Most geologists study processes that occurred millions, even billions, of years ago. Not so geologists such as ASU’s Stanley Williams, who specialize in volcanic eruptions.

“What I’m doing is in a completely different time frame,” Williams says. “It’s the last 30 seconds or the next one day. It’s just a different style of geology.”

That style of geology often puts scientists in harm’s way. Eleven of them have been killed in eruptions since 1991. Williams almost became a victim himself at the Galeras volcano in South America, which killed six of his colleagues in January 1993.

Williams got caught with his colleagues in a small but lethal explosion. He sustained a fractured skull, a broken jaw, and burns to his back. Both his legs were broken. Twelve surgeons operated on him over a period of 15 months following the accident, but his recovery is nearly complete.

Now, Williams is working with two companies to develop instruments that would allow him to collect critical gas-emissions data from an active volcano at a safe distance. His collaborators are Barringer Research Ltd. of Toronto, Canada, and Transducer Research Inc. of Naperville, Ill.

Understanding the behavior of gases at a volcano could potentially lead to better eruption-forecasting techniques. The most dominant volcanic gases are water vapor, carbon dioxide, sulfur dioxide, hydrochloric acid, and hydrofluoric acid.

“When you pick up a champagne bottle, there aren’t any bubbles, and the glass is very heavy,” Williams says. “If you pull the cork out quickly, the champagne will explode as a big frothy, foamy eruption.”

Why? “Because you took the pressure off the fluid, the champagne,” Williams explains. “The gases, mostly carbon dioxide that is dissolved in the champagne, are confined by the heavy, strong bottle and a cork.”

Volcanic eruptions work in a similar fashion. A magma body forms about a hundred miles beneath a volcano. Lighter than the surrounding rocks, the magma begins to rise. The pressure that keeps the gases dissolved drops, allowing them to boil off. Carbon dioxide boils first, then sulfur dioxide.

“We’ve seen various eruptions around the world where very precisely, for 12 to even 48 hours, the sulfur dioxide surges up from a hundred tons per day to a thousand tons per day, and then there comes the eruption,” Williams says.

“That’s ideal, a wonderful thing for forecasting an eruption. If carbon dioxide really is more dependent on pressure, then it should be coming out days or weeks before the eruption. The ratio between carbon dioxide and sulfur dioxide ought be a powerful, better way to forecast an eruption.”

Williams began studying volcanic gases in the late 1970s with his mentor, Dick Stoiber, at Dartmouth College. Stoiber had been measuring sulfur-dioxide emissions at volcanoes with a correlation spectrometer (COSPEC) since 1972.

The instrument was manufactured by Barringer Research to monitor pollution from coal-fired power plants and smelters. Because sulfur dioxide absorbs ultraviolet rays, COSPEC can measure the compound by looking at the sky.

“We’ve proven over and over that we can make remote sensing of gases and see patterns, but they’ve changed with time,” Williams says.

Stoiber and Williams found virtually no sulfur dioxide emitted by Mount St. Helens during its early eruptions in 1980. That led to an uncomfortable situation, because the scientists were taking their measurements from an airplane provided by NBC.

At first they thought the instrument was broken. Not wanting to admit this to the national press, Stoiber and Williams began speaking to one another in Spanish. They stayed up all night making sure the instrument worked. It turned out to be an important measurement. The measurement showed that a volcano’s powerful explosions could be caused by boiling, splashing ground water.

The resourceful Stoiber focused on sulfur-dioxide emissions because that’s what the COSPEC was designed to do. Williams and Barringer now are developing a GASPEC to measure the more important carbon dioxide in the same way.

Williams and Transducer Research had the other instrument, a prototype volcanic gas monitor, tested at a South American volcano in early 1993.

The monitor is contained in a lightweight, inexpensive box that is left inside a volcanic crater. The device instantly monitors and analyzes the gases. Data is sent via radio to a remote site for about six months.

“When you’ve got a nasty, threatening, dangerous volcano, you go to the volcano once. You add this instrument,” Williams says. “You don’t have to go back every other day or every week to collect gas samples. This thing does. When it gets destroyed, well, you’ve lost a $5,000 piece of equipment and not your life.”

Understandably, the gas monitor revealed some problems during its first test in the hellish environment of a volcanic crater.

“The gases are very nasty acids. Lethal content,” Williams explains. “You breathe the gas, it’ll knock you flat on your back. All of the instrument’s components corroded away and failed in a short time.”

The science of volcanoes is young but maturing, thanks in part to instrumentation advances. About 25 years ago, volcanology textbooks contained little more than pretty pictures.

But today’s volcanologists are going beyond simply describing the physical properties of a volcano. Using COSPEC and other instruments, they are trying to create numerical models that explain how volcanic processes work.

“That’s a big deal,” Williams says. “We need numbers.”—Steve Koppes

Go to: Physical Science: Geology

Home Page | Arts | Business | Education | Engineering | Health | Life Sciences | Physical Sciences | Social Sciences

Who we are | Contact Info