Volcanica

Magmatic trees: a method to compare processes between igneous systems

2024-07-17T03:09:50+00:00

This paper presents the motivation, instructions, and applications for a new graphical method to construct ‘magmatic trees’, which summarize the petrologic and geochemical processes that formed a particular igneous rock unit or eruption. The method is motivated by the need to develop new ways to compare and contrast igneous systems to address frontier research questions in volcano science. It is designed to be easily executed with common datasets, compel the integration of different data types, and facilitate cross-disciplinary conversations about the processes that underly these data (e.g. between the volcano remote sensing and petrology communities). There are numerous potential applications of the method, which include, a) motivating process-driven hypotheses, b) examining the frequency of particular magmatic processes within and among volcanic systems, c) building mantle and crustal magmatic processes into event trees for hazard assessment, and d) teaching petrologic methods. For example, constructing magmatic trees for successive eruptions at a volcano or for multiple volcanoes within the same tectonic setting not only helps quantify the probability of individual magmatic processes but leads to addressing higher-level questions, such as what crustal and magma characteristics cause the same set of processes to be repeated in successive eruptions at Mounts Hood, Unzen, Pinatubo, and Soufrière Hills, while different sets characterize magmas erupted at neighboring volcanoes like Mount St. Helens? In addition, one can imagine a future where machine learning removes much of the human error from magmatic process identification, as well as magmatic tree construction, thereby enhancing our ability to identify patterns of magmatic processes.

The radial spreading of volcanic umbrella clouds deduced from satellite measurements

2024-09-13T08:57:13+00:00

Analysis of thermal infrared satellite measurements of umbrella clouds generated by volcanic eruptions suggests that asymptotic gravity current models of the temporal (t) radial (r) spreading (r ~t_f, f < 1) of the umbrella-shaped intrusion do not adequately explain the observations. Umbrella clouds from 13 volcanic eruptions are studied using satellite data that have spatial resolutions of ~4–25 km² and temporal resolutions of 1–60 minutes. The umbrella cloud morphology is evaluated using digital image processing tools in a Lagrangian frame of reference. At the onset of neutral buoyancy, the radial spreading is better explained by a stronger dependence on time of r ~ t, rather than t^2/3, t^3/4, or t^2/9. This flow regime exists on the order of minutes and has not been observed previously in satellite data. This may be of significance as it provides a means to rapidly (within the first 2–3 observations) determine the volumetric eruption rate. A hyperbolic tangent model, r ~ tanh(t) is presented that matches the entire radial spreading time history and has a conserved torus-shaped volume in which the intrusion depth is proportional to sech(t). This model also predicts the observed radial velocities. The data and the model estimates of the volumetric flow rate for the 15 January 2022 Hunga eruption are found to be 3.6–5 × 10¹¹ m³s⁻¹, the largest ever measured.

Offshore evidence for volcanic landslide post Last Glacial Maximum at sub-Antarctic Heard Island, southern Indian Ocean

2024-09-13T08:21:08+00:00

Heard Island, an active sub-Antarctic intraplate volcanic island on the Kerguelen Plateau, is mostly covered by glaciers. The amphitheatre shaped summit of the active volcanic centre, Big Ben (2813 m), has been interpreted to be the product of a significant volcanic landslide. Here we present the first offshore geomorphological and geological evidence supporting a volcanic landslide on Big Ben, including: (1) the seafloor to the southwest of Heard resembling a landslide deposit, covering at least 467 km², (2) the spatial correlation between the onshore landslide scar and the offshore deposit and (3) the consistency in lithologies and compositions of rocks sampled from the deposit with the onshore in situ lithologies. ⁴⁰Ar/³⁹Ar geochronology constrains the maximum age of the volcanic landslide to 18.0 ± 1.4 ka, post the Last Glacial Maximum. Finally, we assess the risk of volcanic landslide at Heard Island in the future.

Eldgjá and Laki: Two large Icelandic fissure eruptions and a historical-critical approach for interdisciplinary researchers working on past nature-induced disasters

2024-10-15T08:57:32+00:00

The integration of archives of societies with archives of nature has led to collaborations between the natural sciences and the humanities. Not all those involved consider these archives equal, which led to some studies featuring explanations promoting nature as the prime agent in history. The field of the history of climate and society is currently experiencing a shift away from monocausal explanations. Cultural factors must be considered and their contribution to disasters must be examined. This paper introduces an easy-to-use step-by-step approach composed of crucial questions that need to be considered to analyze the entanglement of nature and society in relation to nature-induced disasters. The approach was developed by examining two large Icelandic fissure eruptions, Eldgjá (939–940 CE) and Laki (1783–1784 CE). The approach presented in this paper offers increased understanding across disciplinary cultures from the perspective of historians and is intended as a thought-provoking impulse for future studies.

Catastrophic lava flow levee failure: precursors, processes, and implications

2024-05-09T07:05:57+00:00

During an effusive eruption crisis the initial advance of a lava flow is typically the primary focus of model forecasts and hazard management efforts. Flow branching and lateral expansion of lava flows can pose significant dangers within evolving flow fields throughout the duration of an eruption and are an underappreciated hazard. We use field monitoring, infrasound, time lapse imagery, and lidar data collected during the 2018 lower East Rift Zone eruption of Kīlauea (Hawai‘i) to track the origins, progression, and implications of a flow branching event caused by catastrophic levee failure. Our analyses show that surges in effusion rate, rheologic transitions between pāhoehoe and ‘a‘ā flow regimes, slope-breaks, pre-existing topographic highs, and the structure of perched levee walls all played a role in the failure of the levee and subsequent re-routing of the lava flow. Failure of perched lava structures leads to an acutely hazardous situation because lava impounded by the structure can rapidly inundate the landscape. This is the first time a levee failure event has been observed in such detail with numerous monitoring techniques; this unprecedented level of observation provides quantifiable insights into levee failure processes that have important implications for hazard mitigation and an improved understanding of lava flow emplacement dynamics

Plant traits, growth stage, and ash mass load control the vulnerability of potato, corn, and wheat crops to volcanic ashfall

2024-03-06T02:06:38+00:00

Current predictive models of ash impact on crops use ash thickness (or mass load) as the explanatory variable but fail to account for other factors, such as plant traits and growth stage, which also influence impact. We conducted a plot experiment with three common crops (potatoes, corn, and wheat), exposing them to representative ash mass loads (0.5 to 9 kg m⁻² ). We recorded visual impacts on the plants at different intervals and estimated yield loss. Distinct impact mechanisms were identified for each crop, including premature flower abscission, irreversible leaf yellowing, desiccation and senescence, and stalk lodging. Exposure of potato, corn, and wheat plants to ash mass loads >1 kg m⁻² significantly reduced yield, but production quality was largely unaffected. These results were used to develop new vulnerability functions for estimating yield loss in potatoes, corn, and wheat following exposure to an ashfall event.

Graph Neural Network based elastic deformation emulators for magmatic reservoirs of complex geometries

2024-05-23T11:47:43+00:00

Measurements of volcano deformation are increasingly routine, but constraining complex magma reservoir geometries via inversions of surface deformation measurements remains challenging. This is partly due to deformation modeling being limited to one of two approaches: computationally efficient semi-analytical elastic solutions for simple magma reservoir geometries (point sources, spheroids, and cracks) and computationally expensive numerical solutions for complex 3D geometries. Here, we introduce a pair of Graph Neural Network (GNN) based elasto-static emulators capable of making fast and reasonably accurate predictions (error upper bound: 15 %) of surface deformation associated with 3D reservoir geometries: a spheroid emulator and a general shape emulator, the latter parameterized with spherical harmonics. The emulators are trained on, and benchmarked against, boundary element (BEM) simulations, providing up to three orders of magnitude speed up compared to BEM methods. Once trained, the emulators can generalize to new reservoir geometries statistically similar to those in the training data set, thus avoiding the need for re-training, a common limitation for existing neural network emulators. We demonstrate the utility of the emulators via Bayesian Markov Chain Monte Carlo inversions of synthetic surface deformation data, showcasing scenarios in which the emulators can, and can not, resolve complex magma reservoir geometries from surface deformation. Our work demonstrates that GNN based emulators have the potential to significantly reduce the computational cost of inverse analyses related to volcano deformation, thereby bringing new insights into the complex geometries of magmatic systems.

