Flow by Ball Philip;

Author:Ball, Philip;
Language: eng
Format: epub
Publisher: Oxford University Press
Published: 2009-01-15T00:00:00+00:00

SELF-ORGANIZED AVALANCHES

The problem with landslides and avalanches is that you can never quite be sure when they will happen. Sometimes they are set off by unpredictable disturbances such as earthquakes—that is how most tsunamis are generated, when seismic tremors send subsea sediments sliding down a slope. But avalanches seem also to have an intrinsic capriciousness. For simple piles of more or less identical grains, we can feel sure that there is trouble in store once the slope exceeds the angle of repose; but even then it is hard to know how big the landslide will be. If the grains are of many different sizes and shapes, or if they have complex frictional properties (as is the case for wet soil or sticky snow crystals), or if the surface on which they rest is rugged, we cannot be sure what to expect. All we know is that when grains are set in motion, we had better watch out.

Does that mean we can make no useful predictions about the timing or the size of avalanches? Not exactly. Rather, it means that avalanche science, like earthquake science, is necessarily statistical: we can’t say exactly what will happen in a particular event, but we can know about the relative probabilities of what might happen.

And in fact, studying landslides this way has proved to be astonishingly productive. For it seems that the humble pile of sand is analogous to a great many ‘catastrophic’ processes that happen in nature, from forest fires to ecological mass extinctions. The key feature of all these processes is that, while they are unpredictable, they are not fully random in the sense that each event happens independently of the others. There is a subtle but very important statistical regularity in such processes, and it is one that connects these seemingly disorderly, unpredictable phenomena to ones that give rise to well-defined patterns. For it appears that landslides, like most of the patterns we have encountered already, are self-organized.

To see what that means, let’s go back to the simple conical sand pile. In 1987, physicists Per Bak, Chao Tang, and Kurt Wiesenfeld at Brookhaven National Laboratory on Long Island, New York, devised a model to describe the way such a pile behaves as it grows by new grains being poured on to the apex. These researchers initially had no intention of studying piles of sand. Rather, they were investigating an aspect of the electronic behaviour of exotic solids that is so recondite I do not even propose to describe it. What they came to realize is that the behaviour of the electrons in these materials can be represented by the behaviour of sand grains in a pile. That is not to say that the electrons themselves form a pile, or anything of the kind. It is a little like the way one can model oscillating chemical reactions by thinking of foxes eating hares, as I described in Book I: the equations describing both behaviours look the same.

Bak, Tang, and Wiesenfeld then considered a pile of sand grains with a specific angle of repose onto which new grains are steadily dropped.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.