One of the most stubborn arguments from the people who deny the reality of climate change is that the sharp upward trend in U.S. temperatures over the last fifty years is due to poor siting of the surface stations that gather the data and the urban heat island effect. This argument is particularly stupid since the trend over time would not be affected even if a station was located in a place where it might read a little high. Imagine two identical urban weather stations; one in an asphalt parking lot and one in a grassy field - the asphalt one would read a few degrees higher, but the temperature trends over time would be identical. It's like saying that if your home scale weighs you three pounds heavier than your doctors scale it is useless to show if you are gaining or losing weight - as long as the bias is consistent then the trend is not affected.
A new paper from NOAA neatly skewers these arguments, complete with a graph showing the lack of difference between the so-called "best sites" and the rest of the weather stations in the network. As you can see both the annual readings and the smoothed trend line match nearly perfectly.
The NOAA paper also points out that there are many ecological indicators that show temperatures are rising rapidly, so dwelling on the minutiae of individual stations while ignoring the preponderance of evidence for rising temperatures is intellectual dishonesty at its worst.
(thanks again to Climate Progress)