Tiltmeters help scientists predict eruptions at Mount St. Helens
Inside the crater of Mount St. Helens, a lava dome was built between October 1980 and 1986 by seventeen eruptions of viscous dacite lava. Each eruption added between 1 and 22 million m3 of new lava to the dome; most of the growth occurred when magma extruded onto the surface of the dome to form lava flows as thick as 40 m and as large as 400 m in diameter. The repeated rise of magma into the dome before the eruptions produced an exceptional pattern of ground tilt on the crater floor that began 2 to 4 weeks before magma extruded onto the dome. This repeated pattern of ground tilt was one of the most reliable measurements of deformation that we used to accurately predict when the Mount St. Helens lava dome would erupt.
The nearly flat crater floor adjacent to the lava dome provided an excellent "platform" on which to measure a change in inclination or "tilt" of the ground during the first few years of dome growth. Tiltmeters had to be installed within a few hundred meters of the vent (located directly beneath the dome) to record significant tilt changes. Several tiltmeters were needed around the dome because the crater floor did not tilt uniformly at all locations. At each tiltmeter site (black canister in foreground), a radio sent data to the USGS Cascades Volcano Observatory, located about 80 km away in Vancouver, Washington.
Tiltmeters detect increasing rates of deformation before eruption
This graph shows a three-month record of tilt from a tiltmeter located less than 100 m north of the dome before the March 19, 1982, eruption (vertical red line). The tiltmeter recorded a change in ground slope in two directions, radial and tangential to the dome. Note the gradual tilt that began in mid-January and accelerated sharply in mid-March. Then, within 30 minutes of the eruption, the radial ground tilt reversed direction; this reversal indicated the ground stopped tilting away from the dome and began to subside rapidly.
Eruption Warnings Based on Monitoring DataBased on tiltmeter records, measurements of dome deformation, and earthquake activity, scientists of the USGS Cascades Volcano Observatory issued eruption warnings in the form of predictions on March 12, March 15 and again on March 19. The eruption began with a short-lived explosion on March 19 at 7:27 P.M.; lava began erupting on top of the dome during the night of March 20.
March 12, 1982 8:00 am:
"Seismicity beneath Mount St. Helens continues at elevated levels, but individual earthquakes are of low magnitude, but earthquakes are of low magnitude. Earthquakes have been occurring in 1- to 2-day-long episodes separated by 1- to 2-day intervals of decreased activity. The earthquakes are occurring between the surface and a depth of about 6 miles. Rates of ground deformation in the crater area have increased during the last 2 weeks, and they are similar to patterns observed before previous dome building events."
"Based on rates of deformation, an eruption is likely within the next 3 weeks. Deformation is confined to the crater areas, suggesting that renewed dome growth will occur. The current seismic patterns differ from any observed before 1980-81 eruptions, however, and raise the possibility of more hazardous various in eruptive behavior. If there were to be any pyroclastic flows, from either an explosive eruption of collapse of the steep north face of the dome, the possibility of rapid snowmelt would be of concern."
March 15, 1982 7:00 pm:
"Accelerating rates of ground deformation in the crater of Mount St. Helens suggest that an eruption, most likely of the dome-building type, will probably begin within 1 to 5 days. Deep earthquakes have almost ceased, and shallow-earthquakes continue at a moderated rate. A further increase is shallow seismicity is likely before the eruption starts."
March 19, 1982 9:00 am:
"Seismicity at Mount St. Helens has increased significantly during the past day. This indicates that an eruption will begin soon, probably within the next 24 hours. The character of both the seismicity and deformation in the crater area indicated that the most likely type of activity is dome growth."