Friday, December 09, 2005

Bryden won't bet on cooling either

For those who've been asleep, or on Mars, for the past week, I'm talking about this paper which I mentioned here, and the associated press coverage such as this Guardian article. I've deliberately linked to the google cache in preference to the current page because while writing this blog I've noticed that the page has changed subtly. The original version had the sub-head Temperatures in Britain likely to drop by one degree in next decade but this has now been removed. If the googlc cache has changed by the time you read this, you can find the original text quoted here (googling on the exact phrase brings up several copies). The text of the article attributes this prediction to Professor Bryden, but only with an additional qualification: If the current remains as weak as it is, temperatures in Britain are likely to drop by an average of 1C in the next decade, according to Harry Bryden. Even though there's no clear indication there whether he actually expects the circulation to stay this weak or not (and indeed there's no guarantee that he was correctly quoted at all), I thought there was enough of a question-mark to make it worth enquiring as to whether he really did expect such a cooling to occur.

From his rather curt reply, it seems clear that he is not prepared to consider a wager at all.

[I'm going to be pedantic this time and refer to the circulation in question as the North Atlantic Meridional Overturning Circulation (MOC for short). The name "Gulf Stream" is IMO best restricted to the rapid surface current flowing north in the Caribbean/Gulf of Mexico area close to the east coast of the USA. Some of this doesn't get very far north before returning south at or near to the ocean surface (essentially forming a large clockwise gyre in the Atlantic), but the rest continues flowing further north, up to the latitude of the UK and beyond. This bit, sometimes called the North Atlantic Drift, is what keeps NW Europe mild in winter. When the water gets cold enough, it sinks (is mixed with deep water via convection) and returns at depth (there's a diagram here, described in this article by Stefan Rahmstorf). It is this northerly branch that Bryden claims to have decreased markedly, based on estimated changes in how the return flow varies with depth (more of it is coming back near the surface, implying a stronger subtropical gyre and weaker MOC). One possible mechanism for this would be if the surface water is freshened by increased river flows and ice melt near the limit of its northern extent - fresh water is less dense than salty, hindering convection.]

Anyway, there seem to be three cases to consider:

A) The MOC has decreased and will stay weak (or weaken further) in coming years/decades
B) The MOC has decreased but will increase again in the near future
C) The MOC didn't really decrease at all, the measurements (or their analysis) are inaccurate.

I think everyone agrees that case A implies a strong cooling for the UK in the near future. In fact many people have suggested that the reduction, if real, should have already caused some noticeable cooling, given that the decrease is supposed to have happened over the past several decades (half the measured change was in the interval 1957-1992). My money would have to be on case C, but I'm not sure what Bryden and his co-authors think. His paper certainly mentions the possibility of observational inaccuracy, saying that "the observed changes are uncomfortably close to these uncertainties", but follows this immediately with "but [foo] and [bar] represent strong arguments that the observed changes are robust". But although he argues that the MOC really has decreased, he's not prepared to bet that temperatures will drop. I don't think anyone has proposed a plausible argument for the MOC increasing again soon, if it really has decreased (case B), but it seems to be the only option left open to him.


Brian said...

So did you contact him about betting? That part is a little unclear.

James Annan said...

Yes, I did. I got one very brief reply, and no further response to my follow-up email. He did not say what he really thought about the likelihood of cooling, but it seems safe to conclude that it's small to nonexistent.

It might have been useful for the journalists to ask him the same question prior to running the story, don't you think?