Fracture formula yields volcanic forecasts.

Fracture formula yields volcanic forecasts

The equation [Omega]--[infinity][Omega]--A = 0 won't win any awards for easy reading, but it just might save some lives. New research suggests that this mathematical expression can help experts predict when a restless volcano will blow its top.

The equation represents a general law describing how various materials -- ranging from metal to plastic to ice -- fracture under stress. But it also appears to have promise for indicating when an imminent volcanic eruption will occur, report Barry Voight and Reinold R. Cornelius of Penn State University in University Park.

Before volcanoes blow their tops, they generally offer some sort of geophysical clues, such as earthquakes and bulging of the land surface. Voight and Cornelius demonstrate how volcanologists can predict the timing of an eruption by feeding a particular type of earthquake information into the fracturing equation.

In the April 25 NATURE, they apply their technique to four cases studies: two small. nonexplosive eruptions from Mount St. Helens in 1985 and 1986, and two explosive bursts from Alaska's Mount Redoubt in 1989 and 1990. In their analysis, [Omega] stands for average seismic ground movement; A and [infinity] represent empirical constants. plugging in data on earthquake activity preceding the four eruptions, Voight and Cornelius demonstrate that this method could have predicted a narrow "window" for eruption time several days in advance. In the case of Mount Redoubt's violent eruption on Jan. 2, 1990, the technique showed predictive skill with information available a full week before the eruption, they report.

Some volcano experts remain skeptical of the method's usefulness, noting that Voight and Cornelius have tested it only by "hindcasting" past events, and have yet to demonstrate its accuracy before an eruption has occurred. "I think the jury's still out on whether we can use it as a predictive tool," says Alaska State Seismologist John Davids in Fairbanks.

But it does work, the technique promises to make eruption predictions much more rigorous and more accurate, says Stephen D. Malone, a seismologist at the University of Washington in Seattle. Scientists currently use less objective tests for forecasting. "It's a lot of judgment calls -- the seat-of-your pants sort of stuff," Malone says. He adds that he plans to incorporate the new method in his analyses of seismic information for Mount St. Helens.

Voight and Cornelius stress that their approach will not work for all volcanoes under all circumstances, and that researchers will obtain the best predictions by using several different techniques together.

For instances, this particular mathematical method could not have predicted the devastating May 18, 1980, eruption of Mount St. Helens. The blast started after an earthquake triggered a landslide, uncorking the pressure building within a bulging flank of the volcano. Because the landslide short-circuited the normal eruptive process, the volcano didn't provide any of the seismic clues normally used in predicting eruptions--and Voight and Cornelius' method hinges on many of those sam seismic changes.