In the 1960's a peak in the seismic amplitude spectra around 26 s was discovered and detected on stations worldwide. The source was located in the Gulf of Guinea, with approximate coordinates (0,0), and was believed to be generated continuously. A source with similar spectral characteristics was discovered near the Vanuatu Islands, at nearly the antipodal location of the Gulf of Guinea source. Since it was located close to the volcanoes in Vanuatu, this source is commonly attributed to magmatic processes. The physical cause of the 26 s microseism, however, remains unclear.

We investigate the source location and evolution of the 26 s microseim using data from permanent broadband stations in Germany, France and Algeria and temporary arrays in Morocco, Cameroon and Botswana for spectral analysis and 3-C beamforming to get closer to resolving the source mechanism responsible for this enigmatic signal. We find that the signal modulates over time and is not always detectable, but occasionally it becomes so energetic it can be observed on stations worldwide. Such a burst can last for hours or days. The signal is visible on stations globally approximately 30 percent of the time. Our beamforming analysis confirms that the source is located in the Gulf of Guinea, as shown in previous studies, and that the location is temporally stable. Whenever the signal is detectable, both Love and Rayleigh waves are generated. We discover a spectral glide effect associated with the bursts, that so far has not been reported in the literature. 

The spectral glides last for about two days and are observed on stations globally. Although at higher frequencies, very long period tremors and gliding tremors are also observed on volcanoes as Redoubt in Alaska and Arenal in Costa Rica, suggesting that the origin of the 26 s tremor is also volcanic. However, there is no reported volcanic activity in the area where the source appears to be located.

How to cite: Bruland, C., Mader, S., and Hadziioannou, C.: Observation of gliding tremors from the Gulf of Guinea might help solve over 50 year old mystery, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-3447, https://doi.org/10.5194/egusphere-egu21-3447, 2021.

The receiver function analysis (RF) is a commonly used and well-established method to investigate crustal and mantle structures, removing the source, ray-path and instrument signatures. RF gives the unique signature of sharp seismic discontinuities and information about P and S wave velocities beneath a seismic station. In particular, using the direct P wave as a reference arrival time, and the relative arrival time of P-to-S (Ps) conversions and multiple reflections allow constraining the principal crustal structures and studying the effects of dipping interfaces and crustal layering.

We have applied RF analysis to the active volcanic islands of Tenerife and La Palma (Canary Islands). In recent years, both islands have increased their seismic activity and showed variation in geochemical parameters attributed to a magmatic-hydrothermal activity. Previous studies evidenced in La Palma and Tenerife a seismic Moho depth at 14 km and 12 and 15 km, respectively, but it is not clear because there are some others discontinuities under the stations (Lodge et al., 2012). Other RF studies indicated a depth of seismic Moho discontinuity between 16 and 30 km beneath the eastern islands to 11-15 km under the western isles, observing a thinning of the crust towards the west (Martinez-Arévalo et al., 2013). 

We processed 313 teleseisms recorded by 17 stations for Tenerife and 252 teleseisms recorded by six stations for La Palma. Since the receiver functions display a significant complexity, as expected in oceanic volcanic islands, we applied a transdimensional inversion approach to image the 1D velocity structure beneath each station. We observe at least three discontinuities related with the oceanic crust and the overlying volcanic rocks layer. We compare the retrieved crustal structure with the seismicity recorded in recent years, showing how earthquakes have a radically different origin on these two islands. While in Tenerife they seem to be related to the dynamics of a shallow hydrothermal system, in La Palma they are related to magmatic intrusions in the upper mantle beneath the island.

References

Lodge, A., Nippress, S. E. J., Rietbrock, A., García-Yeguas, A., & Ibáñez, J. M. (2012). Evidence for magmatic underplating and partial melt beneath the Canary Islands derived using teleseismic receiver functions. Physics of the Earth and Planetary Interiors, 212, 44-54.

Martinez-Arevalo, C., de Lis Mancilla, F., Helffrich, G., & Garcia, A. (2013). Seismic evidence of a regional sublithospheric low velocity layer beneath the Canary Islands. Tectonophysics, 608, 586-599.

How to cite: Ortega, V., D'Auria, L., Cabrera-Pérez, I., Barrancos, J., Padilla, G. D., and Pérez, N. M.: Crustal structure of La Palma and Tenerife (Canary Islands) from receiver function analysis., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-11140, https://doi.org/10.5194/egusphere-egu21-11140, 2021.

The Klyuchevskoy Volcanic Group (KVG) located in Kamchatka, Russia is one of the World’s most active clusters of subduction volcanoes. In order to investigate its structure and very intense seismovolcanic activity, an international collaboration designed the KISS experiment operating a dense temporary seismic network between August 2015 and July 2016. During this period, the main volcano of KVG – Kyuchevskoy entered into eruption in the spring 2016. The preparation and eruptive periods have been characterized by a large number of volcanic earthquakes and tremors.

We applied in this study three cross-correlations network-based methods to detect and locate seismovolcanic tremor sources. From these three methods we extract simple 1D functions: spectral width (averaging in the 0.5-5 Hz frequency band the width of the network covariance matrix eigenvalue distribution), network response function (performing the 3D back-projection of the inter-station cross-correlations) and correlation coefficient function (averaging correlation coefficient functions computed at single station that characterize the stability in time of the single-station intercomponent cross‐correlation function). The simultaneous application of these network features allowed us to classify the wavefield recorded by the dense seismic network. We then computed inter-station cross-correlations extracted from the first eigenvector filtered covariance matrix and generate time series of 3D spatial likelihood functions. Using output of our classification approach, we stack over time these 3D spatial likelihood functions for time windows containing tremor and we finally obtain a 3D Density Likelihood function imaging the seismovolcanic tremor sources distribution within KVG.

