The problems I encountered with the latest HadCRUT data were really due to a new version 5 which I had missed because this data was only made available on the MET Office site but not on the CRU website, which I have been using to analyse all the raw temperature data. This has now been corrected thanks to Timothy Osborne.
The version 5 analysis goes further than Version 4, and supposedly also does “infilling” of temperature data into regions without measurements. However I don’t trust this procedure, so will use only the “measured” values as shown below.
As world population grew measuring points shifted.
My concern is its all faked into fear. Fear is taxable.
While hexafluoride tearing down rain forests. Exploding populations killing of big grazers and natural cycles are downgraded.
To get a visible change in monthly averaged temperatures of 1C you need to analyse over 120 years of data. In reality the daily average change of temperature in one locality can be over 10C. So in reality a long term global increase of just 1C is hardly noticeable.
The most pertinent observation is what drives the switch Temperature trends. Warming 1910-1940, 1940-1980, 1980-2020.