The NW submarine flank of Pantelleria volcano was investigated through the integrated analysis of multibeam bathymetry and seafloor samplings. It is morphologically dominated by a 4-km wide insular shelf and a large field of volcanic cones, including the vent associated with the last 1891 eruption. Weakly porphyritic lavas, dominated by olivine over clinopyroxene, are found in the shelf-edge/upper slope sector. Porphyritic pyroclastic deposits, characterized by a plagioclase+clinopyroxene±olivine±oxides assemblage, are mainly recovered from the volcanic cones. Submarine lavas and pyroclasts generally encompass the compositional variations of major elements recognized for the mafic subaerial deposits. In particular, lavas show lower TiO2-P2O5 with respect to the pyroclastic products, a chemical feature previously used for distinguishing between young (between ~30 and ~ 10 ka) and old (> ~60 ka) mafic products onland. In this study, these chemical differences have been related to various degrees of partial melting of a mantle source heterogeneously enriched in recycled MORB material, in agreement with recent studies on volcanism in the Sicily Channel. Based on geomorphological and petro-chemical constraints, the submarine lavas can be linked to the effusive activity that occurred onland ~30 ka, with lavas flowing over an emerged shelf at that period. Submarine pyroclastic products have been related to different phases of mild explosive eruptions, as indicated by the morphology of the related cones. More interestingly, pyroclastic products recorded a higher degree of fractional crystallization and/or stronger contribution of recycled MORB material with respect to the submarine lavas, so providing insights on magma ascent pathways at Pantelleria volcano.
The evolution of magma rheological behavior and permeability during the ascent determines the explosivity of volcanic eruptions. In this study, we investigated the evolution of magma permeability under viscous and brittle deformation. We performed uniaxial compression experiments for rhyolite magma at temperatures of 806–955 °C and strain rates of 10–6.5–10–1.3 s−1. We used rhyolite lava collected from Niijima Island (Japan) as the starting material, with 15–25 vol% phenocrysts and 20–35 vol% porosity. The gas permeabilities of samples were measured before and after deformation. Our experiments showed viscous deformation of rhyolite magma at Weissenberg number (Wi) of less than 10–3.4 and brittle deformation at Wi greater than 10–2.5. During viscous deformation, the permeability did not change significantly at low strain, while it decreased slightly at large strain. On the other hand, the permeability increased strongly when the magma failed in the brittle regime, with larger strain (>0.08), originating from the formation of large cracks and their widening. These results imply that efficient outgassing occurred in the conduit with high Wi regions, corresponding to the shallow and outer parts of the conduit. Once wide fractures are formed, the permeability and outgassing rate increase by two orders of magnitude, which is significantly higher than that of the viscous magma beneath the fractured magma in the conduit. We propose that a zoned structure, consisting of a brittle plug with high permeability and porosity, and viscous vesicular magma, may have formed at the top of the volcanic conduit.
Establishing reliable correlations among proximal, medial, and distal deposits is essential for reconstructing the eruptive histories of large caldera-forming eruptions. However, such correlations are often difficult because depositional facies vary markedly with distance from the source. In this study, we re-examined the 106 ka Toya caldera-forming eruption, southwestern Hokkaido, Japan, using lithofacies characteristics, juvenile-material assemblages, and matrix-glass compositions to establish correlations among proximal, medial, and distal deposits and to reconstruct the overall eruptive sequence. The proximal deposits are subdivided into six eruptive units (Units 1–6). Three juvenile-material types were identified: crystal-poor pumice (CP pumice), crystal-rich pumice (CR pumice), and grey-banded pumice (GB pumice). Units 1–3 contain only CP pumice, whereas CR and GB pumices first appear in Unit 4 and become progressively more abundant through Units 5 and 6, providing a distinctive stratigraphic framework throughout the eruption. Correlations among proximal deposits, medial ignimbrites, and distal co-ignimbrite ash deposits were established using juvenile-material assemblages and matrix-glass compositions. The results indicate that Unit 5 generated widespread co-ignimbrite ash deposits distributed more than 400 km from the source, whereas Units 2 and 6 remained relatively restricted compared with Unit 5. Integration of proximal, medial, and distal records indicates that the eruption consisted of two major eruptive stages: Stage 1 (Units 1–3; ~16 km3 DRE) and Stage 2 (Units 4–6; ~59 km3 DRE), separated by a significant temporal break between Units 3 and 4. Stage 1 was characterized by phreatomagmatic activity, abundant accretionary lapilli, and exclusive eruption of CP pumice. In contrast, Stage 2 was characterized by larger eruptive volumes, coarser PDC deposits, the appearance of CR and GB pumices, and widespread ash dispersal. The total erupted volume is estimated at approximately 75 km3 DRE. The large eruptive volumes of Units 5 and 6, together with the occurrence of lag breccia deposits, suggest that the principal phase of caldera collapse occurred during Stage 2. The successful correlation of proximal, medial, and distal deposits provides a more complete reconstruction of the Toya caldera-forming eruption and demonstrates the value of temporal variations in juvenile-material assemblages for correlating deposits and reconstructing large silicic eruptions.