The addition of the temporary seismic stations from the KISS network greatly increased our detection and location resolution and thus allowed us to refine our knowledge about seismovolcanic tremor at KVG. Our results highlight a large distribution of tremor sources connecting different volcanoes and different depth levels. Most of tremor sources are located below the Klyuchevskoy volcano in a narrow zone vertically extended from the surface to the crust-mantle boundary and exhibit a highly intermittent behavior characterized by burst of activities and rapid upward and downward migrations between deep and shallow locations. Several tremor sources are also located along a SW-NE structure extending from Tolbachik to Klyuchevskoy volcanoes. We thus image the near-vertical quasi-open main conduit connecting the deep magmatic reservoir to Klyuchevskoy volcano in which very rapid pressure transfers might occur as well as a possible secondary conduit that links the marginal part of the deep reservoir to the Tolbachik volcanic system in which the system overpressure may be sometimes evacuated.

How to cite: Journeau, C., Shapiro, N., Seydoux, L., Soubestre, J., Koulakov, I., Jakovlev, A., Abkadyrov, I., Gordeev, E., Chebrov, D., Sens-Schönfelder, C., Luehr, B., Tong, F., Farge, G., and Jaupart, C.: Imaging seismovolcanic tremor sources distribution with seismic network-based methods reveals fluid pressure pathways within Klyuchevskoy Volcanic Group magmatic system, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-7871, https://doi.org/10.5194/egusphere-egu21-7871, 2021.

We analyze data from one tiltmeter and twelve broadband seismic stations recorded at the beginning of the 2018 Kilauea eruption, to provide an integrated view of distinct tremor sources that preceded and accompanied this eruption. Studying the beginning of the eruption is challenging because of the diversity and complexity of signals that were recorded during this phase. But such undertaking represents a key aspect for understanding the dynamics of the different processes that took place at the start of the lava lake withdrawal on May 2 and during the twelve major collapses that occurred in Halema‘uma‘u Crater through May 26. The application of a network-based method to automatically detect and locate seismic tremor, combined with physical modeling of the underlying source processes, enables a characterization of these tremor sources in unprecedented detail.

Our analyses document one tremor source active during the period preceding the eruption, which is attributed to the quasi-steady radiation from a shallow hydrothermal system located at the south-southwest edge of Halema‘uma‘u Crater. These analyses further document two newly described sequences of gliding tremor. The first sequence is attributed to progressive jerky intrusions of a rock piston into a shallow hydrothermal reservoir between May 7 and May 17. The second sequence is attributed to the gradual degassing of a bubbly magma within an east striking dike below Halema‘uma‘u Crater, impacted by repeated roof collapses, and resulting in a quasi to totally degassed magma by May 26.

How to cite: Soubestre, J., Chouet, B., and Dawson, P.: Characterizing volcanic tremor sources associated with collapses in Halema‘uma‘u Crater at the beginning of the 2018 Kilauea eruption, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-10723, https://doi.org/10.5194/egusphere-egu21-10723, 2021.

We investigate the geometry of the metamorphic basement of the Santorini volcanic island using ambient noise data to determine the pre-Alpine/pre-volcanic bedrock structure. The geometry of pre-volcanic Santorini is important in order to constrain the recent volcanic history of the island and also to study the site-effect of the volcanic formations on seismic motions. Santorini is the most active volcano of the Southern Aegean Volcanic Arc and is the southernmost island of the Cyclades islands metamorphic complex. As a result, the volcanic material that has accumulated during the last 600+ Kyrs has been superimposed on the pre-volcanic Santorini (Cycladic) island. To map the thickness of volcanic material, we have performed a large number (>200) of single-station noise measurements in the Santorini area.  Measurements were mainly performed using conventional acquisition systems (Guralp-40T 30sec seismometer and Reftek-130A digitizer). We also employed additional single-station noise data from several previous studies (Dimitriadis et al. 2006, PROTEUS Project 2015), as well as permanent stations from the Hellenic Seismological Network in the same region. HVSR curves were calculated using single-station noise data and were used to estimate the fundamental frequency, f₀, as well as the corresponding maximum HVSR amplitude, A₀^HVSR. The majority of HVSR curves showed prominent peaks (A₀^HVSR locally larger than 7-8), indicating a clear impedance contrast between volcanics and metamorphic formations. To map the bedrock depth, we estimated the thickness of the upper volcanic formations using the quarter-wavelength approximation for each site. For this assessment, the average shear-wave velocity (Vs) of the volcanic formations was estimated from the inversion of several passive ambient noise array data, as well as additional constraints from selected MASW measurements. Where possible, the reliability of the spatial variation of volcanic formation thickness was checked with independent geological information. Using the digital elevation model and the volcanic formation thickness for each site of the single-station noise data, we estimated the spatial distribution of the pre-Alpine, metamorphic bedrock depth. The resulting geometry of the pre-volcanic Santorini island shows very deep basins (now filled with volcanic formations) around the pre-Alpine bedrock outcrop in the southern part of Santorini (Profitis Ilias), increasing to 100+ meters in the Kamari-Perissa basin area (southeastern Santorini) and to more than 400+ in the central (Fira-Imerovigli) and the north Santorini areas (Oia), in agreement with recent larger-scale tomographic results (Heath et al., 2019). The results are also in very good agreement with the pre-Alpine bedrock geometry independently inferred from gravity data inversion (Tzanis et al., 2019.)

This research is co-financed by Greece and the European Union (European Social Fund- ESF) through the Operational Programme «Human Resources Development, Education and Lifelong Learning» in the context of the project “Strengthening Human Resources Research Potential via Doctorate Research” (MIS-5000432), implemented by the State Scholarships Foundation (ΙΚΥ), the Hellenic Foundation for Research and Innovation (HFRI) under the “First Call for HFRI Research Projects to support Faculty members and Researchers and the procurement of high-cost research equipment grant” (Project Number: 2924) and the Institute for the Study and Monitoring Of the SAntorini Volcano (ISMOSAV).

How to cite: Chatzis, N., Papazachos, C., Theodulidis, N., Hatzidimitriou, P., Anthymidis, M., Vougioukalakis, G., Panagiotopoulos, D., Hooft, E., Heath, B., Toomey, D., Paulatto, M., Morgan, J., and Warner, M.: Investigation of the bedrock (metamorphic basement) geometry of Santorini island using single-station ambient noise data , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-7968, https://doi.org/10.5194/egusphere-egu21-7968, 2021.

The island of Gran Canaria is located in the Canarian Archipelago, with an area of 1560 km² and a maximum altitude of 1956 m.a.s.l., being the third island of the archipelago in terms of extension and altitude. The island has two very well differentiated geological domains: the southwest domain or Paleo-Canarias, which is the geologically oldest part, and the northeast domain or Neo-Canarias, where are located the vents of the most recent Holocene eruptions. This volcanic island hosted Holocene eruptions. Therefore, apart from being affected by volcanic risk, it potentially hosts geothermal resources that could be exploited to increase the percentage of renewable energy in the Canary Islands.

The main objective of this work is to use Ambient Noise Tomography (ANT) for retrieving a high-resolution seismic velocity model of the first few kilometres of the crust, to improve local earthquake location and detect anomalies potentially related to active geothermal reservoirs. Currently, the 1-D velocity model of the island does not allow a correct determination of the hypocenters, being unable to take into account the substantial horizontal velocity contrasts correctly.

To realize the ANT, we deployed 28 temporary broadband seismic stations in two phases. Each campaign lasted at least one month. We also exploited data recorded by the permanent seismic network Red Sísmica Canaria (C7) operated by INVOLCAN. After applying standard data processing to retrieve Green’s functions from ambient noise cross-correlations, we retrieved the dispersion curves using the FTAN (Frequency Time ANalysis) technique. The inversion of dispersion curves to obtain group velocity maps was realized using a novel non-linear multiscale tomographic approach (MAnGOSTA, Multiscale Ambient NOiSe TomogrAphy). The forward modelling of surface waves traveltimes was implemented using a shortest-path algorithm that allows the topography to be taken into account. The MANgOSTA method consists of successive non-linear inversion steps on progressively finer grids. This technique allows retrieving 2-D group velocity models in the presence of substantial velocity contrasts with up to 100% of the relative variation. Then, we performed a depth inversion of the Rayleigh wave dispersion curves using a transdimensional Bayesian formulation. The final result is a 3-D model of P- and S-wave velocities of the island. The preliminary results show the presence of a low-velocity zone in the eastern part of the island that coincides spatially with anomalies observed in previous geophysical and geochemical studies and which could be related to actual or fossil geothermal reservoirs. Furthermore, the model shows the presence of high-velocity anomalies that are associated with the mafic core of the island.

How to cite: Cabrera Pérez, I., Soubestre, J., D'Auria, L., Cervigón-Tomico, G., Martínez van Dorth, D., Barrancos, J., Padilla, G. D., and Pérez, N. M.: Ambient noise tomography of Gran Canaria island (Canary Islands), EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-10957, https://doi.org/10.5194/egusphere-egu21-10957, 2021.

Understanding important characteristics of Yellowstone's magmatic system such as the melt fraction, composition, and geometric organization of melt are critical for improving our knowledge of volcanic processes and assessing the potential for future eruptions.  While previous tomographic images have provided much insight into the magmatic system, imaging results are complicated by an incomplete understanding of how large crustal magmatic systems affect seismic waveforms. In particular, tomographic studies based on asymptotic methods may underestimate the seismic wave speed anomaly of the magma reservoir because first arriving energy may be diffracted around strong low wave speed anomalies. Here, we present a high-resolution shear wave speed model of Yellowstone’s crust and uppermost mantle structure, based on the most up to date dataset of ambient noise correlation functions from broadband stations deployed in the Yellowstone region over the past two decades. This model serves as the starting point for an adjoint inversion, which has potential to improve resolution by incorporating more accurate sensitivity kernels based on realistic wave propagation physics. We discuss our adjoint tomography methodology and present the first model iterations. Continued iterations promise to sharpen features in the model which can provide new inferences into the present state of Yellowstone’s magmatic system.

How to cite: Maguire, R., Chen, M., Schmandt, B., Jiang, C., Wilgus, J., and Li, J.: Ambient noise waveform imaging of Yellowstone's magmatic system, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-13437, https://doi.org/10.5194/egusphere-egu21-13437, 2021.

In May 2018, a seismic crisis started in the Comoros archipelago, East of Mayotte, which was widely felt on the Island. The related discovery of a new, 800-m high, submarine edifice 50 km East of Mayotte showed that the seismicity was caused by the birth of a volcano. The eruption is still on going at the time of writing and has sparked a large interest in the scientific community.

