In fact there seem to be plenty of opportunities for error of transcription and presumably the subsequent digital transfer (thinking of older records which would not have been entered directly in a computer system).
Makes me wonder just how good the data are for the practical purposes of creating social policy.
Thinking about this further Jim, a better approach for validating or invalidating the greenhouse effect at Death Valley might be to look at ground temperature data rather than (shaded) air temperature data. The idea of the greenhouse effect, as I understand it, is that the ground gets its temperature boosted by 'back-radiation' from greenhouse gases in the atmosphere, so it makes more sense to look at ground temperature. If you look at air temperature, you are looking at the thing which is doing the back-radiating.
I don't think ground temperature is normally measured at all but it might be available in the specific case of Death Valley. According to this webpage:
"Highest ground temperatures
The highest ground temperature recorded was 201° F at Furnace Creek on July 15, 1972. The maximum air temperature for that day was 128° F."
The maximum ground temperature recorded was 201 deg F (93.9 deg C) in 1972. The fact that a specific date is quoted suggests ground temperature is being recorded every day and it is possible the data might be available to the public. I suspect that any ground temperature data might also be less likely to suffer from 'adjustments'.
I'm not sure they publish ground temperatures. I agree with you, the surface temperatures are really SATs (Surface Air Temperatures)--the temperature of the air just above the surface. The climate models always refer to surface temperatures, but they really mean SATs. That's another problem with climate models (and there are so many problems).