Can they know when Mount St. Helens will blow?

Answers to your questions about the news.
Oct. 5 2004 1:15 PM

How To Predict a Volcanic Eruption

Can they know when Mount St. Helens will blow?

Mount St. Helens: Clear your calendar
Mount St. Helens: Clear your calendar

Scientists monitoring Mount St. Helens said last week that there was a 70 percent chance the mountain would erupt within the month. This weekend, serious tremors followed a small explosion Friday that sent steam 10,000 feet into the air, and the team at the Cascades Volcano Observatory now says it expects a significant eruption. How do scientists predict volcanic eruptions? And how did they calculate that 70 percent figure?

Scientists with the U.S. Geological Survey consider a volcano's geological profile and history, data from tests monitoring seismic activity, whether it is releasing gas, and any deformations of the mountain's surface. They're concerned with the frequency, location, and size of earthquakes; whether the earthquakes suggest that magma is moving underneath the mountain; whether the magma involved is old or fresh; whether the volcano is emitting gas, suggesting magma rising to the surface; whether the ground is swelling or compacting ("deforming," if you're a geologist), and so forth. Basically, the more the mountain shakes, the walls warp, and gas rises, the more likely the mountain will erupt.


How did the USGS arrive at that 70 percent figure, then? Scientists gathered as much data as possible to create the most accurate picture of what they believed was happening. They then compared it with a database of past volcanic activity around the world. In 70 percent of cases in which a volcano showed the same kind of activity that Mount St. Helens was displaying last week, it erupted.

The degree of confidence with which scientists make their probability estimates can vary, depending on the intensity and reliability of the data. In the case of Mount St. Helens, they felt that thousands of small earthquakes at a rate of four or five an hour was a very strong indicator of a shift in the volcano's activity, whereas GPS tests showed no significant swelling of the mountain's flanks. "Sometimes these are seat of the pants estimates," said Steve McNutt, a volcano seismologist at the Geophysical Institute of the University of Alaska Fairbanks. The system works, though: The USGS has accurately predicted every eruption of Mount St. Helens since 1980. (This does not, however, mean they can predict exactly when it will explode—though there were clear signs of impending eruption for months in 1980, the explosion came after a brief calm and caught many by surprise, killing 57.)

Now, the data indicates an imminent eruption. After the small eruption on Friday—Mount St. Helens' first in 18 years—the seismic activity has ramped up, with one tremor lasting 50 minutes. This and new test results have prompted the team at the Cascades Volcano Observatory to raise the alert at Mount St. Helens to Level 3—the most severe warning of danger. That hasn't scared away the voyeurs; thousands have flocked to the mountain, hoping to see the show.

Explainer thanks Cynthia Gardner and Jacob Lowenstern  of the Cascades Volcano Observatory and Steve McNutt of the Alaska Volcano Observatory.

Louisa Thomas is on the editorial staff of The New Yorker.