Global Temperature Trends Based On Non-Existent Data
By Paul Homewood
https://www.metoffice.gov.uk/hadobs/hadcrut4/diagnostics.html
We are all too familiar with graphs showing how much global temperatures have risen since the 19thC.
The HADCRUT version above is typical, and also very precise, with fairly tight error bars even in the early part of the record.
One wonders where they got the data to work all this out, because it certainly could not have come from thermometers.
All of the major global temperature datasets rely heavily on the Global Historical Climatology Network (GHCN). Yet as the “Overview of the Global Historical Climatology Network-Daily Database”, published by Matthew Menne et al in 2012, rather inconveniently showed, most of the world had little or no temperature data in the 19thC, and even up to 1950.
Density of GHCN-Daily stations with daily maximum and minimum temperature
http://journals.ametsoc.org/doi/full/10.1175/JTECH-D-11-00103.1
Prior to 1950, there were no more than a couple of hundred or so of GHCN stations outside of North America:
There are many competent scientists and statisticians who believe that even now it is not possible to measure the Earth’s average temperature, indeed that it is a meaningless concept.
Whether they are right or not, no serious scientist would claim to know the global temperature a century or more ago.
Comments are closed.
NOT A LOT OF PEOPLE KNOW THAT 
I’m a scientist who says the Earth does not have a temperature, because temperature is an intensive variable. Whether I am competent, I couldn’t possibly comment.
As for using an anomaly, based on 1961 – 1990, well that implies that the Earth’s temperature is normally the average value it had over those years – complete nonsense of course.
It is all propaganda aimed at deceiving politicians and the general public.
Incidentally, I went on a tour of the Met Office’s grandiose headquarters in Exeter last Saturday. More propaganda, telling us all what a world-leading centre of excellence it is.
Phillip
I’m not a scientist. What I would like to know is how temperature records in the Stephen screen/thermometer era can be relied on.
The variabilities seem enormous. Before the climate change ‘enlightenment’ I would wager that most Stephenson screens wer not well maintained, they were not painted with the same materials, if they were painted at all. They were badly sited with overhanging trees, built up housing etc. (I wont go into all that as Anthony Watts has more than enough detail on the subject) but more. Who checked the thermomoters? To imagine it was highly qualified scientists rather than the local shopkeeper, or his son/wife/cousin etc. who had better things to do with their day is naieve.
The what about the height of the person reading the thermometer? I’m 6’2″, my wife is 5’6″, how would the height difference affect the reading, what if snow had built up a layer of only 1Ft. , would that influence our observations.
Then there’s the very real conditions of data retrevial, can I actually be bothered to go out in the sun/rain/snow/cold/heat etc. to check what feels today, exactly what it felt yesterday. An would it be more productive to send the tea boy.
Sea temperature measurements are much the same. How many science aware individuals were available to clinically inspect a bucket of water retrieved from over the side. How many engineers would want a bothersome scientist prowling their engine room for ‘data’ far less the tea boy sent for the information.
In short, in my opinion, data gathered in the thermometer era should be largely discarded as it is too unreliable.
Your objection appears to be based on a misconception. At no point do the compilers of the various temperature series attempt to work out the actual average global temperature. What they do, in essence, is to work out the average temperature over a common set of base years for each individual station. They then calculate the differences from these averages for each year for each individual station. Next, they grid these differences by averaging, (typically weighted), all the differences for each station within the grid. Where a grid has no stations within it, they interpolate the “likely” results from some sort of further weighted averaging. Finally, they use an area weighted average for all the grids to arrive at global or regional temperature differences.
The errors implicit in any interpolation can be assessed by comparing what values would have been interpolated with the actual measured values.
The main valid objections concern the adjustments to individual station temperatures “justified” by changes to those stations. Typically these are changes of location, instrumentation and so on.
And if they applied correct error bands to all that at every stage , the bands would exceed any ‘anomalies’ by several hundred percentage points. Its all numerology, with no basis on fact or science.
It is a good description, but the interpolation claim is bogus. How many trials do you need to do to test your interpolation? And in how many areas and at how many times of year and over what period of time?
And then, what do you do? Let’s say you have done sufficient trialing – is the “number” always the same, i.e. is the relationship between the “unknown” and the “known” always a fixed ratio? I very much doubt if the algorithm can be tuned to give the right answer for every single day of every single year. So do they use a different algorithm every day? And even then, you still haven’t shown that the areas where you have no measurements are interpolated accurately using algorithms from other areas.
Perhaps I am wrong, but I very much doubt if the methodology is at all sound enough for us to say let’s totally change our economy and spend trillions doing so.
I would say that the Earth does have a temperature, but the only reliable way to measure it would be to go somewhere like Mars so that the Earth would appear, in Carl Sagan’s words, as a pale blue dot. You could then measure its radiative temperature. However, whether one would get the same result by measuring local temperatures on the Earth’s surface and combining them in some fashion is a moot point. I suspect there are many ways of combining local temperatures, all of which would give somewhat different results.
What happened to the plus 0.6°C anomaly in 1938 as shown in 1975? Now it is only plus 0.2°C.
They worked out that those reading were wrong.
Oddly only really the hot ones were wrong.
OMG look at all that warming in 165 years 0.85 C
Yes, and when most of it came when we converted to digital thermometers.
Reblogged this on Wolsten.
In post-modern ‘science’ no data is good. More flexibility for the rigorous application of feelings. And, anyways, numbers are racist.
Ah, modelling is a marvellous thing. The error bars are something else!
Tony Heller posts this information fairly often. Here is his latest, wherein he illustrates the global weather station coverage for 1900 and 1940. https://realclimatescience.com/2017/11/unambiguous-fraud-in-the-national-climate-assessment/
Tony’s GIF graph
2 months ago Anthony watts did a longer analysis
Still trying to wipe out the pause with fake info in that bar chart?
You can’t even begin to correctly model something which doesn’t exist. You can twist statistics and make up extraordinary projections, all you need is a bent scientivist or 97% of 40 odd men and al gore isms.
Carbon dioxide the man made bit and ‘RUNAWAY’ ha, ha, ha, ha! even slightly rising temperatures [modelled um……. guessed at] are evidently only evidence educed out of the fetid imaginations of the liars of climastrology department soothsayers.
The great global warming swindle…………………YES It’s a scam, a tax grab – nothing else.
It is clear that the only meaningful anomalies can be for Europe and North America (USA even). The largest part of the world’s surface ( oceans and polar regions) was / is barely measured: so the anomaly is the anomaly of selected data.
A further point about modern “statistics” is the abuse of confidence limits. Confidence limits in statistics refer to the percentage of data falling within a defined error margin, commonly , 95 % of data lying within two standard deviations. It is now ( notably in IPCC reports and the graph above) used as the believed accuracy of a viewpoint or value: thus is a consensus value, an opinion or a political value.
Also, if one accepts the data, what caused the temperature rise rate in the 1930s to be as high as the post 1980 value? Did CO2 make a rapid appearance and then disappear?
Re the Met Office- take a look at their website page on ‘climate change ‘ and count the untruths of which there are many. That site hasn’t changed in over two years by my count. Despite the mounting contra evidence.
Slightly off topic I complained to the BBC about the poor quality of this article.
http://www.bbc.com/news/business-41967241
Which has both a lack of data causing problems and increasing extreme weather events within a couple of paragraphs. My point was how do you know extreme weather events are increasing if you have no weather data.
I’m expecting the usual BBC response of we’re right but we’ve made a note of your complaint for future guidance (unsaid but nobody is bothered)
The change in average temperature of the world over time is important!
But only important when assessed against exactly what are the changes, over the same time period, in average solar output, average changes in atmospheric chemistry, average atmospheric humidity, average atmospheric pressure, average seismic venting activity, and changes in the average color hue of the world. Plus many other parameters.
All in all, average temperature of the world over time is as really important as knowing the average foot size of the world’s population.
🙂
As my major professor stated: “the plants don’t care about the average temperature, they care about the extremes.” And he was not inferring that plants were “standing” around “wringing their leaves.”
I have 2 points to make, as a scientist:
1. The Scientific Method requires that data collection be done with all things as equal as possible. Therefore, temperatures should be taken in areas which are open and as unaffected as possible by their surroundings. Readings must be comparable to be compared–DUH!
2. There is no such thing as standard temperatures. The earth sees to that. As one with a more than passing amount of paleobotany in my background and who keeps up with it, I can say with certitude that climate changes. Also the position and configuration of the continents changes which also has an important effect on climates over time.
Plants and animals have been though glacial cold times and interglacial warm times. Those we have today have survived both and thus are genetically “predisposed” to survive them again. Think Polar bears.
A final observation is that warmer times are better for people than colder times.
If I look at that Hadcrut graph with red and green lenses, does it come out in 3D ?
Reblogged this on Climate Collections.
Australia makes up a large part of the Sth Hemisphere data 1910-1950. Some of us have been studying this for years, in great detail, station by station and agglomerated. The present state of our comprehension is roughly –
1. The official warming of about 1 deg C per century is hard to see. We can see about 0.45 deg at best.
2. The difference is largely from adjustments made by the BOM, plus their selection of which stations go into the final mix.
3. Adjustments are cosmetic, when the fundamental problem is mainly the screen housing the temperature device, The Stevenson Screen, designed to make stations more equal, turned out to be a nervous little Nellie, which caused changes from small effects. The colour of the lime is an example.
4. Roughly about the 1990s, replacement of liquid in glass thermometers with platinum resistance thermometers was done in a fairly secretive way. Hindsight investigations show that screens remain a problem, The screen and surrounds of the newer Pt devices have been engineered to give slow responses, supposed to mimic the response time of the LIG. To the extent that they can/do not, there is more error.
5. Some of us have made representations to the authorities, but as expected we get fobbed off. There have been some Forums set up for stats experts and others to look at adjustments, but the reports are reminiscent of those after Climategate. ‘Played a blinder’ stuff.
6. For the moment, we are continuing to do more study and submit more reports. The federal political scheme is unstable so there is little point in submissions until that settles down.
7. Overall, data prior to about 1990, with satellite T data to assist interpretation, we regard global and national T estimates as not fit for purpose, especially when that purpose shakes the economic world.
Where did good, clean science go? Geoff.