago 148   20.26 20.26 NIL
Posadas 140 Y 20.86 20.86 NIL
La Rioja 67 Y 18.95 18.95 NIL
Catamarca 88 Y 19.40 19.40 NIL
Ceres 10 Y 18.22 18.22 NIL
San Juan 298 Y 16.63 16.63 NIL
Pilar <10   16.76 16.76 NIL
Parana 160 Y 17.57 17.28 -0.29
Rio Cuarto 110   15.49 15.49 NIL
Jujuy 124 Y 17.92 18.22 0.30
Junin 42 Y 16.28 16.28 NIL
Rivadavia <10   22.93 22.83 -0.10
Santa Rosa 52 Y 15.03 14.64 -0.39
Dolores 20 Y 14.80 14.80 NIL
Mar del Plata 407   13.46 13.36 -0.10
Bahia 321 Y 14.78 14.78 NIL
Bariloche 48 Y 8.84 8.24 -0.60
San Luis 71 Y 15.63 16.23 0.60
San Antonio <10 Y 15.23 15.23 NIL
Esquel 15 Y 9.20 8.30 -0.90
Trelew 52 Y 13.53 13.53 NIL
Comodoro 97   12.78 12.78 NIL
Rio Gallegos 43 Y 6.68 6.68 NIL
Ushuaia 11 Y 5.79 5.79 NIL
La Paz 635   10.04 6.64 -3.40
Manaus 613   26.71 26.29 -0.42
Quixeramobin <10   26.99 25.99 -1.00
Salvador 1496   24.93 24.93 NIL
Cuiaba 167   25.42 25.42 NIL
Sao Paulo 7034   18.26 18.26 NIL
Curitiba 844   17.00 17.00 NIL
Arica 88 Y 19.11 20.01 0.90
Pudahuel 3615 Y 14.55 13.75 -0.80
Punta Arenas <10 Y 6.85 6.05 -0.80
Isla Juan <10   15.51 14.91 -0.60
Asuncion 388 Y 23.88 22.48 -1.40
AVERAGE         -0.27



For the sake of completeness, I have added to the original list a few stations which don’t have complete records back to 1940.

5 Comments leave one →
  1. Paul Matthews permalink
    April 11, 2012 1:43 pm

    I have been looking through the GHCN data files to check your numbers, and they agree pretty well. While checking some of the slight discrepancies I found something interesting – the GHCN adjustments seem to be continually changing, in an apparently random and undocumented way.

    For example, for Mar del Plata, in your table you have a 1940 cooling of -0.1.
    In the file I have from March, there was a cooling of -0.23, but in the latest data there is no 1940 cooling at all, see

    For Santiago, there is NIL adjustment in your table and that agrees with the March file I have. But In January 2012, the Santiago 1940 adjustment was -0.73, and now it is at -0.59, see

    San Juan had a 1940 adjustment of 0.42 in GHCNV3.0, Nov 2011, then 0 in GHCNV3.1 Nov 2011, then 0.48 in Jan 2012 and then back to 0 again.

    For a station called La Quiaca, the adjusted temperature in Jan 1940 has evolved as follows:
    Nov 2011 12.40
    Nov 2011 10.88
    Jan 2012 11.65
    Jan 2012 12.40
    Mar 2012 -99.99
    Apr 2012 12.40
    (the -99.99 indicates that in March this year the GHCN algorithm decided to delete the 1940 data)

    The GHCN changelog file only mentions changes to the system in Nov 2011 and Feb 2012, but it’s clear that in fact the adjustments are changing virtually every week. For example between the April 2nd and April 11th files, 1940 data changed for Jujuy, Rivadavia, Santiago, Ceres, Parana, Mar del Plata, Bariloche and Esquel.

    • April 11, 2012 5:03 pm

      Just checked GISS – not changed there yet. But the latest GHCN version is dated today, so GISS shouls pick it up on their update at the end of the month. I’ll check again then.

      It sounds as if the algorithm is in charge of the asylum at the moment.

    • Paul Matthews permalink
      April 12, 2012 5:26 pm

      Looking back at your previous posts, I see you had already noted that the adjustments were being adjusted in your March 11th post.

      The version currently up at the GISS site matches the GHCN data I downloaded in March 2012, in which La Quiaca doesn’t have data for 1940 (that’s probably why it doesnt appear in your table). In the April data the pre-1940 data for La Quiaca changes by about 1.6 degrees!


  1. Vijesti o globalnom zaglupljavanju (09/04/2012) « Nedjeljni Komentar
  2. This Isn’t About The Climate | suyts space

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

%d bloggers like this: