Comparison Temperatures

Lets take a little look at a few parts of the world as seen in WolframAlpha vs how GIStemp sees them.

The WolframAlpha site is here while NASA GISS is at:

New Zealand

Christchurch NZ Temp WolframAlpha

Christchurch NZ Temp WolframAlpha

Auckland NZ Temp WolframAlpha

Auckland NZ Temp WolframAlpha

Wellington NZ Temp WolframAlpha

Wellington NZ Temp WolframAlpha

GISS Christchurch Homog Feb2011

GISS Christchurch Homog Feb2011

Well, GISS certain sees the world a bit differently… One really does have to wonder how they turn cooling into warming. But an adjustment here, and some homogenizing there… it all works out…

How About Germany?

Berlin Germany Temp WolframAlpha

Berlin Germany Temp WolframAlpha

Oddly, I could not find Berlin in GIStemp… perhaps they’ve lost it? At any rate, here is Munchen:

Munchen Germany GIStemp Feb 2011

Munchen Germany GIStemp Feb 2011

How about the Hawaiian Islands

Bradshaw AAF Hawaii Temp WolframAlpha

Bradshaw AAF Hawaii Temp WolframAlpha

Hilo Hawaii Temp WolframAlpha

Hilo Hawaii Temp WolframAlpha

Honolulu Hawaii Temp WolframAlpha

Honolulu Hawaii Temp WolframAlpha

Kaneohe Bay Marrine Corp Airfield Temp WolframAlpha

Kaneohe Bay Marrine Corp Airfield Temp WolframAlpha

Honolulu Hawaii Airport GIStemp

Honolulu Hawaii Airport GIStemp

Well…. Yes, the Oahu airport does show a bit of warming in WolframAlpha… but the other stations do not. So how does GIStemp turn that into such runaway warming?

A Bit Of Nevada

Center Nevada Temp WolframAlpha

Center Nevada Temp WolframAlpha

Well, I couldn’t find a “center of Nevada” equivalent, but here is Elko as GISS sees it. Then I’ve got the WolframAlpha for Elko after it:

Elko Nevada GISS Feb 2011

Elko Nevada GISS Feb 2011

Elko Nevada WolframAlpha

Elko Nevada WolframAlpha

Well… That’s a bit “different”…

Somehow I think we need a top to bottom end to end brand spanking new global temperature series that has not had all the “usual suspects” monkey with it. Or, to quote Casablanca… “I’m Shocked! Shocked to find that gambling going on in here! … ‘You’re winnings, sir…’ ”

Subscribe to feed

About E.M.Smith

A technical managerial sort interested in things from Stonehenge to computer science. My present "hot buttons' are the mythology of Climate Change and ancient metrology; but things change...
This entry was posted in AGW Science and Background and tagged , . Bookmark the permalink.

22 Responses to Comparison Temperatures

  1. Ken McMurtrie says:

    Pleased to see your continued interest and input.
    Given that the world continues to crush the financial stability of many countries using carbon trading and carbon “pollution” charges, based on the fraudulent claim of CO2 driving global warming, there is an ever increasing need to publicise the issue.
    Are we too late?

  2. Malaga View says:

    I think shocked is the right word…

    Asking myself: Is it for real or just playing to my prejudices?

    WeatherData Source Information

    WeatherData is based on a wide range of sources, with enhancement at Wolfram Research by both human and algorithmic processing. WeatherData is continually maintained with the latest available information, with automatic updating when WeatherData is used inside Mathematica.

    Among current principal sources for WeatherData are:

    Citizen Weather Observer Program. “Citizen Weather Observer Program (CWOP).” 2008. »

    National Oceanic and Atmospheric Administration. “National Weather Service.” 2008. »

    United States National Climatic Data Center. “Global Surface Summary of Day.” 2008. »

    United States National Climatic Data Center. “Integrated Surface Database.” 2008. »

  3. Malaga View says:

    One observation is that presentation is so much of the message:

    Graphed Example: Christchurch NZ

    GISS makes it look so dramatic by using a very small temperature range of 2.5º C

    WolframAlpha is far more laid back by using a far larger temperature range of 20º F

    Which is correct?

    Christchurch NZ has a record high of 99º F in 1982 and a record low of 14º F in 1980.
    That gives a maximum temperature range of 85º F in the temperature record.

