Here’s another gridcell where CRU has truncated high early values. This is the gridcell that contains the important Jacoby Mongolian tree ring series. If untruncated gridcell information is used, then the ring width chronology has a negative correlation to gridcell temperature. What a swamp.
Here’s a plot of the series in the Mongolian grid cell 47.5N 97.5E which contains the Sol Dav tree ring series. Here CRU has retained only values from 1960 to 1983. Values from Anxi station go back as far as 1938, but are higher than later values. I’m sure that CRU has a “good” reason for not using the values.
This plot shows the 20th century portion of Jacoby’s Sol Dav Mongolia chronology (mong003.crn), overlaid by the Osborn and Briffa version (which I’ve diagnosed elsewhere as coming from a Mann and Jones 2003 digitization of a snail version of the Jacoby chronology and is slightly askew). Below are the NOAA gridcell (black) and CRU gridcell (Red). There is a negative correlation between the Jacoby chronology and the gridcell chronologies (Jacoby reported a positive correlation to a gridcell in which the trees were not located and used that as a reference.)