So how bad was the recent U.S. heat wave?
On the right is a figure from NASA, via Climate Central.
A 12°C (22°F) anomaly over 7 days certainly sounds bad. Is it?
I'm not sure, but here's how I tried to think about it:
I've kept records of the weather in Scappose, Oregon, down the road from me, for the last 3.5 years, via this NOAA site.
Like they do, I take the average daily temperature to be the average of the day's high and the day's low.
Over the course of my record keeping, from Sept 10, 2008 to yesterday, the temperature here has been -0.73°F from normal. (The annual average normal here is 52°F, with 37 inches of precipitation. Yes, it doesn't rain huge amounts in Oregon, it just rains hugely often.)
From here on I'll convert to Celsius, so as to compare with the map above.
Over the course of my record keeping, I calculated the average for each 7-day period (rolling), and compared it to the normal. I took the absolute value to get a deviation regardless of sign.
The average 7-day absolute deviation was 1.72°C, with a standard deviation of 1.43°C.
The maximum 7-day deviation was 9.64°C -- that was a cold snap in December 2009 where we had two consecutive nights where it got down to 8°F, with daytime highs both days of 31°F. The pipes running through the attic from my water heater to the bathroom broke, and my landlords had to replace the pipes, redo part of the bathroom ceiling, and replace several square yards of carpeting. I lost a few papers that were on the floor and a bowl of kibble.
So even in this 3.5 year period I had a 7-day deviation of nearly 10°C, almost like the heat wave in the NASA map above.
That was probably an unusual cold snap for here -- the plumber said pipes were breaking all over town. But even in 2011, the maximum 7-day deviation was 6.11°C (a warm spell last August).
Given my average and standard deviation, how unusual would a 7-day heat (or cold) wave of 12°C be? It'd be very unusual: 7.2 standard deviations above the mean. Assuming its a Gaussian distribution, the chances of that occurring are about 3 x 10-13.
Since every day you get a new 7-day interval, this means a 7-day heat/cold wave with a ±12°C anomaly would occur, on average, every 52 billion years.
That's surely too high, probably because my record is very short -- only 3.5 years long. But clearly such a heat wave is very rare. So it was partly fair for McKibben to highlight it, although if you'll notice, there is also a very large deep purple spot on the map above. At the very same time as the USA48 heat wave. So 52 billion years is surely far too large.
On the other hand, who's going to complain about a big heat wave in March? I'm sure almost everyone was saying, if this is global warming, give me more of it! The eastern European cold snap claimed hundreds of lives. Here is saved a lot of people money on heating and salved their cabin fever.
OK, you say, what if it had occurred in July? Well, it didn't, and anyway July is only 1/12th of the year (OK, August is another 12th). Heat waves happen. As Mike Hulme writes in you know where, in 1976 there was a heat wave in France that caused about 6,000 excess deaths that "passed completely unnoticed and the death count was only revealed by later retrospective analysis."
Are we really going to rebuild all of civilization to avoid the occasional heat wave and rising seas and some slightly stronger hurricanes? I was thinking about this last night, and the answer is clearly no. Yes, these aren't the only impacts of climate change, by any means, but we get a lot of value from burning fossil fuels. A lot. There is as yet no viable substitute for them. I was thinking about my personal list of concerns, the big ones, and the honest answer is climate change is pretty far down the list, after health care, income (and so, the economy), clean local air and water, crime, and then things like war and terrorism and then hungry and poor people across the globe and animal suffering and probably a few things that aren't occurring to me now, and then climate change. Sorry, but that's the truth.
So partial apologies to Bill McIibben. But I can't go all the way to "K," because he still picks and chooses his weather events, and because he completely ignored the main finding of the recent Swart and Weaver article on the Canadian tar sands causing only 0.03°C of warming. True to form, he took it as more proof of his position.