    So from my perspective the average temperatures for Christchurch NZ should be graphed in their appropriate context which is with a vertical range of 85º F… and the averages would then appear as a very boring straight line with a few very small wiggles… and that is it… nada mas… global warming is a storm in a teacup… it is an misrepresentation of reality… it is an artefact of presentation… and all efforts of the The Team with all their splicing and dicing data manipulation techniques applied to the GISS data still amount to diddly squat.

    Obviously, this is why The Team are so keen on their anomaly graphs… it helps them misrepresent the data in their climate science world of hype and hysteria… but back in the real world Christchurch NZ still experiences weather than bounces around within their 85º F temperature range.

  4. Malaga View says:

    Ken McMurtrie
    Are we too late?

    I hope I am wrong but I don’t think they really care about whether we are early or late… we are not invited to the party… punto.

  5. For those who have the ability and a peaceful heart for dealing with statistics, that sacred “pamphlet era” science, could get a better perspective of the evolution of temperature through time by considering only, not minimum averaged temperatures, but the actual minimum temperatures.
    Minimum temperature is one of the limits, the other is Maximum temperature, however this one, looks to have been fatally injured by the continuous massaging, making it an unreliable source of data.

  6. Malaga View says:

    @ Adolfo Giurfa
    Minimum temperature is one of the limits, the other is Maximum temperature

    Totally agree… hunting for Global Warming or a New Ice Age in average tempertures is pretty crazy… but then Climate Science is pretty crazy… just like the crazy world of arthur brown

  7. Malaga View says:

    I guess I must have enjoyed a misspent youth

    Guessing because there are times I can’t remember :-)

    And there are times I would prefer to forget :-(

  8. Thanks for educating me about wolframalpha :)
    Three country towns that I know well in Australia all show similar pictures. Try a wolframalpha search for
    “weather trends Echuca-Moama Australia” or
    “weather trends Shepparton Australia” or
    “weather trends Deniliquin Australia”.
    They have zero or negative regressions for temperature vs time.
    Compare these trends with
    “temperature trends Melbourne”
    and you see the Big City UHI effect of +2 degrees per century. The weather observations for Melbourne are taken in the heart of the CBD, see Google maps for 59 Lonsdale St Melbourne VIC 3000 (37.81S 144.97E)

  9. Rob R says:


    The Christchurch temp series caught my eye. There are currently two long-running climate stations in Christchurch. One is Christchurch Gardens (Christchurch Botanical gardens in Hagley Park).

    The “Gardens” site is precisely in the center of the city, an absolute bullseye. It has clearly been subject to a substantial urban heat island effect.

    The other is Christchurch Aero, which is still just outside the city, though development is encroaching. I have looked at the comparison between these two sites (gardens and aero) in detail. The Aero site does not appear to be particularly strongly impacted by UHI effects when compared with rural sites in Canterbury province.

    It appears that GISS has a warming trend but this is grossly exaggerated by the scale on the graph and by having 1951 (a cold year) as a start point.

    The record for ChCh Aero commences in 1953 (NIWA cliflo database). The most recent few years from the official record for ChCh Aero match the GISS graphic nicely but GISS becomes corrupted back through time.For instance in 1960 the Aero temp was 11.3 C, whereas GISS has it at about 11.0 C.

    Prior to 1953 the appropriate splice would be with Wigram Aero, a few km away. Wigram was the NZ Royal Airforce airfield, and was also just ouitside of ChCh (same side ot the city as ChCh Aero) and also doesn’t appear dramatically affected by UHI. In the overlap from 1953 to 1967 Wigram Aero is slightly warmer than ChCh Aero. The Wigram record Wigram commences in about 1938.

    1951 ChCh Gardens 10.8 C
    1951 Wigram Aero 10.8 C
    1951 GISS 10.3 C
    Diff 0.5 C

    Whats up with that? There are no climate stations from the ChCh area that are as cold as 10.3 C during 1953.

    If there is any UHI currently impacting on ChCh Aero, then to correct it the past would need to be warmed (or the present cooled).

    By my eyeballed estimate from the official (NZ) data ChCh Gardens has warmed by about 0.6 to 0.8 C due to UHI since 1950 relative to ChCh Aero.

    So what is it about GISTemp that cools the past relative to the present.

