Through the analysis of the textural, chemical, and isotopic characteristics of eruptive products we can elucidate the inner workings and the architecture of magma plumbing systems, as well as constrain pre- and syn-eruptive processes. Analytical/field observations, laboratory experiments and numerical modelling are fundamental tools for the investigation of pre- and syn-eruptive processes, and for understanding eruptive dynamics. This information is of paramount importance for policymakers in charge of mitigating the risks associated with volcanic activity.
In this session, we welcome petrological, geochemical, geophysical and volcanological studies that investigate the dynamics of magmatic processes within magma reservoirs and volcanic conduits through natural, experimental, and numerical-based approaches. Contributions that investigate the hazards associated with volcanic activity and interdisciplinary works that consider the close and complex interplay between magmatic processes, conduit dynamics, eruptive behaviour, and emplacement mechanisms are encouraged.
Orals: Tue, 5 May, 08:30–12:25 | Room -2.21
The British Palaeogene Igneous Province (BPIP) has long been an area of research, both for British volcanology and in the context of the early stages of opening of the North Atlantic. Previous studies have investigated the petrogenesis of the Mull lavas (e.g. [5]) and the evolution of other igneous centres associated with the BPIP, such as on Arran [3],[4]. However, since the Mull Memoir was published by Bailey et. al. [1], little substantive research has been carried out on the early evolution of the Mull central complex, the focus of this study.
Through combined fieldwork, geochemical and geochronological research, this study has elucidated significant new information in understanding the early development of the central complex on the Isle of Mull.
Volcanological focussed fieldwork, combined with geochemical fingerprinting and petrological examination, has revealed a new model for the post-collapse volcanic infill sequence of the early caldera on Mull. The range of volcanic units within this sequence, from pillow basalts to massive lapilli tuffs and rhyolites, indicates during this period of evolution, the Mull volcano had varying eruption styles and showed a general trend towards higher silica magma compositions, with increasingly explosive eruptive episodes.
Major and trace element analysis, on a wide range of samples, has investigated processes including magma mixing and crustal contamination and links between sub-surface magma conduits and erupted deposits. Isotopic and elemental fingerprinting has revealed the nature of the basement rocks, through which magmas have risen and evolved. Most intrusive and extrusive units associated with the early Mull central complex show a significant upper crustal contamination signature, consistent with contamination by Moine metasediments. A contrasting trend is seen in the basaltic rocks at the base of the caldera infill sequence, which show less contamination than the units later in the stratigraphy, indicating a potentially shorter crustal residence time or a separately fractionating basaltic system to the basalts erupted during earlier events on Mull. Ongoing U-Pb geochronology has determined timescales of activity for the two earliest volcanic centres of the Mull central complex to temporally constrain these models.
This study has resulted in an updated volcanological and petrological model of the early evolution of the Mull central complex. It has also expanded our understanding of the BPIP, through allowing comparison of timescales and volcanic evolution to other centres, such as Arran [3],[4] and Skye (e.g. [2]). More widely, this study has implications for our understanding of explosive volcanism, particularly in young and evolving environments.
References:
[1] Bailey E.B. et.al. (1924) “Tertiary and Post-tertiary Geology of Mull, Loch Aline and Oban”, British Geological Survey
[2] Drake S.M. et.al. (2022) Volcanica 5: 397–432
[3] Gooday R.J. et.al. (2018) Bulletin of Volcanology 80:70
[4] Gooday R.J. (2024) Lithos 488-489, 107789
[5] Kerr A.C. (1995) Journal of the Geological Society, London 152:975-978
How to cite: Goddard, F., Kerr, A., McDonald, I., and Gooday, R.: Early Evolution of the Palaeogene Mull volcano: An integrated volcanological, Geochemical and Geochronological approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7018, https://doi.org/10.5194/egusphere-egu26-7018, 2026.
The architecture of plumbing systems, and the dynamics of magma-mush mobilisation during eruptions have received a great deal of attention in recent years. Magmatism plays a central role in rift dynamics, yet the structure and evolution of magma plumbing systems during continental break-up remain poorly constrained. The Afar Rift offers a rare opportunity to study active plate divergence and associated magma processes. We investigate the 1978 Ardoukoba fissural eruption in the Asal Rift, a syn-rift volcanism archetypal example and the most recent eruption in this segment of the Afar Rift system. Using a comprehensive dataset of melt inclusions and host mineral compositions, volatile contents (H₂O, CO₂, δD) and thermobarometry, we reconstruct the transcrustal plumbing system and track magma storage, transfer and degassing during the eruption. We show that magmas feeding the system derived from heterogeneous mantle sources, whose signatures are preserved in melt inclusions. Our results reveal polybaric magma recharge events destabilizing the system, triggering progressive tapping of increasingly deeper mush zones. The eruption began with shallow, evolved melts and transitioned to deeper, more primitive melts and crystal cargos. These findings offer a high-resolution view of magma dynamics during rift-related eruptions and provide key constraints on the magmatic architecture of incipient oceanic spreading centres.
How to cite: Chazot, G., Pin, J., France, L., Deloule, E., Daoud, M. A., and Le Gall, B.: Reactivation of a Transcrustal Plumbing System During an Eruptive Rifting Event (Asal Rift, Djibouti): volatiles and chemical compositions of melt inclusions and zoned crystals, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4499, https://doi.org/10.5194/egusphere-egu26-4499, 2026.
The Portrush Sill, located on the north coast of County Antrim, Northern Ireland, is a bowl-shaped Paleogene (c. 58.5 Ma) intrusion emplaced into Jurassic sediments rich in disseminated pyrite and pyritised macrofossils. We have undertaken a detailed remapping of the intrusion, coupled with a study of the stratigraphic variation of microstructure and geochemistry. The central part of the sill is characterised by a striking magmatic texture comprising centimetre–decimetre sized, rounded melanocratic regions (globules) set within a leucocratic matrix. The melanocratic globules vary in size and morphology through the stratigraphy. Globule - matrix pairs from individual samples, along with samples of the homogeneous over- and underlying parts of the sill, were geochemically characterised using XRF (major element oxides), ICP-MS (trace elements), and EPMA (mineral compositions). The globules and matrix are composed of the same mineral phases but in differing proportions: there is no indication of chemical disequilibrium, with the minerals having identical compositions in both globules and matrix. The petrographic and geochemical data, in conjunction with field observations, are consistent with the two components representing conjugate immiscible Fe- and Si-rich liquids produced as the evolving parent magma encountered a binode.
Cooling rates determined from the stratigraphic variation of clinopyroxene-plagioclase-plagioclase dihedral angles show that the sill intruded as a single body, in contrast to previous studies arguing for several separate intrusions separated by screens of sedimentary rock. Field observations show the magma intruding along bedding planes of the host rock, as well as evidence of significant anatexis and contamination of the proximal magma. Analysis of Sr–Nd–Pb isotopes indicates an increase in the extent of contamination towards sedimentary screens in the middle and upper parts of the sill. We infer that the onset of immiscibility, and the unmixing of conjugate Fe- and Si-rich liquids within the Portrush sill, was a consequence of assimilation of pyrite-rich country rock. This represents the first documented example of macro-scale assimilation-induced liquid immiscibility, with major implications for our understanding of magmatic evolution.
How to cite: Geifman, E., Stock, M., Holness, M., Cooper, M., Andersen, J., van Acken, D., Huber, C., Carter, E., and Chew, D.: New Observations of Assimilation-Induced Silicate Liquid Immiscibility in the Portrush Sill, Northern Ireland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1078, https://doi.org/10.5194/egusphere-egu26-1078, 2026.
Dome-building eruptions are a complex expression of hybrid, effusive to explosive volcanism at the Earth’s surface. They are characterized by frequent and rapid shifts in eruptive style that make them highly unpredictable and hazardous. This multifaceted behavior is strongly influenced by the thermal/rheological evolution and mechanical response of lavas and rocks in the dome, and the efficiency of gas release through the edifice. Gas escape or entrapment in lava domes depends primarily on the permeability of the porous networks at different scale. Fracturing is crucial in developing connected and permeable pore pathways and thus in enabling and regulating gas escape efficiency in lava domes. In this study, we use X-ray micro-tomographic data obtained on a dacite from a block-and-ash flow deposit from the 1990-1995 eruption of Mt Unzen (Japan). We simulate pore-controlled fracturing in the 3D volume through a watershed-type image analysis, through which we explore the combined role of fractures and pore size distribution on pore connectivity and permeability. We use two different starting pore networks: one consisting of macro-porosity surrounded by micropores and a second set consisting of macropores only, yielding two distinct starting pore size distributions and pore connectivities. We quantify the influence of crack number density and width on the evolution of pore connectivity with porosity for the two pore networks. We then use Lattice Boltzmann simulations to quantify the porosity-permeability relationships of virtually cracked rocks. Our results show that as cracking progresses in a crystal-rich dome rock, connectivity and permeability strongly increase. The initial pore size distribution has a strong impact on crack propagation as well as connectivity and permeability. Our dataset also suggests that connectivity for a given pore network depends mostly on crack-number density, whereas permeability is more sensitive to crack width. Combining measurements of connectivity and permeability may be key when assessing the extent and mode of fracturing in volcanic edifices such as lava domes. We compare our simulations with mechanical tests of uniaxial compressive strength on similar Unzen dacites in order to link the impact of initial porosity and pore size distribution on crack propagation during failure. We finally discuss the implications of our results for the stability and eruptive style of silicic lava domes.