The seismicity is still active and is being continuously monitored thanks to several seismic stations installed on the island of Mayotte. The oceanographic campaigns that were carried out since the beginning of the crisis deployed a number of ocean bottom seismometers directly above the seismicity, to accurately understand the crisis and particularly its location. A new technique of automatic detection based on Machine Learning enabled to considerably increase the number of earthquakes that can be used to constrain the extent of the seismicity. Furthermore, the development of a new velocity model for the region allowed a precise location of these earthquakes.

These new developments permitted to reconstruct the seismicity evolution during two years of this seismic crisis and to complete the seismicity map associated with the new seismic activity. These results provide more details on the active structures to study the evolution in time as well as their precise spacial variations, allowing the analysis of the daily-to-yearly timescales of this unprecedented eruption. This is crucial to understand the dynamics of the volcanic and magmatic processes beneath Mayotte island. Linking these spatial and time variations with the real-time data, as well as the deformation and petrology evolutions, will provide crucial details on the dynamics of submarine eruptions.

How to cite: Lavayssière, A. and Retailleau, L.: Short- and long-term evolution of the seismicity associated with the New Volcanic Edifice offshore Mayotte island, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-2769, https://doi.org/10.5194/egusphere-egu21-2769, 2021.

The EUROVOLC project aims to promote an integrated and harmonised European volcanological community, with one of its main themes focusing on understanding sub-surface processes. Early identification of magma moving towards the surface is very important for the mitigation of risks from volcanic hazards, and joint research activities within the project aim to develop and improve schemes for detecting pre-eruptive unrest. Volcanic tremor is a sustained seismic signal that is often associated with such volcanic unrest, and has been linked to the movement of magmatic fluids in the subsurface. However, signals with similar spectral content can be generated by other surface processes such as flooding, rockfalls or lahars. Hence, one of the best ways of distinguishing between different possible mechanisms for generating tremor is by tracking the location of its source, which is also important for mitigating volcanic risk. Due to its emergent nature, tremor cannot be located using travel-time based methods, and therefore alternatives such as amplitude-based techniques or array analysis must be used. Dense, small-aperture arrays are particularly suited for analyzing volcanic tremor, yet costs associated with installation and maintenance have meant few long-term or permanent seismic arrays in use for routine monitoring.

Given the potential for wider usage of arrays, this work presents a freely available python-based software tool, developed as part of the EUROVOLC project, that uses array data and array processing techniques to analyze and locate volcanic tremor signals. RETREAT utilizes existing routines from the open-source ObsPy framework to carry out analysis of array data in real-time and performs either f-k (frequency-wavenumber) analysis, or alternatively Least-Squares beamforming, to calculate the backazimuth and slowness in overlapping time windows, which can help track the location of volcanic tremor sources. A graphical, or web-based, interface is used to configure a set of input parameters, before fetching chunks of waveform data and performing the array analysis. On each update the tool returns several plots, including timeseries of the backazimuth and slowness, a polar representation of the relative power and a map of the array with the dominant backazimuth overlaid.

The tool has been tested using real-time seismic data from the small-aperture SPITS array in Spitsbergen, as well as on data from a small aperture seismic array deployed during the 2014 eruption of Bárðarbunga volcano, Iceland. Although designed specifically for seismic array data (with a particular focus on volcanic tremor), RETREAT can also be used with infrasound sensors and has been successfully tested on infrasonic array data of explosive activity recorded at Mt. Etna, Italy, in 2019.

Although RETREAT has been designed for deployment as part of volcano monitoring systems and provides the ability to track tremor sources in real-time, it also has the capability to analyse existing datasets for testing, comparison and research purposes. However, RETREAT is primarily intended for use in real-time monitoring settings and it is hoped that it will facilitate wider use of arrays in tracking volcanic tremor or infrasonic sources in real-time, thereby enhancing monitoring capabilities.

How to cite: Smith, P. and Bean, C.: A REal-time TREmor Analysis Tool for seismic and infrasonic arrays, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15902, https://doi.org/10.5194/egusphere-egu21-15902, 2021.

The topic of my work is a seismic tomography which has as object the investigation of Southern Tyrrhenian. This tomography has been obtained by means of inversion of teleseismic data to investigate subduction zones in the Southern Tyrrhenian oceanic back-arc basin. The subducting lithosphere has been mostly consumed along the Tyrrhenian-Apennine system has been consumed with the exception of the Calabrian arc sector. This kind of inversion could provide a good resolution to depth of 500-600 km, whereas previous local tomographies of Southern Tyrrhenian show results to depth of 250-300 km. The adopted database consists of 1929 teleseisms  recorded in period 1990-2012 by 122 southern Italian seismic station directly connected to ISC (International Seismological Centre). The software FMTT was employed for the inversion of these arrival times. I have implemented a grid of 0-500 km in depth, 7°E-20°E in longitude and 35°-48° in latitude, with a grid spacing of 50 km in depth, 0.8 degrees in longitude and 0.4 degrees in latitude. I have made 10 horizontal sections of final model from 50 km of depth to 500 km of depth, with an interval of 50 km of depth from each other. I have made 8 vertical sections, 4 NS vertical sections at fixed longitude and 4 WE vertical sections at fixed latitude. Finally, I have made 3 transversal sections. Summarising, the horizontal sections show an evolution of the high velocity body that represents the Ionian slab. It is visible both at depth of 50 km and at depth of 100 km, beneath the Calabrian arc and extends to northern Sicily beneath the Aeolian arc with a maximum of 0.6-0.8 km/s. At depth of 250 km, the tomography evidences a sort of “transition” due to the absence of the Southern Tyrrhenian HVA and the occurrence of a low velocity region with maximum of -0.5 km/s scattered between the Aeolian Islands and Calabria. In the depth interval from 250 km to 400 km, there are two impressive high velocity areas in northern Sicily and along southern Campania with a value of 0.3 km/s, separated by a low velocity area (LVA) along the Calabrian arc and the Aeolian Islands in the range [0.4 ; 0.6] km/s. Extensions of HVAs and LVAs previously mentioned have been estimated by means of vertical and transversal sections. This evidence could be interpreted as the effect of a three-dimensional circulation of astenospheric flow provoked by slab roll-back. A new evidence from the tomography is the presence of a LVA in the [250 ; 400] km depth interval with an extension of 100-150 km that practically splits the Tyrrhenian slab into two parts, in Neapolitan region and in the southern Calabria-northern Sicily region. The presence of this “window slab” could be interpreted as a tear in which unperturbed mantle insert itself.

How to cite: Pucciarelli, G.: Seismic Tomography of Southern Tyrrhenian by means of teleseismic data, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-1752, https://doi.org/10.5194/egusphere-egu21-1752, 2021.

During volcanic unrest, multiple subsurface processes can happen simultaneously and may lead to an eruption. The analysis of seismic records in an unrest period before an eruption reveals information about the pre-eruptive processes and might be able to provide hints for a possible future eruption.

The 2014–2015 Holuhraun eruption was the largest one in Iceland in 230 years. It was extensively monitored and studied in a variety of multidisciplinary research approaches. Intense seismicity and ground deformation were interpreted as magma propagation from Bárðarbunga volcano 48 km laterally at ∼6 km depth over two weeks before an eruption started at Holuhraun. Different processes including vertical and lateral magma migration, dike propagation, caldera subsidence, and subglacial eruptions happened in this period and some models linking these processes are suggested. In the two-week interval preceding the eruption, there is still no clear connection between the observed tremor and pre-eruptive processes. Both the tremor source location and tremor generation process are not well understood yet. While cauldrons as a sign of subglacial eruptions were identified on the glacier surface from aerial photos, these cauldrons might have been formed earlier and there is hence an uncertainty of a few days. A tremor location might help to constrain these dates. However, the simultaneous occurrence of intense seismicity and tremor hinders the study and location of tremor. Here, we use a recent volcanic tremor extraction algorithm (Zali et al., 2020) and extract pre-eruptive tremor signals in order to better locate them using the Seismic Amplitude Ratio Analysis (SARA) method. Furthermore, the occurrence of the tremor could open new insights into ascending magma and fluid migration as well as the timing and duration of the subglacial eruptions.

We also observed short-lived tremors before the eruptions on August 29 and 31, which could be considered as eruption precursors. The primary investigation on the extracted tremor signals is promising while further analysis is on-going.

How to cite: Zali, Z., Eibl, E., Ohrnberger, M., and Scherbaum, F.: Investigation of the pre-eruptive processes of the 2014/15 Holuhraun eruption based on extracted volcanic tremor signals, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-12812, https://doi.org/10.5194/egusphere-egu21-12812, 2021.

The 2018 rift zone eruption of Kilauea volcano was accompanied by a remarkable and episodic collapse of its summit. Between May-August the eruption and collapse sequence included over 70,000 earthquakes (M≥0) and 54 major earthquakes (M≥5). We analyzed the seismicity in the Kilauea summit region and estimated seismic full moment tensors with their uncertainties for the 54 M≥5 events. These events occurred at almost daily intervals and were accompanied by intense seismicity which was concentrated between 0-3 km depths beneath the Halema‘uma‘u pit crater. The hypocenters reveal partial elliptical patterns (map view) that migrated downward by ∼200 m. The moment tensors reveal remarkably consistent mechanisms, with negative isotropic source types and localized uncertainties, and vertical P-axis orientations. From the moment tensors we derived Poisson’s ratios which are variable (ν = 0.1 − 0.3) for the first half of the collapse events and converged to ν ∼ 0.28 from June 26 onward.

How to cite: Alvizuri, C., Matoza, R., and Okubo, P.: Episodic earthquake mechanisms and intervening seismicity during the 2018 summit collapse at Kilauea caldera , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-14433, https://doi.org/10.5194/egusphere-egu21-14433, 2021.

The plate boundary between the American and Eurasian plates runs in southwest Iceland along a 5-10 km wide seismicity zone on the Reykjanes Peninsula. There, tectonic spreading events take place as continuous seismic release and seismic episodes (swarms and individual large events) with recurrence interval of about 40 years and volcanic episodes at intervals of 800-1000 years. The crust in Reykjanes is, therefore, particularly thin and hot and geothermal energy is currently harnesses in two areas on the western part of the peninsula in Reykjanes and Svartsengi.

Since January 2020, earthquake swarms with larger events up to M5.6 have been occurring frequently over the entire Reykjanes Peninsula, accompanied by unusual uplift (up to 12 cm) and subsidence cycles in the Svartsengi-Eldvörp fissure swarm. This raises the question whether we might be at the beginning of a new volcanic episode. In order to classify such processes at an early stage, multidisciplinary geophysical measurements are particularly valuable.

The Icelandic Meteorological Office (IMO), University of Iceland as well as ISOR and several partners responded immediately after the unrest began. As soon as January 2020, GFZ proposed a rapid response field campaign (MAGIC: MultidisciplinAry imaGIng and Characterization of the magma/fluid reservoir beneath Svartsengi). Only one week after the uplift start and first earthquake swarm, we connected a Distributed Acoustic Sensing interrogator to a 21 km long telecommunication fibre optic cable which crosses the uplift and swarm area. In addition, while we complied to strict constraints due to the Covid-19 pandemic, the rapid response activities comprised deployment of several additional sensors including broadband seismology, rotational seismology and we performed repeated surveys including gas-, gravity-, electromagnetic-, airborne and ground magnetic- measurements.

We present preliminary results from various techniques and discuss their role in discriminating different scenarios aiming at explaining the magma-tectonic unrest phase. In particular, we analyze how the combination of airborne snapshots of ground morphology can be combined with the high temporal and spatial resolution deformation fields along the fibre optic cable.

How to cite: Jousset, P., Hersir, G. P., Shevchenko, A., Agustsson, K., Gudnasson, E. A., Milkereit, C., Morschhauser, A., Eibl, E., Walter, T., Reinsch, T., Erbas, K., Wollin, C., Schantz, A., Samrock, F., Agustsdottir, T., Dahm, T., Flovenz, O., and Krawczyk, C.: Volcano-seismic 2020 unrest in Reykjanes Iceland: The MAGIC multi-parametric rapid response during Covid-19 lockdown, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-16134, https://doi.org/10.5194/egusphere-egu21-16134, 2021.

Kīlauea Volcano, Hawai’i, is one of the world’s most active volcanoes. From 1983 to 2018 the magmatic system was in near continuous eruptions. This eruption ended on 30 April 2018 when the deflation of Kīlauea caldera began and a dike intrusion from the Middle East Rift Zone of Kīlauea Volcano downrift towards the Lower East Rift Zone (LERZ) was observed in seismic data. On 3 May 2018, the first of final 24 eruptive fissures opened at the LERZ. This was the beginning of the largest effusive event of the last two centuries at the LERZ. Here, we present Time-Averaged Discharge Rate (TADR) and lava eruption volume estimations based on a joint analysis of multi-sensor infrared (IR) Visible Infrared Imaging Radiometer Suite (VIIRS) and Moderate Resolution Imaging Spectroradiometer (MODIS) satellite Earth observation data together with laboratory viscosity measurements to investigate this large eruption event at the LERZ. First, the TADR measurements were performed independently for each sensor data to cross-check the results against each other. Second, a joint timeseries of the VIIRS and MODIS TADR estimates was created to obtain more frequent measurements. This joint analysis of VIIRS and MODIS data resulted in an erupted lava volume of 0.924 ± 0.462 km³. Independent measurements based on airborne Synthetic Aperture Radar Interferometry (InSAR) and LIDAR topography changes are within the range of the IR data-based estimates of the erupted lava volume. The 2018 LERZ eruption could be differentiated into four main phases based on major element compositions of the eruptive products. The VIIRS and MODIS-based TADR estimation showed a relatively low Mean Output Rate (MOR) of 2.82 ± 1.41 m³/s during early Phase I. MOR then almost doubled to 4.94 ± 2.47 m³/s in late Phase I. A strong increase of MOR to 64.97 ± 32.48 m³/s occurred during Phase II. In Phase III, MOR again doubled to 137.67 ± 68.83 m³/s. This strong increase of the MOR during the different phases of the 2018 LERZ eruption agrees well with the evolution of the lava from low-temperature, highly differentiated sluggish ‘a‘ā lava flows in the beginning to high-temperature mafic more fluid pāhoehoe lava from Phase II onwards, as observed in the field by the USGS.

How to cite: Plank, S., Massimetti, F., Soldati, A., Hess, K.-U., Nolde, M., Martinis, S., and Dingwell, D. B.: Combining multi-sensor infrared satellite and laboratory measurements to estimate the lava discharge rate of 2018 Kilauea Volcano, Hawai'i eruption, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-276, https://doi.org/10.5194/egusphere-egu21-276, 2020.

Vulcanian explosions are hazardous and are often spontaneous and direct observations are therefore challenging. Ebeko is an active volcano on Paramushir Island, northern Kuril Islands, showing characteristic Vulcanian-type activity. In 2019, we started a comprehensive survey using a combination of geophysical field station records and repeated unoccupied aircraft system (UAS) surveys to describe the geomorphological features of the edifice and its evolution during ongoing activity. Seismic data revealed the activity of the volcano and were complemented by monitoring cameras, showing a mean explosion interval of 34 min. Digital terrain data generated from UAS quadcopter photographs allowed for the identification of the dimensions of the craters, a structural architecture and the tephra deposition at cm-scale resolution. The UAS was equipped with a thermal camera, which in combination with the terrain data, allowed it to identify fumaroles, volcano-tectonic structures and vents and generate a catalog of 282 thermal spots. The data provide details on a nested crater complex, aligned NNE-SSW, erupting on the northern rim of the former North Crater. Our catalog of thermal spots also follows a similar alignment on the edifice-scale and is also affected by topography on a local scale. New analysis are included in this presentation as well as a long term change analysis based on remote sensing data.

How to cite: Walter, T. R. and Belousov, A. A. M.:  The 2019 Eruption Dynamics and Morphology at Ebeko Volcano Monitored by geophysical instrument networks and Unoccupied Aircraft Systems (UAS), EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-687, https://doi.org/10.5194/egusphere-egu21-687, 2021.

The highly productive high temperature geothermal fields in Iceland are located within active volcanic systems on the plate boundaries. When an earthquake swarm or an unusual surface uplift or subsidence occur, it is important to assess the hazards and whether the unrest is triggered or controlled by volcanic or anthropogenic processes, or a combination of both.

On January 22nd, 2020, a rapid, large-scale uplift (14 km x 12 km) started at the Svartsengi geothermal field on the plate boundary of the Reykjanes Peninsula, along with an intense earthquake swarm that began simultaneously about 3 km east of the centre of uplift. The centre of uplift was located about 1 km west of Mt. Thorbjörn, in the middle of the Svartsengi geothermal field, close to the reinjection wells. Over a period of 6 months, three such uplift cycles occurred, each lasting for several weeks and followed by periods of relatively rapid subsidence. The duration and timing of the uplift-subsidence cycles appears to follow a clear trend where the successive inflation episodes lasted longer but with lower inflation rate.

The centres of uplift and the deflation cycles are the same and remained stationary. The accompanied intense earthquake swarms migrated along the 40 km long oblique plate boundary of the Reykjanes Peninsula, demonstrating a major plate tectonic event. The maximum depth of earthquakes was close to 4.5 km directly above the centre of uplift but extending to 6-7 km in the surroundings where the maximum magnitudes reached M_W 4.8.

A few weeks after the onset of the unrest, nine additional seismic stations were deployed to densify the local seismic network in place. In addition, complimentary data from an existing 21 km long fibre optics cable were used to monitor high-frequency linear strain rates. Both measures led to a significant improvement in the earthquake detection and location which predominantly occurred in swarms. Likewise, InSAR data analysis of temporal uplift cycles was performed, repeated gravity measurements at permanent sites were performed, and resistivity was remeasured at chosen sites.  

Multiple different elementary models were developed and tested to explain the cyclic excitation of the uplift, subsidence, and seismicity. While the individual unrest episodes might be controlled by possible magma intrusions into the lower crust, our favoured model explains the spatio-temporal pattern of ground uplift by the rise and diffusion of pore pressure in a 4-5 km deep geothermal aquifer. To distinguish between different models, we use multi-disciplinary geophysical datasets, such as deformation, seismicity, and gravity.

How to cite: Flóvenz, Ó., Wang, R., Hersir, G. P., Ágústsson, K., Vassileva, M., Drouin, V., Heimann, S., Isken, M., Hainzl, S., Gudnason, E. Á., Ágústsdóttir, T., Magnússon, I., Horalec, J., Milkereit, C., Motagh, M., Walter, T., Rivalta, E., Jousset, P., and Dahm, T.: The possible role of magma and geothermal fluid in the episodic uplift cycles and intense seismicity beneath the Svartsengi high temperature geothermal field, Iceland , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-7870, https://doi.org/10.5194/egusphere-egu21-7870, 2021.

Villarrica is a basaltic volcano with an active lava lake in South Central Chile. The lava lake displays a variety of degassing styles from gentle seething to more violent Strombolian explosions. This activity is accompanied by sequences of transient seismic waveforms suggesting the presence of discrete gas bubbles in the upper magma column. Gas bubbles flow through liquid-filled pipes according to distinct patterns depending on viscosity of the liquid and volumetric gas flow rate. Laboratory experiments indicate that these regimes are characterized by distinct frequency distributions of bubble sizes and spacings. By assuming that these parameters are reflected by the magnitude of the transients and the time between them, we compared their statistical distributions to infer a flow regime for the shallow conduit of Villarrica. The approximately log-normal distributions indicate a sustained slug flow regime in which the gas ascends in trains of conduit-wide gas slugs. The event catalog for our analysis contained about 20,000 events and was generated from 12 days of seismic data from March 2012 acquired by a dense local network. A well-known problem in earthquake statistics is the incompleteness of event catalogs towards low magnitudes due to decreasing detectability in the ambient noise. We estimated the actual distribution of magnitudes by using a Monte Carlo simulation of the event detection based on the statistical properties of the observed seismic noise. The unknown source depth and mechanism introduce further ambiguity regarding the distributions. Nevertheless, we hope to refine the degassing model by taking into account degassing rates, magma properties and more detailed analysis of the nature of the seismic events.

How to cite: Lehr, J., Bredemeyer, S., and Rabbel, W.: Inferring a shallow degassing model for Villarrica Volcano from seismic explosion signals and SO2 flux, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-8796, https://doi.org/10.5194/egusphere-egu21-8796, 2021.

Unzen Volcano is located in Shimabara Peninsula, Nagasaki, Japan. After 198 years of dormancy, the volcano erupted throughout 1990-1995 and resulted the emergence of new lava dome called Heisei-Shinzan. Following the eruption, numerous studies have been intensively conducted in Unzen volcano to assess the eruption mechanism and the magma plumbing system. Regarding to the magmatic system, the most preferred model is that the primary supply of magma is stored beneath Chijiwa bay. This magma chamber is located about 15 km west of the active dome at vertical depth approximately 15 km, and followed by subordinate shallower magma chambers beneath the volcano (e.g. Nakamura 1995; Kohno et al 2008). Upon the eruption, the magma ascended obliquely towards the summit in east direction (e.g. Umakoshi et al 2001). However, how main magma chamber  and shallower chambers are connected to the summit via oblique pathway is poorly imaged in terms of structure.
As widely known, Magnetotelluric method is highly sensitive to low resistivity zone caused by interconnected fluids. Low resistivity zone detected in the volcanic area usually can be interpreted as hydrothermal/magmatic fluid and or magma chamber containing partial melt (e.g. Aizawa et al 2014; Hill et al 2015). Thus, by using broadband Magnetotelluric method, we aim to investigate resistivity structure of Unzen volcano associated with magmatic system and its controlling structure (e.g. pathway and faults).
Although the shallow structures around Unzen volcano are estimated by the 2017-2019 campaigns (Triahadini et al 2019; Hashimoto et al 2020), we are unable to image deeper structure around the proposed location of magma chambers and magma pathway. To achieve our goals, during November-December 2020, we installed 35 new sites to cover whole area in Shimabara Peninsula. In total, deployed 99 Magnetotelluric stations covering Unzen volcano and Shimabara Peninsula. On this meeting, we would like to present our resistivity structure derived from all dataset.

