Session NH2.3

NH2.3 | Technologies for Forecasting Volcanic Hazards: Enhancing Risk Mitigation through Observations and Models

Technologies for Forecasting Volcanic Hazards: Enhancing Risk Mitigation through Observations and Models

Co-organized by GMPV9, co-sponsored by AGU

Convener: Ciro Del Negro | Co-conveners: Alessio Alexiadis, Eleonora AmatoECSECS, Silvia Massaro, Leonardo Mingari, Pablo TierzECSECS, Federica TorrisiECSECS

Orals

| Tue, 29 Apr, 14:00–15:45 (CEST)

Room 1.14

Posters on site

| Attendance Wed, 30 Apr, 10:45–12:30 (CEST) | Display Wed, 30 Apr, 08:30–12:30

Hall X3

When a volcano erupts, providing information on hazardous volcanic phenomena, their effects to communities and enviornmentes, and the eruption's duration is crucial to inform risk mitigation strategies. However, eruptions are complex phenomena governed by interactions of many processes, which are often nonlinear and stochastic. Numerous uncertainties in the involved parameters make precise predictions of specific events in time and space usually unattainable; that is, volcanic eruptions can be intrinsically unpredictable. Despite these limitations, significant progress has been made in forecasting volcanic hazards and, in specific circumstances, in making predictions. Understanding and predicting volcanic phenomena requires a comprehensive approach that integrates satellite observations, field measurements, and advanced modelling techniques. This has led to the increased use of data-driven approaches, including artificial intelligence (AI) techniques, to address volcanic hazards. Looking to the future, AI models can be combined with physical constraints to bridge the gap between data-driven methods and physical modeling, thereby increasing the interpretability of AI predictions. This offers an alternative approach to dealing with the strongly nonlinear and time-dependent character of volcanic phenomena.
This multidisciplinary session seeks to bring together contributions focusing on enhancing traditional ground-based volcano monitoring systems through technological innovation in satellite remote sensing and computational methods, integrating deep-learning, data-driven, physical and statistical modelling approaches, to better understand and forecast volcanic hazards. By fostering discussions and sharing insights, we aim to drive forward the development of more comprehensive and integrated approaches to volcanic hazard assessment and risk mitigation.

Orals: Tue, 29 Apr, 14:00–15:45 | Room 1.14

Chairpersons: Eleonora Amato, Pablo Tierz, Federica Torrisi

14:00–14:10

EGU25-6355

On-site presentation

Modern Deep Learning Techniques for Volcanic Unrest Monitoring using InSAR Data

Nantheera Anantrasirichai, Juliet Biggs, Robert Gabriel Popescu, Xuan Wern Joshua Kong, and Tianqi Yang

Satellites provide essential capabilities for widespread, regional, or global volcano surveillance, often offering the first indications of volcanic unrest or eruptions. Here, we focus on Interferometric Synthetic Aperture Radar (InSAR), a technology detecting surface deformation that is statistically strongly linked to volcanic activity. Recent technological advancements have enabled the generation of vast amounts of monitoring data—e.g., LiSC system currently provides over 3.4 million raw interferograms. Clearly, manual analysis of such a large dataset is no longer feasible. This talk presents several modern, learning-based techniques for ground deformation monitoring using InSAR data, including supervised, semi-supervised, and unsupervised learning approaches.

Supervised learning methods have successfully detected fringes in wrapped interferograms. We improved our CNN-based detection process [1,2,3] by incorporating state-of-the-art Transformers. However, these methods may miss ground deformations with characteristics differing from the training data. To address this limitation, we explore the potential of using semi-supervised learning [4]. In this approach, a global feature representation of InSAR data is learned through unsupervised contrastive learning [5], and the detection task is subsequently fine-tuned on a limited number of labelled samples. For unsupervised learning, our model identifies samples that deviate from the norm of the data as anomaly detection. It is performed in the feature space of unwrapped interferograms [6] and employs a statistical-based approach, Patch Distribution Modelling [7]. The results show that this method outperforms existing supervised learning techniques when the characteristics of deformation are unknown.

Interferograms capture deformation signals and atmospheric effects, which can distort detection accuracy. While GACOS provides atmospheric corrections, it may fail to fully remove effects and sometimes introduces artifacts. To address these limitations, we enhance our system with learning-based denoising techniques to mitigate atmospheric effects. Two approaches are presented: Transformer-based and diffusion model-based denoising. The first method adapts the state-of-the-art image denoising model, Reformer [8], but replaces the feed-forward network with multi-layer perceptron. The second method leverages Denoising Diffusion Probabilistic Models [9], incorporating turbulence noise in the forward diffusion process. Initial results, evaluated against GPS data, demonstrate that this method outperforms traditional time-series processing in mitigating atmospheric effects.

