Search

Magnetotelluric (MT) data image the bulk resistivity of the subsurface which can be used to infer magma reservoir conditions beneath volcanoes. The bulk resistivity of magma depends primarily on the melt volume fraction, temperature, and water content. These variables are coupled thermodynamically, yet mixing relations for bulk resistivity implicitly treat them as independent. Here, we use a parameterization of the rhyolite‐MELTS thermodynamic model to constrain relationships between melt fraction, temperature, dissolved water content and bulk resistivity for rhyolitic magmas. This method results in MT interpretations which are (a) thermodynamically consistent at near‐equilibrium conditions, (b) independent of temperature and water content estimates derived from erupted products, and (c) able to consider saturated melts containing a volatile (i.e., aqueous fluid) phase. The utility of the method is demonstrated with three case studies of silicic systems: Mono Basin, Newberry volcano and the Laguna del Maule Volcanic Field (LdMVF). The moderately conductive feature at Mono Basin can be explained by under‐saturated partial melt (6–15 vol%) at <775°C, indicating relatively stable magma storage conditions since the last eruption. However, a relatively resistive feature at Newberry Volcano requires lower temperatures (<750°C) than previous estimates, suggesting that the system has cooled since the last eruption. A conductive feature at the LdMVF cannot be explained by saturated or under‐saturated rhyolitic melt and requires additional conductive phases. These results demonstrate the potential of this new method to reduce uncertainty in MT interpretations and highlight the need for additional coupling strategies between petrology, geophysics, and thermo‐mechanical models to better understand magmatic systems. Plain Language Summary: Determining the magma storage conditions beneath a volcano can give insights into the volcanic hazards. The magnetotelluric (MT) method measures the electrical resistivity of the subsurface, which is valuable because molten rock (i.e., melt) is less resistive than solid rock (or crystals). A mixture of melt and crystals will have a particular resistivity value which means that MT data can infer the relative proportion of melt (i.e., melt fraction). Magmas with higher melt fractions are more mobile and more likely to erupt. In addition, hotter melt with more dissolved water is more conductive than cooler, drier melt, so the electrical resistivity can also be used to infer the temperature and water content. However, there are more unknown variables (temperature, melt fraction, and water content) than known variables (observed resistivity), so more constraints are needed. Thermodynamic models indicate that hotter, wetter conditions have higher melt fractions than cooler, drier conditions, so the unknown variables are not independent. We use thermodynamic models to provide additional constraints on the interpretation of magma reservoir conditions from bulk resistivity, which enables us to provide realistic bounds on the range of plausible conditions and better explain the electrical resistivity observed at three different volcanoes. Key Points: A parameterization of the thermodynamic rhyolite‐MELTS model is coupled to empirical relations for the bulk resistivity of rhyolitic magmaThe coupling can significantly alter previous interpretations of magma reservoir conditions from magnetotelluric dataAllows for a comparison between past magma reservoir conditions inferred from erupted products and current magma reservoir conditions [ABSTRACT FROM AUTHOR]

Imaging silicic systems using geophysics is challenging because many interrelated factors (e.g., temperature, melt fraction, melt composition, geometry) can contribute to the measured geophysical anomaly. Joint interpretation of models from multiple geophysical methods can better constrain interpretations of the subsurface structure. Previously published resistivity and shear wave velocity (Vs) models, derived separately from magnetotelluric (MT) and surface wave seismic data, respectively, have been used to model the restless Laguna del Maule Volcanic Field, central Chile. The Vs model contains a 450 km3 low‐velocity zone (LVZ) interpreted as a region with an average melt fraction of 5–6%. The resistivity model contains a conductor (C3) interpreted as a region with a melt fraction >35%. The spatial extents of the LVZ and C3 overlap, but the geometries and interpretations of these features are different. To resolve these discrepancies, this study investigates the resolution of the MT data using hypothesis testing and constrained MT inversions. It is shown that the MT data are best fit with discrete conductors embedded within the larger LVZ. The differences between the MT and seismic models reflect resolution differences between the two data sets as well as varying sensitivities to physical properties. The MT data are sensitive to smaller volumes of extractable mush that contain well‐connected crystal‐poor melt (C3). The seismic data have lower spatial resolution but image the full extent of the poorly connected crystal‐rich magma storage system. The combined images suggest that the LdMVF magma plumbing system is thermally heterogeneous with coexisting zones of warm and cold storage. Key Points: A published velocity model derived from surface wave inversion is used to constrain the 3‐D magnetotelluric inversion at Laguna del MauleDiscrepancies between the velocity and resistivity models are likely due to magma storage conditions rather than inversion nonuniquenessThe MT images a crystal‐poor melt volume within a larger crystal‐rich mush imaged by seismic indicating a thermally heterogeneous system [ABSTRACT FROM AUTHOR]

