Sorry tried to read the link, but the picture irked me. Straight away i thought why bother.
Why cant we look at all raw data per site. Then say on average there has been a 0.5 degree Celsius raise per site over the last xx years.
If there has been anomalies in the individual sites i am sure that would average out. Exclude the outliers if required.
THAT i would feel a lot more comfortable with.