References:

[1] N Anantrasirichai et al., Application of Machine Learning to Classification of Volcanic Deformation in Routinely Generated InSAR Data. JGR Solid Earth, 2018

[2] N Anantrasirichai et al., A deep learning approach to detecting volcano deformation from satellite imagery using synthetic datasets, RSE, 2019

[3] N Anantrasirichai et al., The application of convolutional neural networks to detect slow, sustained deformation in InSAR time series, GRL, 2019

[4] N Anantrasirichai et al., Semi-supervised Learning Approach for Ground Deformation Detection in InSAR, Fringe, 2023

[5] T Yang et al., A Semi-supervised Learning Approach for B-line Detection in Lung Ultrasound Images. ISBI, 2023

[6] R Popescu et al., Anomaly detection for the identification of volcanic unrest in satellite imagery, ICIP, 2024

[7] T Defard et al., A Patch Distribution Modeling Framework for Anomaly Detection and Localization, ICPRW, 2021

[8] N Kitaev et al., Reformer: The Efficient Transformer, ICLR, 2020

[9] J Ho et al., Denoising diffusion probabilistic models. NIPS, 2020

How to cite: Anantrasirichai, N., Biggs, J., Popescu, R. G., Kong, X. W. J., and Yang, T.: Modern Deep Learning Techniques for Volcanic Unrest Monitoring using InSAR Data, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6355, https://doi.org/10.5194/egusphere-egu25-6355, 2025.

14:10–14:40

EGU25-3620

solicited

On-site presentation

Forecasting the evolution of the 2021 Tajogaite eruption, La Palma, with TROPOMI/PlumeTraj-derived SO2 emission rates

Mike Burton, Ben Esse, Catherine Hayer, Giuseppe La Spina, Ana Pardo Cofrades, María Asensio Ramos, José Barrancos Martínez, and Nemesio Pérez

As global populations grow, the exposure of communities and infrastructure to volcanic hazards increases every year. Once a volcanic eruption begins it becomes critical for risk managers to understand the likely evolution and duration of the activity to assess its impact on populations and infrastructure. Here, we report an exponential decay in satellite-derived SO₂ emission rates during the 2021 eruption of Tajogaite, La Palma, Canary Islands, and show that this pattern allows a reliable and consistent forecast of the evolution of the SO₂ emissions after the first third of the total eruption duration. The eruption ended when fluxes dropped to less than 6% of their fitted maximum value, providing a useful benchmark to compare with other eruptions. Using a 1-D numerical magma ascent model we suggest that the exponentially decreasing SO₂ emission trend was primarily produced by reducing magma chamber pressure as the eruption emptied the feeding reservoir. This work highlights the key role that satellite-derived SO₂ emission data can play in forecasting the evolution of volcanic eruptions and how the use of magma ascent models can inform the driving mechanisms controlling the evolution of the eruption.

How to cite: Burton, M., Esse, B., Hayer, C., La Spina, G., Pardo Cofrades, A., Asensio Ramos, M., Barrancos Martínez, J., and Pérez, N.: Forecasting the evolution of the 2021 Tajogaite eruption, La Palma, with TROPOMI/PlumeTraj-derived SO2 emission rates, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3620, https://doi.org/10.5194/egusphere-egu25-3620, 2025.

14:40–14:50

EGU25-13414

On-site presentation

Forecasting volcanic activity onset and eruption with the next generation of thermal infrared data

Michael Ramsey and Claudia Corradino

Thermal infrared (TIR) imaging of volcanic activity has become common over the past quarter century with the advent of smaller, inexpensive, ground-based cameras and greatly expanded orbital coverage. Because of these advances, TIR data are also now integrated into the standard set of monitoring tools at many volcano observatories. These data are acquired using permanent ground-stations, less frequent campaign mode deployments from the ground and air, as well as orbital remote sensing. However, the ability to forecast a new eruption using orbital TIR data remains unrealized despite decades of data acquisition, modeling, and analysis. Fundamentally, these data are limited due to the design metrics of the sensors such as spatial and/or temporal resolution. One endmember group of these instruments is defined by lower spatial, higher temporal resolution whose data can detect large-scale thermal change such as new lava on the surface. Sensors in this class are used to rapidly identify a new eruption and monitor its evolution, for example. The other endmember has sensors with higher spatial, lower temporal resolution data with sensitivity to detect subtle temperature changes (1-2 degrees) over small spatial scales. Our work examines decades of TIR data from this second endmember class to identify precursory thermal eruption signals. By including all data (day and night) screened for clouds, we produce a larger statistical dataset from which to extract thermal signal deviations from a standard baseline. This long time series orbital TIR data enable a unique opportunity to quantify low-level anomalies and small eruption plumes over long periods. Most significant is the finding that the smaller, subtle detections served as precursory signals in ~81% of eruptions for our five test locations, which we have now expanded to a wider range of volcanoes and activity styles. The results also serve as training for machine learning based modeling that is applied to different targets for this study. This model learns to identify discriminant thermal trends associated to unrest conditions preceding eruptions. Over the next decade, several high spatial (~ 60 m) resolution orbital sensors are planned will provide near-daily TIR data at every volcano, vastly improving thermal baselines and detection of new activity. One of these, the Surface Biology and Geology (SBG) TIR mission, contains an infrared instrument and a planned higher-level data product called the Volcano Activity (VA), which will be crucial for accurate daily monitoring of volcanic temperatures and degassing rates. However, despite the promise of SBG data, the next fundamental step-change in orbital volcanology will not come until high-speed, spaceborne data are possible. A proposed “hypertemporal” TIR mission would acquire these data at sub-minute scales to determine mass and thermal flux rates of gas emissions, eruptive ash plumes, and lava flows. With such a mission, data now acquired by current ground-based cameras will become possible from orbit for the first time.

