I don’t really understand how this discussion at the Reference Frame is relevant to anything. A study says that the clustering of high temperatures at the end of the last 100 years cannot be just random statistical chance, while Lubos argues that the chance of it happening is low but not nearly as low as the authors state.
I guess this may be an interesting exercise in probability theory for autocorrellated functions, but that is about it. I mean, does anyone really doubt that there has been some sort of upward trend in world temperatures?
More relevent are the questions
- Is this upward trend unique or is it within the range of normal natural fluctuations?
- Is this trend exaggerated by upwards biases in surface temperature measurements?
- What proportion of this trend is due to anthropogenic factors (e.g. Co2) and what proportion is due to natural variations (e.g. unusually strong solar cycles in late 20th century)?
- How can one reconcile the trend over the last 100 years with climate models, which use assumptions that imply that past warming should have been much higher?