Textural complexity and geochemistry of the last millennium pyroclastic deposits from Puyehue-Cordón Caulle Volcanic Complex

2024-09-24T15:10:26+00:00

The component variability in Puyehue-Cordón Caulle Volcanic Complex (PCCVC) products reflects the inherent complexity of volcanic processes. We examine pyroclastic deposits from Cordón Caulle (2011 and 1960 eruptions) and Puyehue (MH tephra) in a profile ∼20 km windward of the PCCVC. All levels have comparable components (pumice, scoria, glass shards, crystals), but their proportions vary according to the dominant eruptive style in both vent sources. The particle microtextures combined with mineralogy and geochemistry differentiate juvenile from non-juvenile particles in macroscopically undifferentiated components, questioning prior assumptions. Highly vesicular pumice is the dominant juvenile component indicating decompression-driven gas exsolution processes. Juvenile blocky glass shards/obsidians, frequently associated with lithics, now provide insights into the potential higher involvement of magma in the phreatomagmatic phases of the MH deposit. Nevertheless, the variability of tephra components is a characteristic of the PCCVC, regardless of the juvenile or lithic character. This research refines tephrochronological tools and deepens our understanding of volcanic processes and deposits in the PCCVC.

Fe-rich filamentary textures reveal timescales of magmatic interaction before the onset of high-energy explosive events at basaltic volcanoes

2024-12-11T13:10:01+00:00

Fe-rich filamentary textures are almost ubiquitous in products from explosive eruptions at basaltic volcanoes and, in particular, they characterize the groundmass of ash and lapilli emitted during high-energy events. Here, we present a multidisciplinary study integrating petrological analyses with computational fluid dynamics simulations to propose a new mechanism responsible for their formation. Detailed textural and compositional features of Fe-rich filaments were examined in the products of explosive eruptions with different intensities from Stromboli and Etna (Italy) volcanoes. Results reveal that they represent compositional boundary layers developed at the plagioclase-melt interface in response to the interaction between magmas with different compositions and volatile contents. Numerical simulations indicate that boundary layers can detach from crystals and disperse into resident melts due to their higher density and can survive as metastable melts for some days under magmatic conditions. We suggest that Fe-rich filaments testify to the recharging of deep magma a few days before high-energy explosive events at basaltic open-vent volcanoes, even when primitive magmas are not erupted.

VolcanoVR: A virtual reality environment for volcanic data visualisation and communication

2025-03-05T14:33:32+00:00

With the increasing size and complexity of geological datasets relating to volcano monitoring and research, effective visualisation can be challenging. Here, we demonstrate the possibilities of volcanic data visualisation utilising virtual reality (VR) and 3D game engine technology to create a robust and adaptable program called VolcanoVR. VolcanoVR can display multiple complex datasets and has been developed to investigate volcanoes in the Taupō Volcanic Zone, New Zealand. To assess the usability and effectiveness of VolcanoVR a survey was conducted, involving 33 participants, ranging in education level and previous experience with volcanic data and VR. Results indicate VolcanoVR is easy to use with high immersion ratings and acceptable system useability and mental demand scores, with minor improvements made following the survey. Limited variability across user experience levels indicates the program is useable for a broad range of geoscientists. We have made the source code for VolcanoVR freely available so that it can be easily adapted and applied worldwide to a range of different volcanoes and geological settings.

Fast or slow: An evaluation of Ti-in-quartz diffusion coefficients through comparisons of quartz and plagioclase diffusion times

2024-11-14T05:14:36+00:00

Diffusion geochronometry using Ti-in-quartz has become a valuable method in understanding the evolution of silicic magmas. However, four different options for Ti diffusivity (D_Ti) currently exist, spanning three orders of magnitude, resulting in substantially different estimated times and interpretations. We present Ti-in-quartz diffusion times for the Cerro Galán Ignimbrite using the Cherniak et al. [2007] (10.1016/j.chemgeo.2006.09.001), Audétat et al. [2021] (10.1130/g48785.1), Audétat et al. [2023] (10.1038/s41467-023-39912-5), and Jollands et al. [2020] (10.1130/g47238.1) D_Ti value and (1) compare these against plagioclase diffusion times derived from the same samples, (2) consider evidence for Ti diffusion in quartz under relevant magmatic timescales, and (3) compute derived quartz growth rates for crystals from the Cerro Galán Ignimbrite. On all accounts, we find that the Cherniak et al. [2007] diffusion coefficient yields diffusion times that agree much better with independent evidence than those derived using slower D_Ti values [Jollands et al. 2020; Audétat et al. 2021; 2023].

