It's a slow news day so far today, so I'll go with a Guardian-talking-drivel-on-climate story, which is a bit like a Pope-is-Roman-Catholic story, but can occasionally provide some light relief. Today's headline from the fount of foolishness is
Extreme weather already on increase due to climate change, study finds.
Unfortunately for the Guardian, the study in question, by Fischer and Knutti, is actually nothing to do with observations of extreme weather at all. Instead it is about their attribution to humankind. You have to wonder whether the headline writers even read the paper.
And if you look at the study, it turns out to be just an extension of the use-shonky-GCMs-to-blame-humankind approach adopted by others in the past. I'm hugely amused by its suggestion that GCMs, which have precisely zero ability to predict precipitation, can be used to show that "18% of moderate daily precipitation extremes over land are attributable to the observed temperature increase since preindustrial times". Particularly since in the IPCC's view it's hard to find any evidence of changes in extreme weather anyway.
It's like...magic.