The combination of Computational Fluid Dynamics (CFD) and Artificial Intelligence (AI) enables to expand the scope of fluid modeling and improve simulation performance. Eulerian methods have already been extensively integrated with AI, providing high-fidelity and reliable results (e.g., weather prediction); the application of AI to Lagrangian methods remains less consolidated. Smoothed Particle Hydrodynamics (SPH) is a Lagrangian mesh-less CFD numerical method with well-established advantages for the simulation of highly dynamic free-surface complex fluids. However, reliable SPH simulations require long execution times and large computational resources. This downside can be resolved using emulators, which can reproduce the dynamics of a physical system, speeding-up the simulations. In detail, an emulator is a model in which AI algorithms join or replace the equation-based mathematical representation of physics. Thus, the emulator learns from CFD simulations the behavior of the CFD reference model and reproduces it to solve fluid dynamics problems in shorter times. For the emulator to be trusted, it is important to assess the ability to generalize and correctly predict the behavior of the fluid in conditions that have not been presented during training phase. Here, we present the generalizability of an AI-based emulator for SPH, utilizing an Artificial Neural Network (ANN) to estimate the hydrodynamic forces between SPH particles. We demonstrate the capability of the emulator to reproduce particle interaction quantities, as opposed to their specific spatial configuration. Therefore, when a different problem is treated, as long as the variables values are in a range with physics meaning presented during training, the emulator will be capable to produce physically meaningful results. We discuss the salient points of designing this emulator, such as the choice of the AI model, i.e., the ANN, and the choice of the features. We show applications to different test cases, highlighting the emulator ability to generalize for problems with varying levels of complexity, and for changes in the model resolution.

How to cite: Amato, E., Zago, V., and Del Negro, C.: Generalizability of AI-based Emulators for CFD Lagrangian methods , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-563, https://doi.org/10.5194/egusphere-egu24-563, 2024.

Nowadays, several satellite missions provide thermal infrared data at various spatial resolutions and revisit time, enabling nearly continuous monitoring of thermal volcanic activity worldwide. In addition, computer vision techniques, based on Machine Learning (ML) and Deep Learning (DL) models, offer unique advantages in automatically extracting valuable information from large datasets. Here, we propose a Machine Learning approach that leverages the capabilities of such models, combined with nearly continuous high spatial resolution images (20 m) acquired from the Sentinel-2 MultiSpectral Instrument (S2-MSI), to detect high-temperature volcanic features and to quantify volcanic thermal emissions. The impact of the volcanic activity is assessed based on the spatial distribution of the erupted products. Spatially characterizing the detected thermal anomalies allows us to both highlight significant pre-eruptive thermal changes and estimate the areal coverage, length of the erupted products, and the lowest altitude reached by them. We utilize the entire archive of high spatial resolution Sentinel-2 data, which comprises more than 6000 S2-MSI scenes from ten different volcanoes around the world. By employing a “top-down” cascading architecture that integrates two distinct Machine Learning models, a scene classifier (SqueezeNet) and a pixel-based segmentation model (Random Forest), we achieve very high accuracy, specifically 95%. This result comes from overcoming the limitations of the scene-level DL classification model, which compresses the entire spatial and spectral information into one unique label, by relying on the pixel-level Random Forest (RF) model. The use of multiple models allows to create more robust and powerful predictors making this tool suitable for near real time volcanic activity detection. These results demonstrate that the cascading system processes any available S2-MSI image in near-real time, providing a significant contribution to the monitoring, mapping, and characterization of volcanic thermal features worldwide.

How to cite: Cariello, S., Corradino, C., Torrisi, F., and Del Negro, C.: How Machine Learning and Satellite Data Enhance Near-Real Time Detection of Volcanic Activity Worldwide, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-564, https://doi.org/10.5194/egusphere-egu24-564, 2024.

During an explosive eruption, a major hazard to population can be represented by the ejection in the atmosphere of gases and ash, with the consequent creation of volcanic clouds, which can compromise aviation safety. The combined use of a variety of satellite data in different spectral ranges with diverse spatial and temporal resolutions allows us to continuously monitor volcanic ash clouds in an efficient and timely manner. Specifically, the latest generation of high temporal resolution satellite sensors, such as EUMETSAT MSG Spinning Enhanced Visible and InfraRed Imager (SEVIRI) and GOES18 Advanced Baseline Imager (ABI), provide almost continuous radiometric estimates to track the entire evolution of volcanic clouds produced by worldwide volcanoes. Therefore, leveraging the strengths of these satellite sensors, which provide frequent observations at a wide variety of wavelengths, can provide critical information to help understand volcanic processes and extract eruptive products. Nevertheless, the satellite data volume is too large for ad hoc processing and analysis especially when considering daily global-scale observations. Deep learning (DL), a fastest-growing technique of artificial intelligence in remote sensing data analysis applications, has an excellent ability to learn massive, high-dimensional image features and has been widely studied and applied in classification, recognition, and detection tasks involving satellite imagery.

Here, we developed a new DL model, based on Deep Convolutional Neural Networks (CNNs), which exploits a variety of spatiotemporal information mainly coming from geostationary satellite sensors. It is trained on a combination of Thermal Infrared (TIR) bands acquired by MSG-SEVIRI and GOES18-ABI. The proposed model aims to extract complex spectral and spatial patterns autonomously to recognize a volcanic ash cloud. Preliminary capabilities and limitations of this model will be presented here. Thanks to the wide area covered by these satellite sensors, it is possible to apply this model to different volcanoes. Specifically, this model has been applied to the paroxysmal events that occurred at Mt. Etna (Italy) between 2020 and 2020 and at Shishaldin (Alaska, USA) in 2023.

How to cite: Torrisi, F., Corradino, C., Cariello, S., Lopez, T., and Del Negro, C.: Joint use of machine learning and geostationary satellite data for volcanic ash cloud detection, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-822, https://doi.org/10.5194/egusphere-egu24-822, 2024.

Nowadays, near-real time volcano monitoring at global scale is made possible thanks to the thermal infrared sensors on board of several satellite platforms providing accurate estimates of the volcanic thermal emissions. In particular, they are able to provide reliable estimates of the Volcanic Radiative Power (VRP), i.e. the heat radiated during the volcanic activity. In addition, Remote Sensing Data Fusion (RSDF) techniques allow to combine data from multiple satellite sensors to improve the potential values of the single source and to produce a high-quality data representation. Fusion techniques are useful for a variety of applications, ranging from object detection, to object tracking, change detection. In particular, we aim to use them to integrate the different satellite data acquired with different spatial and spectral resolutions to produce fused data that contains more detailed information than each of the data sources. We introduce a novel RSDF algorithm deployed in a Cloud Computing environment to monitor VRP worldwide from multiple multispectral satellite sensors, namely the polar MODIS, SLSTR and VIIRS and the geostationary SEVIRI. The RSDF algorithm demonstrates heightened sensitivity in detecting high-temperature volcanic features and thus VRP monitoring compared to conventional already processed Level 2 products available online. Specifically, the overall accuracy has improved in terms of omitted rate and false detections, reducing from 78% to 5.3% and from 6.5% to 4.7%, respectively. The decision to combine the use of different satellite sensors stems from the need to offer complete continuous monitoring of each volcanological phenomenon, taking advantage of each sensor's own characteristics.

How to cite: Di Bella, G. S., Corradino, C., Cariello, S., Torrisi, F., and Del Negro, C.: A Remote Sensing Data Fusion algorithm for Near Real time monitoring of Volcanic Radiative Power, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1030, https://doi.org/10.5194/egusphere-egu24-1030, 2024.

In a perspective of volcanic hazard assessment, it is fundamental to be able to determine as early as possible whether, where and when the magma that has started to propagate from the storage zone will reach the surface. The propagation phase is generally rapid, lasting from a few hours to a few months, but it induces seismicity and deformation signals recorded by continuous sensors and InSAR data. Furthermore, dynamic numerical models can be used to calculate the trajectory and the velocity of magma propagation as a function of the physical properties of the magma and the crust, and the initial conditions (local stress field and magma reservoir location).

Data assimilation is a method that combines a dynamic model with current and past observations based on error statistics and predicts the future state of the observed system.This method therefore appears to be an appropriate tool for addressing the need to predict the position and timing of a volcanic eruption based on available models and observations.

The particle filter is particularly noteworthy for its ability to handle nonlinear models and non-Gaussian error statistics. This method is based on a representation of the probability density of the dynamic model by a discrete set of model states (particles) and relies on Bayes' theorem.

In order to assess the potential of the particle filter for tracking magma propagation at depth, we implemented this assimilation strategy by considering, in two dimensions, the case of magma propagating beneath a caldera in an extensional stress field. The input parameters of the propagation model are the initial position of the magma at depth, its viscosity and driving pressure, the volume of magma injected, the crustal rigidity, and the local stress field characterized by the balance between tectonic extension and caldera unloading. Surface displacements induced by magma propagation are estimated using an Okada dislocation model. We first validate our assimilation strategy with synthetic data in order to take into account geodetic data recorded on volcanic systems in the future.

How to cite: Zuccali, L., Pinel, V., and Yan, Y.: Assimilation of geodetic data for volcanic hazard assessment in near-real time by means of a particle filter, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2013, https://doi.org/10.5194/egusphere-egu24-2013, 2024.

Soil gas monitoring proves to be a powerful tool for studying volcanoes and their subsurface dynamics. Various gas species are examined as indicators of volcanic activity, with a focus on understanding the processes governing gas emissions. Carbon dioxide (CO₂) is of particular interest due to its significant release during both active volcanic periods and periods of dormancy. Hydrogen (H₂) concentration is also analysed to gain insights into the oxygen fugacity of magmatic gases, a parameter that influences the redox state and the distribution of elements among solid, fused, and gas phases.

To effectively monitor volcanic gases, automated geochemical stations are deployed to simultaneously measure multiple gas species. This study presents a comprehensive analysis of reducing capacity monitoring data collected at Stromboli during the 2021-2023 time window. This station records data on molecular hydrogen (H₂) concentration, CO₂ flux, CO₂ concentration, soil water content, atmospheric pressure, temperature, and relative humidity.

The dataset underwent normalization of variables and filtering processes of spurious data to facilitate Principal Component Analysis (PCA), allowing for the identification of linear dependencies among the variables. The results of the PCA reveal insights into the relationships of the monitored parameters, shedding light on potential factors influencing volcanic activity.

These data reveal variations in reducing capacity and soil CO₂ flux that correlate with changes in Stromboli's eruptive activity. The delay in CO₂ variations following shifts in reducing capacity is attributed to the advective-diffusive transport of gases through the volcanic rock formations. This research contributes to a deeper understanding of the volcanic degassing of Stromboli and shows the efficacy of combining multiple regression analysis methods for filtering geochemical datasets. This research enhances our understanding of volcanic behaviour and aids in volcano monitoring efforts.

How to cite: Messina, M., Di Martino, R., and Paonita, A.: Advancing volcano surveillance through soil gas monitoring: insights from Stromboli Island, Italy., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5254, https://doi.org/10.5194/egusphere-egu24-5254, 2024.

Satellite-based thermal remote sensing is a useful tool for monitoring volcanoes. It involves detecting thermal anomalies, called 'hotspots,' and calculating the radiative energy emitted by volcanic activity. Various volcanic hot spot detection algorithms already exist in the literature. However, every algorithm has its advantages and disadvantages, as they are limited depending on the tradeoffs made during algorithm development, the sensor used for aquisition, and  the geometry of acquisition. Depending on the algorithm used, different results are obtained from the same data and, hence, different interpretations can be made in terms of, e.g., energy emitted, effusion rates, and eruption duration. In the present work, we aim at creating a new hotspot detection algorithm using MODIS and VIIRS imagery, which allows us to efficiently look at the dynamics of thermal emissions coming from persistent lava lakes, i.e., bassins of lava maintained molten through thermal convection and outgassing. We investigate the applicability of sensor fusion ideas, using multiple bands, and incorporating cloud cover information. We expect that by combining all available data the robustness of the detection process will increase.

How to cite: Rolain, S., Peumans, D., and Smets, B.: Thermal remote sensing of lava lakes: a comparison of approaches, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5284, https://doi.org/10.5194/egusphere-egu24-5284, 2024.

Ultraviolet (UV) cameras provide advantages in spatial and temporal resolution for the measurement of sulphur dioxide (SO₂) emission rates, when compared to the more common differential optical absorption spectroscopy (DOAS) instruments. Up to this point, however, most instances of UV camera usage are restricted to discrete campaigns, rather than permanent installations. Notable exceptions include Stromboli, Etna, and Kīlauea (2013-2018). The reasons for this are largely related to cost, with commercially available UV cameras often being prohibitively expensive, and also to the challenges associated with developing robust algorithms for automated retrieval of emission rates from the collected imagery. The PiCam, developed at the University of Sheffield, is a new UV camera system designed for permanent installation in the field and provides automatic measurements of SO₂ emission rates (for full details see Wilkes et al., 2023, doi: 10.3389/feart.2023.1088992). The PiCam is currently installed at six volcanoes globally: Kīlauea (USA), Cotopaxi and Reventador (Ecuador), Lascar and Lastarria (Chile), and Merapi (Indonesia). Each location has unique requirements and setup, for instance the UV cameras at Reventador and Lastarria operate autonomously but require user visits for data download. Those at Kīlauea and Merapi are integrated within existing telemetry solutions, while at Lascar and Cotopaxi we have begun to integrate Starlink satellite data telemetry alongside our cameras. We present some early results from measurements at Kīlauea where emission rates are compared to more traditional measurements using DOAS. Since the PiCam installation in 2022, the system has recorded multiple styles of activity, from low-level degassing during periods of quiescence to syn-eruptive emissions exceeding 10,000 t/d. Our work highlights continuing challenges of processing of UV camera data, where automated protocols and real-time processing for reliable emission rate retrievals are still in their infancy.

How to cite: Pering, T., Wilkes, T., Nadeau, P., Hidalgo, S., Aguilera, F., Layana, S., Aisyah, N., Humaida, H., Sakti, A., and Kern, C.: Application of a low-cost ultraviolet camera solution for permanent sulphur dioxide measurements at six volcanoes worldwide., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5555, https://doi.org/10.5194/egusphere-egu24-5555, 2024.

Identifying changes on volcano unrest condition and tracking the evolution of eruptive activity are fundamental for volcanic surveillance and monitoring. Under this perspective, magmatic gases play a key role; therefore, monitoring changes in volcanic plume composition is essential.

Between and during each eruption large systematic variations in the volcanic plume composition can be observed on Mt. Etna. Specifically, gas emissions implicitly contain critical information on the state of the volcano in terms of magma dynamics in the plumbing system.

New technologies have improved the ability to identify gas emissions through remote sensing. Since March 2000, the remote sensing group of INGV-OE Catania has regularly measured the volcanic gas in the plume emitted from Mt. Etna by solar occultation FTIR on average 2-3 acquisitions per week. These measurements allow to quantify the column amounts of SO2, HCl and HF in the path between the instrument and the Sun, and obtain SO2/HCl, HCl/HF and SO2/HF ratios

Extracting meaningful information and gaining new insights from 23 years data collection is a challenging task. Here, we perform a data driven investigation exploiting the relationships between volcanic activity and volcanic gas emissions from 2000 to 2023 period. A general decrease in SO2/HCl ratio, due to an increase in the halogens emission rate (HCl and HF), is observed during eruptions that results closely connected with the central conduit of Etna (e.g. 2004 and 2008).This evidence suggests that magma residence time is required to sustain an efficient halogen degassing via summit craters and the lava flow draining gradually led to a break in the magma convective overturn within a shallow reservoir. Whereas, eruptions not connected with the central conduit of Etna (e.g. 2002-03) are preceded by several weeks of elevated SO2/HCl ratio, produced by a relative reduction in HCl emission consistent with inefficient magma circulation that partially inhibits ascent of magma thought the central conduit of Etna enhances the probability of lateral eruptions.

How to cite: La Spina, A., Burton, M., Bonfanti, P., and Murè, F.: Unveiling magma dynamics behind geochemical data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19259, https://doi.org/10.5194/egusphere-egu24-19259, 2024.

Tephra fallout is a common feature of different styles of eruptions and is the most widespread of volcanic hazards. Fallouts containing substantial amounts of fine-grained particles pose a concern for human health since exposure to respirable PM (sub-10 μm, i.e., PM₁₀) is associated with adverse health effects. The products of effusive or poorly explosive events (e.g., lava fountains, strombolian eruptions) and their impacts are less researched as they typically generate rather coarse-grained deposits. Yet, reworking of tephra deposits by wind, traffic or other human activities can potentially alter the initial grain size distribution of a deposit and generate finer material. Remobilisation of such reworked deposits can affect ambient air quality (i.e., PM₁₀ levels), thus leading to an increased exposure hazard.

In this study, we conducted in situ experiments on the slopes of Etna volcano, Italy, which frequently covers neighbouring urban areas with coarse-grained basaltic tephra. We aimed to understand the changes in tephra grain size distribution and airborne PM₁₀ concentration associated with vehicular activity. For this purpose, we drove a small SUV-type car over an area of a road that we covered with tephra, and investigated the outcomes as a function of 1) the number of car passages (between 10 and 70), 2) the starting thickness of the tephra deposit (between 2 and 10 mm) and 3) vehicle speed (between 20 and 50 km/h). The results show that the grain size of the original tephra deposit decreases with the number of car passages, most notably with higher vehicle velocity (50 km/h) and increasing deposit thickness. Airborne PM₁₀ increased with a higher number of car passages, but also with increased tephra thickness and increased vehicle speed. Our observations have important implications for the management of tephra fallouts in urban areas. We have shown that vehicles will change the grain size distribution of basaltic ash by comminution so local communities can expect that, after an eruption, concentrations of PM₁₀ may increase with time and affect exposures close to roads.

How to cite: Tomašek, I., Tournigand, P.-Y., Andronico, D., Eychenne, J., Horwell, C., Kueppers, U., Taddeucci, J., Claeys, P., and Kervyn, M.: Tephra comminution by vehicles and resuspension of volcanic ash: impact on ambient air quality in urban areas, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16876, https://doi.org/10.5194/egusphere-egu24-16876, 2024.

Volcanic eruptions are enigmatic, not in the least because of volcanic lightning during eruptions, which is still not fully understood. In an effort to better constrain conditions under which electrical discharges occur, we use data from a multicomponent geophysical network installed in 2019 at Sakurajima volcano, Japan. The network includes one vertically scanning Doppler radar to resolve the internal structure of the eruption column at a low temporal resolution, and two fixed Doppler radar systems to resolve the dynamics of the eruption near the vent at high temporal resolution. Eruption onset times were determined using a total of 5 infrasound stations. Electrical discharges were detected by three electric field mills (EFM), a thunderstorm detector from Biral (BTD), a lightning mapping array sensor (LMA) as well as a fast antenna (FA). In addition, meteorological conditions are monitored by a local weather station and eruptions were recorded by two different cameras.

Sakurajima volcano is known to exhibit frequent complex eruptions/explosions often exhibiting a sequence of single pulses at intervals down to a few minutes. From our network, we generated an eruption catalog spanning 7 months of activity between June and Dec 2019. Out of this catalog, we select representative examples of multiple pulse events of different strength in terms of erupted mass, plume height, and eruption velocities. For each pulse within such a sequence, we analyzed the electrical activity as well as extracted eruption velocities, eruption volumes, plume heights, and eruption onsets among others. We find that the timing of the electrical discharges is correlated with the onsets of single pulses within a sequence. Furthermore, we can detect changes in the electrical activity ranging from times of CRF signals from small streamer discharges up to volcanic lightning while being able to track the erupted mass and plume extend at the same time. While electrical activity can increase together with eruptive strength, we also observe quite “strong” eruptions with little detectable discharges. On the other hand, less energetic eruptions are observed to produce electrical discharges, especially when the plume of a pulse rises through pre-existing ash clouds in the crater region.

How to cite: Geisler, A., Hort, M., Behnke, S., Edens, H., Yamada, T., Iguchi, M., and Nakamichi, H.: On the occurrence of volcanic lightning during vulcanian eruptions at Sakurajima volcano, Japan, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18729, https://doi.org/10.5194/egusphere-egu24-18729, 2024.

Vulcano Island, an active closed conduit volcano in south Sicily, hosts from 2018 the Vulcano International Training Summer School of Geochemistry. The basic idea is to share scientific knowledge and experiences in a multi-aspect geochemical community, using local resources with a low-cost organization in order to allow as many students as possible access to the school. The Vulcano International Training Summer School of Geochemistry is addressed to students, graduate students, senior graduate students and  fellows, coming not only from Italy but also from European and non-European countries, aiming to bring together the next generation of researchers active in studies concerning the geochemistry and the budget of volcanic gases. The main aim of the school is to bring the next generation of researchers, active in studies concerning geochemistry in volcanic environments, and to introduce them with innovative direct sampling and remote sensing techniques. Furthermore, it gives young scientists an opportunity to experiment and evaluate new protocols and techniques to be used on volcanic fluid emissions, covering a broad variety of methods. The teaching approach includes practical demonstrations and field applications, conducted by teachers of international level. Students therefore have the opportunity to learn by practicing sampling techniques. Practical exercises include: direct sampling of high-temperature fumaroles and plume measurement techniques in the fumarolic crater field; direct sampling of low-temperature gas emissions and submarine emissions; measurement of diffuse soil gas fluxes of endogenous gases at Levante Bay Beach; sampling of groundwater and dissolved gases, and measurements of physical-chemical parameters by using multiparametric probes, in thermal waters in the Vulcano village. Also this year the school will take place from 17 to 21 June 2024.

How to cite: Pecoraino, G., Tassi, F., Calabrese, S., Cabassi, J., Vaselli, O., Rouwet, D., Tamburello, G., Capecchiacci, F., and Venturi, S.: Vulcano International training Summer School of Geochemistry , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19253, https://doi.org/10.5194/egusphere-egu24-19253, 2024.

The origin of volcanic lightning is still a matter of intense research because it may complement other monitoring techniques in detecting volcanic eruptions at remote or not well monitored volcanoes. Various methods have been used to detect electrical discharges over the last decades and here we compare four different methods for the detection of electrical discharges which were operated in late 2019 at Sakurajima volcano. The electrical activity was observed by an electric field mill (EFM, which measures the static electrical field at 10 S/s (samples/second)), a thunderstorm detector from Biral (BTD, also measures the static electrical field at 100 S/s), as well as a lightning mapping array station (LMA, measuring VHF electromagnetic radiation) and a fast antenna FA (time constant τ = 10^-4 s⁻¹, 180MS/s). We developed special algorithms to identify electrical discharges in the EFM and BTD data sets because those were the most continuous data sets available to us. From a period of high volcanic activity during which also data from the other instruments were available, we then picked several eruptions characterized by different types of electrical discharges. We analyzed the number of discharges detected by each instrument as well as how well the amplitudes (electrical field changes) are resolved by the different instruments. We find good correlations of detected signal strength between the FA and EFM data, the correlation between these data and the BTD is especially for larger events not as perfect. In order to analyze this problem, we use the FA measurements to downsample and synthetically reconstruct the BTD data set, showing deviations for larger amplitudes as well. Overall, the instruments show good agreement on detecting electrical discharges within volcanic eruptions with differences on the detection of „assumable weaker“ signals. We highlight the advantages and disadvantages of the different techniques in terms of detecting electrical discharges during volcanic eruptions. We finish discussing how the different instruments can be used to determine electric field changes as well as further investigate the type of electrical discharges detected.

How to cite: Hort, M., Geisler, A., Heck, F., Behnke, S., Edens, H., and Iguchi, M.: Detecting volcanic lightning with different methods: a comparison, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-21214, https://doi.org/10.5194/egusphere-egu24-21214, 2024.

Geo-electric methods such as Electrical Resistivity Tomography (ERT) and Induced Polarization (IP) have become increasingly important in the characterization of volcanic and geothermal systems. The methods rely on the electrical properties of the subsurface. In volcanic settings, the main influences are temperature, gas content (i.e. saturation), mineralizations, and the presence of alteration clays.

The ERupT project aims to assess the suitability of ERT to visualize the dynamics in hydrothermal systems. With the long-term aim of improving hazard assessment associated with phreatic/hydrothermal eruptions. In that context, a semi-permanent ERT setup was installed at the Reykjanes geothermal area in Iceland. In October 2022, the monitoring system was installed on-site, automatically measuring one profile per day, with currently 15 months of uninterrupted data, except for 5 days due to technical issues. The profile is 355 meters long and has a depth of investigation of 30 to 50 meters.

The 2021 eruption at Fagradalsfjall has marked a new age of volcanism in the Reykjanes peninsula, 2023 was marked by two eruptions. The first eruption happened in the Fagradalsfjall system on July 10^th. The second eruption started on 18 December, North of Grindavik, in the Svartsengi system. The eruption sites are located at respectively 30 and 10 km from the field site. Although the ERT system is located at a considerable distance from the eruption sites and the investigation depth is quite shallow, we observed signals possibly related to both eruptions and accompanying unrest, manifesting as a significant increase in resistance (figure1).

The first peak in resistance happened between 19 and 26 June with an increase of more than 100%, shortly after that, the Fagradalsfjall system erupted. A second peak is observed at the end of August, here the increase happens slowly, as opposed to the sudden peak in June, with a steady increase starting on August 8^th, reaching a peak on August 30^th. Contradictory to the first example, no immediate eruption occurred after this second peak.

In this context, an increase in resistance is likely caused by a drop in saturation due to high gas levels, which can be caused by magma degassing during the uprise. This raises the question of the time difference between the peaks and the subsequent eruptions, possible factors are the difference in morphological context (sill vs dyke intrusion) and preferential flowpaths relative to our monitoring site.  It should also be noted that this behavior is not observed in all data points, hence advanced processing is needed combined with interpretation using data from other methods. Tremor and soil temperature data are available along the ERT profile, together with one C0₂ sensor in the center of the profile.

To our knowledge, this is the first time that ERT has been used for daily monitoring of a volcanic system. With joint interpretation to deduce the signal origin, we believe that ERT can be a valuable addition to volcanic monitoring networks.

Figure 1: Resistance evolution in one measurement point, eruptions are indicated by the red lines.

How to cite: Vanhooren, L., Hermans, T., and Caudron, C.: Is ERT suited for the continuous monitoring of volcanic systems? A case study during the 2023 unrest of the Reykjanes system., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15372, https://doi.org/10.5194/egusphere-egu24-15372, 2024.