Geodetic observations during the 2009 eruption of Redoubt Volcano, Alaska

• Published in: Journal of volcanology and geothermal research 2013-06, Vol.259, p.115-132
• Main Authors: Grapenthin, Ronni, Freymueller, Jeffrey T., Kaufman, Alexander Max
• Format: Article
• Language: English
• Subjects: Crystalline rocks; Earth sciences; Earth, ocean, space; Eruption; Exact sciences and technology; Geodesy; Igneous and metamorphic rocks petrology, volcanic processes, magmas; Plume detection; Redoubt; Source modeling; Volcano
• ISSN: 0377-0273; 1872-6097
• DOI: 10.1016/j.jvolgeores.2012.04.021

In March 2009 Redoubt Volcano, about 160km to the SW of Anchorage, Alaska, began its most recent explosive eruption. Deformation induced by this event was recorded by a GPS campaign network consisting of 14 benchmarks, which had been established in 1991 after the previous eruption. The network was partially reoccupied in 2001 and 2008 and no volcanic deformation was detected during that period. In response to precursory unrest starting in January 2009, the Alaska Volcano Observatory temporarily deployed continuously recording GPS instruments at four of the campaign benchmarks only days before the onset of explosive activity in March 2009. The only GPS instrument recording continuously during the months prior to the eruption was the Plate Boundary Observatory (PBO) station AC17, about 28km northeast of the volcano's summit. Data from this station reveals subtle motion radially outward from the volcano beginning as early as May 2008, which reversed with the onset of explosive activity. Using simple analytical models we link the precursory activity to a point source intrusion of 0.0194 0.03400.0092km3 in volume at 13.50 17.3310.17kmbelowsealevel (bsl, superscripts and subscripts refer to upper and lower ends of confidence intervals at the 95% level). During the explosive phase about 0.05 >0.10.028km3 of magma was evacuated from a prolate spheroid with its centroid at 9.17 15.176.92kmbsl, a semimajor axis of 4.50 >10.001.25km length and a semi-minor axis of 0.475 >4.00.3km. The effusive activity is inferred to come from the same source, decreasing in volume by 0.0167 0.02280.0106km3. Including observations from seismology and petrology, we hypothesize a mid-crustal two reservoir system with material sourced from >20km flowing in at about 13.5km depth and reheating residual material in the proposed spheroid. The mixture migrated to shallower depth (2–4.5kmbsl) and reheated material there. As this residual magma erupted, it was replaced by the material from the spheroidal reservoir at 7–11.5km depth, which renders the shallow source undetectable for geodetic instruments. In addition to long term displacements we investigate sub-daily kinematic positioning solutions and find that large, short-term positioning offsets correlate with large explosive events. Spikes in phase residuals plotted along the sky tracks of individual satellites can be related to individual plumes given favorable satellite–station-geometry, which may be of use in volcano monitoring. ► Deform