For example, the satellite and
radiosonde measurements indicate something like 1/3 as much, and in fact even this is only true after they have been corrected.
Not exact matches
They have gone through a number of types of
radiosondes and the satellite data would
indicate the
measurement change with the new device since the new
radiosondes wouldn't match the satellite data and the old
radiosondes would This is the same problem — mandatory objective environmental test standards would give historic continuity.
Trends reflect the mean change in temperature (in K per decade) between 20 ° N and 20 ° S for the period 1979 — 2005, obtained from
radiosonde temperature
measurements 5 (blue and green colours), climate models 8 (dashed orange, with grey shading
indicating 2 - sigma range) and the new reconstructions from
radiosonde winds 4 (pink, with error bars
indicating 2 - sigma range).
Surely these differences would be more significant if some showed cooling, other Global temperature
measurements such as
Radiosonde, SST's and the proxy atmospheric water vapor also
indicate warming, so inconsistancies between the indexs surel are more a technical issue.