Precursory velocity changes prior to the 2019 paroxysms at Stromboli volcano, Italy, from coda wave interferometry

2025-01-16T19:57:46+00:00

Open-conduit basaltic volcanoes are susceptible to sudden transitions from mild activity to violent explosive eruptions with little to no warning. Such was the case at Stromboli in the summer of 2019, when two paroxysmal explosions occurred within approximately two months (July 3 and August 28). We apply coda wave interferometry to identify possible transitions in behavior in the build-up to these events, computing seismic velocity changes using five broadband seismic stations on the volcano between 2013–2022. This timeframe encompasses a range of volcanic activity including effusive activity, major explosions and paroxysms. Cross-correlation functions are computed both between pairs of stations and single-station cross-components in multiple frequency bands that allow the sampling of different depths (between approximately 100–1000 m) within the plumbing system. Shallow velocity changes (1–2 Hz and 2–4 Hz) reveal mid-to-long term precursors prior to the paroxysms in 2019. For example, we observe that 2–4 Hz velocities recorded at the station closest to the active crater show an increase of 0.2–0.3 % relative to velocities recorded at other stations. This increase is largely accumulated from mid-2017, coinciding with previously observed heightened activity at the volcano, peaking approximately one month prior to the first paroxysm. A long-term decrease is also observed in deeper velocity changes (0.5–1.0 Hz) during the same time interval. It is hypothesized that these changes represent greater magma overpressure from increased volatile input from depth. The different response in the shallow subsurface may reflect a local response due to the same source within the vicinity close of the crater terrace. These findings illustrate how coda wave interferometry can provide meaningful insights into the evolving dynamics of open-conduit basaltic volcanoes.

First hydroacoustic recording of ebullition events in an active volcanic lake – Poás, Costa Rica

2024-09-13T05:38:55+00:00

This study demonstrates how the sounds of subaqueous gas seeps can be used to measure the volume of volcanic gas being released into an active crater lake, weeks before an eruptive period. A hydrophone placed in Poás crater lake recorded changes in the subaqueous soundscape over the course of one month. Using new passive acoustic inversion techniques, we were able to measure the volume of gas being released from the lakebed at a sample rate of 5 min, far higher than traditional sub aerial gas sampling techniques. Comparing these findings to local seismic measurements allowed us to observe variations in gas flux driven by both volcanic and non-volcanic factors. Non-volcanic causes consist of small-scale diurnal variations of ~2 L min^-1 driven by local atmospheric pressure conditions. We also see a large and abrupt aseismic mass bubbling event releasing 18,000 ± 3000 L of gas in just 15 hours (compared to a daily average of 3600 ± 500 L) likely resulting from the collapse of gas pocket(s) in the sediment underlying the lake. Alongside an even larger mass bubbling event releasing 30,000 ± 5000 L of gas in 24 hours correlated with local seismic activity, presumed to be triggered by excess volatiles being released from deeper within the volcano, which preceded a new eruptive period at Poás volcano. This work paves the way for future studies to quantify subaqueous volcanic gas emissions via hydroacoustics, a potential new volcano monitoring technique.

Thermal remanence of the ∼0.6 kya Rangitoto volcano eruption, Auckland volcanic field (New Zealand) inferred from self-potential and CO₂ flux measurements

2024-11-08T12:33:05+00:00

Rangitoto volcano is the most recent (⁓0.6 ka) and voluminous volcano in New Zealand's Auckland Volcanic Field (AVF). In this study, we investigate the status of its hydrothermal system using a combination of self-potential (SP) and CO₂ gas flux measurements along the west-east Rangitoto–Motutapu main road. SP data revealed a "W"-shaped signal near the main crater, indicating an active hydrothermal system. In contrast, the CO₂ flux data showed diffuse emissions peaking ⁓620 m east of the SP anomaly peak, suggesting they originated from different sources. The SP anomaly is likely due to a hydrothermal system heated by a shallow, cooling basalt body, and CO₂ emission is from deeper crustal or mantle sources. An SP electric potential offset was also detected at the Islington Bay Fault under the Rangitoto–Motutapu bridge without a corresponding CO₂ flux anomaly.

