Discussion about this post

User's avatar
bobbert's avatar

This is excellent. It might be worth pointing out that the axes of the second graph are "temperature anomaly" -- that is, deviation from the mean -- on the X, and probability on the Y.

The most significant feature (you know this, I'm just emphasizing it), isn't the shift in the mean value, but the fact that in recent decades the distribution has become less and less Gaussian; it is now much wider than Gaussian.

So the chance of a given value (say,any Wonker's local temperature) being 3 standard deviations from the mean is now maybe 3%, whereas in 1980 it was only a tenth of a percent. That is, in 1980 (or before), you might expect one day every couple of years to be 3 sigma hotter or colder than the average; today, you'd expect ten days per year each way. Of course, the mean itself has also increased, so the probability of extremely <i>cold</i> days (compared to 1980) is actually slightly reduced.

Anyway, fewer really cold snaps, moar and longer hot spells. Oh, and the 5 sigma days will be another 2 or 3 degrees F hotter than the 4 sigma days, I imagine.

Of course, this all comes from James Hansen, so algore is fat, lalalala I can't hear you.

Expand full comment
bobbert's avatar

I would like to stick a carefully selected shrubbery up each of their asses.

Expand full comment
39 more comments...

No posts

Ready for more?