Eruption Clues from a Humming Volcano

Shishaldin Volcano is a snow-covered cone with a point that reaches 9,414 feet above sea level. Located on Unimak Island in the Aleutians, Shishaldin did a curious thing before it exploded in 1999—it hummed.

Shishaldin hummed for half a day in April 1999 before the volcano blasted an ash plume ten miles into the sky. The voice of the volcano may have much to tell people who study volcanoes.

Jackie Caplan-Auerbach is one of those people. She's a post-doctoral researcher with the Alaska Volcano Observatory in Fairbanks. She came to town to interpret Shishaldin's mutterings. Another researcher, Milton Garces, now of the University of Hawaii, had installed an industrial microphone called a pressure sensor on the northern flank of Shishaldin in 1997; the device recorded the volcano's wheezes, belches and groans during its eruption two years later.

Caplan-Auerbach hunkered down and studied the recordings of the industrial microphone, which captures sound waves on and around the mountain. Shishaldin hummed at a frequency of about 2-3 Hertz, a buzz too low for the human ear to detect but not for the microphone. The inner workings of the volcano caused the hum by governing the steady release of gas.

"Whatever was generating the sound was doing it for 13 hours," Caplan-Auerbach said. "Then it stopped humming. The pressure sensor didn't hear anything for four minutes."

When the hum stopped, the ground started shaking. Shishaldin erupted explosively, blowing ash thousands of feet into to sky. Wind blew most of the ash to the south of the mountain and a flow of hot mud just missed the microphone and a seismometer next to it, allowing the equipment to continue sending radio signals to Cold Bay, then on to the Fairbanks and Anchorage offices of the Alaska Volcano Observatory.

Is Shishaldin's hum a dependable predictor of an explosive eruption or a one-time event? Scientists don't know, because so few microphones are installed on volcanoes, so there's not much volcano noise for comparison.

Scientists primarily use seismometers to detect shaking within volcanoes and satellites to measure the heat of volcano vents. Alaska Volcano Observatory researchers have installed seismometers on 23 volcanoes in Alaska. Because seismometers have produced such dependable results and data lines on remote Alaska volcanoes are limited, volcanologists have installed microphones on only two Alaska volcanoes, Shishaldin and Pavlof.

Steve McNutt of the Alaska Volcano Observatory, a research professor at the Geophysical Institute, thinks the untainted signals of industrial microphones make them deserve a greater role in Alaska volcano monitoring.

"When measuring the vibration of the ground with seismometers, waves start to move away through different earth material and you get a fairly complicated group of waves," he said. "Signals through the air are much cleaner and much more representative of what's happening at the source."

In the past, acoustic monitoring of volcanoes was what Caplan-Auerbach called "icing on the cake." This summer, Alaska volcanoes will get a bit more icing, when Caplan-Auerbach, McNutt, and others install a microphone along with a seismic network on Okmok Volcano.