Surface deformation derived from satellite interferometry and GNSS observations represents a fundamental tool for investigating magma dynamics beneath active volcanoes. However, geodetic inversion results may depend on the quality, quantity, and accuracy of the datasets employed. Therefore, the assessment of model robustness is essential for deriving reliable interpretations of subsurface processes. Campi Flegrei caldera (Italy) has been experiencing progressive uplift since 2005, associated with increased seismicity and gas emissions. Modelling of the 2007–2023 deformation from COSMO-SkyMed and GNSS data suggested the presence of a shallow inflating source at 3.9 km depth since 2015 and a deeper deflating reservoir. This interpretation was obtained using a complex 3D heterogeneous numerical modelling framework. In this work, we evaluate the model robustness by repeating the geodetic inversion using multi-mission satellite datasets, including Sentinel-1 observations and a different COSMO-SkyMed analysis, together with GNSS measurements. We show that, although discrepancies affect part of the COSMO-SkyMed descending dataset, their impact on the inversion remains limited because the data are analysed over discrete time windows, the discrepancies remain moderate relative to the overall deformation, uncertainties are accounted for, and the inversion integrates multiple geodetic datasets. The results show a stable geometry and depth of the shallow source, together with comparable estimates of volume change. These findings demonstrate that the source parameters are reproducible across alternative InSAR datasets, providing an explicit cross-validation. More broadly, they show that first-order constraints on magma-driven deformation at Campi Flegrei are robust and can support interpretations of volcanic unrest and hazard assessment.
A high-resolution reconstruction of volcanic evolution is crucial for understanding magma processes and eruptive mechanisms. Establishing robust correlations between tephra units, which provide temporal constraints, and proximal edifice-building deposits is key to this reconstruction. Rather than correlating tephra and edifice deposits solely on the basis of major-element compositions, we first defined the magma type of each tephra unit based on the established tephra stratigraphy and clarified the temporal variation in magma types. We then assigned previously uncorrelated edifice deposits to specific tephra units using magma-type affinities, thereby reconstructing the eruptive history of the volcanic edifice. Magma types were distinguished using multiple petrological parameters, including whole-rock major and trace element compositions and SrNd isotopic ratios. Applying this approach to Akita-Komagatake volcano in the Northeast Japan arc, we reassessed its volcanic evolution from the late Pleistocene to the present. Thirteen tephra units were classified into five magma types based on SrNd isotopic compositions and large-ion lithophile element abundances (K, Rb, Ba). Edifice evolution was reconstructed as three major eruptive stages that are further subdivided into seven substages according to vent locations, eruption styles, and stratigraphic relationships. Each stage is characterized by a distinct magma type. This integrated correlation enabled a refined reconstruction of the chronology, vent locations, erupted volumes, and eruption styles of each stage. A revised magma-discharge step diagram integrating both tephra and edifice deposits reveals a three-stage evolutionary sequence: (1) an initial explosive phase with high magma discharge rates and caldera formation, (2) a subsequent phase dominated by vigorous effusive eruptions that built the volcanic edifice, and (3) a final stage of declining magma discharge rate. Temporal changes in magma types correspond closely to successive eruptive stages and substages, suggesting that each eruptive phase was supplied by a compositionally distinct magma. The integrated approach presented here provides a robust framework for reconstructing volcanic evolution and can be applied to other complex stratovolcanoes.
The 2018–2019 eruption of the Planchón-Peteroa Volcanic Complex (Chile-Argentina) provided a rare opportunity to investigate syn-eruptive degassing in peripheral hydrothermal valleys. During February 2019, we acquired a multi-parameter dataset combining diffuse soil CO₂ fluxes, free- and dissolved-gas chemistry, carbon isotopes (δ13C-CO₂, δ13C-CH₄), and radioactive tracers (222Rn, 22⁰Rn) in two hydrothermal valleys: Valle Baños del Azufre (VBA) and Valle del Peñón (VP). Soil CO₂ fluxes show contrasting statistical and spatial patterns. VBA (n = 171) displays a broader flux range and higher maximum values (0.5–514.9 g m−2 d−1), whereas VP (n = 63) shows a narrower and smoother distribution (2.7–60.2 g m−2 d−1). Free gases are CO₂-dominated (61–94 vol%) but differ in redox-sensitive components and isotope signatures. VBA is characterized by elevated 222Rn activity (up to 5 × 104 Bq m−3), higher CH₄ contents, and isotopically light methane (δ13C-CH₄ down to −54‰), whereas VP shows lower 222Rn activity (≤5 × 103 Bq m−3), stronger air-water interaction, and markedly heavier methane (up to +11‰), consistent with methane oxidation and/or mixing during lateral migration. Dissolved CO₂ reaches 11.8 mmol L−1, while dissolved CH₄ is mainly restricted to the most reducing VBA samples, reaching up to 0.06 mmol L−1. Overall, the dataset reveals marked differences in degassing style and redox-sensitive gas signatures between the two peripheral valleys. These results provide new constraints on the behaviour of peripheral hydrothermal systems during eruptive activity and highlight their potential relevance for integrated volcano monitoring.
Pele's hair is a cylindrical pyroclast traditionally formed by the elongation of low-viscosity (≤100 Pa s) mafic lava droplets during Hawaiian style eruptions. Current impulse experiments simulating volcanic lightning have been shown to produce Pele's hair in starting material of varying compositions (48–76 wt% SiO2). No naturally produced lightning-induced hairs have yet been discovered, as no current method exists to distinguish them from those formed by fragmentation. Here, we provide a comprehensive comparison between experimentally produced lightning-induced hairs of mafic (48–49 wt% SiO2) and felsic (72–76 wt% SiO2) composition to fragmentation-induced hairs through direct diameter analysis using scanning electron microscopy and established calculations. Lightning-induced hairs ranged in diameter from 0.03 to 5.4 μm with a geometric mean of 0.49 μm. Fragmentation-induced hairs ranged from 0.3 to 763.6 μm with a geometric mean of 31.5 μm. The calculated velocities show that those forming lightning-induced hairs (97.6–1227.4 m s−1) are consistent with velocities calculated from current impulse shock fronts, while those forming fragmentation-induced hairs (7.8–379.7 m s−1) are consistent with the spurting velocities of Hawaiian style fire fountains. Viscosity calculations used to constrain the initial temperature of lightning-induced hairs reveal that lightning temperatures are capable of decreasing the viscosity of felsic melt droplets to a suitable range for hair formation. Results of this study indicate that any Pele's hair observed in ash samples with a bulk felsic composition (>70 wt% SiO2) are likely to be lightning induced, however, diameter alone cannot reliably distinguish them in more mafic samples.
This study uses ambient seismic noise processing methods as the primary tool to investigate suspected maar volcanic structures at Nežichov village and Sepuska hill in West Bohemia, Czechia. The research employs ambient seismic noise surveys, gravimetric surveys, electrical resistivity tomography (ERT), and numerical modeling to analyze these structures. At Nežichov, the ambient seismic noise survey identified fundamental mode amplification frequencies, with particle motion trends indicating a complex 3D resonance structure. The gravimetric survey revealed a negative anomaly, suggesting a volcanic breccia fill. ERT results showed low resistivity zones attributed to maar lake sediments, with a high resistivity intrusion dividing these zones. At Sepuska, similar methods identified fundamental mode frequencies, with a circular amplification distribution. The gravimetric survey indicated a negative anomaly, consistent with a maar structure. The study concludes that both Nežichov and Sepuska are likely maar volcanic structures, with complex subsurface features revealed through combined geophysical methods. These findings contribute to understanding the volcanic history and geological evolution of the Bohemian Massif.
Cerro Machín Volcano (CMV), central Colombia, is considered one of the most hazardous magmatic systems in northwestern South America, due to its eruptive history, highly explosive character and the presence of active fumaroles in the central dome. Although characterization of the shallow structure of fumarolic zones is essential for understanding hydrothermal dynamics and potential pressurization mechanisms during volcanic unrest, a lack of geophysical data has limited our ability to assess the current state of the system. To address this, we performed a geoelectrical survey focused on the area encompassing the CMV dome’s main fumarole. Our acquisition integrated Electrical Resistivity Tomography (ERT), Time-Domain Induced Polarization (IP), and Transient Electromagnetic (TEM) soundings. The ERT sections reveal a coherent high-resistivity anomaly, which we interpret as a vapor-dominated region resulting from phase separation processes within a predominantly liquid-dominated hydrothermal system. TEM soundings provide complementary information consistent with the presence of a deeper resistive interval. Additionally, both ERT and IP profiles delineate a subsurface body extending laterally for over 150 m and to depths of at least 50 m, characterized simultaneously by high chargeability and very low resistivity. We interpret this feature as an extensive zone of ongoing hydrothermal alteration. In combination with available gas geochemistry, which indicates CO2-rich fluids with a predominantly meteoric signature, these results indicate that the persistent fumarolic activity at the CMV dome is primarily sustained by circulation through a fractured hydrothermal system that is locally capped or sealed by an argillic alteration zone. Although our findings do not completely rule out a deeper magmatic contribution to the broader volcanic system, they do not support interpreting the present-day fumarolic field as direct evidence of shallow magmatic unrest.
Lascar volcano, located in the Central Andes Volcanic Zone, is one of the most active volcanoes in Northern Chile, characterized by recurrent minor explosive events. We have analyzed the volcanic degassing process at Lascar between August 2017 and August 2021 (Continental Chile: Alternates between UTC-4 (winter time) and UTC-3 (summer time)). using a multiparametric approach that integrates seismicity, visual observations, satellite thermal data, and ground-based SO2 flux measurements, which were deseasonalized to remove meteorological artifacts.This correction revealed intrinsic volcanic variations that allowed us to identify three distinct patterns: i) Pattern I – Overpressure and System Sealing (August 2017–January 2018), characterized by low deseasonalized SO2, emission rates despite frequent valid measurements, minimal thermal activity, and increasing long-period seismicity, indicating progressive sealing of the shallow system and internal pressurization, ii) Pattern II - Pressurization, Fracturing, and Partial Unsealing (February 2018–November 2018), involves pressurization and mechanical failure of the conduit, marked by the appearance of tornillo and volcano-tectonic seismicity, followed by partial unsealing and increased gas release, iii) Pattern III - Explosion, Decompression, and Cycle Reset (November 2018–February 2020), correspond to explosive decompression, starting in the 22 November 2018 with ash emission event, followed by peak SO2 fluxes and transient thermal anomalies.Post-eruptive observations show that this event did not establish a stable equilibrium but reset the volcanic cycle, with the repetition of sealing and pressurization phases between 2020 and 2021. These results indicate that conduit sealing and rupture are recurrent processes that control Lascar's recent activity and highlight the importance of integrated gas, seismic, and thermal monitoring for identifying pressurization phases preceding minor explosive events.
Petrological, geochemical and geophysical data have been integrated to characterize the magmatic processes driving the volcanic activity at the South East Crater of Mt. Etna between 2022 and 2023. The studied period includes two phases of dominant effusive activity that occurred in May–June 2022 and November 2022–February 2023 and a subsequent explosive phase characterized by two lava fountaining episodes of 21 May and 13–14 August 2023. Petrological constraints on olivine zoning and diffusion timescales enabled the identification of some dominant phases of magma intrusion and emplacement at various crustal depths, which temporally correlate with the records of bulk plume SO2 flux and volcanic tremor amplitude and source location associated to magma/gas migration through the upper conduits. The onset of effusive activity in May 2022 is consistent with the ascent of fresh volatile-rich magma towards the shallowest portion of the plumbing system (<2 km below the summit). Several months later, magmatic inputs from deeper reservoirs remobilized the previously emplaced residual melt, prompting the second effusive phase in November 2022. Paroxysmal eruptions during 2023 developed in response to the activation of the deepest magmatic environments (>9 km below the summit), and the rapid migration of Fo80 to Fo86 magmas releasing large amounts of volatiles during ascent. This study demonstrates that petrological records of magmatic processes can be linked with real-time monitoring signals in the lead-up to eruptions to ensure a more accurate interpretation of the conditions and processes that can cause changes in the state of volcanic activity.
Between December 2020 and February 2022 a series of more than 60 paroxysmal events occurred on Etna. For this eruptive sequence, previous studies demonstrated that the evolution of magma composition was mostly driven by the mixing of a basaltic and trachybasaltic magma. To investigate the role of magma composition and of syn-eruptive conditions on magma ascent dynamics during paroxysmal activities at basaltic volcanoes, we performed a series of numerical simulations using a conduit model. Our data show that, across most of the investigated combinations of syn-eruptive conditions, both the more evolved trachybasaltic magma and the more primitive basaltic magma exhibit an equivalent style of activity. Thus, it is the different syn-eruptive conditions (i.e. pressure, temperature, volatile and crystal content) of the two end-members at the inlet of the conduit that mainly control the eruptive style and not just magma composition. Numerical results indicate that sustained fountaining activities of this paroxysmal sequence require a temperature at the inlet of the conduit (4 km below the vent) greater than 1090–1100 °C and an inlet crystal content lower than ∼25 vol%. On the contrary, effusive eruptions in agreement with those observed during the 2020–2022 eruptions require an inlet temperature lower than 1110–1120 °C, a water content lower than ∼3 wt% and an inlet crystal content greater than ∼20 vol%. We also found the presence of critical viscosity values, that, when exceeded, result in a transition in eruptive style with a sudden increase in mass eruption rate of more than an order of magnitude. These transitional viscosity values are not influenced by the thermodynamic conditions of the magmatic system, but they are affected by conduit geometry, outgassing efficiency and timescales of crystallisation and ascent.
CO₂ flux measurements have become an essential tool in volcanic monitoring, yet standardized methodologies remain critical for cross-study comparability. Here, we present a newly acquired CO₂ flux dataset from the Escarot mofette field, processed using a partially automated geostatistical web application that minimizes user subjectivity. We calculate a CO₂ emission rate of 1065 ± 110 t yr−1 over an area of 35,000 m2. A direct comparison with a CO2 flux study by Gal et al. (2018) initially suggests a substantial decrease in emissions by roughly a factor of eight (8100 ± 1800 t yr−1). However, this apparent decrease largely reflects differences in data treatment and output calculation. Threshold definition and outlier treatment strongly affect these estimates and must be applied consistently to ensure comparability between studies. Recalculating the 2018 dataset with the same geostatistical methodology as applied here yields a CO2 output of 692 t yr−1, revealing an increase of ∼370 t (∼54%) over eight years. In addition, air-corrected 3He/4He ratios of 5.93–6.38 Rₐ confirm a subcontinental lithospheric mantle (SCLM) signature, while δ13CCO2 values of −1.67 to −1.62‰ suggest secondary chemical and physical fractionation processes potentially driven by enhanced deep-sourced fluid flux. Together, these results indicate that structural control at Escarot governs both the origin of the gases and the spatial distribution and degassing rates of surface CO2 emissions. Based on the results we assume that the system may have been in geochemical disequilibrium over the past decade, potentially coupled with the contemporaneous regional seismic activity. Continuous, long-term fluid monitoring is needed to reliably distinguish real system change from natural background variability. Plain language summary A volcanic system that appears dormant, like the Monts Dore volcanic province, which shows no clear signs of increased activity, is not necessarily truly inactive. By applying a new, semi-automatic geostatistical tool to both an older (2018) and a more recent (2025) CO₂ flux dataset measured at the Escarot degassing site in the French Massif Central, we show that the system's CO₂ output increased by approximately 54% over the past eight years. Furthermore, the geochemical fingerprint of the gases, defined by both their carbon and helium isotopic signatures, points to a subcontinental lithospheric mantle source. Thus, gases must have migrated over long distances (>25 km) through the Earth's crust, along permeable structures, before reaching the surface. This is further supported by the widespread degassing observed across Escarot, most clearly expressed by bubbling mofettes and, at times, accompanied by the sound of escaping gas.
Pantelleria island is the southernmost active volcanic system in Italy and hosts intense hydrothermal activity associated with significant geothermal potential. To improve seismic monitoring and investigate the shallow crustal structure of the island a temporary seismic network, integrated with the permanent INGV stations, was deployed. In this study, we present the first 1D shear-wave velocity (Vs) model for Pantelleria Island down to ∼1 km depth, based on ambient seismic noise analysis. Seismic array technique was applied to ambient noise recordings collected across the volcanic island. Stratigraphic information from deep wells, together with data from geophysical and geological surveys, was used to support the geological interpretation of the retrieved velocity model. The results provide new constraints on the shallow structure of Pantelleria and contribute to a better understanding of the dynamics of the volcanic-hydrothermal system. In addition the velocity model may improve earthquake localisation and support future studies of local seismicity potentially related to geothermal resources.