How to cite: Ramsey, M. and Corradino, C.: Forecasting volcanic activity onset and eruption with the next generation of thermal infrared data, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13414, https://doi.org/10.5194/egusphere-egu25-13414, 2025.

14:50–15:00

EGU25-3167

On-site presentation

Integration of Sentinel 2-MSI and Landsat 8/9-OLI data for detecting and mapping sea-water discoloration around submarine volcanoes

Emanuele Ciancia, Francesco Marchese, Simon Plank, and Nicola Pergola

Shallow eruptions of submarine volcanoes can hamper navigation of ships and alter the biological response of marine ecosystems. Hydrothermal vents and ash-laden plumes can spread on sea-surface for weeks affecting the optical properties of the water column. Systematic in situ observations (i.e., underwater observations, hydro-acoustic and seismic arrays) are usually time-consuming, expensive, and difficult to carry out before and during an eruptive event. On the other hand, satellite remote sensing can provide timely and continuous information about volcanic activities around dangerous sites contributing to the assessment on the pre-, syn- and post-eruptive phenomena. Among these, sea-water discoloration is one of the most significant indicators of underwater volcanic activity as its accurate and timely detection may support in revealing possible precursor processes of submarine volcanic eruptions. Most of the published studies have been performed to characterize discolored water patches after huge eruptions through the assessment of their reflectance patterns by using multispectral ocean color data acquired by MODIS, VIIRS and Sentinel-3 OLCI. Although these sensors enable a timely detection of submarine eruption features, their coarse spatial resolution makes them unsuitable for mapping discolored patches whose size and spatial dynamics are at ten- or hundred-meter scale. The improved spatial resolution offered by Sentinel 2-MSI and Landsat 8/9-OLI data (10-60 m) can ensure an accurate mapping of sea-water discoloration. Moreover, their joint use would allow for monitoring discolored plumes at unprecedented rates with a potential revisit time of 2-3 days at global scale. In this study, we aim at assessing the potential of the Sentinel 2-MSI and Landsat 8/9-OLI integrated datasets in characterizing sea-water discoloration around a selected test case, namely the Kavachi submarine volcano (Solomon Islands, South Pacific Ocean).

By exploiting a 3-year (2020-2022) MSI-OLI combined dataset, we developed a novel spectral-derived method to detect and map discolored patches before potential subaerial eruptions. The proposed work is expected to provide a first contribution in better investigating the possible precursor signs of submarine volcanic eruptions.

How to cite: Ciancia, E., Marchese, F., Plank, S., and Pergola, N.: Integration of Sentinel 2-MSI and Landsat 8/9-OLI data for detecting and mapping sea-water discoloration around submarine volcanoes, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3167, https://doi.org/10.5194/egusphere-egu25-3167, 2025.

15:00–15:10

EGU25-13887

On-site presentation

Magma Depletion: An alternative to time-homogeneity for forecasting vent distribution in volcanic fields

Mark Bebbington, Melody Whitehead, and Gabor Kereszturi

For small volume eruptions, such as those common for volcanic fields, the location of an eruptive vent controls the hazards, their intensities, and ultimately the impact of the eruption. An eruption through water can result in a highly explosive event, and an eruption beneath a hospital or critical infrastructure can cause significant long-term impacts. We look here at long-term probabilistic assessments, the outputs of which inform evacuation plans, the (re)location of vital infrastructure, and inform the placement of early-warning monitoring equipment.

Current estimates of future vent locations are based on point-process methods with probability surfaces built from patterns, clusters, and/or lineaments identified from previous vent locations. These all assume that locations with more past-vents are more likely to produce future-vents, or in other words a null hypothesis

H_o: The likelihood of an eruption at the location of an existing vent is a local maximum of the spatial density surface.

Critically, under this model the occurrence of an eruption does not change the likelihood of further eruptions at that locality. We investigate here an alternative (but not necessarily better) hypothesis of magma depletion, i.e., that after an eruption, the magma source at depth is depleted by the volume of the eruption in this area, lessening the likelihood by creating a local depression in the probability surface. More formally we consider the alternative hypothesis

H_a: The likelihood of an eruption at the location of an existing vent is a local minimum of the spatial density surface.

We present the mathematics and code for various alternatives to current kernel density estimates, and then set out to try and disprove our null hypothesis by examining goodness of fit to data, all using the exemplar of the Auckland Volcanic Field, New Zealand

How to cite: Bebbington, M., Whitehead, M., and Kereszturi, G.: Magma Depletion: An alternative to time-homogeneity for forecasting vent distribution in volcanic fields, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13887, https://doi.org/10.5194/egusphere-egu25-13887, 2025.

15:10–15:20

EGU25-15508

On-site presentation

Probabilistic hazard maps of pyroclastic density current at Vesuvius volcano (Italy): A new strategy for risk reduction

Daniela Mele, Pierfrancesco Dellino, Fabio Dioguardi, and Roberto Sulpizio

The hazard of pyroclastic density currents (PDCs) at Vesuvius is investigated based on past eruptions. The analysis is extended to all eruptions that left substantial deposits on the ground.

The currents are bipartite, with a basal highly-concentrated part, which was fed from the impact of the eruptive fountain on the ground, and an overlying part generated by the squeezing of the collapsed material that fed a dilute and turbulent shear flow.

Dynamic pressure, particle volumetric concentration, temperature and flow duration are hazardous characteristics of PDCs that can impact buildings and populations and are defined here as impact parameters. They have been calculated through an implementation of the PYFLOW code, which uses the deposit particle characteristics as input. The software searches for the probability density function of impact parameters. The 84th percentile has been chosen as a safety value of the expected impact at long term (50 years). Maps have been constructed by interpolation of the safety values calculated at various points over the dispersal area, and show how impact parameters change as a function of distance from the volcano. The maps are compared with the red zone, which is the area that the National Department of the Italian Civil Protection has declared to be evacuated in the impending of an eruption. The damaging capacity of currents over buildings and population is discussed both for the highly concentrated part and the diluted one.

How to cite: Mele, D., Dellino, P., Dioguardi, F., and Sulpizio, R.: Probabilistic hazard maps of pyroclastic density current at Vesuvius volcano (Italy): A new strategy for risk reduction, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15508, https://doi.org/10.5194/egusphere-egu25-15508, 2025.

15:20–15:30

EGU25-5433

On-site presentation

Prioritization and Interconnectivity Analysis of Critical Facilities for Volcanic Ash Risk Management in South Korea: An ANP Approach

Yoo Jung Kim, Sungsu Lee, Byung Cheol Park, and Sanghoon Yoon

Volcanic ash from large eruptions in and around the Korean Peninsula poses significant risks to critical facilities. This study employs the Analytic Network Process (ANP) to evaluate the relative importance and interconnectivity of different facility sectors vulnerable to volcanic ash impacts. The analysis focused on 12 facility categories grouped into three main sectors: transportation, infrastructure, and public facilities.

Historical volcanic damage cases were analyzed using data from vHub and the Global Volcanic Program (GVP), revealing that 53.8% of volcanic eruption cases involved ash-related damage. Based on this analysis and expert consultation, a network model was developed to capture the complex relationships between facility sectors. Volcanic disaster experts participated in a survey to assess the relative importance and influence relationships between different facility categories.

The results showed that transportation facilities had the highest importance (0.509), followed by infrastructure (0.354) and public facilities (0.137). Among all subcategories, aviation emerged as the most critical sector with an importance value of 0.246, significantly higher than other facilities. This was followed by electricity (0.117), broadcasting and communication (0.110), and ships and ports (0.103). The high ranking of aviation reflects South Korea's particular vulnerability to long-range ash dispersion effects, similar to the impacts observed during the 2010 Eyjafjallajökull eruption in Europe.

Interconnectivity analysis using a weighted super-matrix revealed significant cascade effects between sectors. Road damage showed substantial influence on medical facilities (42.8%), aviation (27.1%), and railways (15.2%). The electricity sector demonstrated broad impacts across all facilities, with particularly strong influences on broadcasting and communication (23.1%), medical facilities (20.4%), and railways (16.6%). Medical facilities emerged as highly dependent on other sectors, being significantly affected by disruptions to roads, water supply, and electricity.

These findings provide valuable insights for volcanic ash risk management in South Korea, where the threat primarily comes from distant volcanoes like Mount Baekdu. The results highlight the need for targeted mitigation strategies focusing on aviation and electrical infrastructure, while also considering the complex interdependencies between different facility sectors. This study contributes to the development of more effective disaster response planning and risk assessment methodologies tailored to South Korea's specific volcanic hazard context.

How to cite: Kim, Y. J., Lee, S., Park, B. C., and Yoon, S.: Prioritization and Interconnectivity Analysis of Critical Facilities for Volcanic Ash Risk Management in South Korea: An ANP Approach, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5433, https://doi.org/10.5194/egusphere-egu25-5433, 2025.

15:30–15:45

Discussion

Posters on site: Wed, 30 Apr, 10:45–12:30 | Hall X3

Display time: Wed, 30 Apr, 08:30–12:30

Chairpersons: Alessio Alexiadis, Ciro Del Negro, Silvia Massaro

X3.1

EGU25-481

ECS

Combining numerical CFD models and AI to enhance lava flow simulations

Eleonora Amato, Vito Zago, and Ciro Del Negro

Lava flows are complex, non-Newtonian fluids with visco-thermal dependencies that can overcome barriers, form tunnels, and significantly impact surrounding areas. Understanding and predicting these flows are critical for quantifying volcanic hazards. Computational Fluid Dynamics (CFD) models are indispensable tools for simulating lava dynamics, but they often entail high computational costs, limiting their real-time applicability. To address these challenges, we propose an AI-enhanced CFD emulator for lava flows, designed to improve modeling efficiency while preserving accuracy. Our approach integrates AI with CFD to capture the visco-thermal properties of lava and its intricate dynamics, including phase transitions, particle solidification, and the influence of air on thermal behavior. The emulator has been validated through simulations of diverse physical scenarios, demonstrating its capability to generalize across varying conditions. Additionally, we conducted a sensitivity analysis, exploring the influence of key parameters, such as effusion rate, on lava flow evolution and eruption styles. By incorporating satellite-derived estimates, we provide insights into eruptive behaviors while minimizing the risks of field observations. Our results showcase the potential of combining AI, numerical models, and remote sensing to enhance traditional volcanic monitoring approaches. This hybrid methodology enables faithful, near real-time simulations of lava flows, offering valuable tools for hazard assessment and risk mitigation.

How to cite: Amato, E., Zago, V., and Del Negro, C.: Combining numerical CFD models and AI to enhance lava flow simulations, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-481, https://doi.org/10.5194/egusphere-egu25-481, 2025.

X3.2

EGU25-532

ECS

Spatiotemporal Tracking of the Volcanic Cloud Dispersion Using ConvLSTM Models and SEVIRI Imagery

Federica Torrisi, Claudia Corradino, and Ciro Del Negro

Explosive volcanic eruptions inject a variety of particles and gases into the atmosphere, forming volcanic clouds that significantly impact human health, climate, and aviation safety. Accurately capturing the temporal evolution of these clouds is essential for understanding their dynamics and improving predictive capabilities. Due to the rapid and unpredictable nature of explosive eruptions, volcanic clouds can form, expand, and disperse in short timeframes. For this reason, high-temporal-resolution geostationary satellite data are indispensable for near-real-time monitoring. SEVIRI (Spinning Enhanced Visible and InfraRed Imager), onboard the Meteosat Second Generation (MSG) geostationary satellite, provides high-frequency radiometric data essential for tracking volcanic clouds on a global scale. SEVIRI's ability to acquire images at intervals of 5–15 minutes enables the identification of patterns in cloud formation and dispersion, supporting timely warnings and informed decision-making during crises. Here, we propose a novel approach using a convolutional long short-term memory (ConvLSTM) model, a type of recurrent neural network designed to handle spatiotemporal data, for effectively tracking the spread of volcanic clouds using satellite imagery. By training the model on a dataset of Ash RGB images derived from SEVIRI data, we analyze volcanic events at Mt. Etna (Italy) to demonstrate the model's capability to capture both spatial and temporal dynamics. Our findings show that ConvLSTM models excel in addressing complex spatiotemporal challenges, providing robust segmentation and reliable tracking of volcanic clouds over time. This approach delivers timely information that enhances aviation safety, emergency response, and public health monitoring, contributing to more effective management of volcanic crises.

How to cite: Torrisi, F., Corradino, C., and Del Negro, C.: Spatiotemporal Tracking of the Volcanic Cloud Dispersion Using ConvLSTM Models and SEVIRI Imagery, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-532, https://doi.org/10.5194/egusphere-egu25-532, 2025.

X3.3

EGU25-17985

AI-driven insights into the volcanic processes and dynamics of explosive episodes inferred by satellite-based SO2 estimates, ground-based gas measurements, and petrological data

Claudia Corradino, Alessandro La Spina, Lucia Miraglia, Federica Torrisi, and Ciro Del Negro

Identifying changes in a volcano's unrest and tracking the evolution of its eruptive activity are crucial for effective volcanic surveillance and monitoring. Variations in gas composition and amount can be associated with pre-eruptive changes in the volcano plumbing system. When combined with petrological studies, the emitted Sulphur dioxide (SO2) reflects the amount of magma involved (erupted or degassed), making it a useful parameter for constraining volcanic processes, dynamics, and the volume of magma. This work proposes an Artificial Intelligence (AI) strategy to provide new insights into the volcanic processes and dynamics of explosive episodes using a multidisciplinary approach. Through advanced machine learning (ML) algorithms, we investigate the spatio-temporal relationships among the SO2 satellite image time series (SITS), ground-based gas measurements, and petrological data associated with volcanic pre- and syn-eruptive phases. SO2 emissions are estimated via satellite ultraviolet remote sensing, i.e. TROPOspheric Monitoring Instrument. Both the quiescent/pre-eruptive and syn-eruptive/explosive gas phases are constrained from ground-based infrared remote sensing data i.e Fourier Transform InfraRed (FTIR). Rock compositions and textural features (e.g. crystallinity and vesicularity) of volcanic products are estimated by petrological study. The ML algorithm allows to both discover pre- and syn-eruptive patterns indicative of future eruption and better characterize volcanic processes. Unsupervised ML techniques are considered to explore previously unknown relationships without any external bias. We have tested this approach on recent volcanic activity that occurred on Mt Etna.

How to cite: Corradino, C., La Spina, A., Miraglia, L., Torrisi, F., and Del Negro, C.: AI-driven insights into the volcanic processes and dynamics of explosive episodes inferred by satellite-based SO2 estimates, ground-based gas measurements, and petrological data, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17985, https://doi.org/10.5194/egusphere-egu25-17985, 2025.

X3.4

EGU25-14603

ECS

Revisiting Wickman (1966): Forecasting eruption onset and periods of activity at Popocatépetl volcano (Mexico)

Daniela Hernández Villamizar and Hugo Delgado Granados

Delgado Granados et al. (1988) forecasted the initiation of an eruption at Popocatépetl volcano of any kind in 1997 with a 97% confidence and with a recurrence of eruptions every ~70 years. For this, they used the methods developed by Wickman (1966a-e) and Thorlaksson (1967) using the repose period concept. The current eruption at the volcano was initiated in 1994, three years in advance to the forecasted year. In this work we used the same expressions to calculate a new forecast using the same data. Interestingly, the forecast at 95% confidence indicates 1994 as the initiation of the next eruption after the end of the last eruptive period (1927) with a recurrence time of ~67 years. Further, we made the calculation adding more dates found in the recorded history of the volcano. We obtained a forecast of the initiation of the next eruption, after 1927, in the year 1994 at 95% confidence and a recurrence period of ~67 years. Using the same tools, but now for the duration of the activity intervals, we obtained a period of activity duration at ~43 years. Using this timing, the current eruption could be ending in 2037 at 95% probability.

REFERENCES

Delgado Granados H., Carrasco Núñez G., Urrutia Fucugauchi J., Casanova Becerra J.M., 1988, Analysis of the Eruptive Records of the Popocatépetl Volcano, Mexico, Kagoshima International Conference on Volcanoes, Proceedings Volume 1988, pp. 510-513.

Thorlaksson J.E.,1967, A probability model of volcanoes and the probability of eruptions of Hekla and Hekla and Katla, Bull. Vol., 31, 97-106.

Wickman F.E., 1966, Repose period patterns of volcanoes. I. Volcanic eruptions regarded as random phenomena. Ark. Mineral. Geol., 4, 7, pp. 291-301.

Wickman, F.E., 1966, Repose period patterns of volcanoes. IV. Volcanic eruptions regarded as random phenomena. Ark. Mineral. Geol., 4, 10, pp. 337-350.

How to cite: Hernández Villamizar, D. and Delgado Granados, H.: Revisiting Wickman (1966): Forecasting eruption onset and periods of activity at Popocatépetl volcano (Mexico), EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-14603, https://doi.org/10.5194/egusphere-egu25-14603, 2025.

X3.5

EGU25-13863

GNSS network and volcanic deformation patterns: Somma-Vesuvius case study.

(withdrawn)

Umberto Tammaro, Mario Dolce, Giuseppe Brandi, Antonio Iorio, Giovanni Scarpato, and Prospero De Martino

X3.6

EGU25-5139

From Magma to Eruption: Modeling VolcanicProcesses with Diffusion Theory

Cataldo Godano, Massimiliano Semeraro, Giuseppe Gonnella, Giovanni Macedonio, Francesco Oliveri, Patrizia Rogolino, and Alessandro Sarracino

X3.7

EGU25-20727

Long-term probabilistic hazard assessment posed by gas dispersion at Vulcano island (Aeolian archipelago, Italy)

Silvia Massaro, Antonio Costa, Domenico Granieri, Manuel stocchi, giovanni macedonio, fabio dioguardi, alejandra guerrero, and arnau folch

Persistently active volcanoes emit gas continuously and may present a long-term hazard depending upon the gas specie, concentration levels and exposure time.

Following the last gas crisis occurred at Vulcano island (Aeolian archipelago, Italy) during 2021-2022, the surveillance activities carried out by the personnel of the Istituto Nazionale di Geofisica and Vulcanologia of Palermo Branch and the Etnean Observatory, let us to investigate the state of the island's hazard concerning volcanic gases by considering two degassing scenarios for CO₂ and SO₂ dispersion (background and unrest). To do this, we used the recently released version of VIGIL workflow (1.3.8) able to run automatically passive or gravity-driven gas dispersion simulations using DISGAS (2.6.0) and TWODEE-2 (2.6.0) models, respectively. Both models are interfaced with DIAGNO simulator (1.5.0) that require daily meteorological data.

Our results are based on 1000 simulations using averaged wind profiles from the ECMWF ERA5 database which constitute a representative sample of meteorological variability over the past 30 years (1993-2023). Long-term hazard maps are related to the probabilities of exceedance (PE) at 5% and 10% of the simulated CO₂ and SO₂ concentration at a height of 1.5 m above the ground (referring to the average height of a person). Persistence maps are built considering different thresholds for human exposure in accordance with the regulations of the European Union and the World Health Organization. Additionally, we present ongoing efforts to address current limitations in the VIGIL workflow, including improvements in handling source uncertainty (location and intensity), a more user-friendly interface, and the integration of a new wind simulator.

How to cite: Massaro, S., Costa, A., Granieri, D., stocchi, M., macedonio, G., dioguardi, F., guerrero, A., and folch, A.: Long-term probabilistic hazard assessment posed by gas dispersion at Vulcano island (Aeolian archipelago, Italy) , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20727, https://doi.org/10.5194/egusphere-egu25-20727, 2025.

X3.8

EGU25-16114

ECS

Making a volcanic point: certain subduction zones worldwide accumulate highly hazardous, pointy stratovolcanoes

Pablo Tierz, Teresa Ubide, John Caulfield, Philippa White, Fabrizio Ponce, Roberto Mérida, Susan Loughlin, and Eliza Calder

Volcanic landscapes are amongst the most breathtaking visual features on Earth. Volcano morphologies can be extremely varied, including negative-relief topographic depressions (e.g. calderas) as well as many different configurations of positive reliefs (e.g. shields or stratovolcanoes). These volcano morphologies provide information about magmatic and eruptive processes and, therefore, represent invaluable sources of data, especially for data-scarce volcanic systems. Volcano morphology also modulates volcanic hazard, for example, by providing potential energy (edifice height, mean flank slope, etc.) for propagation of volcanic mass flows (lava flows, pyroclastic density currents, lahars, etc.); and/or in relation to potential instability of the volcanic edifice which, upon gravitational collapse, can generate large-volume, long-runout debris avalanches and debris flows.

Here, we quantify volcano morphology at several hundred volcanic systems worldwide, using a metric derived from an innovative, data-driven method to search for analogue volcanoes (VOLCANS). The metric simplifies volcano morphology by combining: (1) edifice height, (2) mean flank slope, (3) crater diameter and (4) degree of truncation of the edifice (i.e. ratio between the width of the summit area divided by that of the whole edifice). This makes it possible to distinguish between high, steep, pointy volcanoes with small craters and low, gentle-slope, truncated volcanoes with large craters/depressions. The VOLCANS metric indicates that high, steep and pointy (i.e. non-truncated) stratovolcanoes (henceforth referred to as ‘pointy volcanoes’) do not occur at random. Instead, pointy volcanoes tend to accumulate within specific subduction zones worldwide. Some of the most striking examples of subduction segments with high proportions of pointy volcanoes include Guatemala, which hosts all the pointy volcanoes in the entire Central American region (e.g. Fuego, Agua, Atitlán, Santa María, Tacaná) and Kamchatka, Russia, which hosts around 20% of all the pointy volcanoes identified worldwide (e.g. Klyuchevskoy, Vilyuchik, Kronotsky, Koryaksky). Other pointy-rich subduction segments include: the Alaskan Peninsula and the Cascades, USA (e.g. Pavlof or Mt Baker), Ecuador (e.g. Sangay), Java, Indonesia (e.g. Semeru, Merapi) or Central and Southern Chile (e.g. Lanín, Osorno, Villarrica).

We postulate that, in order to build such extreme volcano morphologies, frequent eruptive activity of mildly-evolved magmas (with low-to-intermediate viscosities), plus a limited spatial variability in the location of the eruptive vent(s), are necessary to maintain vertical growth of the volcanic edifice. Moreover, sparsity of large-explosive eruptions safeguards the ‘pointiness’ of the volcano, avoiding truncation of the edifice and/or mantaining small craters. We acknowledge that volcano morphology represents just a snapshot in time within the geological evolution of any volcanic system. Interestingly, however, some pointy volcanoes have experienced gravitational collapse(s) of their edifices in the past (e.g. Acatenango-Fuego, Guatemala), and have managed to rebuild their pointy edifices through subsequent eruptive activity. Currently, we are exploring several datasets of: (i) subduction kinematics, (ii) magma geochemistry and (iii) eruptive fluxes, to try to tie our morphological observations to their possible causative processes. Such an analysis is extremely relevant, not only to improve our understanding of how volcanic systems operate but also to quantify volcanic hazard at subduction zones and their volcanic systems.

How to cite: Tierz, P., Ubide, T., Caulfield, J., White, P., Ponce, F., Mérida, R., Loughlin, S., and Calder, E.: Making a volcanic point: certain subduction zones worldwide accumulate highly hazardous, pointy stratovolcanoes, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16114, https://doi.org/10.5194/egusphere-egu25-16114, 2025.

X3.9

EGU25-18986

Latest developments in measurement and geodetic monitoring techniques for shallow water volcanic areas subjected to vertical deformation phenomena (application on Campi Flegrei caldera).

(withdrawn after no-show)

Sergio Guardato, Rosario Riccio, Rebecca Sveva Morelli, Francesco Chierici, Stefano Caliro, Giovanni Macedonio, and Giovanni Iannaccone

X3.10

EGU25-16933

ALERTACO2 Project update: An extensive monitoring network for monitor and mitigate the CO2 hazard of indoor and outdoor air CO2 at the inhabited areas of Puerto Naos and La Bombilla, La Palma (Canary Islands)

Germán D. Padilla, Carmen López, Nemesio M. Pérez, Rubén López, Pedro A. Hernández, David Moure, Luca D'Auria, Pedro Torres, Gladys Melián, Daniel D'Nardo, Carla Méndez, Alexis González, and Juan A. Bermejo

As a result of the Tajogaite eruption (2021), La Palma island, anomalous volcanic CO₂ emissions were observed by the end of November 2021 in the neighborhoods of La Bombilla, Puerto Naos, and some banana plantations, where appear daily many dead fauna (insects, birds, lizards and small mammals), located at about 6 km southwestern from Tajogaite eruption vents. These urban areas, not directly damaged by lava flows, were included in the exclusion zone due to the strong volcanic-hydrothermal CO₂ concentrations (>5-20%). CO₂ enters into the homes and premises through hydraulic and electrical conduits and the vertical structure of the buildings itself, causing an accumulation of CO₂ indoor that reaches high or very high concentrations. CO₂ is an asphyxiating and toxic gas in very high concentrations, as it implies a corresponding reduction in the oxygen (O₂) content. Immediate evacuation of indoor spaces is recommended if the CO₂ concentration excedes 1.5% (15,000ppm).

During the last two years after the eruption, several institutions deployed indoor and outdoor own gas networks, to try to delimitate the CO₂ anomalies where CO₂ air concentration exceed hazardous thresholds, but with an insufficient number of CO₂ sensors (less than 100) to cover all homes, garages, basements and stores in real time. These studies aim to understand the dynamics of CO₂ emission to delimitate the CO₂ anomalies where CO₂ air concentration exceed the hazardous thresholds, and help the authorities’ decision-making of people's return to their homes and stores.

The ALERTACO2 project, participated by IGN and INVOLCAN institutes, was financed by the Spanish Government with an amount of 3M€ during a period of 4 years (2023-2026), and has the goals of implementing a much more extensive network of CO₂ sensors (around 1,200 NDIR sensor developed by Sieltec Canarias) in real time in most of the building of both inhabited areas, the creation of a 24-hour monitoring room and an information and awareness campaign for the population about this volcanic hazard.

At the present time, 1294 sensors are installed (1,287 indoor and 7 outdoor), of which 147 are in La Bombilla and 1,133 in Puerto Naos and 7 moving stations and 7 outside these places. Each sensor has a color light code to indicate the CO₂ concentration (green, yellow, orange and blue if the sensor is not working), and a QR code to view the information remotely. Each sensor sends the data to the 24-hour monitoring room via a gateway installed at the roof of each building. Thanks to ALERTACO2, many families have been able to return to their homes in safety conditions since December 2023, because their homes average CO₂ concentrations were below 1,000 ppm.

How to cite: Padilla, G. D., López, C., Pérez, N. M., López, R., Hernández, P. A., Moure, D., D'Auria, L., Torres, P., Melián, G., D'Nardo, D., Méndez, C., González, A., and Bermejo, J. A.: ALERTACO2 Project update: An extensive monitoring network for monitor and mitigate the CO2 hazard of indoor and outdoor air CO2 at the inhabited areas of Puerto Naos and La Bombilla, La Palma (Canary Islands) , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16933, https://doi.org/10.5194/egusphere-egu25-16933, 2025.

X3.11

EGU25-4140

The contribution of multidisciplinary petrological and geochemical framework (PGF) to assess the influence of plumbing architecture on volcano dynamics and monitoring strategies

(withdrawn after no-show)

Andrea Di Muro, Andrea Rizzo, Marco Liuzzo, Fausto Grassa, and Bhavani Benard