How to cite: Colombier, M., Kendrick, J. E., Birnbaum, J., Vasseur, J., Lamur, A., Lavallée, Y., Dobson, K. J., Miwa, T., Scheu, B., and Kueppers, U.: The influence of pore size distribution on fracturing in lava domes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13652, 2026.
Arc magmas commonly undergo many mineral-melt reactions during changes in P-T-X conditions (e.g. during ascent through the crust, magma mixing, or CO2 flushing). Constraining these reactions can be key to unlocking details of the magmatic systems that underlie high-threat arc volcanoes, such as those of the Cascades Arc, USA. However, identifying “lost” phases remains a challenge. Clues can come from geochemical methods and phase-equilibria modelling, but direct textural evidence is often assumed to be eradicated upon completion of the mineral-melt reactions.
In this study, we demonstrate that even mineral phases that have been completely reacted out of the magma can still leave behind distinctive microstructures. Using electron backscatter diffraction (EBSD) and detailed petrographic and geochemical characterisation, we describe the reaction textures produced by the breakdown of olivine, amphibole and biotite in intermediate rocks from a variety of high-threat volcanoes in the Cascades. We trace how the microstructures evolve throughout the course of the reaction and continue to evolve during magma storage, even after the original phase is completely consumed. We outline key features that can distinguish reaction-generated glomerocrysts from other polycrystalline aggregates such as mush fragments.
We find that distinctive crystal lattice distortion occurs in the products of all of the reactions. We infer that this distortion arises due to the strain from volume changes and lattice mismatches. Reaction textures also commonly feature epitaxial relationships, and regions where neighbouring crystals share similar orientations. Other clues can include unusual mineral assemblages, and intergrowth textures such as symplectites.
Textural re-equilibration during magma storage can change originally unmistakeable reaction textures into much less distinctive textures that could be mistaken for mush fragments. Common changes include grain growth, progressive equilibration of grain shapes, the recrystallisation of metastable phase assemblages, and the consolidation of lattice strain into subgrain-like structures. However, some markers, including the distinctive orientation relationships between minerals, and the remnants of lattice distortion, remain relatively robust throughout re-equilibration. Application of such microstructural indicators helps to reveal the true diversity of phases present in these magmatic systems, enabling better reconstruction of the pre-eruptive histories of complex arc magmas.
How to cite: Gordon, C., Wieser, P., Till, C., and Kent, A.: Ghosts of phases past: The microstructural markers of completed mineral-melt reactions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15724, 2026.
Microlite crystals, 1-30 microns across, are common in the products of explosive basaltic eruptions. They represent precious tools to investigate magma evolution immediately before or during eruptions, and their presence affects eruption dynamics by changing the physicochemical properties of magma. Nanolite crystals, 0.03-1 microns across, are less studied than microlites but increasingly recognized as key modifiers of magma rheology and eruption dynamics. Growing microlites (and crystals in general) pushes incompatible elements into the surrounding melt, forming a so-called compositional boundary layer, or CBL. Here we focus on micro- to nano-scale features of CBLs around microlites that are either intact or broken during magma fragmentation. Study cases include lapilli of mafic composition from Etna and Stromboli (Italy), Xitle (Mexico), and Cumbre Vieja (Spain) volcanoes. Samples were investigated by using Transmission Electron Microscopy for high-resolution imaging, EDS analysis and mapping, SAED, and 4D-STEM. In the CBL around plagioclase microlites and inside glass-filled fractures within, we found evidence of liquid immiscibility, with droplets of a denser phase dispersed in a lighter phase. The denser phase is enriched mostly in Fe and variably in Ti, Mg, Ca, while the lighter phase is depleted in the above elements. The size of the droplets of the denser phase decreases away from the CBL. At the microlite surface, the denser phase often crystallizes into Fe-oxides (magnetite) nanolites, mostly a few tens of nm in size, and, occasionally, into clinopyroxene (augite) nanolites. Incipient crystallization of the denser liquid droplets into nanolites suggest that CBL development and consequent liquid immiscibility are key steps leading to local nanolite enrichment of basaltic melts, which ultimately can affect the bulk viscosity of the magmatic suspension and its rheological response to deformation during fragmentation.
How to cite: Taddeucci, J., Pontesilli, A., Di Fiore, F., Roddatis, V., Schreiber, A., Nazzari, M., and Scarlato, P.: A TEM look at microlite and nanolite growth in erupting basalts., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16705, 2026.
In early October 2023, earthquake swarms occurred near Izu-Torishima in the southern Izu arc. On October 9, tsunamis reaching up to 60 cm were observed at Hachijojima and along the Pacific coast of the Japanese archipelago where the height of the tsunamis were disproportionate to the earthquake magnitudes (Sandanbata et al., 2023). T-wave analysis identified the epicenter near Sofu Seamount, bathymetric high west of Sofugan volcano, and subsequent bathymetric surveys by JAMSTEC and Japan Coast Guard independently revealed significant topographical changes, including a concave depression near the western summit of Sofu Seamount (Fujiwara et al., 2024; Minami & Tani, 2024). This significant bathymetric change suggests an intense deep-sea eruption even though the water depth of the pre-eruption summit was ~800 m.
We conducted seafloor geological surveys using ROV and dredges during cruises KM25-02Leg2 (March 2025) and KM25-09 (September 2025) by the R/V Kaimei. Sofu Seamount is an east-west trending bathymetric, which was not previously recognized as an active volcano. Its western edifice features a caldera-like structure approximately 5 km in diameter, with a central cone in the northern section now containing a 1.5 km wide crater formed during the 2023 activity. The crater floor lies at a depth of ~1200 m, with the rim at ~900 m. Recent shallowing within the outer caldera was detected, suggesting significant tephra deposition in the proximal area. Rock samples were collected from the inside of the central cone, inside of the outer caldera, and northern slope of the seamount. Around the central cone crater, dark-colored volcanic rocks were collected, some of which featured brown-colored coatings identified as iron oxyhydroxides. Bacterial mats were observed on both sides of the northeastern crater rim, suggesting an active hydrothermal venting. White pumice (clasts to boulders) was found deposited on the upper most surface of the caldera interior, likely originating from the October 2023 event. Large dark-colored volcanic rocks were observed outside the caldera, notably lacking iron oxide coatings.
The collected samples ranged from basaltic andesite to rhyolitic pumice, with SiO2 = 52 – 73 %. Mafic rocks contain olivine (Fo70-75), clinopyroxene, Ca-rich plagioclase, and rare orthopyroxene. Some dredged samples contained xenoliths with less-differentiated olivine (~Fo85) and well-preserved basaltic volcanic glass. Olivine-melt composition suggests P-T conditions of ~1050 C and ~160 MPa. In contrast, pumice samples contain clinopyroxene, orthopyroxene, and Na-rich plagioclase, with the estimated conditions of ~900 C and 135-165 MPa. Thermodynamic modelling indicates the compositional trend cannot be explained by fractional crystallization. As such, the mixing of high-temperature mafic magma and low-temperature felsic magma within a pressure range of 134-165 MPa beneath Sofu Seamount triggered an explosive eruption in October 2023.
How to cite: Yoshida, K., Sato, T., Hamada, M., Tada, N., Hanyu, T., Akamatsu, Y., Ichihara, H., Nakano, M., McIntosh, I., Hagiwara, Y., Chang, Q., Gautreau, L.-M., Pank, K., Ishibashi, H., Tamura, Y., and Ono, S.: Petrographic and geochemical characteristics of Sofu Seamount, a recently-erupted deep submarine volcano in the Izu arc, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18897, 2026.
Mafic intrusions provide critical insights into the emplacement dynamics of upper-crustal magma reservoirs, yet their formation remains debated, with two competing models: (i) large, liquid-dominated magma chambers, in which fractionation is driven by crystal accumulation and separation, or (ii) crystal-rich mush systems, incrementally assembled by multiple small intrusions. Constraining which model best describes the development of any mafic intrusion is central to addressing fundamental questions of cumulate formation and subvolcanic chamber replenishment.
The Carlingford Complex (Co. Louth, Ireland) preserves the shallow crustal architecture of a volcanic centre active during Paleogene plume-related rifting, presenting an ideal case study to evaluate the two proposed models of mafic intrusion assembly. Here, we present integrated field, microstructural, and geochemical data to develop a model for the emplacement of the Carlingford Complex.
High-resolution sampling of outcrop and drill-core material reveal four stratigraphic zones, each characterised by distinct mineralogical and geochemical signatures. The basal sequence records pulsed replenishment and accumulation within an open, liquid-dominated environment, later transitioning into a phase of high melt flux approaching a closed system, facilitating large-scale crystal settling, flotation, and convection. The upper sequence preserves evidence of complex liquid-solid interactions, including late-stage infiltration into existing mush zones, and sporadic intrusion of laterally discontinuous sills of varied composition. Together, these zones preserve evidence of both liquid- and crystal-rich modes of magma replenishment, spanning the behavioural endmembers exemplified by other Paleogene intrusions. Our work demonstrates that subvolcanic systems can shift between contrasting emplacement regimes over relatively short spatial and temporal intervals, highlighting the sensitivity of magma systems to changes in melt flux and thermal state.
