Natural disasters, such as volcanic eruptions, pose significant challenges to daily life and incur considerable global economic losses. The emergence of next-generation small-satellites, capable of constellation-based operations, offers unparalleled opportunities for near-real-time monitoring and onboard processing of such events. However, a major bottleneck remains the lack of extensive annotated datasets capturing volcanic activity, which hinders the development of robust detection systems. This paper introduces a novel dataset explicitly designed for volcanic activity and eruption detection, encompassing diverse volcanoes worldwide. The dataset provides binary annotations to identify volcanic anomalies or non-anomalies, covering phenomena such as temperature anomalies, eruptions, and volcanic ash emissions. These annotations offer a foundational resource for developing and evaluating detection models, addressing a critical gap in volcanic monitoring research. Additionally, we present comprehensive benchmarks using state-of-the-art models to establish baselines for future studies. Furthermore, we explore the potential for deploying these models onboard next-generation satellites.
Volcanoes emit large amounts of CO2, directly influencing human lives. Mapping volcanic gas emissions helps to forecast eruptions and understand the impact of volcanoes on climate and the environment. Drone-based gas sensing significantly reduces risks in volcanic monitoring but faces technical limitations when measuring gas, as rotor downwash disperses the gas plume before detection. Gas Tomography using remote gas sensing addresses this challenge. At the Salinelle dei Cappuccini mud volcanoes, we demonstrate that while drone-mounted in-situ sensors failed to detect CO2 emissions due to aerodynamic disturbance, open-path sensing successfully enabled remote gas distribution mapping. We present a novel model-based gas tomographic reconstruction approach that incorporates a Lagrangian model to compensate for wind-induced advection. The resulting gas distribution maps align with manually collected in-situ measurements, confirming that model-based gas tomography effectively overcomes downwash limitations and enables accurate mapping of volcanic emissions.
Forecasting volcanic activity is critical for hazard assessment and risk mitigation. Volcanic Radiative Power (VPR), derived from thermal remote sensing data, serves as an essential indicator of volcanic activity. In this study, we employ Bayesian Regularized Neural Networks (BRNN) to predict future VPR values based on historical data from Fuego Volcano, comparing its performance against Scaled Conjugate Gradient (SCG) and Levenberg-Marquardt (LM) models. The results indicate that BRNN outperforms SCG and LM, achieving the lowest mean squared error (1.77E+16) and the highest R-squared value (0.50), demonstrating its superior ability to capture VPR variability while minimizing overfitting. Despite these promising results, challenges remain in improving the model's predictive accuracy. Future research should focus on integrating additional geophysical parameters, such as seismic and gas emission data, to enhance forecasting precision. The findings highlight the potential of machine learning models, particularly BRNN, in advancing volcanic activity forecasting, contributing to more effective early warning systems for volcanic hazards.
Modern seismic and volcanic monitoring is increasingly shaped by continuous, multi-sensor observations and by the need to extract actionable information from nonstationary, noisy wavefields. In this context, machine learning has moved from a research curiosity to a practical ingredient of processing chains for detection, phase picking, classification, denoising, and anomaly tracking. However, improved accuracy on a fixed dataset is not sufficient for operational use. Models must remain reliable under domain shift (new stations, changing noise, evolving volcanic activity), provide uncertainty that supports decision-making, and connect their outputs to physically meaningful constraints. This paper surveys and organizes recent ML approaches for seismic and volcanic signal analysis, highlighting where classical signal processing provides indispensable inductive bias, how self-supervision and generative modeling can reduce dependence on labels, and which evaluation protocols best reflect transfer across regions. We conclude with open challenges for robust, interpretable, and maintainable AI-assisted monitoring.
Volcanic gas emissions are key precursors of eruptive activity. Yet, obtaining accurate near-surface measurements remains hazardous and logistically challenging, motivating the need for autonomous solutions. Limited mobility in rough volcanic terrain has prevented wheeled systems from performing reliable in situ gas measurements, reducing their usefulness as sensing platforms. We present a legged robotic system for autonomous volcanic gas analysis, utilizing the quadruped ANYmal, equipped with a quadrupole mass spectrometer system. Our modular autonomy stack integrates a mission planning interface, global planner, localization framework, and terrain-aware local navigation. We evaluated the system on Mount Etna across three autonomous missions in varied terrain, achieving successful gas-source detections with autonomy rates of 93-100%. In addition, we conducted a teleoperated mission in which the robot measured natural fumaroles, detecting sulfur dioxide and carbon dioxide. We discuss lessons learned from the gas-analysis and autonomy perspectives, emphasizing the need for adaptive sensing strategies, tighter integration of global and local planning, and improved hardware design.
