Curry: Detection and attribution of extreme weather events requires:
a very long time series of historical observations of the extreme event at a particular location (which is rarely available)
an understanding of the variability of extreme weather events associated with multi-decadal ocean oscillations (which requires at least a century of observations)
climate models that accurately simulate both natural internal variability on timescales of years to centuries and the extreme weather events (a pipe dream; we are not even close)
IMO the best, and most important, paper on extreme event attribution is a paper by Hall et al. [discussed previously at this link]. This paper by Hall et al. is not even referenced in the new National Academies report.
