The latest exchanges over the Marvel et al paper make for fascinating reading. Over at RealClimate, Gavin Schmidt writes a rather thin response to Nic Lewis's critique. Lewis has responded at length at Climate Audit.
Gavin, as might be expected, has made heavy use of his standard, "paraphrase, don't quote" technique, creating a series of strawmen that he can knock down with ease. For instance, at one point Lewis set out a great deal of evidence that suggested that land-use changes may have been omitted from a calculation. He mused about whether there was a rational explanation. Gavin paraphrased this as [my emphasis]:
Lewis in subsequent comments has claimed without evidence that land use was not properly included in our historical runs, and that there must be an error in the model radiative transfer.
It's perhaps worthy of note that Schmidt doesn't actually provide any explanation of what Lewis observed.
Nic's piece has an amusing sting in the tail as well. In their original paper, Marvel et al said that:
TCR and ECS are calculated by regressing ensemble-average decadal mean forcing or forcing minus ocean heat content change rate against ensemble-average temperature change.
Unfortunately this wasn't quite right. They actually did the regression the other way round.
And they didn't regress ensemble averages.
And they got their uncertainty calculations wrong.
By 100%.
Except for one instance.
Where it was more than 100%.
;)