How to cite: Beckwith, J., Stock, M., Holness, M., Cooper, M., Andersen, J., Huber, C., Chew, D., Higgins, O., Carter, E., and Broom-Fendley, S.: From melt to mush: dynamic assembly of the Carlingford Complex layered intrusion, Ireland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-823, https://doi.org/10.5194/egusphere-egu26-823, 2026.
All magmas transition through a crystal mush stage during solidification, and long-lived crystal mushes are thought to be common in crustal magmatic systems. The efficacy of many crystal mush processes depends primarily on the permeability and porosity of the mush, and thus on the shape and size of crystals in the mush framework. Plagioclase is one of the most common minerals to form in igneous rocks and is commonly a framework-forming phase.
Crystal shape is determined by the relative growth rates on different crystal faces and can be readily measured in thin section. Holness (2014) showed a relationship between plagioclase crystal shape and cooling conditions, but without a mechanistic underpinning. Here, we show experimentally that plagioclase 3D crystal shape evolves from prismatic to tabular (platy) with the thermodynamic driving force for crystallisation (undercooling) that is experienced by the crystal. This occurs because crystallisation on the long and intermediate axes transitions to faster, higher-energy growth mechanisms with increasing undercooling, relative to the short axis. This is a general crystallographic processes and we anticipate that similar relationships can be found for other anisotropic minerals.
We combine our experiments with numerical forward modelling to produce a quantitative measure of the evolving undercooling at any instant that is experienced by the crystal. This instantaneous undercooling varies over the course of a single experiment or stage of crystallisation. Its maximum value controls the majority of growth and correlates with 3D crystal shape.
An evolution from more tabular to more prismatic crystal shape occurs with decreasing undercooling (or cooling rate). This framework can also be extended to decompressing systems. As the thermal state (maturity) of the crust is a critical control on the local cooling rate, magma intruded into cooler crust and in smaller batches will have more tabular crystals than large batches of magma intruded into warmer crust. Repeated episodes of intrusion, resorption and remobilisation will also affect crystal shape by decreasing crystal aspect ratio and increasing the crystallinity at which the mush becomes immobile.
Overall, crystal shape is a strong control on igneous rock texture, and quantitative investigation of crystal shape has great potential to uncover the details of magmatic plumbing and volcanic processes.
How to cite: Humphreys, M., Lindoo, A., Brooker, R., Geifman, E., Gordon, C., Llewellin, E., Mangler, M., and Wadsworth, F.: The crystal-scale characteristics of crystal mushes: crystal shape as a record of dynamic magmatic processes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19129, 2026.
The generation of eruptible magma from long-lived, crystal-rich mush reservoirs remains a fundamental challenge in volcanology. Magma stored in the crust commonly resides as a high-crystallinity mush below the eruptibility threshold, yet many volcanoes exhibit frequent eruptions that require rapid remobilization of stored melt. In this study, we investigate the physical mechanisms by which hot melt recharge reorganizes a colder, partially crystalline reservoir to produce localized zones of mobile, eruptible magma. We present a fully coupled three-dimensional thermo-poroelastic model that simulates hot melt injection into a porous magma mush, incorporating Darcy flow, heat transfer with phase change, and poroelastic deformation. Magma properties, including melt fraction, viscosity, density, and mush permeability, evolve dynamically as functions of temperature, pressure, and composition. We utilize thermodynamic models, specifically MagmaSat and MELTS, to simulate decompression-driven H₂O-CO₂ exsolution, melt-crystal phase development, and crystallization processes in magmatic systems. The resulting volatile budgets and phase equilibria are then used to parameterize our coupled finite element model, providing melt fraction, density, viscosity, and compressibility inputs to the fully coupled thermo-poroelastic deformation model. We explore three initial mush storage temperatures (800, 850, and 900 °C) and a range of recharge temperatures from 900 to 1300 °C. These conditions are implemented in a fully coupled 3D finite element model that resolves Darcy melt migration, heat transfer, and thermo-poroelastic deformation within a mush reservoir embedded in a linear elastic half-space. Our results show that low temperature recharge produces only small, isolated melt pockets, while hotter injections generate channels with high melt fraction. These channels grow upward and outward, forming vertically connected networks where melt fractions exceed ~50 vol%, a threshold for eruptibility. We constrain reservoir and injection parameters to yield realistic surface deformation, and we find that incorporating temperature- and volatile-dependent feedbacks does not alter the overall surface deformation pattern. These results provide a physical framework for understanding magma rejuvenation, channelized melt transport, and eruption triggering in crystal-rich reservoirs.
How to cite: Alshembari, R., Hickey, J., Mantiloni, L., and Mccormick Kilbride, B.: From Crystal Mush to Eruptible Magma: Thermo-Poroelastic Controls on Melt Channelization, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12677, 2026.
The transformation of hot, pyroclastic deposits into dense, coherent magma is a primary way that permeability evolves in shallow conduits, governing sealing timescales, plug formation, and ultimately shifts in eruptive style. This densification occurs via sintering, modulated by competing processes that include diffusive outgassing and vesiculation. Vesiculation introduces hysteretic volume and rheological changes within particles that can ultimately inhibit sintering of hydrous particle packs. While the role of temperature, water content, and grain size distribution in governing viscous sintering kinetics are well constrained, natural deposits commonly contain rigid crystals and experience compressive stresses, whose effects on sintering timescales remain poorly quantified. Here, we experimentally investigate how sintering of coarse ash-to-lapilli (0.50-2.50 mm), hydrous, rhyolitic clasts is modulated with increasing crystal fraction and by the application of stress up to 1-3 MPa. Our results can be divided into two different suites. First, we show that in the absence of applied load increasing crystal content systematically reduces sintering efficiency, preserving permeable pathways for longer. But second, under an applied stress, we induce extremely rapid densification and suppress vesiculation. Textural analysis shows that under stress, grain boundaries are erased, and near-vesicle-free obsidian is formed, even if crystallinities are as high as 60%, which is in stark contrast to the relatively poor sintering of crystal-bearing samples achieved in the absence of applied stress. We thus demonstrate the effects of crystal cargo and shallow stresses on densification, suggesting they exert first-order controls on permeability evolution in the shallow conduit - such as in tuffisites - and providing a framework for interpreting the variable degrees of sintering seen in silicic volcanic environments.
How to cite: Schunke, J., Kendrick, J., Birnbaum, J., Lamur, A., Wadsworth, F., Brauneis, K., and Lavallée, Y.: Sintering timescales of crystal-bearing pyroclast mixtures under stress , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-628, https://doi.org/10.5194/egusphere-egu26-628, 2026.
Understanding the controls on the explosive-effusive volcanic eruption transitions is central to hazard management and fundamental to building a more complete picture of eruption dynamics. While traditional models suggest that explosive and effusive eruptions represent very distinct behaviours and timescales, most silicic eruptions are complex, often exhibiting hybrid explosive-effusive behaviour and/or explosive ash-venting through effusive lavas. Dominantly explosive and dominantly effusive phases of eruptions can occur simultaneously or be separated in time by years. A recent conceptual eruption model proposes that the fragmentation and subsequent sintering of material within the conduit may provide a framework that explains both explosive and effusive eruptions, thereby blurring the boundaries of eruptive styles previously established. Here, we investigate this by measuring the material properties – porosity and permeability – of 150 samples from more than ten iconic eruptions representing both explosive (ignimbrites) and effusive (domes lavas) events. We find that across all porosities, the properties of dome lavas are strikingly similar to those of ignimbrites, suggesting that dome lavas may be a lava-like, thoroughly sintered product of pyroclasts. We also present macro- and micro-textures, with a focus on pore-network textures, to support the possibility that the lava samples are sintered products. Textural similarities include: (1) broken phenocrysts; (2) convolute pore networks typical of sintering; (3) cuspate vesicles at low porosity, indicative of the end-stage of sintering; (4) juxtaposed textures with very different groundmass crystallinity and mineralogy; and (5) direct textural evidence that some lavas are clastic. Our results provide visual and empirical evidence of sintering dynamics in dome rocks, supported by a large database revealing shared characteristics between effusive and explosive magma samples. We identify that, as for the blurred boundary between eruption styles, there is a significant overlap in physical properties across both types of samples and the presence of sintering relic textures in dome rocks, features previously associated primarily with explosive products. We provide evidence here that sintering is a key process governing eruptive transitions. By quantifying the similarities between effusive and explosive volcanic rocks, these discoveries and new datasets contribute to our understanding of eruption styles, paving the way for a new interpretation of conduit processes operating before, during, and after volcanic eruption.
How to cite: Theurel, A., Wadsworth, F. B., Llewellin, E. W., Humphreys, M. C. S., Kendrick, J. E., Tuffen, H., Lavallée, Y., Heap, M. J., and Lamur, A.: Blurring the boundaries between explosive and effusive eruption styles: New Evidence for Sintering-Driven Eruption Transitions , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5084, https://doi.org/10.5194/egusphere-egu26-5084, 2026.