The Laguna del Maule (LdM) volcanic field comprises the greatest concentration of postglacial rhyolite in the Andes and includes the products of ~40 km3 of explosive and effusive eruptions. Recent observations at LdM by interferometric synthetic aperture radar and global navigation satellite system geodesy have revealed inflation at rates exceeding 20 cm/year since 2007, capturing an ongoing period of growth of a potentially large upper crustal magma reservoir. Moreover, magnetotelluric and gravity studies indicate the presence of fluids and/or partial melt in the upper crust near the center of inflation. Petrologic observations imply repeated, rapid extraction of rhyolitic melt from crystal mush stored at depths of 4–6 km during at least the past 26 ka. We utilize multiple types of surface‐wave observations to constrain the location and geometry of low‐velocity domains beneath LdM. We present a three‐dimensional shear‐wave velocity model that delineates a ~450‐km3 shallow magma reservoir ~2 to 8 km below surface with an average melt fraction of ~5%. Interpretation of the seismic tomography in light of existing gravity, magnetotelluric, and geodetic observations supports this model and reveals variations in melt content and a deeper magma system feeding the shallow reservoir in greater detail than any of the geophysical methods alone. Geophysical imaging of the LdM magma system today is consistent with the petrologic inferences of the reservoir structure and growth during the past 20–60 kyr. Taken together with the ongoing unrest, a future rhyolite eruption of at least the scale of those common during the Holocene is a reasonable possibility. Key Points: Results from a major deployment of seismic instruments at Laguna del Maule volcanic center are presentedWe use a novel application of multiple types of surface‐wave dispersion data to infer 3D shear velocity structureWe present a unified interpretation of results from seismology, geodesy, gravity, magnetotellurics, and petrochronology [ABSTRACT FROM AUTHOR]

Magnetotelluric (MT) data were collected at the Laguna del Maule volcanic field (LdMVF), located in central Chile (36°S, 70.5°W), which has been experiencing unprecedented upward ground deformation since 2007. These data were used to create the first detailed three-dimensional electrical resistivity model of the LdMVF and surrounding area. The resulting model was spatially complex with several major conductive features imaged at different depths and locations around Laguna del Maule (LdM). A near-surface conductor (C1; 0.5 Ωm) approximately 100 m beneath the lake is interpreted as a conductive smectite clay cap related to a shallow hydrothermal reservoir. At 4 km depth, a strong conductor (C3; 0.3 Ωm) is located beneath the western edge of LdM. The proximity of C3 to the recent Pleistocene-to-Holocene vents in the northwest LdMVF and nearby hot springs suggests that C3 is a hydrous (>5 wt% H 2 O), rhyolitic partial melt with melt fraction >35% and a free-water hydrothermal component. C3 dips towards, and is connected to, a deeper conductor (C4; 1 Ωm). C4 is located to the north of LdM at >8 km depth below surface and is interpreted as a long-lived, rhyolitic-to-andesitic magma reservoir with melt fractions less than 35%. It is hypothesized that the deeper magma reservoir (C4) is providing melt and hydrothermal fluids to the shallower magma reservoir (C3). A large conductor directly beneath the LdMVF is not imaged with MT suggesting that any mush volume beneath LdM must be anhydrous (<2 wt% H 2 O), low temperature and low melt fraction (<25%) in order to go undetected. The presence of large conductors to the north has important implications for magma dynamics as it suggests that material may have a significant lateral component (>10 km) as it moves from the deep magma reservoir (C4) to create small, ephemeral volumes of eruptible melt (C3). It is hypothesized that there may be a north–south contrast in physical processes affecting the growth of melt-rich zones since major conductors are imaged in the northern LdMVF while no major conductors are detected beneath the southern vents. The analysis and interpretation of features directly beneath the lake is complicated by the surface conductor C1 which attenuates low-frequency signals. The attenuation from C1 does not affect C3 or C4. At 1 km depth directly beneath LdM, a weak conductor (C2; <10 Ωm) is imaged but is not required by the data. Forward modeling tests show that a relatively large (30 km 3 ), high melt fraction (>50%), silicic reservoir with 5 wt% H 2 O at 2 to 5 km depth beneath the inflation center is not supported by the MT data. However, a smaller (10 km 3 ) eruptible volume could go undetected even with relatively high melt fraction (>50%). The location of large melt regions to the north has important implications for long-term volcanic hazards at LdMVF as well as other volcanoes as it raises the possibility that the vent distribution is not always indicative of the location of deeper source regions of melt. [ABSTRACT FROM AUTHOR]

