I have read papers by Christy (I think) where the authors used change points to
adjust radiosonde data (where they corresponded with instrument changes) and found a better fit with satellite data.
Not exact matches
These problems are sufficiently serious that the US National Weather Service (NESDIS)
adjusts satellite data every week to match
radiosondes, in effect relying upon
radiosondes as a reference instrument.
The paper discusses four different sets of data on satellite atmospheric monitoring (all producing slightly different end products), two
radiosonde data sets (from UK Hadley Centre and University of Vienna, both
adjusted for inhomogeneities — and that opens another can of worms), four different surface temperature data sets (based on reconstructed sea surface temperature data sets from Hadley Centre, again, and Climate Research Unit).
And, since IPCC assumed essentially constant relative humidity with warming to arrive at the water vapor feedback and it appears that RH decreases with warming (Minschwaner + Dessler 2005, NOAA
radiosonde and satellite humidity records), the water vapor feedback is very likely too high by around 0.3 C to 0.6 C, bringing the overall
adjusted ECS to roughly 0.9 C to 1.2 C.
I suspect the next chapter in this saga is that the remaining
radiosonde datasets that still do not show substantial warming will be the next to be «
adjusted» upward.