Cordilleran Section - 103rd Annual Meeting (4–6 May 2007)

Paper No. 6
Presentation Time: 8:00 AM-6:00 PM

ESTIMATES OF GAS FLUX FROM INFRASONIC SIGNALS AT AUGUSTINE VOLCANO DURING THE JANUARY 2006 ERUPTION: IMPLICATIONS FOR ERUPTION PLUME TYPES


FERNANDES, Jennifer J., BELLESILES, Anna K. and CAPLAN-AUERBACH, Jacqueline, Geology Department, Western Washington University, 516 High St. MS 9080, Bellingham, WA 98225, sinfathisar@comcast.net

Augustine volcano, located in the Cook Inlet, Alaska, is an active stratovolcano with six major past eruptions (1883, 1935, 1963-1964, 1976, 1986, and 2006) of primarily andesitic to dacitic composition. The 2006 eruption was extensively monitored by the Alaska Volcano Observatory and yielded a range of data including pressure sensor data for the entirety of the eruption. Most previous studies of volcanic infrasound focus primarily on strombolian-type eruptions, whereas Augustine's eruptive style typically includes a vulcanian phase followed by an effusive phase. The explosive phase of the eruption began 11 January 2006, followed by a period of discrete eruptive blasts ending 28 January 2006.

This study examines the infrasonic record during these eruptive blasts. Two signal types observed in the acoustic record are characterized as impulsive events and emergent events. We use the infrasonic records to estimate the gas flux of specific eruptive phases during the eruption. The method involves first calculating the acoustic power of the signal, from which gas velocity may be determined. Gas velocity can then be used to find the rate of gas outflow. Estimated gas velocities range from 50-300 m/s at the vent with gas flux on the order of 103 to 104 m3/s. Preliminary estimates of gas flux for the explosive phase of the eruption are consistent with those determined for the 1999 eruption of Shishaldin volcano in the Aleutian Islands. These values assume the observed infrasonic signal represents a purely volcanic source, therefore any contribution from wind or coeruptive processes such as pyroclastic flows will result in an overestimate of gas velocity and flux. In general, maximum gas velocity and gas flux estimates yield higher values than those calculated from plume heights assuming a sustained eruption column. This suggests that the majority of eruptive blasts were discrete events and that these plumes were not continuously fed at the source.