Another smoking gun that USHCN adjustments are garbage, and inverted from reality.
The weather station at Fort Collins, CO is a classic study in UHI, yet NCDC adjusts Fort Collins temperatures by massively cooling the past – the exact opposite of what they should be doing.

In 1937, the station was located in the middle of a farm,

By 1950, the area was starting to get built up.

By 1969, the city had surrounded the weather station.

Now it is in the middle of a parking lot.

The Fort Collins trend massively diverges from nearby Boulder, CO. Fort Collins is warming rapidly, while Boulder is cooling. Homogenization should warm the past, not cool it.

What about TOBS? The Fort Collins temperature has always been read near sunrise, which adds an additional cold bias to the data. Every factor in the equation should cause the NCDC adjustments to warm the past, rather than cool it.
Because of TOBS, all older temperatures should be adjusted upwards, not downwards. And because of UHI, all recent temperatures should be adjusted increasingly downwards. The actual adjustments are exactly the opposite of what they should be – according to the USHCN documentation.