    Perhaps in this case GISS have actually taken the ChCh Gardens temp series, and cooled the present so that it matches ChCh Aero, then carried virtually the same cooling adjustment back in time right through to 1950, but solely on the ChCh Gardens data? That would produce a pretty good match for the GISTemp graphic presented above.

    By the way, the ChCh Gardens site goes back in time unbroken to 1905. It was also used from 1864 to 1880. In the gap it was closed due to NZ Government budget cuts. But it was replaced by another long-lived site at Lincoln about 20 km outside Christchurch. Lincoln has a continuous record from 1881 – 2010 though there were some site moves in the 1980’s/1990’s. So it is possible to splice together a continuous temperature record for Christchurch Gardens/Lincoln from 1864 to 2010 (a project that is on my to-do list). Spliced records of similar length can be obtained for Hokitika, Dunedin, and Nelson. Spliced records can also be constructed for Ashburton, Timaru, Invercargill, Gore and other sites from around 1905 to 1910. (I have about 13 spliced records constructed for South Island localities with up to 10 more waiting to be done). There are lots of good rural records from the South Island that are available for comparison as well.

    The trend in the GISS graphic is clearly exaggerated.

  10. Joseph says:

    What do you enter into the Wolfram search engine to get the charts you posted?

  11. Jason Calley says:

    Joseph, easy deal. Go to Wolframalpha and in the search box enter, for example, “weather honolulu”. Do check to make sure that when the screen for Honolulu comes up, it is actually what you are looking for. If you enter a choice that is unavailable, the search engine will bring up whatever location is as close as it can find.
    The response will be divided into several boxes, one of which will be named “Weather history and forecast”. Immediately to the right of that name will be another box that says: “current week.” That is a drop down box; just choose “all” and it will display the long term (since 1940) trend.

    Hope that helps. :)

  12. E.M.Smith says:


    What Jason said ;-)

    Then I just “right click” the graph and choose the ‘save as’ to get an image I can upload / save / post.

    I’ve run into one or two cases where it reaches several hundred miles to find a ‘live station’ so you do have to watch that. I’ve also had it give two very different graphs but list the same station ID when doing that “long reach” (but not on things ‘close up’) so I think there is some blending of stations when it reaches a long ways, and sometimes it may be reaching a long ways for the data and a different long ways for the identifier… I’ve sent in a ‘suspected bug report’ on the one I could document. Individual and closer up stations seem to check against Wunderground OK. I’ve mostly run into this on isolated islands a long ways from nowhere…

  13. E.M.Smith says:

    @Sandy McClintock:

    Nice set…

    FWIW, that seems to be the pattern. A lot of surrounding small places will show “not much going on” or in a few of them “down trend” then the large urban area with the thermometer surrounded by miles of tarmac shows “warming trend”….

    Looks to me like “Global Warming” has discovered that concrete and tarmac are warmer than grass….

    @Malaga View:

    I think I resemble that remark ;-)


    I started to do this at one point, but the public MIN and MAX data were, er, “sparse” and the presentation is a bit wacked.

    @Rob R:

    What you are experiencing is the “magic” of GIStemp “homogenizing” and “Wrong Way UHI”. It creates a “splice artifact” from homogenizing together different stations (so it would splice the present downtown station onto the older rural station) then it will “UHI adjust” the series. This is done by holding the present constant and adjusting the past. Now you would expect that the past would be warmed up to correct that UHI, but in about 1/2 the cases it cools the past instead. Personally, I think it is “a convenient 1/2 that cools”… but proving it is a bit hard.

    An example is here:

    Where a nice Swiss mountain is used to “correct” Pisa Italy and gets it the wrong way. Basically, this needs to be done for EVERY SINGLE SITE in GIStemp to show what they have got wrong. (Otherwise they just say “that’s just one, the average is good.”)

    The “magic sauce” is scattered all through the GIStemp code, and not in any one spot. There are a lot of places where things that superficially look reasonable are done, but the result is broken. Then the question becomes:

    1) Programming bug?
    2) Bad design?
    3) Data artifacty?
    4) This Behavior Is By Design?
    5) FUBAR?

    and trying to tease that out of 40 year old crappy FORTRAN makes your brain cry for mercy…

  14. Joseph says:

    E.M. and Jason,

    Thanks guys. I will be showing that off to school kids tomorrow (off today for a holiday), and I just know they will love it.