Fracture and healing cycles in magma exert important controls on outgassing, permeability, and stability in dynamically evolving volcanic systems. Understanding these cycles requires constraint of the mechanical behaviour of viscoelastic melts. Here, we explore the role of shear stress on fracture mechanics using standard borosilicate melt as an analogue for magma.
Experiments were conducted in a biaxial press, where glass rods at high temperature (above Tg) were subjected to fracture-healing cycles by: 1) axially forcing two glass rods against each other at controlled temperatures (hence viscosity) and normal stresses of 0.1-10 MPa, and 2) holding contact for variable time between 1 to 9000 s, before 3) applying a controlled simple shear rate until rupture of the fracture healing pane, to 4) quantify shear strength recovery. The shear stress required to break the two rods apart again increased the longer the rods were held in contact before shearing. Here, strength recovery is defined as the shear stress measured upon rupture compared to that required to rupture a single intact rod. The initial strength of the glass (δ0) and the recovered strength along the healed artificial fracture (δ1) give the strength recovery (δ1/δ0). These results are compared to previous tests rupturing samples in a tensile regime, with samples showing lower strength recovery when broken in tension than at equivalent conditions in torsion.
We show that higher normal stress during healing accelerates healing by improving surface-surface contact area along the simulated fracture plane. Additionally, at higher temperatures and resultingly lower viscosities, surface-surface contact is improved, leading to more efficient strength recovery. We also show that smoother initial surfaces heal more efficiently, and relate this to strength recovery through time. Finally, we formulate a semi-empirical but physically-grounded model to explain the relationship between strength recovery and the variables: normal stress, viscosity, and time.
Our results highlight how shear strength of fractured melt recovers through time, and that recovery is both normal stress, surface roughness, and viscosity dependent. This suggests that fractures in volcanic systems may heal at different timescales and depths within the system. Resultingly, this leads to differences in the efficacy of degassing that could facilitate localised pressure-increase and contribute towards dictating eruptive style.
How to cite: James, H., Kendrick, J. E., Lamur, A., Birnbaum, J., Wadsworth, F. B., and Lavallee, Y.: Mechanical controls on magma fracture-healing cycles: Experimental insights , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-992, https://doi.org/10.5194/egusphere-egu26-992, 2026.
Deformation of magmas and hot rocks occurs at a range of strain rates in natural systems including rapid deformation as magma is sheared against the conduit wall upon ascent or during collapse of parts of the volcanic edifice. The initiation of cracks and fractures in magma is crucial to the development of permeable pathways through which volatiles may degas and alleviate overpressure in the system. Yet, experimental data on the deformation of hot magmas at high strain rates remains sparse, with the majority of tests conducted at strain rates on the order of 10-5 s-1. Using a drop tower equipped with a furnace, we subject high temperature (890 – 950 °C) rhyolitic obsidians to high strain rate impacts (100-102 s-1) at various impact energies. Our results indicate a strong effect of both temperature and strain rate on the peak stress recorded in the melts. Despite being far above their glass transition temperature (717 °C), the samples all deform in a brittle manner, owing to the ratio between relaxation to observation timescales which is expressed as the dimensionless Deborah number (De). At colder temperatures (890 – 930 °C), samples behave predominantly elastic-brittle whereas at higher temperature (950 °C) the increasing viscous component of deformation weakens the melt, causing lower peak stresses and more comprehensive fragmentation. Our findings provide insights into how changes in temperature, energy and strain rate affect the rupture behaviour of melts, thereby improving our understanding of dynamic magmatic processes such as magma-conduit interaction upon magma ascent.
How to cite: Heinrigs, K., Kendrick, J. E., Hess, K.-U., Lamur, A., Birnbaum, J., and Lavallée, Y.: Drop it like it’s hot - viscoelastic deformation of rhyolitic melt at high strain rates, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-515, https://doi.org/10.5194/egusphere-egu26-515, 2026.
Understanding the physico-chemical conditions governing magma plumbing systems is one of the central objectives in volcanological research, as eruptive styles and associated phenomena are strongly influenced by pressure, temperature, and volatile content in magmatic reservoirs. However, precisely constraining pre-eruptive pressure conditions remains challenging. Experimental investigations on phase relationships and stability fields of pressure- and volatile-sensitive mineral phases can provide insightful information to address this issue. Recent geological surveys on Ventotene Island (Pontian Islands, Tyrrhenian Sea) revealed the presence of primary analcime in the Cala Battaglia Unit (UCB), a sequence of pure Plinian fallouts lacking pyroclastic density currents (PDCs) deposits. In contrast, analcime is absent in both Plinian fallout and PDC deposits related to the Parata Grande caldera-forming eruption. Since analcime stability is generally constrained to PH₂O > 200 MPa, these observations point to significant differences in pre-eruptive storage pressure conditions, highlighting the fundamental role of pressure in controlling phase relations, mineral stability, and therefore eruptive style. Phase equilibria experiments were performed using a piston-cylinder apparatus to investigate the role of pressure and volatile content (H₂O) on phase relations in differentiated alkaline magmas. Two starting compositions representative of two eruptive units were selected for the experimental runs: a tephriphonolite (MD1, Parata Grande) and a trachy-phonolite (UCB2, Cala di Battaglia). Experiments were performed under H2O-undersaturated to H2O-oversaturated conditions at pressures of 150, 300, and 600 MPa, and temperatures between 700 and 1000 °C. For the MD1 tephriphonolite at 600 MPa, the mineral assemblage consists of clinopyroxene + apatite + oxides at 1050 °C, followed by biotite, plagioclase, and K-feldspar with decreasing temperature, whereas under H₂O-oversaturated conditions at 950 °C the assemblage is dominated by biotite, clinopyroxene, oxides, and apatite. At 300 MPa, all experiments were conducted under H₂O-saturated conditions, and the mineral assemblages are dominated by clinopyroxene, biotite, and oxides. For the UCB2 trachy-phonolite, experiments at 600 MPa show the crystallization of a hydrous feldspathoid associated with plagioclase and biotite at 900 and 850 °C, followed by K-feldspar at lower temperature (750 °C). In contrast, hydrous feldspathoids do not crystallize at 150 MPa, where the mineral assemblage is limited to K-feldspar, plagioclase, biotite, and oxides. At 300 MPa, the assemblage is dominated by K-feldspar and plagioclase, with subordinate biotite and oxides. These results suggest that hydrous feldspathoid cannot crystallize from H₂O-saturated trachy-phonolitic magmas at pressures ≤150 MPa, emphasizing how pressure variations can affect phase equilibria. This evidence supports the hypothesis of a polybaric differentiation path for the Cala Battaglia plumbing system, leading to pure Plinian events, in contrast to a shallower and isobaric evolution for the Parata Grande system, leading to under-pressure caldera-forming events.
How to cite: Pavese, R., Perinelli, C., Palladino, D. M., Fabbrizio, A., Masotta, M., Colle, F., Monaco, L., and Gaeta, M.: Magma storage conditions as key factors in the genesis of pure Plinian vs. caldera-forming eruptions: Experimental constraints, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18967, 2026.
Most volcanoes on Earth display a range of eruptive styles, alternating between effusive and explosive activity. Understanding the factors controlling these transitions is critical for volcanic hazard assessment but remains debated. While both pre-eruptive storage conditions and syn-eruptive conduit processes influence eruption style, their relative roles are often difficult to disentangle, particularly in alkaline volcanic systems. Here we investigate the transition between effusive and explosive behaviour recorded in the historical eruptive activity of Ischia Island (Phlegraean Volcanic District, southern Italy), where trachytic magmas produced both small volume lava domes and explosive eruptions of variable intensity. We focus on three eruptions from the Montagnone volcanic complex whose activity started in the 6th century B.C.: the Cretaio Tephra (sub-Plinian), the Bosco di Conti Tephra (sub-Plinian to lava fountaining activity), and Montagnone–Maschiata lava dome. We investigated magma storage conditions and processes through phenocrysts’ zoning patterns and compositions, including clinopyroxenes, plagioclases, alkali-feldspars, and apatites. Volatile content (H2O, CO2, F, Cl) was measured in clinopyroxene-hosted trachytic melt inclusions and groundmass glass. Pre-eruptive temperature, pressures, and water content were additionally calculated through clinopyroxene-liquid thermobarometry and alkali feldspar-liquid hygrometry. The phenocrysts content is low in both explosive (pumice clasts) and effusive (lava) products. The mineral assemblage is identical across the different eruptions, characterised by unzoned sanidine phenocrysts with plagioclase crystals as core, normal-zoned plagioclase crystals and oscillatory sector-zoned clinopyroxenes. We also observe the presence of sanidine microlites in the lava, whilst pumice clasts are characterised by a microlite-free groundmass. Clinopyroxene-liquid geothermometer indicates a pre-eruptive temperature of 915±18ºC. Water and CO2 contents in melt inclusions and their solubility in the Montagnone trachytic magmas calculated using the MagmaSat model indicate minimum pre-eruptive pressures between 100 and 200 MPa. Our results suggest crystallisation of a trachytic magma as the dominant process in the magma reservoir prior to the eruption, and extraction of melt from the resulting crystal-rich environment. The three eruptions from the Montagnone complex were fed by those extracted magma batches, with low crystal content and indistinguishable pre-eruptive state from one eruption to another. Our petrological and geochemical data indicate that magma mixing or mingling with a more mafic magma prior to eruptions is unlikely. The study of Montagnone eruptions give new constraints for models of Ischia’s plumbing system, trachytic magmas existing at larger depths than previously thought. Our study highlights that trachytes can erupt effusively or explosively with similar magma reservoir conditions, and that the observed differences in eruptive styles will be induced by conduit dynamics. This must be considered for volcanic risk management in trachyte-dominated volcanic areas like Ischia or the nearby Campi Flegrei.
