Magma injection into a long-lived reservoir to explain geodetically measured uplift: Application to the 2007-2014 unrest episode at Laguna del Maule volcanic field, Chile

Large-scale demonstration of machine learning for the detection of volcanic deformation in Sentinel-1 satellite imagery

Radar (SAR) satellites systematically acquire imagery that can be used for volcano monitoring, characterising magmatic systems and potentially forecasting eruptions on a global scale. However, exploiting the large dataset is limited by the need for manual inspection, meaning timely dissemination of information is challenging. Here we automatically process ~ 600,000 images of > 1000 volcanoes acquired by the Sentinel-1 satellite in a 5-year period (2015-2020) and use the dataset to demonstrate the applicability and limitations of machine learning for detecting deformation signals. Of the 16 volcanoes flagged most often, 5 experienced eruptions, 6 showed slow deformation, 2 had non-volcanic deformation and 3 had atmospheric artefacts. The detection threshold for the whole dataset is 5.9 cm, equivalent to a rate of 1.2 cm/year over the 5-year study period. We then use the large testing dataset to explore the effects of atmospheric conditions, land cover and signal characteristics on detectability and find that the performance of the machine learning algorithm is primarily limited by the quality of the available data, with poor coherence and slow signals being particularly challenging. The expanding dataset of systematically acquired, processed and flagged images will enable the quantitative analysis of volcanic monitoring signals on an unprecedented scale, but tailored processing will be needed for routine monitoring applications.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s00445-022-01608-x.

ADTC-InSAR: a tropospheric correction database for Andean volcanoes

Monitoring geophysical hazards requires a near real-time response and precise interpretation of InSAR data, typically recording minute surface deformations. Accurate tropospheric adjustment is an essential aspect of InSAR processing. This study provides a free database of ready-to-use Tropospheric Correction for InSAR for the three volcanic zones from north to south of the Andes. Average Daily Tropospheric Correction for InSAR (ADTC-InSAR) is a collection of average daily tropospheric delay matrices created using ECMWF re-analysis of the global atmosphere and surface conditions (ERA5) as atmospheric data and TRAIN software. The construction method and annual variation according to the climatic zones are provided, and its effectiveness is evaluated. ADTC-InSAR facilitates the generation of tropospheric corrections in InSAR with easy access, fast application, and accuracy comparable to TRAIN. Its purpose is to serve as a starting point for tropospheric correction in the event of emergency response to extreme occurrences and as a reference for other research and academic objectives.

Measurement(s)	Average daily tropospheric delay
Technology Type(s)	Matlab m file
Factor Type(s)	atmospheric pressure • atmospheric temperature • relative humidity • Digital Elevation Model
Sample Characteristic - Environment	area of barren land
Sample Characteristic - Location	Andean Region

A free-geometry geodynamic modelling of surface gravity changes using Growth-dg software

Globally there is abundant terrestrial surface gravity data used to study the time variation of gravity related to subsurface mass and density changes in different geological, geodynamical and geotechnical environments. We present here a tool for analysing existing and newly acquired, 4D gravity data, which creates new findings from its reuse. Our method calculates in an almost automatic way the possible sources of density change responsible for the observed gravity variations. The specifics of the new methodology are: use of a low number of observation points, relatively small source structures, low signal/noise ratio in the data, and a free 3D source geometry without initial hypothesis. The process is based on the non-linear adjustment of structures defined by aggregation of small cells corresponding to a 3D section of the sub-floor volume. This methodology is implemented in a software tool, named GROWTH-dg, which can be freely downloaded for immediate use, together with a user manual and application examples.

The dynamics of large silicic systems from satellite remote sensing observations: the intriguing case of Domuyo volcano, Argentina

Silicic magmatic systems are the most dangerous volcanoes on Earth, capable of large and catastrophic eruptions, yet their low eruptive frequency makes it challenging to interpret their short-term unrest. Here we present a decade-plus analysis that integrates, for the first time, time series of satellite interferometric synthetic aperture radar (InSAR) surface deformation and satellite thermal infrared edifice-scale surface warming at a large silicic system: Domuyo volcano, in Argentina. We find that deformation and warming are highly correlated, and depending on the sign and lag between the time series, either shallow sealing or magma influx could drive Domuyo’s ongoing inflation (~ 0.15 m/year; from an InSAR-derived tabular source, ~ 11 × 8 × 1 km; ~ 6.5 km depth; ~ 0.037 km³/year volume-change rate) and warming (0.3–0.4 °C/year). This study shows the potential that combined satellite surface deformation and edifice-scale surface warming time series have on assessing the physical mechanisms of silicic volcanic systems and for constraining deterministic models.

Thoughts on the criteria to determine the origin of volcanic unrest as magmatic or non-magmatic

As our ability to detect volcanic unrest improves, we are increasingly confronted with the question of whether the unrest has a magmatic origin (magma on the move) or a non-magmatic origin from a change in the hydrothermal system (fluids that are not magma on the move) or tectonic processes. The cause of unrest has critical implications for the potential eruptive hazard (e.g. used in constructing Bayesian Event Trees), but is frequently the subject of debate, even at well-studied systems. Here, we propose a set of multi-disciplinary observations and numerical models that could be used to evaluate conceptual models about the cause of unrest. These include measurements of gas fluxes and compositions and the isotopic signature of some components (e.g. H₂, He, C, SO₂, H₂ O, CH₄ and CO₂), the spatial and temporal characteristics of ground deformation, thermal output, seismicity, changes in gravity, and whether there is topographic uplift or subsidence spanning hundreds to thousands of years. In several volcanic systems, both magmatic and non-magmatic unrest is occurring at the same time. While none of these observations or models is diagnostic on its own, we illustrate several examples where they have been used together to make a plausible conceptual model of one or more episodes of unrest and whether eruptions did or did not follow the unrest. This article is part of the Theo Murphy meeting issue 'Magma reservoir architecture and dynamics'.

The lateral growth and coalesence of magma systems

Thermal and mechanical models of magma reservoir growth need to be reconciled with deformation patterns and structural relationships observed at active magma systems. Geophysical observations provide a series of short time-scale snap-shots (10⁰-10² years) of the long-term growth of magmatic bodies (10³-10⁶ years). In this paper, we first review evidence for the growth of magmatic systems along structural features and the associated deformation patterns. We then define three distinct growth stages, (1) aligned melt pockets, (2) coalesced reservoirs, (3) highly evolved systems, which can be distinguished using short-term surface observations. We use two-dimensional thermal models to provide first-order constraints on the time scales and conditions associated with coalescence of individual magma bodies into large-scale reservoirs. We find that closely spaced intrusions (less than 1 km apart) can develop combined viscoelastic shells over time scales of 10s kyr and form laterally extensive mush systems over time scales of 10-100 kyr. The highest temperatures and melt fractions occur during a period of thermal relaxation after melt injection has ceased, suggesting that caldera-forming eruptions may preferentially occur long after the main intrusive activity. The coalescence of eruptible melt-rich chambers only occurs for the highest melt supply rates and deepest systems. Thus, these models indicate that, in most cases, conductive heat transfer alone is not sufficient for a full coalescence of magma chambers and that other processes involving mechanical ruptures and mush mobilization are necessary; individual melt lenses can remain isolated for long periods within growing mush systems, and will only mix during eruption or other catastrophic events. The long-term history of the magmatic system is therefore critical in determining rheological structure and hence short-term behaviour. This framework for the development of magmatic systems in the continental crust provides a mechanical basis for the interpretation of unrest at the world's largest volcanoes. This article is part of the Theo Murphy meeting issue 'Magma reservoir architecture and dynamics'.

Geomorphic expression of rapid Holocene silicic magma reservoir growth beneath Laguna del Maule, Chile

Large rhyolitic volcanoes pose a hazard, yet the processes and signals foretelling an eruption are obscure. Satellite geodesy has revealed surface inflation signaling unrest within magma reservoirs underlying a few rhyolitic volcanoes. Although seismic, electrical, and potential field methods may illuminate the current configuration and state of these reservoirs, they cannot fully address the processes by which they grow and evolve on geologic time scales. We combine measurement of a deformed paleoshore surface, isotopic dating of volcanism and surface exposure, and modeling to determine the rate of growth of a rhyolite-producing magma reservoir. The numerical approach builds on a magma intrusion model developed to explain the current, decade-long, surface inflation at >20 cm/year. Assuming that the observed 62-m uplift reflects several non-eruptive intrusions of magma, each similar to the unrest over the past decade, we find that ~13 km³ of magma recharged the reservoir at a depth of ~7 km during the Holocene, accompanied by the eruption of ~9 km³ of rhyolite. The long-term rate of magma input is consistent with reservoir freezing and pluton formation. Yet, the unique set of observations considered here implies that large reservoirs can be incubated and grow at shallow depth via episodic high-flux magma injections. These replenishment episodes likely drive rapid inflation, destabilize cooling systems, propel rhyolitic eruptions, and thus should be carefully monitored.

Regional electrical structure of the Andean subduction zone in central Chile (35°-36°S) using magnetotellurics

A profile of broadband magnetotelluric stations was acquired between 2009 and 2016 at 35°-36°S in the Southern Volcanic Zone of the Chilean Andes to image the subduction zone and its relation with the volcanic arc at this latitude. This transect extends from the Coastal Cordillera across the Central Valley and the volcanic arc of the Principal Cordillera to the Argentine border. Two active volcanic complexes are found along this profile: Tatara-San Pedro is located on the modern volcanic front, and the Laguna del Maule volcanic field is found approximately 30 km to the east. The latter exhibits considerable signs of unrest, such as uplift rates of up to 25 cm/year, and has produced a high concentration of silicic eruptions in the last 25 ky. The data covered the period range from 0.001 to 1000 s. Robust processing techniques were used, including remote reference, and dimensionality was investigated by estimation of geoelectric strike, skew and analysis of the induction arrows. The data were modeled using a 2D inversion algorithm to produce a resistivity model which was consistent with surface geology and seismicity. The final resistivity model shows a generally resistive fore-arc structure, coincident with the tectonic environment, and a wide conductive region from the volcanic front to the east. This suggests a broad region of magmatism throughout the arc, related to three distinct magma bodies, associated with the Tatara-San Pedro and Laguna del Maule volcanic complexes and the Mariposa Geothermal System.