Monday, May 09, 2011

Facebook and Forest Fire Prevention

Less is more when it comes to preventing forest fires, according to a new analysis that explains why aggressive fire fighting in the US may make the risk of severe forest fires worse.

Many people have speculated that land management policies in the US could be contributing to the severity of wildfires that seem to plague the country's broad swaths of wilderness each year. The argument is that routinely putting out small fires leads to the accumulation of unburnt fuel - leaves, sticks, and fallen trees - just waiting for that one, big, late season fire to come along.

A new model presented in a paper published in the journal Physical Review E shows that it's not so much the amount of fuel that's the problem, but instead the connections between flammable areas that are to blame.

University of California researchers (Mark Yoder, Donald Turcotte, and John Rundle) studied forest fires in the same way that other researchers study the flow of rumors through Facebook connections, or the spread of diseases in populations.

If you consider Facebook-spread rumors, for example, the more connections between a group of people, and the larger the number of people in the group, the faster and farther rumors travel. One way to put a stop to a rumor, at least in theory, is to sever some connections (i.e. make some people de-friend a certain number of folks in their Facebook friend list). It's fairly easy to determine how many connections have to be cut in order to make sure rumors peter out and die rather than spreading like . . . well, like wildfire.

Similarly, if you could remove sections of flammable forest litter in a park, you could make sure that the fires can't spread very far. As it turns out, the researchers' model shows that letting small fires burn during the moist seasons each year gets rid of little patches of fuel, which breaks connections between portions of forest that may burn during the dry season.

How do the researchers know their model is accurate? By comparing wildfires in two similar, adjacent regions that have very different land management systems in place: Southern California (wildfire central in the US) and Baja California Norte, Mexico. In Baja, there's rarely any attempt to put out small fires that occur through out the year, and there are very few massive wildfires. Just across the border, small fires are quenched immediately, but raging forest fires make national news nearly very year. In both cases, the frequency and severity of the fires fit well in the new physics model of fires spreading across highly-connected and poorly-connected networks.

Will the new model have any effect on US land management policies? Who knows, but maybe I'll send a complimentary copy of the April issue of Physical Review E to the Bureau of Land Management, considering that they probably don't already subscribe.

No comments:

Post a Comment