I just want to round up a few recent stories regarding “climategate”, as I’ve not had time to do the more in-depth work I’m planning yet.
Firstly, RealClimate look at the integrity of the CRU data set, with an analysis comparing raw data with the adjusted quality controlled data:
The key points: both Set A and Set B indicate warming with trends that are statistically identical between the CRU data and the raw data (>99% confidence); the histograms show that CRU quality control has, as expected, narrowed the variance (both extreme positive and negative values removed).
Thus if their analysis is sound, it would seem that the warming trend was present in the raw data, and the adjustments did not introduce it. Note that in contrast to Willis Eschenbach, Real Climate take a random sample of the whole data set rather than focus on a few stations.
Meanwhile, apparently the US Department of Energy has issued a “litigation hold” notice to CRU employees asking them to preserve documents, suggesting that some form of legal action might be being prepared for.
Also, the Russian Institute of Economic Analysis claims that the CRU have cherry picked the warmest stations in the HADCRUT data set (which is joint work of the Hadley Centre for Climate Change and the CRU). The article is a bit confused blaming the Hadley Centre rather than the CRU (who provided the land-based data in the HADCRUT data set) however. I’m not sure whether Real Climate’s analysis of the CRU data linked to above would account for cherry picking of the weather stations. It depends on whether the full raw data was in the data set with the adjusted data affected by the cherry picking or both.
Finally, Megan McCardle speculates that climate scientists may have been calibrating against each other, this would be more subtle than conspiring to fake the data, and they might not even realise they were doing it. The only way one can address these concerns is to analyse the raw and adjusted datasets, and any code and methods used for adjustment for signs of bias.