Earthquake Forecasts Getting Better?

from the we-hope-it's-true dept

Predicting earthquakes has never been what you would call an exact science, so claims of newfound accuracy in this area simultaneously fill us with hope and doubt. Last year some researchers said they'd developed a method that could predict when and where a quake would hit within a nine-month range. The results have been inconclusive: It apparently worked with two major quakes, but failed a third time. Now the U.S. Geological Survey is getting into the act, giving Californians probability forecasts through the USGS website for 24-hour periods. The goal, they say, is not to predict earthquakes as much as warn people of the probability. It will be interesting again to see how well these statistics bear out in reality (this system appears different than the one discussed last year). Still, it's hard to imagine people checking the USGS site all the time for their local earthquake forecasts, let alone knowing what to do with the various warning levels (terror alerts, anyone?). Ironically, the system might be more valuable after an initial quake, when people can log on and find out the potential for aftershocks. Even better, the USGS could improve the dissemination of forecasts with automated notifications, maybe via email or text message. After that, the only thing missing would be a Google Maps hack.