Earlier I showed that the USHCN stations which are losing data since 1990, were much cooler stations than the ones which aren’t losing data. NOAA is replacing cold station data with warm station data.
So now lets look at what has happened since 1990, when the station data started disappearing. The stations with complete data are warming much faster than the ones which are losing data.
There are two ways this could be interpreted.
- They are replacing missing cold temperatures with warm ones, and creating a fake warming trend.
- The stations losing data are by some spectacular statistical fluke selectively losing warm data. That gross impossibility is the fundamental assumption behind the FILNET adjustment.
I did an experiment (below) to see what happens to the absolute temperature if you randomly remove 30% of the months. Not surprisingly, it has very little effect on the trend, which is what I was trying to explain to Zeke.
This leaves the only real possibility – NOAA is losing cold station data, and doing a large warming adjustment when they should be doing a large cooling adjustment.
Bottom line is that USHCN data alterations are either fraudulent – or just plain stupid. Take your pick.