Monday, July 23, 2007

Britain's floods and the virtual reality of climate science

U.K.'s The Independent reported breathlessly today that the extraordinary flooding now underway in Britain is happening because George W. Bush withdrew the United States from the Kyoto Treaty.

Well, not really. I should really stop exaggerating like that. I wouldn't be surprised, though, if somewhere out there someone has blamed Bush for this.

What The Independent did report is the claim that "scientists" have "confirmed" a link between Britain's summer floods and "man-made global warming".

Let's look at the opening paragraphs of the article, which is riddled with superlatives:

It's official: the heavier rainfall in Britain is being caused by climate change, a major new scientific study will reveal this week, as the country reels from summer downpours of unprecedented ferocity.

More intense rainstorms across parts of the northern hemisphere are being generated by man-made global warming, the study has established for the first time ­ an effect which has long been predicted but never before proved.

The study's findings will be all the more dramatic for being disclosed as Britain struggles to recover from the phenomenal drenching of the past few days, during which more than a month's worth of rain fell in a few hours in some places, and floods forced thousands from their homes.

So, how were the scientists able to pin the blame so unambiguously on AGW? Not by the hopelessly old-fashioned method of empirical data collection and analysis, but through the use of -- you guessed it! -- computer simulations. Climate models -- which supposedly are not useful for predicting individual weather events (the article admits as much) -- allegedly yielded something akin to this summer's individual weather events when simulations both with and without greenhouse gases were run.

The article turns out to be quite confusing. In places we are confidently assured that AGW Did It, and then in other places the confidence wavers (emphasis added):
Meteorologists agree that the miserably wet British summer of 2007 has generally been caused by a southward shift towards Britain of the jetstream, the high-level airflow that brings depressions eastwards across the Atlantic. This is fairly normal. But debate is going on about whether climate change may be responsible for the intensity of the two freak rainfall episodes, which have caused flooding the like of which has never been seen in many places.
So, rainfall events enabled by a normal shift in the jet stream are now the fault of AGW because of "unprecedented" flooding that resulted. Remember, no empirical link has been established -- just computer simulations and human inferences.

The new study, carried out jointly by several national climate research institutes using their supercomputer climate models, including the Hadley Centre of the UK Met Office, does not prove that any one event, including the rain of the past few days in Britain, is climate-change related.

But it certainly supports the idea, by showing that in recent decades rainfall has increased over several areas of the world, including the mid-latitudes of the northern hemisphere, and linking this directly, for the first time, to global warming caused by human emissions of greenhouse gases.

In other words, "It looks awfully like what we think might happen." More:
"Some people would argue that you can't take a single event and pin that on climate change, but what happened in Britain last Friday fits quite easily with these conclusions. It does seem to have a certain resonance with what they're finding in this research."
Yikes! This is what passes for the scientific method among adherents of the CoGW.

The study, to be published in the journal Nature later this week, may end up being a lot less sloppy than the reporting here presents it to be, but I'm not willing to bet money on that.