Multidisciplinary Constraints on Magma Compressibility, the Pre‐Eruptive Exsolved Volatile Fraction, and the H 2 O/CO 2 Molar Ratio for the 2006 Augustine Eruption, Alaska

• Published in: Geochemistry, geophysics, geosystems : G3 Vol. 22; no. 9
• Main Authors: Wasser, Valerie K., Lopez, Taryn M., Anderson, Kyle R., Izbekov, Pavel E., Freymueller, Jeffrey T.
• Format: Journal Article
• Language: English
• Published: 01.09.2021
• ISSN: 1525-2027; 1525-2027
• DOI: 10.1029/2021GC009911

Abstract Geodetically modeled reservoir volume changes during volcanic eruptions are commonly much smaller than the observed eruptive volumes. This discrepancy is thought to be partially due to the compressibility of magma, which is largely controlled by the presence of exsolved volatiles. The 2006 eruption of Augustine Volcano, Alaska, produced an eruptive volume that was ∼3 times larger than the geodetically estimated syn‐eruptive subsurface volume change. In this study, we use a multistep methodology that combines constraints from geodetic, volcanic gas, geologic, and petrologic data together with equations relating physical processes to observable parameters. We apply a Monte Carlo approach to quantify uncertainties. Ultimately, we solve for the exsolved volatile volume fraction and the magma compressibility. We estimate Augustine's 2006 pre‐eruptive exsolved volatile phase to be ∼5.5 vol% of the magma at storage depths, yielding a bulk magma compressibility of ∼3.8 × 10 −10  Pa −1 . We develop a novel approach to estimate the H 2 O/CO 2 ratio of the syn‐eruptive gas emissions in the absence of direct H 2 O emission measurements which are hard to obtain due to the high background levels in ambient air. We find a best‐fit H 2 O/CO 2 molar ratio of 29. We also investigate the effects of applying different equations of state to our model. We find that the Ideal Gas Law might be used as a first approximation due to its simplicity; however, it overestimates volatile density and compressibility significantly at storage depths. This project capitalizes on the insights that can be gained by integrating multidisciplinary data with models of physical processes. Plain Language Summary Estimates of the reservoir volume change (deflation of the magma chamber) occurring during volcanic eruptions are often smaller than the volume of magma that ultimately erupts. This discrepancy can be explained in part by the presence of gas bubbles in the magma chamber which make the magma compressible. Compressible magma can hide the full extent of magma moving within a volcano. In this study, we used gas emissions, deformation, petrologic, and geologic observations of the 2006 eruption of Augustine Volcano, Alaska, to calculate that gas bubbles made up ∼5 vol% of magma in the chamber. Even though the volume fraction of gas bubbles is small, they caused the magma to be compressible enough for the volume of eruptive products to be three times larger than the reservoir volume change estimated from deformation data. In addition, we were able to calculate the total mass of water emitted by the volcano during the eruption. This is an important parameter that is challenging to directly measure due to large quantities of water in ambient air. This project underlines the importance of including magma compressibility and its effects on eruption size to better forecast the characteristics of impending eruptions. Key Points Even small magma compressibilities can lead to eruption volumes several times larger than syn‐eruptive geodetic volume changes We present a new technique to estimate syn‐eruptive surface H 2 O/CO 2 composition when in situ H 2 O emission measurements are unavailable We calculate an exsolved volatile phase of ∼5.5 vol% and a molar H 2 O/CO 2 ratio of 29 for the 2006 Augustine eruption