Magmatic volatiles in geothermal settings ascend through permeable pathways developed from various tectonic and volcanic processes, resulting in structurally complex fracture networks controlling emissions of CO₂ and other magmatic volatiles. To better understand these controls, we examine spatial patterns and isotopic signatures of fumaroles and diffusely degassing CO₂ in the Menengai Geothermal Field, Kenya. Sequential Gaussian simulation analyses support a total CO₂ emission likely exceeding 389.7 t d−1, with heavier δ13C-CO2 values suggesting primarily mantle-derived carbon contributions. Degassing pattens mimic major rift structures, including NNW- to NNE-trending faults and secondary E- to ENE-trending structures. Fumarole lineaments, however, exhibit substantial scattering, with some observed lineament patterns falling outside the typical rift-related structural fabric. We suggest that the broad distribution of CO₂ degassing (10s of km2 scale) is influenced by a large, ≥6.5 km-deep igneous body, while deep volcanic and tectonic structures allow the upward migration of volatiles when hydraulically linked to this source. Lineaments observed in CO₂ flux patterns (1000s of m2 scale) support shallow CO₂ transport along blind rift-parallel faults that terminate within post-caldera lavas. From this depth, volatiles likely ascend through a permeability network derived from volcanic structures developed in post-caldera lava flows, governing the local-scale (100 s of m2 scale) patterns of CO₂ degassing. We conclude that a range of tectonic- and volcanic-related factors, operating and interacting across multiple scales, control the spatial patterns of CO₂ degassing at Menengai, with possible implications for other geothermal systems within the East African Rift System (EARS), particularly the Kenyan Rift.
Ambient seismic noise interferometry is increasingly used for crustal imaging, but its application to intraplate volcanic systems remains limited because seismic networks are often sparse, ambient-noise sources are variable, and strong scattering in volcanic terrains makes stable reflectivity retrieval challenging. We analyze 70 days of continuous borehole seismic records from the VORiSA network in Harrat Al-Madinah, western Saudi Arabia, to investigate the crustal structure beneath a large basaltic volcanic field. Eight three-component seismometers deployed at approximately 120 m depth provide stable recordings with reduced near-surface noise. We apply conventional autocorrelation and phase autocorrelation with phase-weighted stacking (PAC–PWS) to the vertical component. Conventional autocorrelation provides an amplitude-preserving reference based on linear correlation and stacking, whereas PAC–PWS reduces the influence of amplitude fluctuations and emphasizes phase-coherent arrivals. Their combined use allows evaluation of reflector consistency across complementary processing schemes. The results identify six principal reflectors above approximately 50 km depth. These include boundaries corresponding to the Conrad discontinuity at approximately 22 km and the Moho at approximately 38 km, consistent with previous seismic studies, as well as additional reflectors near 10 km, within the mid-crust at 10–25 km, and near 46–47 km. Comparison with magnetotelluric imaging suggests that several mid-crustal reflectors may be associated with conductive zones interpreted as possible magma reservoirs. These results indicate that borehole ambient-noise autocorrelation, supported by complementary phase-based processing, can provide useful constraints on crustal structure beneath intraplate volcanic regions.
Large parts of the Arctic, including Svalbard, have experienced recurrent episodes of uplift and erosion over the last 15–20 million years, coinciding with localized magmatism and seafloor spreading. This has left sparse onshore stratigraphic records of the Miocene period, limiting our understanding of the tectono-sedimentary and paleo-landscape evolution across this vast region. The volcanic Seidfjellet Formation in northwestern Spitsbergen thus provides a rare e onshore Miocene record. To determine emplacement mechanisms, volcanic architecture, and the pre-volcanic paleoenvironment, we combined detailed field-based facies analysis with digital outcrop models, thin-section petrography, and palynology. We document 10–20 basaltic lava flows, totaling ~400 m in cumulative thickness, now exposed at 600–1350 m elevation. The lava sequence rests on a gently undulating Devonian peneplane that locally is incised by several-hundred-meters-wide, up to 150-m-deep depressions interpreted as paleo-valleys. These valleys likely formed by stream rejuvenation associated with Miocene peneplane uplift. At valley bases we report previously unrecognized pockets of unconsolidated sediment, evidence of pre-volcanic sedimentation. Valley margins are locally draped by volcaniclastic deposits that contain in-situ charcoal-lined tree molds, indicating forested margins before eruption. The main valley fills are hyaloclastite deposits up to 150 m thick, arranged in steeply dipping forests that record southward progradation of a lava-fed delta after base-level rise and drowning of the valleys. Palynological data indicate a conifer-dominated wetland prior to lava emplacement, while the dinocyst assemblages appear partly reworked and are consistent with a Late Oligocene–Early Miocene signal but more likely reflect a Late Miocene age. This study provides the first systematic description of the Seidfjellet Formation volcanic facies and architecture, constrains its spatial extent, and, through its pre-volcanic sedimentary records, offers new paleoenvironmental reconstructions. Finally, our findings highlight the interplay between magmatism and uplift during Svalbard’s Miocene evolution
The Callaqui volcano (south-central Chile) currently exhibits significant fumarolic activity, and its remains indicate that it has had substantial eruptive activity in the past. However, there is no prior knowledge about the fluids discharged by this volcano, which is critical for volcanic monitoring and subsequent understanding of associated volcanic hazards. This work presents the chemical and isotopic compositions of fumarolic gases from the Callaqui volcano (period 2017–2024) to identify the fluid source(s) and evaluate the physicochemical conditions operating at depth. Gases were collected by direct sampling, using Giggenbach bottles and condensation procedures. The temperature of fumarolic gases ranged from 86.3 to 215 °C, with a composition dominated by water vapor, CO2, and minor amounts of acid species (SO2, HCl, HF). Helium isotope ratios (5.13 to 6.69 Ra) suggest a significant contribution from MORB-like fluids, while δ13C-CO2 isotopes (−10.8 to −9.11‰ vs. V–PDB) suggest assimilation of carbon-bearing crustal compounds. These features suggest a magmatic-hydrothermal origin of Callaqui emissions. The magmatic source corresponds to a degassing magma chamber likely hosted in an intermediate portion of the crust, whereas the hydrothermal system is fed mainly by meteoric waters. Despite the apparent large size of the hydrothermal system, since the volcano is in a rainy and snowy area, magmatically derived species still reach the surface. As a result, partial scrubbing of magma-derived species is assumed at this volcano. Following geothermometric approaches, fumarolic gases indicate the presence of high-temperature magmatic-hydrothermal fluids, with a vapor phase equilibrated at 375–415 °C, as well as super-heated vapors up to 350 °C. Callaqui gases show an increase in the magmatic signal by the end of the study period, consistent with incandescent volcanic episodes and anomalous LP earthquakes reported in late 2021 and early 2022.
Campi Flegrei, the largest restless caldera in Europe, has experienced increasing activity since 2004. This unrest is characterized by ground uplift, seismicity, intensified degassing, and opening of new fumaroles and mud pools. Quantifying carbon dioxide (CO2) emissions is particularly challenging at this volcano because sulfur dioxide (SO2) is efficiently scrubbed by the hydrothermal system, making conventional CO2 emission rate estimates that rely on remote sensing-based SO2 measurements not possible. As a result, CO2 emission rates from fumaroles at Campi Flegrei are measured by ground-based MultiGAS. However, the escalating level of activity could render such measurements impossible in the near future, highlighting the need for another approach to measure fumarolic CO2 emissions. Here, we present the first CO2 emission rate measurements at Campi Flegrei's main fumarolic site, Pisciarelli, using drone-based MultiGAS instrumentation, supported by simultaneous ground-based MultiGAS data. The drone-based average CO2 emission rate (218 ± 118 tons/day) closely matches the ground-based value (263 ± 90 tons/day), validating the drone approach. Our drone-based approach yields gas ratios of CO2/H2S = 295 ± 36 and CO2/H2O = 0.27 ± 0.06; combining these ratios with CO2 emission rates, we derive H2S and H2O emission rates of 0.57 ± 0.30 and 330 ± 193 tons/day, respectively. These results also closely agree with ground-based and direct sampling methods. Our 2024 measurements reveal a decline in CO2, H2S, and H2O emission rates compared to 2019, likely due to redistribution of the gas output pathways at Campi Flegrei through an expansion of the diffuse degassing zone. This study demonstrates that drone-based MultiGAS is a reliable, safe alternative for volcanic gas monitoring. We provide recommendations for further deployments at Campi Flegrei and other challenging volcanic environments.
