My GHCN/USHCN software calculates a non-gridded average of all US daily or monthly temperature readings (depending on the data source.)
I was wondering how closely my numbers agree with NCDC’s gridded numbers. The graph below shows the difference between my calculated USHCN final numbers and NCDC reported US temperatures. As you can see, the match is very good. There is a 0.25F offset, but the trend is zero. My calculation method does not affect the trend, and is thus valid for how I am using it.