Impulse-response Experiments for Integrating Space-borne, Field, and Laboratory Measurements of Magmatic Systems

Friday, 12 January 2018: 09:00
Salon Quinamavida (Hotel Quinamavida)
Tim Masterlark, Sui Tung, Gokce K Ustunisik and Michael Baranowski, South Dakota School of Mines and Technology, Rapid City, SD, United States
Abstract:
The upward migration of magma drives the eruption cycle of an active volcano. This migration (the impulse) and surrounding structure of the volcano produce a deformation signature (the response). The characteristics of this impulse-response system may be inferred independently from space-borne, field, or laboratory measurements. The ability to integrate multiple types of information in a self-consistent model is the key to advancing our understanding of magmatic systems. We examine an active magmatic system using numerical impulse-response experiments to investigate how geodetic, seismic, and petrologic data contribute to our understanding of the system, when considered independently or simultaneously. The long repose times of silicic volcanoes prevent us from capturing a complete eruption cycle with space-borne geodetic data. However, we can appeal to the shorter repose times of basaltic systems to examine co-, post-, and inter-eruption behaviour with geodetic data. We present interpretations of geodetic data spanning the most recent eruption cycle of Okmok volcano, Alaska. If we consider only the geodetic data and customary elastic models, the observed deformation linearly tracks magma flux for a relatively shallow magma chamber. Previous petrologic studies of melt inclusions and volatiles also suggest a shallow magma chamber. Adding seismic tomography provides a similar interpretation, but with a different linear function because the shallow low velocity zone requires a deeper magma chamber. New petrologic studies of melt inclusions in plagioclase, which account for volatiles (CO2) in the shrinkage bubbles, also suggest a deeper magma chamber. Finally, models that include a thermally-weakened material surrounding the magma chamber predict a nonlinear response to magma injection. In this case, the transient deformation is the result of a series of discrete magma impulses modulated by a viscoelastic country rock, having a distribution of material properties and a deep chamber, as required by seismic tomography. Preliminary results suggest that the observed deformation is caused by five discrete magma flux episodes. These results will be verified against forthcoming characterizations of plagioclase zoning, which records the timing and nature of perturbations in the system (see figure).