Yet another problem with surface temperature measurement

Here.

Excerpt:

This means that a significant warm bias exists in the 2007 IPCC Report on the trends of the global average radiative forcing, since they base their estimate on a surface air temperature trend that includes minimum temperatures over land. This bias, which occurs whenever the nighttime surface boundary layer is stably stratified over land and the winds are light is, therefore, very significant at high latitudes in the winter, where much of the warming is reported to have occurred. This error in the IPCC construction does not mean that no warming has occurred, but it does mean that they have significantly overstated global warming in their report.