The location, volume and physical states of magma reservoirs are primary controls on the eruptive behavior of volcanic systems. Fundamental to understanding and monitoring these systems is the ability to identify reservoir size and physical properties, in particular melt fraction which plays an important role in the rheology and stability of a magmatic system. Large silicic volcanic eruptions in the geological record suggest that extensive pockets of melt-rich silicic magma must exist in the subsurface but such melt pockets have not been detected by geophysics. This has led to the question of whether the reservoirs that feed large volcanic eruptions are only melt-rich for a short time and thus would only be detected by geophysics shortly prior to an eruption. Magnetotelluric data measure the electrical resistivity of the subsurface and are sensitive to subsurface fluids and partial melts making it a powerful tool for imaging subvolcanic magma reservoirs. This study examines the ability for magnetotelluric data to accurately estimate melt fraction using both stochastic Bayesian inversion and deterministic regularized inversion. Results from synthetic modelling indicate that magnetotelluric data are best able to predict the melt fraction for the thick melt-rich layer using both inversion methods, though both methods under-estimate the true amount of melt. In addition, magnetotelluric data can accurately detect changes in melt fraction from crystal-rich mush (0.1 melt fraction) to melt-rich magma (0.9 melt fraction) for thick layers. Thickness is a key parameter which provides a method to assess the total volume of melt present, but it is difficult to estimate using smooth regularized inversions. Fixed-dimension Bayesian inversions provide estimates of layer thickness and their uncertainties and it is shown that estimates of total volume of melt are more accurate for both thin (0.2 km) and thick (1 km) layers than individual estimates of melt fraction or thickness alone. The melt fraction of the thin melt-rich layer is under-estimated using both the deterministic inversion method and the most probable solution of the Bayesian inversion. The Bayesian distribution of solutions has long tails which includes the true solution and better represents the uncertainty of the modelled parameters. As we consider past large melt-rich caldera eruptions at Taupo (New Zealand), Toba (Indonesia), Yellowstone and Long Valley (USA), our analysis suggests that thin melt-rich, potentially eruptible zones involved in large (e.g. >500 km3) eruptions could be systematically misinterpreted as thicker non-eruptible crystal-rich mush zones. Testing the sensitivity of MT to detect melt accumulation in active magmatic systems has important implications for volcanic hazard assessment and the use of Bayesian inversions allows for a better understanding of the range of possible interpretations. • Bayesian inversion is used to estimate melt fraction from synthetic MT data. • MT data can detect melt fraction from 0.1 to 0.9 (immobile to mobile) for thick layers. • Smooth inversions often underestimate melt fractions especially for thin layers. • Melt volume and melt fraction is better characterized using Bayesian inversion. • Lack of geophysical evidence for thin melt-rich lenses does not preclude them. [ABSTRACT FROM AUTHOR]