We present a novel method of constraining volcanic activity on extrasolar terrestrial worlds via characterization of circumstellar plasma tori. Our work generalizes the physics of the Io plasma torus to propose a hypothetical circumstellar plasma torus generated by exoplanetary volcanism. The quasi-steady torus mass is determined by a balance between material injection and ejection rates from volcanic activity and corotating magnetospheric convection, respectively. By estimating the Alfvén surfaces of planet-hosting stars, we calculate the torus mass-removal timescale for a number of exoplanets with properties amenable to plasma torus construction. Assuming a uniform toroidal geometry comparable to Io's "warm" torus, we calculate quasi-steady torus masses inferable from the optical depth of atomic spectral features in torus-contaminated stellar spectra. The calculated quasi-steady masses can be used to constrain the volcanic outgassing rates of each species detected in the torus, providing quantitative estimates of bulk volcanic activity and interior composition with minimal assumptions. Such insight into the interior state of an exoplanet is otherwise accessible only after destruc
Explosive volcanic eruptions play a fundamental role in the evolution and observability of rocky exoplanets, serving as a key mechanism for injecting volatiles into planetary atmospheres and potentially modifying their climate and composition. This process may be particularly important for close-in exoplanets where tidal forcing can drive substantial internal heating, analogous to (but often exceeding) Io's volcanism. In this work, we adapt and extend a classic 1D volcanic plume model originally developed in IDL by Glaze and Baloga for Venus and Mars applications, and port it into a flexible, open Python framework suitable for exoplanet studies. The model explicitly couples vent thermodynamics, buoyant entrainment, and vertically varying static stability to predict plume rise, neutral-buoyancy height, and overshoot for a wide range of planetary and atmospheric conditions. We first benchmark the Python implementation against the original IDL code and analytic scaling laws to ensure adequate momentum budgets and strict mass conservation. We then apply the model to a suite of exoplanet-relevant background states, including CO2-rich atmospheres under strong irradiation and diverse surf
Fluid movement beneath volcanic regions can influence earthquake activity, but the processes linking seismic and volcanic systems are not fully understood. In early 2025, an unusual seismic sequence occurred close to Santorini, providing new insight into these interactions. Here we show that the sequence was likely initiated by the accumulation and migration of fluids beneath the volcanic complex. Seismic and ground deformation data reveal a progression from deep fluid buildup and microfracturing to the concentration of shallow earthquakes beneath Columbo volcano. This culminated in a four-day seismic episode that behaved like a single, slow-propagating rupture along a 16-kilometer fault, releasing energy equivalent to a magnitude 6.2 earthquake. The rupture was followed by a typical aftershock sequence. These observations suggest that fluid-driven processes can generate large earthquakes and redistribute stress in ways similar to tectonic mainshocks. This challenges conventional views on how seismic and volcanic hazards are connected and assessed.
Assessing the prevalence of atmospheres on rocky planets around M-dwarf stars is a top priority of exoplanet science. High-energy activity from M-dwarfs can destroy the atmospheres of these planets, which could explain the lack of atmosphere detections to date. Volcanic outgassing has been proposed as a mechanism to replenish the atmospheres of tidally-heated rocky planets. L 98-59 b, a sub-Earth transiting a nearby M dwarf, was recently identified as the most promising exoplanet to detect a volcanic atmosphere. We present the transmission spectrum of L 98-59 b from four transits observed with JWST NIRSpec G395H. Although the airless model provides an adequate fit to the data based on its $χ^2$, an SO$_2$ atmosphere is preferred by 3.6$σ$ over a flat line in terms of the Bayesian evidence. Such an atmosphere would likely be in a steady state where volcanism balances escape. If so, L 98-59 b must experience at least eight times as much volcanism and tidal heating per unit mass as Io. If volcanism is driven by runaway melting of the mantle, we predict the existence of a subsurface magma ocean in L 98-59 b extending up to $R_p\sim 60-90\%$. An SO$_2$-rich volcanic atmosphere on L 98-5
Forecasting volcanic eruptions remains a formidable challenge due to the inherent complexity and variability of volcanic processes. A key source of uncertainty arises from the sporadic nature of volcanic unrest, which is often characterised by intermittent phases of quiescent deceleration and sudden acceleration, rather than a consistent, predictable progression towards eruption. This seemingly erratic pattern complicates volcano forecasting as it challenges conventional time-to-failure models that often assume a simple smooth power law acceleration. We propose a log-periodic power law singularity model, which effectively captures the intermittent and non-monotonic rupture dynamics characteristic of reawakening volcanoes at the site scale. Mathematically, generalising the power law exponent by extending it from real to complex numbers, this model captures the partial break of continuous scale invariance to discrete scale invariance that is inherent to the intermittent dynamics of damage and rupture processes in heterogeneous crustal systems. By performing parametric and nonparametric tests on a large dataset of 34 historical eruptions worldwide, we present empirical evidence and th
Volcanic regions have shaped human settlements for millennia, placing over 500 million people worldwide within close proximity to active volcanoes. Predicting eruptions that threaten both lives and property remains a critical challenge, given the complexity and diversity of seismic signals and recording formats. Effective early warning systems require tools that are not only powerful but also user-friendly, computationally efficient, and capable of handling large seismic datasets and complex mathematical and probabilistic models used in Machine Learning and Artificial Intelligence. We present a Shannon Entropy Estimator for the Characterization of Seismic-Volcanic Signals using Python designed to meet these demands. Featuring intuitive graphical user interfaces and a low learning curve, the system enables real-time analysis of seismic data using four key metrics: Shannon Entropy, Kurtosis, Frequency Index, and Energy. These metrics provide robust insights into a volcanoes state across different stages, independently of the type of event (VT, VLP, Tremor, Hybrid, Explosions) or the data format (SAC, MSEED, SEISAN, etc.). In this work, the efficacy of this tool is demonstrated throug
Satellite images have the potential to detect volcanic deformation prior to eruptions, but while a vast number of images are routinely acquired, only a small percentage contain volcanic deformation events. Manual inspection could miss these anomalies, and an automatic system modelled with supervised learning requires suitably labelled datasets. To tackle these issues, this paper explores the use of unsupervised deep learning on satellite data for the purpose of identifying volcanic deformation as anomalies. Our detector is based on Patch Distribution Modeling (PaDiM), and the detection performance is enhanced with a weighted distance, assigning greater importance to features from deeper layers. Additionally, we propose a preprocessing approach to handle noisy and incomplete data points. The final framework was tested with five volcanoes, which have different deformation characteristics and its performance was compared against the supervised learning method for volcanic deformation detection.
Precisely classifying earthquake types is crucial for elucidating the relationship between volcanic earthquakes and volcanic activity. However, traditional methods rely on subjective human judgment, which requires considerable time and effort. To address this issue, we developed a deep learning model using a transformer encoder for a more objective and efficient classification. Tested on Mount Asama's diverse seismic activity, our model achieved high F1 scores (0.930 for volcano tectonic, 0.931 for low-frequency earthquakes, and 0.980 for noise), superior to a conventional CNN-based method. To enhance interpretability, attention weight visualizations were analyzed, revealing that the model focuses on key waveform features similarly to human experts. However, inconsistencies in training data, such as ambiguously labeled B-type events with S-waves, were found to influence classification accuracy and attention weight distributions. Experiments addressing data selection and augmentation demonstrated the importance of balancing data quality and diversity. In addition, stations within 3 km of the crater played an important role in improving model performance and interpretability. These f
The distribution of Io's volcanic activity likely reflects the position and magnitude of internal tidal heating. We use new observations of Io's polar regions by the Juno spacecraft Jovian Infrared Auroral Mapper (JIRAM) to complete near-infrared global coverage, revealing the global distribution and magnitude of thermal emission from Io's currently erupting volcanoes. We show that the distribution of volcanic heat flow from 266 active hot spots is consistent with the presence of a global magma ocean, and/or shallow asthenospheric heating. We find that Io's polar volcanoes are less energetic but about the same in number per unit area than at lower latitudes. We also find that volcanic heat flow in the north polar cap is greater than that in the south. The low volcanic advection seen at Io's poles is therefore at odds with measurements of background temperature showing Io's poles are anomalously warm. We suggest that the differences in volcanic thermal emission from Io's poles compared to that at lower latitudes is indicative of lithospheric dichotomies that inhibit volcanic advection towards Io's poles, particularly in the south polar region.
Early detection of volcanic eruptions is of major importance for protecting human life. Ground deformation and changes in seismicity, geochemistry, petrology, and gravimetry are used to assess volcanic activity before eruptions. Studies on Mt. Etna (Italy) have demonstrated that vegetation can be affected by pre-eruptive activity before the onset of eruptions. During two consecutive years before Mt. Etna's 2002/2003 flank eruption, enhanced vegetation index (NDVI) values were detected along a distinct line which later developed into an eruptive fissure. However, the mechanisms by which volcanic activity can lead to changes in pre-eruption tree growth processes are still not well understood. We analysed $δ^{13}$C, $δ^{18}$O and $^{14}$C in the rings of the survived trees growing near to the line where the pre-eruptive increase in NDVI was observed in order to evaluate whether the uptake of water vapour or fossil volcanic CO2 could have contributed to the enhanced NDVI. We found a dramatic decrease in $δ^{18}$O in tree rings formed before 2002/2003 in trees close to the eruption fissure, suggesting uptake of volcanic water by trees during pre-eruptive magma degassing. Moist condition
Frequencies of volcanic eruptions in the past 270 years are compared with variations of solar activity and summary curve of principal components of the solar background magnetic field (SBMF).Frequency analysis with Morlet wavelet reveals the most pronounced period of volcanic eruptions of 22 years. There is a strong correlation (0.84) between volcanic frequencies and the summary curve of SBMF for 11 cycles after 1868. The maxima of volcanic eruptions are shown to occur during solar activity cycles with the southern magnetic polarity. The next anticipated maximum of volcanic eruptions is expected to occur during cycle 26, when SBMF have a southern magnetic polarity.
The existence of volcanic lightning and alteration of the atmospheric potential gradient in the vicinity of near-vent volcanic plumes provides strong evidence for the charging of volcanic ash. More subtle electrical effects are also visible in balloon soundings of distal volcanic plumes. Near the vent, some proposed charging mechanisms are fractoemission, triboelectrification, and the so-called "dirty thunderstorm" mechanism, which is where ash and convective clouds interact electrically to enhance charging. Distant from the vent, a self-charging mechanism, probably triboelectrification, has been suggested to explain the sustained low levels of charge observed on a distal plume. Recent research by Houghton et al. (2013) linked the self-charging of volcanic ash to the properties of the particle size distribution, observing that a highly polydisperse ash distribution would charge more effectively than a monodisperse one. Natural radioactivity in some volcanic ash could also contribute to self-charging of volcanic plumes. Here we present laboratory measurements of particle size distributions, triboelectrification and radioactivity in ash samples from the Grímsvötn and Eyjafjallajökull
Volcanism is a major and long-term source of volatile elements such as C and H to Earth's atmosphere, likely has been to Venus's atmosphere, and may be for exoplanets. Models simulating volcanic growth of atmospheres often make one of two assumptions: either that atmospheric speciation is set by the high-temperature equilibrium of volcanism; or, that volcanic gases thermochemically re-equilibrate to the new, lower, temperature of the surface environment. In the latter case it has been suggested that volcanic atmospheres may create biosignature false positives. Here, we test the assumptions underlying such inferences by performing chemical kinetic calculations to estimate the relaxation timescale of volcanically-derived atmospheres to thermochemical equilibrium, in a simple 0D atmosphere neglecting photochemistry and reaction catalysis. We demonstrate that for planets with volcanic atmospheres, thermochemical equilibrium over geological timescales can only be assumed if the atmospheric temperature is above ~700K. Slow chemical kinetics at lower temperatures inhibit the relaxation of redox-sensitive species to low-temperature thermochemical equilibrium, precluding the production of t
Ground deformation measured from Interferometric Synthetic Aperture Radar (InSAR) data is considered a sign of volcanic unrest, statistically linked to a volcanic eruption. Recent studies have shown the potential of using Sentinel-1 InSAR data and supervised deep learning (DL) methods for the detection of volcanic deformation signals, towards global volcanic hazard mitigation. However, detection accuracy is compromised from the lack of labelled data and class imbalance. To overcome this, synthetic data are typically used for finetuning DL models pre-trained on the ImageNet dataset. This approach suffers from poor generalisation on real InSAR data. This letter proposes the use of self-supervised contrastive learning to learn quality visual representations hidden in unlabeled InSAR data. Our approach, based on the SimCLR framework, provides a solution that does not require a specialized architecture nor a large labelled or synthetic dataset. We show that our self-supervised pipeline achieves higher accuracy with respect to the state-of-the-art methods, and shows excellent generalisation even for out-of-distribution test data. Finally, we showcase the effectiveness of our approach for
We use chemical equilibrium calculations to model the speciation of alkalis and halogens in volcanic gases emitted on Io. The calculations cover wide temperature (500-2000 K) and pressure (10^-6 to 10^+1 bars) ranges, which overlap the nominal conditions at Pele (T = 1760 K, P = 0.01 bars). About 230 compounds of 11 elements (O, S, Li, Na, K, Rb, Cs, F, Cl, Br, I) are considered. We predict the major alkali and halogen species in a Pele-like volcanic gas and the major alklai and halogen condensates. We also model disequilibrium chemistry of the alkalis and halogens in the volcanic plume. Based on this work and our prior modeling for Na, K, and Cl in a volcanic plume, we predict the major loss processes for the alkali halide gases are photolysis and/or condensation onto grains. On the basis of elemental abundances and photochemical lifetimes, we recommend searching for gaseous KCl, NaF, LiF, LiCl, RbF, RbCl, CsF, and CsCl around volcanic vents during eruptions. Based on abundance considerations and observations of brown dwarfs, we also recommend a search of Io's extended atmosphere and the Io plasma torus for neutral and ionized Li, Cs, Rb, and F.