Graph of the day: global warming is real
A statistically (and according to my experience: historically) very robust rule of thumb to investigate if a time series has an upward or a downward trend is to investigate if minima in one period are higher than maxima in a previous period/if maxima in one period are lower than minima in a previous period. Less robust but still useful: instead of looking at a moving average (or its complicated nephew: a HP-filter) you can look at a line drawn along the maxima or the minima of a series. Applying these methods to the leaked data of the draft of the IPCC’s fifth assessment report yields that the hypothesis that temperatures have been rising fast up to and including 2011 can not be rejected. The Fabius Maximus blog is, using these data, wrong to state that this isn’t the case (but there does seem to be a tendency that differences between maxima and minima are declining).
The Fabius Maximus guys might indeed do well to change their motto: ‘to reignite the spirit of a country grown cold’,