When it Rains it Pours

Harvey Taylor sends along a link to an article entitled "Magnitude of Pakistan Floods is Unprecedented". If we consider the Russia and Pakistan catastrophes as part of the same event, we have easily thirty million people directly affected and loss of life in the tens of thousands.

We also have more ammunition for claiming a very high weirdness index for the event.

Two especially interesting quotes:

We had an extraordinary event in northern Pakistan, where ten times the normal annual rainfall for one of those areas fell in four days. This is unprecedented. And, of course, you see the similar sorts of things happening in Africa, in Asia, in Central America. The farmers no longer know when to plant, because the rains don’t come at the time they used to come. All these are the kind of effects of climate change that we’ve feared and are beginning to come to pass, I’m afraid. - John Holmes, UN Under-Secretary-General for Humanitarian Affairs and Emergency Relief Coordinator

Just to give you a sense of how bad the rain was that caused this flooding initially, on the 28th of July, there was 318 millimeters of rain just on one day. To put that into context, the record, all-time record, for rain in Peshawar, which is where this number is from, for one month, the month of July, was 217 millimeters. So it rained more in one day than it had ever rained in an entire month for the monsoon season. The floods that have ensued—and there’s been two waves of these floods now—there’s no government in the world that could have prepared for this or that could have responded to this in the way that we would have liked it to. - Mosharraf Zaidi columnist for The News of Pakistan and Al-Shorouk of Egypt

Rolling a thirteen

Australian Prof. Steven Sherwood of UNSW says the same sort of thing I have been trying to say very nicely in a statement recently quoted on Dot Earth:

The “loading the dice” analogy is becoming popular but it misses something very important: climate change also allows unprecedented (in human history) things to happen. It is more like painting an extra spot on each face of one of the dice, so that it goes from 2 to 7 instead of 1 to 6. This increases the odds of rolling 11 or 12, but also makes it possible to roll 13. What happens then? Since we have never had to cope with 13’s, this could prove far worse than simply loading the dice toward more 11’s and 12’s. I’m not sure whether or not what is happening in Russia or Pakistan is a “13″ yet, but 13’s will eventually arrive (and so will 14’s, if carbon emissions continue to rise).

I think we have just rolled our first thirteen. I don't know how to prove it. (update - or maybe our second, if you count Australia last year)

I can define a fourteen easily. We will have rolled a fourteen when there is no controversy at all about whether the given event was in the range of unforced natural variability.

My attempt to raise this issue at Kloor's went like this:

As the climate, i.e., the distribution of weather events, changes rapidly, the space occupied by the tails of the distribution sweep out territory that were not meaningfully in the distribution before.

So we will occasionally see things we could not have seen before, and some of them may be things we have not imagined before. We will see not only known but previously unlikely conditions (e.g. extreme heat), but also previously unsuspected vulnerabilities (smoke).

It is impossible to predict what these weird events will be. The simulation models are too coarse and too conservative, and we wouldn’t know what to look for in their output anyway. You can’t really do statistical attribution on single events, and causality is pretty complicated in a tightly coupled system. So it’s hard to say much about this beyond that we should not only expect the unexpected, we should expect a great deal more of it.

Sherwood's analogy is very succinct and expresses the same point.