For example, they show that
the modern instrumental record averaged only over their 14 sites captures the full Northern Hemisphere mean temperature variations remarkably well over the available (approximately 150 years) interval.
Guiot's attempt to create a regression formula against
the modern instrumental record has promise for the right variables.
Be careful — even the author of that graph on Wikipedia admits that past temperature records are naturally much smoother than
modern instrumental records.
The paper must be wrong because you can't use
the modern instrumental record to compare with paleo reconstructions.
In order to be able to make an easy visual comparison, in the next graph we have removed the 1990 IPCC graph and rebased CET, so the anomalies match with Dr Mann's reconstruction which — until
the modern instrumental record is inserted at the end - runs between minus 0.2 and minus 0.5 C anomaly - somewhat cooler than CET.