How to cite: Maingault, L., Arzilli, F., Pelullo, C., Balcone-Boissard, H., Popa, R.-G., Arienzo, I., Carroll, M. R., Sansivero, F., Chakraborty, S., De Vita, S., and De Hoog, C.-J.: Effusive and explosive eruptions of trachytic magmas shared common pre-eruptive storage conditions and processes at Ischia volcano, Italy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-849, https://doi.org/10.5194/egusphere-egu26-849, 2026.
Style transitions are common during volcanic eruptions, but the processes that trigger them remain poorly constrained. Understanding these transitions is crucial for hazard assessment, as changes in eruptive style can alter associated risks. Vesiculation plays a fundamental role in magma ascent and eruption dynamics, and the content, size, shape, and distribution of vesicles in pyroclasts record key information on conduit processes and subsequent eruptive behaviour. In this work, we combine 3D textural analyses of basaltic and basaltic-andesite pyroclasts with conduit-flow numerical modelling to investigate the processes driving style transitions at Villarrica, Chile’s most active and highest-risk volcano. Villarrica is characterised by a constant, background Strombolian activity. In March 2015, Strombolian activity progressed quickly into a 1.5-km-high lava fountain. After that event, the activity continued shifting intermittently between Strombolian explosions and open-conduit degassing from the active lava lake.
We analysed 12 pyroclasts erupted between the March 2015 paroxysm and 2024, using X-ray microtomography (XCT) and SEM imaging. Vesicularity ranges from 0.61–0.88 in paroxysm clasts and 0.44–0.93 in post-paroxysm clasts. All samples are characterised by vesicle number densities of ~1012 m-3. Permeability parameters, such as vesicle tortuosity (1.54–2.16) and pore-throat ratios (0.20–0.35), show limited variation and lack clear patterns that differentiate eruptive contexts. This overlap could relate to high decompression and ascent rates, which may restrict outgassing, or suggest that other processes are likely influencing the resultant eruptive style. We used XCT-derived parameters and published eruptive conditions (composition, pressure, temperature, crystal content) to model conduit flow during the 2015 paroxysm and background activity. The 2015 lava fountaining is reproduced with a 4-m-radius conduit and 4 wt.% H₂O, showing no fragmentation within the conduit and gas ascending coupled to the melt. Background, mild-explosive eruptions were reproduced with 2.5 wt.% H₂O, showing fragmentation at ~900m depth and lower decompression rates. Our integrated results indicate that the eruptive style at Villarrica is governed not by a single permeability parameter, but by the complex interplay among outgassing, volatile content, and ascent rate.
How to cite: Rojas Guzmán, F., Polacci, M., Biagioli, E., Bonechi, B., La Spina, G., Evans, E., Romero, J., Neave, D., and Basualto, D.: Combining 3D vesicle textural analysis and numerical modelling of magma ascent to understand eruptive style transitions at Villarrica volcano (Chile), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1303, https://doi.org/10.5194/egusphere-egu26-1303, 2026.
The dynamics of eruptions are controlled by volcanic outgassing through interconnected bubble networks. The question of whether frameworks formed by crystals facilitate gas escape remains unresolved, with the relative spatial arrangements of bubbles and crystals yet to be elucidated. In the presentation, we will show multi-resolution X-ray computed tomography (CT) imaging conducted to examine vesicle-crystal spatial relationships in a dacite bomb, a lava block in a pyroclastic flow deposit, and spine lavas from the 1990-1995 Unzen eruption. The acquisition of micro-CT, computed laminography (CL), and nano-CT images with progressively higher resolution was undertaken. Reconstructed 3D images demonstrate that large vesicles are consistently connected to crystals across all sample types and analysis scales. Size distribution analysis demonstrates preferential connectivity between large vesicles and large crystals. Vesicles in the bread crust bomb that appear isolated are found to form interconnected networks. In contrast, vesicles in shear-deformed dome samples are found to occupy narrow inter-crystal gaps as sheet-like structures. The findings of this study indicate that interconnected bubble networks facilitate efficient initial outgassing at depths of 0.5-0.8 km. The findings of calculations of compaction timescales and gas flow modelling corroborate this assertion. Crystal supported pathways facilitate the subsequent transport of ascending gas through shallow conduit regions. The compaction of crystal-bearing, interconnected bubbles causes the subsequent development of crystal frameworks. These phenomena serve as pathways for outgassing during the final ascent, leading to the formation of the dome.
Reference:
Saito, K., Hoshino, M., Goto, A., Namiki, A. A computed tomography observation of the Unzen lava reveals the frequent existence of vesicles and crystals in proximity. Sci Rep 16, 81 (2026). https://doi.org/10.1038/s41598-025-28770-4
How to cite: Saito, K., Hoshino, M., Goto, A., and Namiki, A.: A computed tomography observation of the Unzen lava reveals the frequent existence of vesicles and crystals in proximity, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2676, https://doi.org/10.5194/egusphere-egu26-2676, 2026.
As the largest and most active intracontinental volcanic system in Northeast Asia, the Changbaishan volcano (CBSV) complex encompasses multiple active edifices. Although all these volcanoes are influenced by asthenospheric upwelling triggered by Pacific Plate subduction, they display marked contrasts in magma composition and eruptive behavior. This study presents the first high-resolution, crust-to-mantle three-dimensional (3D) electrical resistivity model of the CBSV, constructed through densely sampled Magnetotelluric (MT) surveys and 3D inversion. The model reveals a trans-lithospheric magmatic network and a dynamic plumbing architecture, supporting a "deep-source homology and shallow differentiation" evolutionary paradigm for the volcanic field. Integrated analysis of electrical structures, seismic activity, and geodetic deformation further demonstrates ongoing magma recharge beneath the volcanic field. Our results highlight heterogeneous eruption potentials among the CBSV: Tianchi Volcano (TCV), the most prominent edifice, possesses a well-developed shallow magma reservoir and a complex, multi-level plumbing system, indicating a higher likelihood of large-scale eruptions and associated hazards. In contrast, Longgang Volcano (LGV), lacking a major shallow magma chamber, is more susceptible to smaller-scale, fault-controlled fissure eruptions. These findings provide critical theoretical insights and practical guidance for volcanic hazard assessment in the CBSV. More broadly, we propose a conceptual framework explaining the diversity of intracontinental eruption styles far from plate boundaries, emphasizing that fault architecture and topographic loading exert dominant control over magma transport pathways and eruption dynamics. This work advances the understanding of intracontinental volcanism from a "magma-composition-centered" perspective toward a "tectonic-melt coupling" framework, with far-reaching implications for global volcanic research.
How to cite: Zhao, L., Zhan, Y., Hu, Y., Cao, C., Wang, Q., and Sun, X.: Three-Dimensional Electrical Structure Beneath the Changbaishan Volcano (NE China): Implications for Magmatic Plumbing Systems and Diverse Eruption Styles, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1475, https://doi.org/10.5194/egusphere-egu26-1475, 2026.
The Huangzuei complex is a volcanic complex within the 25 km-wide Tatun Volcanic Group (TVG), which is situated on the northern tip of Taiwan and poses significant potential hazards to the Taipei metropolitan area. The Huangzuei complex has well-preserved landforms with three sets of lava flows that radiate northwestward, northeastward and southeastward up to several kilometers from Mt. Huangzuei, the 250-m-high cratered summit of the Huangzuei complex.
High-resolution topography from a LiDAR-based bare-earth DEM enables detailed demarcation and characterization of individual flows, including volume, thickness, and profiles. Results of our geomorphological mapping and 40Ar/39Ar dating of volcanic matrix reveal that eruption of the Huangzuei complex began at 139 ka, was most vigorous from ~129 ka to 133 ka and resumed at ~116 ka. The south-eastern branch of lavas partially overlap ~264-298 ka lavas that may have erupted from Dajianhou, another neighboring volcanic complex. It is likely that the Huangzuei volcanism ended in the formation of domes and a summit crater at 85 ka or later.
Disequilibrium textures, including oscillatory zoning, reverse zoning, partial resorption, and development of reaction rims, are common in samples from the Huangzuei complex. These textures, together with the wide ranges of mineral compositions (Mg# of pyroxenes and An content of plagioclase), indicate that the magma that formed the Huangzuei complex underwent intense magma hybridism, likely involving periodic replenishment by more primitive magmas during crystallization differentiation. We propose the latter as the cause of the voluminous lava emplacement from ~129 to ~133 ka.
How to cite: Li, W.-C., Sieh, K., Jicha, B., Pang, K.-N., Wang, Y., Nguyen, T. T., and Chan, Y.-C.: Geomorphological mapping, 40Ar/39Ar geochronology, and geochemistry of the Huangzuei complex (Tatun Volcanic Group, Taiwan): Implications for volcanic architecture and magma generation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7723, https://doi.org/10.5194/egusphere-egu26-7723, 2026.
Fragmentation is an essential process in explosive eruptions breaking a continuous magma into small pieces, generating pyroclastic materials, and enabling the rapid expansion. Fragmentation of basaltic and basaltic andesitic magmas with low viscosity produces pyroclasts in various shapes, such as scoria, pumice, volcanic ash, and Pele’s hair. The shapes of pyroclasts from low-viscosity magma may preserve their deformation history. Brittle fracture of low-viscosity magma is unlikely; instead, fluid-dynamical deformation during the eruption tears off the magma. Crystallization in ascending magma increases its effective viscosity, thereby extending the relaxation time, and potentially leading to brittle fragmentation. However, rheology measurements of crystal-bearing magma indicate that large strains can unjam the crystals, leading to deformation instead of brittle fracture.
To understand how the shapes of pyroclasts generated from low-viscosity magma are determined, uniaxial tensile experiments were conducted using a magma analogue with three phases: liquid, solid, and gas. These experiments simulate processes in a lava fountain, an elongation of a magma parcel. In our experiments, fluids with particle volume fractions >0.3 tend to tear off at a low strain and exhibit rough fracture surfaces. A high volume fraction of undeformable solid particles reduces the thickness of the deformable liquid region, resulting in easy rupture of thin liquid films. Therefore, the entire fluids are broken at a small strain. On the other hand, bubbly fluid without solid particles generates thin threads of the liquid phase. Bubbles can coalesce into larger bubbles and deform, elongating vertically, splitting the fluid longitudinally to form fibers. The fracture mechanism associated with high particle fraction may generate scoriae with irregular shapes. The mechanism forming thin threads due to bubble deformation may produce Pele’s hair. Our experimental results indicate that bubbles and crystals included in magma affect the fracture manners of magma as well as its physical properties, and that this determines the shapes of pyroclasts variously.
Reference
Oda, S. & Namiki, A. An analogue experiment showing varying shapes of pyroclasts by including crystals and bubbles. Journal of Volcanology and Geothermal Research, in press.
How to cite: Oda, S. and Namiki, A.: An analogue tensile experiment of particle and bubble suspension producing various shapes of pyroclasts, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8534, https://doi.org/10.5194/egusphere-egu26-8534, 2026.
Magmatic volatiles shape eruption dynamics, but a substantial fraction of magmatic CO2 can outgas into crustal hydrothermal systems during shallow storage. Here we combine time-resolved measurements of eruptive gas compositions from the 2021–2025 Reykjanes eruption sequence with chemical monitoring of geothermal fluids at Svartsengi to track CO2 transfer from magma into the upper crust. While the 2021 Fagradalsfjall eruption initially emitted CO2-rich gas, subsequent eruptions at Fagradalsfjall in 2022–2023 and Sundhnjúkur in 2023–2025 emitted persistently CO2-poor gases from the onset of activity. Although the Sundhnjúkur eruptions took place in close proximity to the nearby Svartsengi geothermal power plant, elevated CO2 in geothermal steam only emerged several months later, indicating extensive pre-eruptive degassing of stored magma and a delayed, time-integrated hydrothermal response. Across the Sundhnjúkur eruptions, erupted volumes scale with the CO2 content of eruptive gas, with the smallest eruptions associated with the most CO2-depleted magmas. These observations are consistent with pre-eruptive CO2 loss modulating eruption magnitude and highlight the role of shallow crustal magma reservoirs as dynamic filters that redistribute magmatic carbon from magma to hydrothermal systems during basaltic rifting episodes.
How to cite: Scott, S., Pfeffer, M., Mandon, C., Oppenheimer, C., Caraciollo, A., Ranta, E., Matthews, S., Bali, E., Halldórsson, S., Pedersen, G., Lanzi, C., Huntingdon-Williams, A. G., Mesfin, K., Burton, M., and Stefánsson, A.: Pre-eruptive CO2 loss during shallow magma storage and its impact on eruption volumes at Reykjanes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9486, 2026.
The recent evolution (last 700 years) of La Fossa cone (Vulcano island) has long intrigued scientists and sparked debates regarding the origin and timing of products belonging to its various eruptive units. Although very recent, the stratigraphy of these products, the timing of the main eruptions and their characters, and the identification of the source area(s) still remain partly unclear. In the present work, we aim (i) to reconstruct in detail the stratigraphy of Pietre Cotte pyroclastic succession, representing the result of the activities from the XIV century up to the latest AD 1888-90 eruptive cycle, (ii) to define the processes that control the opening phases of explosive eruptions in intermediate–sialic systems characterized by closed conduits and intense interaction with active hydrothermal systems, and (iii) to understand the mechanisms responsible for variations in eruptive style during the eruptions. Stratigraphic fieldwork, lithofacies analysis and volcanological interpretation have been carried out, together with laboratory analyses of representative sample components, volcanic-glass geochemistry, thin-section petrography, and morphoscopic SEM analyses. EPMA glass analyses define two distinct compositional domains (trachytic and rhyolitic), with the gap bridged by the products of the 1888–90 eruptive cycle, indicating effective mixing/mingling processes within the shallow magmatic system. Juvenile fragment morphologies are consistent with phreatomagmatic fragmentation and magmatic degassing, with localized hydrothermal alteration. These studies have led to the definition of an updated volcanic succession result of recurring hydromagmatic to magmatic eruptions, with vent-opening phreatic phases, that produced multiple depositional units from fallout and pyroclastic density currents. The Pietre Cotte succession includes distinctive pumice-fallout layers and the well-known rhyolitic lava flow, reflecting a complex eruptive evolution. Based on our stratigraphic data and the re-interpretation of the available historical reports, combined with the available paleomagnetic ages, the rhyolitic lava flow is most likely dated to AD 1739, whereas the pumice fallout layers were likely emitted slightly after in AD 1771. Constraining the precise timing of the main explosive and effusive events is crucial to better understand the dynamics of La Fossa’s shallow magmatic–hydrothermal system and the evolution of its most recent eruptive activity, thus providing key insights for volcanic hazard assessment on Vulcano Island.
How to cite: Panelli, G., Roverato, M., Forni, F., De Astis, G., Natale, J., Sulpizio, R., Tranne, C. A., and Lucchi, F.: Unravelling the Stratigraphy and Eruptive Dynamics of the Pietre Cotte and the 1888–90 eruptions activities, Vulcano (Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-600, https://doi.org/10.5194/egusphere-egu26-600, 2026.
Explosive eruptions represent the most powerful and hazardous manifestations of volcanic activity on Earth. During such events, a high-velocity gas/pyroclast mixture is injected in the atmosphere, producing ash columns that can reach altitudes of tens of kilometres. Depending on atmospheric conditions and eruption intensity, these ash clouds can travel thousands of kilometres, potentially disrupting the air-traffic and the climate worldwide.
Jet flow dynamics of explosive eruptions are affected by several parameters, including pressure gradient, temperature of the magmatic mixture, particle mass and size distribution, and vent geometry. However, most of these parameters cannot be measured directly during an eruption. Conversely, some of the characteristics of the volcanic jets, such as exit velocity, jet dimension, and acoustic signals, can be collected by volcanic monitoring systems.
To correlate jet flow characteristics with magmatic conditions below the vent of the conduit, we investigated jet flow dynamics using a combination of shock-tube experiments and numerical simulations. Data from laboratory experiments were collected using a high-speed camera to capture the evolution of the jet at high temporal resolution, as well as acoustic signals from microphones. Schlieren shadow photography has also been adopted to visualise shock waves and density contrasts within the gas during the experiments. Using this setup we investigated the role of particles on jet flow characteristics. Preliminary results indicate that in the supersonic regime, both the amplitude of acoustic signals and the spectral properties of the signal are influenced by solid loading. Finally, transient numerical simulations of the shock-tube experiments were also performed to correlate the evolution of the jet features with the internal thermodynamic conditions of the gas/pyroclast mixture.
How to cite: La Spina, G., Spina, L., Taddeucci, J., Pennacchia, F., Posati, A., Perugini, D., and Scarlato, P.: Jet flow dynamics of explosive eruptions: laboratory and numerical investigation of shock-tube experiments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10109, 2026.
The Campi Flegrei caldera in southern Italy represents one of the most hazardous volcanic systems on Earth, owing to its potential for explosive eruptions and its proximity to a densely populated metropolitan area. Volcanism at Campi Flegrei spans a spectrum of eruptive styles, from effusive dome-building events to high-intensity explosive eruptions, and is dominated by trachytic to trachy-phonolitic magmas, with subordinate latitic compositions. While trachytic eruptions are typically associated with high explosivity, latitic magmas are less frequent and generally produce lower-energy eruptions. Constraining the pre-eruptive processes and magma residence timescales prior to eruption is critical for assessing hazards in this high-risk volcanic setting. However, the difference between pre-eruptive processes and timescales prior to highly explosive trachytic and less explosive latitic eruptions are poorly constrained. Here we integrate chemical analyses of natural samples, geothermometric constraints, thermodynamic simulations and crystallisation experiments (cooling and decompression) to constrain the processes, conditions, and timescales that lead to highly explosive and Strombolian eruptions of trachytic and latitic magmas at Campi Flegrei.
How to cite: Arzilli, F., Bamber, E. C., Appiah, E., Stabile, P., Morgavi, D., Abeykoon, S., Piccini, V., Krasheninnikov, S. P., Almeev, R. R., Holtz, F., and Carroll, M. R.: Pre-eruptive processes and timescales of highly explosive and Strombolian eruptions at Campi Flegrei: The case studies of Agnano Monte-Spina and Fondo Riccio, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14473, 2026.
Temporal variations in magma plumbing architecture and magmatic processes can modulate eruption priming, with direct implications for the interpretation of pre- and syn-eruptive signals. However, the mechanisms in which these processes operate in low-flux volcanoes remains poorly constrained, limiting our understanding of eruption precursors. Here we investigate the temporal evolution of magmatic processes at La Palma (Canary Islands), a low-flux ocean-island basaltic system, by examining clinopyroxene zoning from three historical eruptions that record a transition from tephritic to basanitic lava compositions: El Charco 1712, Teneguía 1971, and Tajogaite 2021. By integrating major and trace element data from clinopyroxene crystals and carrier melts with textural observations, thermobarometry, quantitative trace element mapping, and cluster analysis, we reconstruct the magmatic processes and storage conditions preceding these eruptions. Both tephritic and basanitic magmas were stored in the upper mantle (18–25 km depth) together with an evolved tephritic-to-phonolitic crystal mush, preserved in clinopyroxene antecryst cores. This phonolitic mush was stored at lower temperatures and likely formed through >80% fractional crystallization of a basanitic melt. Prior to each eruption, repeated injections of basanitic recharge melts progressively eroded and remobilized the mush, after which the recharge magma experienced ~10–20% fractional crystallization, generating the tephritic melt. Despite its central role in priming the system, mafic recharge did not act as the immediate trigger for the historical La Palma eruptions. The lack of recharge-related signatures in inner rims of early-erupted tephrite-hosted clinopyroxenes shows that eruption onset was likely controlled by internal reservoir processes rather than by mafic recharge events.
How to cite: Petrelli, M., Caracciolo, A., Ubide, T., Ágreda-López, M., Herrera, R., Marquez, A., González-García, D., Huertas, M. J., Ancochea, E., Chicharro, N., and Coello-Bravo, J. J.: Deep mafic recharge priming ocean island basalt volcanoes: clinopyroxene evidence from La Palma, Canary Islands, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19721, 2026.
The 2023–2024 eruption sequence of Shiveluch volcano (Kamchatka) demonstrates that complex dome-building volcanic systems can rapidly activate and evacuate magma storage zones at markedly different crustal depths. The sequence began on 10 April 2023 with a major explosive eruption (VEI ~4) of Young Shiveluch, the principal Holocene eruptive center of the volcano. The eruption destroyed the active lava dome and produced a 15–18 km-high ash column [1], widespread tephra fallout, and pyroclastic density currents extending up to 19–20 km from the vent. Juvenile products are amphibole–plagioclase andesites typical of modern Young Shiveluch activity, derived from a shallow upper-crustal reservoir at ~5–6 km depth [e.g., 2]. As in previous modern eruptions [3], the 2023 andesites contain olivine-bearing mafic enclaves, recording interaction between resident evolved magma and an injected mafic component.
Only one year later, in April 2024, a new eruptive center formed on the western flank of the volcano, ~5.5 km from the Young Shiveluch crater. The newly formed lava dome produced amphibole-rich basaltic andesites [4], a rare magma type in arc volcanoes and previously unknown at Shiveluch. These rocks contain up to 30 vol.% amphibole, sparse pyroxene, reaction-rimmed olivine, and no plagioclase phenocrysts. Amphibole thermobarometry indicates crystallization at 912–948 °C and 410–632 MPa, corresponding to mid- to lower-crustal storage at ~16–24 km and high pre-eruptive H₂O contents (7.4–8.7 wt%).
Petrological data therefore show that the 2023 and 2024 eruptions tapped two distinct magma storage zones that were activated within only one year. Two end-member mechanisms may explain this rapid succession: (1) evacuation of the shallow reservoir in 2023 induced decompression and stress redistribution, triggering ascent of a volatile-rich deep magma batch; or (2) increased mafic input from the mantle pressurized several crustal reservoirs nearly simultaneously.
Seismic observations support aspects of both models [5]. The 2023 eruption was mainly accompanied by shallow seismicity associated with the upper-crustal reservoir, whereas earthquakes consistent with deep mafic magma input have been recorded since 2021. A distinct cluster of seismicity at depths of ~20–26 km developed immediately after the April 2023 eruption, indicating renewed magma transport in the lower crust prior to the 2024 flank eruption.
Overall, the Shiveluch case demonstrates that deep magma storage zones can be rapidly mobilized following major explosive eruptions, generating new vents outside the main crater area and magmas with contrasting compositional properties. This behavior complicates hazard assessment, as future eruptions may involve deeper, volatile-rich magma capable of producing unexpected eruptive styles, highlighting the importance of integrating petrological and seismic constraints into monitoring of large, long-lived dome-building volcanoes.
The work was supported by the Russian Science Foundation, project no. 25-27-20039 (https://rscf.ru/en/project/25-27-20039/)
References: [1] Girina et al. (2023) Mod. Probl. Remote Sens. Earth Space, 20, 283–291; [2] Dirksen et al. (2006) J. Volcanol. Geotherm. Res., 155, 201–226; [3] Goltz et al. (2020) Contrib. Mineral. Petrol., 175, 115; [4] Gorbach et al. (2025) J. Volcanol. Seismol., 19, 1, S44-S54; [5] Shakirova, Chemarev (2025) J. Volcanol. Seismol., 19, 1, S110-S117.
How to cite: Gorbach, N., Shakirova, A., and Chemarev, A.: Multi-level magma storage zones feeding the 2023–2024 eruption sequence at Shiveluch volcano (Kamchatka): petrologic and seismic constraints, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14684, 2026.
Multiple magma storage levels are commonly recognized beneath magmatic systems and play
an important role in the processes leading to the build-up of large silicic magma chambers in the crust, with possible
critical implications for the occurrence of explosive eruptions. Within such reservoirs, interactions between
different magmas due to new recharge events are common processes, as demonstrated by the presence of mafic
enclaves, which also reveal the occurrence of magma immiscibility conditions.
At Nisyros volcano (Greece), the two most recent eruptive events, the caldera-forming explosive eruption of the
Upper Pumice (UP) and the following effusive activity of the Post Caldera Domes (PCD), emplaced a
thick pyroclastic deposit and six main lava domes, both hosting mafic juvenile products like crystal-rich clasts (CRCs)
and enclaves, respectively. These two eruptions show differences in the abundance, petrographic characteristics,
mineral chemistry, and geochemical and isotopic signatures of their mafic components, as well as in the extent
of the mingling processes, indicating that the magma interaction conditions were different, possibly related to
a change in the magma chamber dynamics and/or in the deeper feeding system structure.
In this work, we investigated the textural characteristics and mineral chemistry of the products erupted by
these two eruptive episodes, exploring their crystallization histories and the possible variations in physical conditions
to reconstruct the structure of the plumbing system throughout the two phases of activities. Our results
revealed the occurrence of evident mineral disequilibria within CRCs and enclaves related to their rapid crystallization
due to the undercooling within the host. In the PCD systems, mineral disequilibria are also related to
the extensive crystal transfer from the host to the enclaves and vice versa, generating a microscale mingling,
which increases with time. The application of geothermobarometers highlight a progressively increase in pressure from
the UP to the PCD under similar temperature conditions. This indicates a deepening of the main eruptible reservoir,
sampled by the PCD activity, after the UP–caldera collapse. We infer that between the two periods, an interconnected
evolved magma-rich system developed through new inputs of mafic melts that refilled and reheated the system,
progressively mingling with the host and generating new conditions for the eruption.
How to cite: Braschi, E., Giannetti, F., Mastroianni, F., Orlando, A., Avanzinelli, R., Tommasini, S., Vougioukalakis, G. E., and Francalanci, L.: Changing magma dynamics and plumbing system architecture at an explosive–effusive transition: the case of Nisyros volcano (Greece), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14849, 2026.
The current study presents a detailed investigation of the field relationships, whole-rock geochemistry, Nd isotope analysis and zircon geochronology of less-studied mafic volcanic–subvolcanic extrusive rocks from the Bengpal Group, southern Bastar Craton. This study reports the first U-Pb zircon date (~2.25 Ga) for a coarse-grained mafic rock, providing a precise crystallization age for these mafic rocks of the Bengpal Group. These extrusive mafic rocks exhibit tholeiitic characteristics and are geochemically classified as basalt to basaltic andesite. Moreover, these rocks exhibit primitive mantle normalized enrichment in LILEs and relative depletion in HFSEs (Nb, Ta, P, and Ti). Four cumulate samples show high-Ti nature (TiO2 = 2.09–2.75 wt%) and are geochemically distinguished by the enrichment of Fe-Ti, Nb, Ta, and REEs, hence classified as high-Ti basaltic rocks.
The initial εNd values and other immobile trace element concentrations of the Bengpal mafic rocks suggest that these rocks are derived from lithospheric mantle (at shallow depth) with a depleted mantle source, accompanied by minor input of crustal assimilation and crystal fractionation. Based on the association of volcano-sedimentary rocks and geochemical signatures, we propose that these preserved volcanic–subvolcanic mafic rocks were emplaced in a stable continental shelf-like environment within an intraplate setting.
The Bengpal mafic rocks are correlated with previously reported coeval continental intraplate mafic magmatic events in other cratons in the Indian subcontinent and worldwide. This global correlation indicates a potential connection with global mafic magmatic events associated with intracontinental rifting events. Therefore, the new precise U-Pb dating of zircon unravels that this magmatic event is most likely associated with early Paleoproterozoic supercontinent break-up events driven by global-scale intracontinental volcanic activities.
How to cite: Verma, A. and Dey, S.: Paleoproterozoic Continental Intraplate Mafic Magmatism and its Contemporaneous Global Links: Evidence from the Southern Part of The Bastar Craton, Central India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15859, 2026.
Since 2005, Campi Flegrei (Italy) has been experiencing volcanic unrest characterized by earthquake swarms with magnitudes up to 4.6 and ground uplift rates of 10–30 mm/month, amounting to approximately 160 cm since the onset of unrest. With ~500,000 residents living within the red zone, it is critical to assess how the ongoing crisis may evolve and to identify potential eruption-triggering scenarios. Here, we apply clinopyroxene-only thermobarometry based on supervised machine learning [1] to pyroclastic samples collected from the opening and upper units of emblematic eruptions spanning a wide range of ages, eruption styles and intensities, and locations within the caldera.
Clinopyroxenes from the opening units display Mg# values (0.7–0.95) comparable to those of the upper units, except for Triflisco, but show stronger bimodality. In contrast, Mg# distributions in the upper units are generally more homogeneous. Thermobarometric estimates indicate that all eruptions were preceded by magma storage at shallow depths of 1–2 kbar (~4 km). Except for Monte Nuovo and Agnano Monte Spina, clinopyroxenes from the opening units also record the extraction of hot magma (~1100 °C) from depths exceeding 2.5 kbar (>8 km). In general, high-intensity eruptions (i.e., Neapolitan Yellow Tuff and Triflisco) show bimodality in the crystallisation temperature estimates separated by a clear gap at 900–1000 °C. Among the studied eruptions, only Triflisco shows clear petrological evidence for magma recharge as a triggering mechanism, whereas Agnano Monte Spina and Monte Nuovo were likely triggered by external processes, such as elastic crustal weakening [2]. Strikingly, the Campanian Ignimbrite displays continuous temperature estimates similar to those observed in low-intensity eruptions (e.g., Solfatara, Santa Maria delle Grazie, Averno 1), suggesting that reservoir configuration alone does not control eruption magnitude or intensity, and that surface deformation characteristics, mostly controlled by dynamics in the shallow portion of the plumbing system, might not be directly linked to the magnitude of a future eruption.
The shallowest depths of magma emplacement correspond to the present-day source of seismicity and ground deformation, while deeper storage levels match the deep reservoir imaged by seismic and recent magnetotelluric surveys [3], highlighting strong links between petrological and geophysical observations. We infer that the volume of eruptible magma present within the mid-lower crustal reservoir is a key parameter in estimating the intensity of a future eruption. This study provides new constraints on preferred magma pathways that may precede eruptive activity, as well as on plausible eruptive scenarios, both of which are essential for volcanic hazard assessment and risk mitigation.
[1] Ágreda-López et al. (2024) Computers & Geosciences, 193
[2] Kilburn et al. (2023) Commun Earth Environ 4, 190
[3] Isaia et al. (2025) Commun Earth Environ 6, 213
How to cite: Lormand, C., González-Ilama, G., Vuadens, S., Isaia, R., Giordano, G., Stock, M., and Caricchi, L.: Clinopyroxene thermobaromatry uncovers trigger mechanisms and reservoir configuration preceding eruptions at Campi Flegrei, Italy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17710, 2026.
The Veiðivötn 1477CE (V1477) eruption in Iceland occurred along a ~70km long discontinuous fissure between the Torfajökull and Bárdarbunga volcanic systems in the Southern highlands of Iceland. The southwestern end of the fissure, where it intersected the silicic Torfajökull system, produced only minor silicic tephra, while small basaltic lava flows were emitted from the northeastern segment. Basaltic tephra (Sio2: 49.99 ±1.35 wt.%) erupted from a 20 km long fissure segment and deposited over half of the country (> 50,000 km2), with a total volume of 10.8 ± 0.5 km3 (or 4.3 km3 DRE), corresponding to a Volcanic Explosivity Index of 5.
The eruptive sequence comprises 32 tephra fall units, with little to no variation in groundmass glass composition. We divide the eruptive sequence into three stages: (i) stage I exhibits the lowest vesicularity, a unimodal grain-size distribution (GSD), and shows no evidence of interaction with external water; (ii) stage II has the highest vesicularity, multimodal GSDs, and evidences of instability in conduit processes and interaction with external water limited to passive secondary fragmentation; and (iii) stage III shows rhythmic instabilities while returning to initial eruptive conditions.
Field data coupled to physical models of volcanic plumes show that the maximum column height reached 30.7 ± 3 km during the eruption peak. 1D plume modelling using the Paris Plume Model yields a mass eruption rate of 2.6 ± 1.3 × 10⁸ kg/s under windy conditions, placing the eruption within the Plinian to Ultra-Plinian range. Model results suggest that a volatile content of >2 wt.% would be required to sustain such a plume height, exceeding the typical volatile contents (< 1wt%) of Veiðivötn basaltic magmas. While magma-water interaction can be invoked to explain this discrepancy, vesicularity data (71–84%) indicate predominantly magmatic fragmentation, with magma reaching full vesiculation prior to fragmentation. Thus, external water likely played only a passive role, contributing to secondary fragmentation that increased the proportion of fine ash visible in GSDs. Furthermore, no evidence of microlite-rich pyroclasts was observed.
The V1477 eruption expands the record of basaltic Plinian eruptions and highlights the need to reassess conventional assumptions regarding the driver mechanisms of such eruptions.
How to cite: Payet--Clerc, M., Thordarson, T., Moreland, W. M., Höskuldsson, Á., and Carazzo, G.: The Veiðivötn 1477CE (Iceland) Plinian basaltic eruption: A review of eruptive processes and plume dynamics , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20446, 2026.
The viscosity of hydrous volcanic melts exerts a primary control on magma ascent, degassing efficiency, and fragmentation, yet its experimental determination is often affected by time-dependent melt structure changes at the nanoscale during measurements. This issue remains poorly constrained for highly polymerised, alkali-rich trachytic magmas, despite their key role in explosive volcanism at caldera systems such as Campi Flegrei (Italy).
We investigate the anhydrous and hydrous viscosity of a trachytic melt from the Agnano–Monte Spina (AMS) eruption (Campi Flegrei) by combining differential scanning calorimetry (conventional and flash DSC), micropenetration viscometry (MP), Brillouin light scattering (BLS), and in situ high-temperature Raman spectroscopy. This integrated approach allows us to directly link viscosity behaviour to nanoscale structural evolution during thermal treatments. Glass transition temperature (Tg) decreases from ~632 to ~349 °C with increasing water (0–4.45 wt.%), while melt fragility increases systematically with hydration, independently constrained from BLS elastic moduli measurements.
In situ Raman spectroscopy reveals that nanoscale melt structure reorganisation initiates within minutes, slightly above Tg = 632 °C. These processes lead to a viscosity overestimate of up to ~1 log unit in standard viscometry experiments. Using the glass transition temperatures derived from DSC measurements and BLS-derived melt fragilities, we develop a composition-specific viscosity model for the AMS trachytic magma that avoids nanostructuration-induced artefacts.
As the AMS trachytic magmas are crystal-poor, its rheology is dominated by the melt phase, making melt viscosity a primary control on magma ascent. Our results show that viscosity is highly sensitive to dehydration, with relatively low initial viscosity at high water content, followed by rapid rheological stiffening during ascent. The new model indicates a 105-fold increase in melt viscosity associated with dehydration from 5 to 0 wt.% H2O, relative to the 104-fold increase calculated using previous experimental and empirical models. As a result, commonly used empirical viscosity laws likely underestimate both the magnitude and rate of viscosity evolution during decompression of hydrous trachytic melts, with significant consequences for degassing efficiency, fragmentation depth, and eruptive style. The spectroscopically guided approach developed in this study is readily applicable to other volatile-bearing magmas and offers a more physically robust rheological framework for modelling viscosity in volcanic systems.
How to cite: Abeykoon, S., Calabrò, L., Di Genova, D., Kurnosov, A., Bamber, E. C., Bondar, D., Valdivia, P., Vona, A., Carroll, M. R., Romano, C., and Arzilli, F.: Constraining trachytic melt viscosity by integrating rheological and in situ spectroscopic analyses: Insights from the Agnano-Monte Spina eruption (Campi Flegrei, Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14624, 2026.
