Thursday, October 29, 2009

The rain in Spain falls mainly on the grain

We have several times in this blog pointed out the problem of too much hubris in science, the attitude that we already know everything important and therefore can be very sure of what we say. But maybe, like Eliza Doolittle in My Fair Lady, we scientists could use a bit of polishing before we speak so confidently, if we think we deserve to be taken seriously. Maybe this is an example (at any rate, it's interesting):

The Oct 21 episode of the BBC World Service radio program, Discovery, included a discussion of 'biopreciptation' (we'd link you to the podcast, but it has already been taken down; here's a link to an older version of the story on BBC Radio 4). The term 'bioprecipitation' was coined some 25 years ago by Dr David Sands at Montana State University, though it pretty much fell on deaf ears. With increasing evidence now, however, the idea that the biosphere may have a significant impact on climate is gaining followers.

Most rain and all snow falls as ice, but the water in clouds doesn't form frozen crystals at 32 degrees F, but rather some degrees colder, and almost always around ice-nucleating particles. These particles can include dust and pollutants, which have been the text-book explanation. But Dr Sands' and colleagues found, as reported in a Science paper last year (Ubiquity of Biological Ice Nucleators in Snowfall, Chirstner et al., Science 29 February 2008: Vol. 319. no. 5867, p. 1214), and recently on BBC radio, that some bacteria serve the same function, but they do so at warmer temperatures.

Christner et al. have examined snow and rain for the presence of nucleators. As reported in the Science paper, they found that 70 to 100% of the nucleators in snow that were active at higher temperatures were biological, and a majority of those were live bacteria. It turns out that most of these bacteria are plant pathogens, specifically Pseudomonas syringae. They infect, but don't necessarily kill plants, spending much of their time there, but they can be swept up into the atmosphere when conditions are favorable, where they drift with the winds aloft, later to form the core of rain drops or snowflakes, and then fall back down to Earth. Those that land on plants reproduce there, and then can be blown back up into the atmosphere, where they can become part of the precipitation cycle all over again.

So, these plant pathogens may contribute to rainy -- or dry -- conditions anywhere in the world, and there may be feedback loops. If an area is suffering from drought, vegetation may be sparse, and thus populations of these rain-inducing bacteria may be sparse, thus leading to less rain, less vegetation, and so on. It's possible that this kind of feedback is in part responsible for the droughts in Africa and Australia, according to Sands and colleagues. That means they can have continental, and hence perhaps in turn even global climate effects.

Christner et al. conclude their paper by saying:
Unearthing a role for biological [ice-nucleation] in the precipitation cycle has implications for deciphering feedbacks between the biosphere and climate, improving climate forecast models, and understanding atmospheric dissemination strategies of plant pathogens and other microorganisms.
This is a beautiful example of the interconnection of the biota on Earth in unexpected ways. But there's a further twist to the story. The ice-nucleating property of this bacterium is known; it's a protein on the cell-surface of the P. syringae bacterium. Intact bacteria, then, at the right temperature and given wet conditions, can nucleate ice formation directly on the surface of plants, causing potentially costly frost damage to crops. A form of the bacterium without the ice-nucleating surface protein, called 'ice-minus', occurs in the wild, but a lot more of them have been genetically engineered and introduced into fields where it's hoped that will have a man-made selective advantage and out-compete their ice-nucleating cousins--to do less crop damage.

Nice, neat story and a triumph for science and (we hate to admit it) big agribusiness, too. Right? Not necessarily!

Given the role of P. syringae in precipitation cycles worldwide, in ways not yet fully understood or even characterized, the possibility of disturbing it is rather alarming. If we seed the atmosphere with ice-minus bugs, and they out-compete the currently-prevalent strain, then there will be less crop damage but there will also be less cloud and rain formation. Cloud and rain formation affects climate, climate affects areas of drought and surfeit of rain, which can affect both climate on a large scale, and of course agricultural productivity. And if the patterns are altered the effect could, like the famous Lorenz butterfly effect of chaos theory, proliferate on a much larger scale.

How large? Enough to affect global climate or climate cycles? Could crop patterns already have had some effects on this interacting system, that we never suspected? Could this go back into history? Could global warming or other recent climate changes be due not to greenhouse gases alone but to agricultural paterns or technology in some significant way? These effects could, of course, turn out to be trivial, "a lot o' nowt wi' no clout" as Eliza Doolittle might say. It could even turn out that 'ice-minus' bacteria really do have a good effect without the bad. On the other hand....

The fact that bacteria are frequently the condensation nuclei around which ice crystals form and hence that are the source of cloud and precipitation illustrates the concept of 'dark matter' that first arose in physics and refers to things not suspected by science until in some way (usually by chance) they are discovered. Before that, we have our theories, that are shoes into which science tries to force the feet of data that we already know about, sometimes causing serious bunions. The lessons of the history of science are a repeated warning to us. We can't develop theories resting on evidence we don't know about, of course, and not everything we discover turns out to be important. But we can't be too sure of ourselves.

Like Eliza Doolittle, we can't win a prized companion if we can't speak the right language. Her pompous tutor Henry Higgins was cock-sure of what that right language was, but the outcome was ambiguous--as it often is in science, where the right language may always be changing, and we need to look constantly to Nature to learn what that is.

No comments: