Mean temperatures rose just as data station numbers fell

Warmists are very fond of saying the correlation between recorded rising temperatures and the increase of CO2 in the atmosphere is strong evidence that Co2 is causing the warming. Well, what’s good for the goose and all that…

Old news, yes. But worth repeating for those who are not familiar with it.  In the graph below (hat tip: Digital Glebe) we see another correlation, this time between recorded rising temperatures and the falling number of data stations that record temperature. So using the warmist approach, it must follow that the reason we are seeing rising temperatures is that there are fewer data stations being used to collect temperature data.

Unless of course CO2 forced an incredibly huge jump in the greenhouse effect at exactly the same time as half of the data stations were removed from the set…

You can read about the graph and its origins here.  Can we trust the temperature data?  You decide.

Update: Now you can read why I put up this post.

12 Responses to “Mean temperatures rose just as data station numbers fell”

  1. 1 Tufty 13/01/2011 at 3:16 pm

    There is a useful SPPI paper by Joe D’Aleo and Anthony Watts on this subject here

  2. 2 Geoff, Worcester, England 13/01/2011 at 3:51 pm

    Lies, damned lies and statistics. They might as well get one warmista to stick a thermometer in a kettle straight after it’s been boiled, and use the reading as ‘proof’ of global warming.

  3. 3 Barry 13/01/2011 at 4:19 pm

    By way of comparison the HADCruT ‘long term mean’ used for many anomaly charts is 1961-1990 and pegged at 14.0 C.

  4. 4 microdave 13/01/2011 at 6:59 pm

    There are times when I really don’t know which version of the “Science” to believe. But when you see that the raw data is this suspect, it suddenly becomes much easier to smell a rat…..

  5. 5 Steve 13/01/2011 at 7:49 pm

    Unfortunately, this graph has been misinterpreted and is a good illustration why the phrase “lies, darn lies etc.” is appropriate.

    The “temperature” in this graph is an average of all the temperature stations, and does not account for the fact that in some areas there are more stations than in others.

    For example, if England and Wales had twice as many stations per square mile than Scotland, then the average reading for Great Britain would reflect English/Welsh temperatures more than Scottish. It would be more correct in my view to give each English/Welsh station only half the weight of each Scottish station.

    In the above graph what appears to have happened is that more stations in cooler places have been lost than those in warmer stations so causing the average temperature to rise. But if the weighting had changed as the stations were removed then the results would not look the same.

    If one does not know “which version of the science to believe” then I suggest googling “gistemp uah hadcrut3 rss” and selecting the woodfortrees link. This shows 4 different temperature metrics including the one done by the “sceptic” scientist Roy Spencer (UAH). You will see they all show similar patterns of warming (though UAH shows the least).

  6. 6 Steve 13/01/2011 at 7:50 pm

    Here is the wood for trees link:

  7. 7 Cthulhu 13/01/2011 at 8:28 pm

    The chart is actually a load of balls and the people who made it are incompetent at station temperature analysis. That chart is far worse than any mistake climate scientists have ever made. Put it that way.

    And it’s so obvious it’s wrong too. That’s what I don’t get. How can people fall for this chart? Look at it! “Average T” doesn’t look remotely like any of the global temperature products. That should be the first alarm bell right there that something is wrong. A big loud alarm bell.

    It amazes me that when this graph is mucked out for skeptic audiences none of them go “hang on a second”. Neither GISTEMP, HadCRUT, UAH or RSS global temperature records show temperature suddenly spiking upwards in 1990 so why on Earth do so many skeptics interpret that graph as if it means they do? If this sudden warming effect is supposed to happen in 1990 – the fact it *doesn’t* happen in the temperature records should be all you need to know.

    Professional temperature records apply area weighting and average relative temperature change at each station. That’s the correct method.

    This incompetent chart doesn’t bother with area weighting and averages absolute temperatures at each station. So it gets a silly answer.

  8. 8 Steve 13/01/2011 at 9:19 pm

    The comments on this self-styled lukewarmer website are an interesting discussion of the validity or otherwise of the temperature data.

    Many of the regulars at that site have reanalysed the data and found nothing particularly wrong with it.

    (A lukewarmer is, I think, someone who believes in the physical basis for CO2-induced warming but think the hype is overdone).

  9. 9 GP 14/01/2011 at 12:06 am


    The chart is indeed a load of balls if you take it as a blunt statement about global average temperatures or anything serious realeted to assessing changes in climate.

    However that wss not (and is not as far as I know) its purpose.

    If you follow the link the AM provided you will find the discussion posted that considers the potential for the effects that such changes might have on the results depending on how the ‘adjustments’ are made.

    It could all go either way – there is absolutely no certainty about the veracity of historic data or methods of collection other, perhaps, than in for a few ocations around the world with a long history af activity and changes that can form a basis for ‘adjustments’ in any direction. Stations with a long and accurate (or as accurate as we are likely to get) are mostly in Europe and the USA.

    When discussing a catastrophic change over 100 years or more that relates to a measurement that would have been within the range of daily reading error of the measuring equipment not so long ago (and indeed still may be today for all we know) I feel we are getting into the realms of the surreal.

    The graph raises a question. Has that question yet been answered by more than assertion and handwaving? Or have people tacitly agreed that there might well be problems and simply moved on to ‘we can rely on the satellite data now we have 30 years worth ….’?

  10. 10 Orde 14/01/2011 at 11:47 am

    I wouldn’t believe the temperature records from stations, even if none of them had been removed. The unsatisfactory nature of weather station monitoring is covered well on W U W T. It is entirely beyond the practical capability of mankind to measure temperatures over large areas over decades of time, and that’s the end of it.

  11. 11 Steve 14/01/2011 at 1:53 pm

    Entirely beyond the practical capability of mankind!!!???

    No doubt geology and palaeontology are beyond us too and we should defer to the Old Testament?

    On the other hand, is it fair of WUWT people to criticise the current climate scientists for the siting of weather stations that took place a) before they had become scientists and b) before anyone thought that decadal temperature trends were worth measuring? Read the above rankexploit links which gives good evidence that they are at least doing the best they can in an honest manner.

  12. 12 JohnRS 14/01/2011 at 5:24 pm

    So in the US we take out the expensive to service (and generally difficult to access/colder place) weather stations; we run down the maintenance and site inspections of those that are left so they are blatantly misreporting; we watch half the weather stations in Europe and Asia go offline; we have no data for most of the Southern hemisphere….and then (ignoring any bias or errors in how this is processed) still claim that there is enough credible data on which to to base trillion-dollar, multi-decade spending decisions.


Comments are currently closed.

Enter your email address below

The Harrogate Agenda Explained

Email AM

Bloggers for an Independent UK

AM on Twitter

Error: Please make sure the Twitter account is public.

STOR Scandal

Autonomous Mind Archive

%d bloggers like this: