It is time to stop quivering in our boots in pointless fear of the future and just roll up our sleeves and build it.
- Ray Pierrehumbert

Thursday, June 11, 2015

The sort of thing that needs explaining

Washington Post:
Down south, 8.81 inches fell on average across the entire state of Texas, which made May the wettest month on record for the state — blowing away the previous record of June 2004 by over 2 inches. The rest of the top five statewide records are all within less than a ¼-inch of each other.
emphasis added. This sort of outlier can't be attributed to any cause, but it sure doesn't look normal. How many of this sort of thing do we need before we get the message?

Climate change is not just about severe events. It's also about unprecedented events. For which, by definition, there are no statistics at all.

It was also the wettest month on record for the contiguous states, but not by such a strange margin.



2 comments:

Rob Ryan said...

How often has such an event (that is, a rain total over a month or similar exhibiting such an extreme delta over previous records) happened in the past if you look at all 50 states? Is this an example of the "Texas sharpshooter fallacy?"

Michael Tobis said...

I don't know. Events that are outside the normal (in both senses) distribution are events that can;t be accounted for in normal ways.

It's possible that hurricane Irene (which most people outside Vermont who don't say the word "Irene" every day may have forgotten) put Vermont well outside its normal monthly rain range. But here we have at least half an explanation - a hurricane hit Vermont! What are the odds of that?

Is that climate change? Well, unless and until a few more hurricanes hit northern New England, we have no statistical basis for that. But it is damned peculiar. Either way, though, "hurricane" is a sort of explanation.

Why this pattern of immense downpours started in this part of the country in early May (another one coming next week apparently, though a bit north of us here in Austin) is harder to put a short answer to. The increased moisture content of the atmosphere is too easy of an answer - the atmosphere as a whole is not that much moister than it was last year, when it was dryish, or in 2011, when we had the sort of drought here that is now hitting California.

It seems to me as I think about it that attribution of severe weather events is a very messy business. That it's by far the most effective "news hook" to get climate change into the press is not doing much for the credibility of the climate issue among the more rational segments of the population. sigh...

On the other hand, it is likely that unprecedented events will do most of the damage. Ecologists are discovering that species viability is much more about severe events than about the normal range of events. I think human infrastructure is going to respond similarly.

So it's a mess...