  15. Keith Hill says:

    The graphs of the 66 ‘greenfields’ surface Stations John Daly lists in the Appendix of his 10th May 2000 “Whats Wrong With The Surface Record” are ideal for comparison.

    Central England Composite – 290 years of records.
    Vardo – Norway – 170 years of records
    Valentia – Ireland – 141 years of records

    are good starters on the trail to show anyone there has been no unprecedented Global Warming but it only takes a few minutes to click the 66 stations.

    What John said 10 yrs ago about ‘Station Records and Climate Models’ is just as applicable today. Quote:

    “Pending an independent review of the GISS-CRU surface record (essential given the policy implications), it is valuable nevertheless to look at individual station records, particularly those which are known to be rural, have continuous and consistent data, and are known to be properly supervised.

    The `ideal’ stations are those which have everything – a long-term record, no breaks, scientifically supervised, completely rural (ie. `greenfields’), and set in a climatically strategic location.”

    (Appendix – Station Records with data to 1998 or 1999)

    “Note particularly those stations in the Arctic and Antarctic regions. The greenhouse theory and the models both suggest that the greatest warmings of all will take place in the coldest of places on earth, for reasons connected with the absorption-emission wavelength characteristics of CO2.”

    It highlights the astounding gall of James Hansen when, on the same GISS: Surface Temperature Analysis site in which he makes his outrageous claims such as ‘2010 hottest ever’, one can check the latest graphs of most of the John Daly listed ‘greenfields’ stations and find in nearly every case, temperatures have tumbled over the last 3-4 years.

  16. Keith Hill says:

    Sorry Chiefio. Forgot. Berlin is under Berlin-Tempel on GIStemp and shows a 2.0C fall over the last 3 years in common with the trend of falling temperatures all over the world in the same period.

    Cheers from Tassie where we had snow again on the mountains just behind Hobart the other day!

  17. Jackie Healy rae says:

    Valentia Ireland

    Aerial 2005 some encroachment and minor UHI,445707,578830,6,0

    Map 2011,445707,578830,6,2

    Aerial 2000,445707,578830,6,4

    Aerial 1995 …even more rural,445707,578830,6,5

    Map circa 1907 even more rural again,445707,578830,6,9

    The station was on an Island when first constructed, co located with the Atlantic Telegraphs constructed there initially in 1858 , it has been at its current location (actually Caherciveen not Valentia Island) since 1892

  18. Jackie Healy Rae says:

    Valentia. Location and aerial map.,445707,578830,6,3

    Not encroached upon dreadfully by Urbanisation,

  19. beng says:

    E.M., I think it would be really interesting to see what the daily max-temp trends are (ignore the min-cold temps). I realize that would be extra work, but Pielke Sr’s work suggests this as the min-temps measured near the ground are easily affected by disruption of the shallow inversion layer and introduce alot of unqualifiable bias. Thermals during the day disrupt any inversion, usually (polar regions might maintain the inversion all the time when sunless).

  20. E.M.Smith says:


    I’d love to do it, but last I looked at the min / max temps I was not happy with the sparcity of the data. There was some discussion that indicated you need to somehow merge them with something else to recover the whole set (i.e. that it’s not just “here is the min” but ‘here is the min for times that it is different” or some such, but I was off to other things by then.

    It is on my “important things to do” list to try again to sort out the min and max data into straight “just the temps” usable series, but I’ve been “busy elsewhere”…

    And yes, what I HAVE seen is that “We’re not getting warmer, we’re getting less cold” in that the tops don’t go up, but the cold excursions get clipped. IMHO, that’s a very good thing, and nothing to be afraid of. I also think it’s largely and artifact of putting the thermometers at airports in recent years (the present GHCN set is over 90% at airports in recent years… measure that tarmac…)

  21. George says:

    “under Berlin-Tempel” is Tempelhof which is in the city proper. Berlin-Schönefeld and Berlin-Tegel are the other two and are in more outlying areas.

    Tegel, back when I lived in Berlin, was pretty much out in the country (such as there was “country” within the wall surrounding West Berlin at the time).

  22. E.M.Smith says:

    As I understand it, the Tempelhof site was shut down when it was transformed from an airport to a shopping mall. Is that not true?

Comments are closed.