> How does an average even makes sense, in such discrepancy in temperatures observed worldwide. > Also, how do they account for urbanization effects (which in

> How does an average even makes sense, in such discrepancy in temperatures observed worldwide. > Also, how do they account for urbanization effects (which in itself causes warming), and placement of thermometers around these areas. Related to your worries, 11 years ago a group of physicists from outside the meteorology and climate science community, had similar worries. Maybe meteorologists and climate scientists haven't done a good analysis of the existing land surface temperature measurements? What about the urban heat island effect, growing cities, and moving measurement locations? These physicists started the Berkeley Earth Surface Temperature Project, went back to the original raw data, and reanalyzed everything, developing their own new statistical methods as needed. https://en.wikipedia.org/wiki/Berkeley_Earth https://en.wikipedia.org/wiki/Berkeley_Earth The result? The same observed warming rate as in the previous studies. Maybe this kind of reanalysis needs to be done again every 10-20 years? Kind of like rewriting software from scratch.