Résumé

Le volcan Rangitoto est le plus récent (⁓0,6 ka) et le plus volumineux du AVF en Nouvelle-Zélande. Nous examinons l'état de son système hydrothermal en utilisant une combinaison de mesures de SP et de flux de CO₂ le long de la route principale ouest-est Rangitoto–Motutapu. Les données SP a révélé un signal en forme de « W » près du cratère principal, indiquant un système hydrothermal actif. Par rapport à, les données de flux de CO₂ ont montré des émissions diffuses culminant à ⁓620 m à l'est du pic de l'anomalie SP, suggérant des sources différentes. L'anomalie SP est probablement due à un système hydrothermal chauffé par les matériaux basaltiques peu profond en refroidissement, tandis que l'émission de CO₂ provient de sources plus profondes de la croûte ou du manteau. Un décalage de potentiel électrique SP a également été détecté à la faille d'Islington Bay sous le pont Rangitoto–Motutapu sans anomalie correspondante de flux de CO₂.

Behaviours of pyroclastic and analogue materials, in dry and wet environments, for use in experimental modelling of pyroclastic density currents

2024-09-24T15:08:01+00:00

Modelling pyroclastic density currents (PDCs) is a challenging yet essential element of hazard assessment. PDCs are unpredictable and internal processes are often difficult to measure directly. Analogue experiments have been an important tool for investigating internal PDC dynamics. Typically, analogue experiments have removed moisture from experimental materials to limit cohesion. However, this does not well represent natural systems, where moisture can be introduced into a PDC. In this study, we investigate pyroclastic and analogue materials in dynamic (i.e., flowing), static (i.e., stationary), wet and dry experiments to explore fundamental behaviours. The addition of moisture can lead to fundamental changes in material properties resulting in significant impacts on geomechanical behaviours (size, density, internal friction angle), fluidisation and flowability. This work highlights the importance of validating the material choice used in modelling experiments, especially in wet conditions, and provides insights into flow dynamics of PDCs and depositional architecture of their deposits.

A generalized deep learning model to detect and classify volcano seismicity

2025-01-13T20:26:53+00:00

Volcano seismicity is often detected and classified based on its spectral properties. However, the wide variety of volcano seismic signals and increasing amounts of data make accurate, consistent, and efficient detection and classification challenging. Machine learning (ML) has proven very effective at detecting and classifying tectonic seismicity, particularly using Convolutional Neural Networks (CNNs) and leveraging labeled datasets from regional seismic networks. Progress has been made applying ML to volcano seismicity, but efforts have typically been focused on a single volcano and are often hampered by the limited availability of training data. We build on the method of Tan et al. [2024] (10.1029/2024JB029194) to generalize a spectrogram-based CNN termed the VOlcano Infrasound and Seismic Spectrogram Neural Network (VOISS-Net) to detect and classify volcano seismicity at any volcano. We use a diverse training dataset of over 270,000 spectrograms from multiple volcanoes: Pavlof, Semisopochnoi, Tanaga, Takawangha, and Redoubt volcanoes\replaced (Alaska, USA); Mt. Etna (Italy); and Kīlauea, Hawai`i (USA). These volcanoes present a wide range of volcano seismic signals, source-receiver distances, and eruption styles. Our generalized VOISS-Net model achieves an accuracy of 87 % on the test set. We apply this model to continuous data from several volcanoes and eruptions included within and outside our training set, and find that multiple types of tremor, explosions, earthquakes, long-period events, and noise are successfully detected and classified. The model occasionally confuses transient signals such as earthquakes and explosions and misclassifies seismicity not included in the training dataset (e.g. teleseismic earthquakes). We envision the generalized VOISS-Net model to be applicable in both research and operational volcano monitoring settings.

The use of UAV-based visible and multispectral thermal infrared data for active volcano monitoring and analysis: test of a low-cost solution applied to the 2022 Meradalir eruption

2025-01-29T10:27:07+00:00

Timely analysis of active lava flow dynamics and emplacement are typically limited by current ground, UAV, and satellite-based observational capabilities. The Python Miniature Thermal Instrument for Uncrewed Aircraft Systems (PyMTI-UAS) is a relatively inexpensive, low-mass, low-power multispectral thermal infrared instrument capable of measuring rapid changes in thermal and gas dynamics of lava flows to at high resolution. The 2022 Meradalir effusive eruption in Iceland offered an opportunity to acquire visible and multispectral thermal infrared data with PyMTI-UAS of recently emplaced lavas. A successful deployment occurred during the end of the 2022 eruption and the resulting thermal infrared data provide insights into lava surface texture relationships, post-emplacement alteration, and gas and thermal flux during cooling. This study demonstrates that PyMTI-UAS offers the framework to provide accurate multispectral thermal infrared data at low cost from small UAVs to provide data vital for monitoring volcanic activity and aiding hazard response.