How to cite: Triahadini, A., Aizawa, K., Hashimoto, T., Uchida, K., Yamamoto, Y., Chiba, K., Muramatsu, D., Miyano, K., Kawamura, Y., and Satoru, A.: Understanding Magmatic System of Unzen Volcano (Nagasaki, Southwest Japan) Inferred from Broad-band Magnetotelluric Observation, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-1866, https://doi.org/10.5194/egusphere-egu21-1866, 2021.

Modern UAS (unmanned aircraft system), light weight sensor systems and new processing routines allow us to gather optical data of volcanoes at a high resolution. However, due to the typically poor colorization, our ability to investigate and interpret such data is limited. Further, the information stored in the red, green and blue channel (RGB) is correlated. This makes any analysis a 3 dimensional task. Principal Component Analysis (PCA) helps us to overcome these problems by decorrelating the original band information and generating a variance representation of the original data. Therefore PCA is a suitable tool to detect optical anomalies, as might be caused by volcanic degassing and associated processes.

Applied in a case study at La Fossa Cone (Vulcano Island - Italy), the PCA showed a high efficiency for the detection and pixel based extraction of areas subject to hydrothermal alteration and sulfur deposition. We observed a broad alteration zone surrounding the active fumarole field, but also heterogeneities within, indicating a segmentation. Systematic variations in color and density distribution of sulfur deposits have implications for structural controls on the degassing system.

Combining the efficiency of PCA with the high resolution of UAS derived data, this methodology has a high potential to be employed in the spatio-temporal monitoring of volcanic hydrothermal systems and processes at surface.

How to cite: Müller, D., Bredemeyer, S., Zorn, E., De Paolo, E., and Walter, T.: Detection and monitoring of hydrothermal alteration by Principal Component Analysis applied on UAS derived optical data, Vulcano Island - Italy, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-4291, https://doi.org/10.5194/egusphere-egu21-4291, 2021.

The ability to accurately and reliably obtain images of shallow subsurface anomalies within the Earth is important for hazard monitoring at many geologic structures, such as volcanic edifices. In recent years, the use of machine learning as a novel, data-driven approach to addressing complex inverse problems in the geosciences has gained increasing attention, particularly in the field of seismology. Here we present a physics-based, machine learning method to integrate disparate geophysical datasets for shallow subsurface imaging. We develop a methodology for imaging static density variations at a volcano with well-characterized topography by pairing synthetic cosmic-ray muon and gravity datasets. We use an artificial neural network (ANN) to interpret noisy synthetic datasets generated using theoretical knowledge of the forward kernels that relate these datasets to density. The deep learning model is trained with synthetic data from a suite of possible anomalous density structures and its accuracy is determined by comparing against the known forward calculation. 

In essence, we have converted a traditional inversion problem into a pattern recognition tool, where the ANN learns to predict discrete anomalous patterns within a target structure. Given a comprehensive suite of possible patterns and an appropriate amount of added noise to the synthetic data, the ANN can then interpolate the best-fit anomalous pattern given data it has never seen before, such as those obtained from field measurements. The power of this approach is its generality, and our methodology may be applied to a range of observables, such as seismic travel times and electrical conductivity. Our method relies on physics-based forward kernels that connect observations to physical parameters, such as density, temperature, composition, porosity, and saturation. The key benefit in using a physics-based approach as opposed to a data-driven one is the ability to get accurate predictions in cases where the amount of data may be too sparse or difficult to obtain to reliably train a neural network. We compare our approach to a traditional inversion, where appropriate, and highlight the (dis)advantages of the deep learning model.

How to cite: Cosburn, K. and Roy, M.: Using artificial neural networks with joint muon-gravity datasets for shallow subsurface density prediction at volcanoes, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-717, https://doi.org/10.5194/egusphere-egu21-717, 2021.

Recent volcanic activity of La Palma island, the fifth in extension (706 km²) and the second in elevation (2,423 m a.s.l.) of the Canarian archipelago, has taken place exclusively in the last 123 ka at the southern part of the island, where Cumbre Vieja volcano, the most active basaltic volcano in the Canaries, has been constructed. A total of seven volcanic eruptions have been reported along the main north-south rift zone of Cumbre Vieja in the last 500 years. On October 7^th and 13rd, 2017, two remarkable seismic swarms interrupted a seismic silence of 46 years in Cumbre Vieja volcano with earthquakes located beneath Cumbre Vieja volcano at depths ranging between 14 and 28 km with a maximum magnitude of 2.7. Five more seismic swarms were registered in 2020.

³He/⁴He ratio has been monitored at Dos Aguas cold mineral spring in La Palma Island since 1991 to date as an important volcano monitoring tool able to provide early warning signal of future volcanic unrest episodes. Magmatic helium emission studies have demonstrated to be sensitive and excellent precursors of magmatic processes occurring at depth. The highest ³He/⁴He ratio reported to date from the Canarian archipelago has been measured at Dos Aguas: 10.24 R_A (being R_A the ratio in atmospheric helium) (Padrón et al., 2015). This value is higher than any value found either in the lavas or terrestrial fluid in the Canary Islands, and indicates an important mantle contribution.  According to the temporal evolution of the magmatic component of helium at Dos Aguas, we suggest the occurrence of aseismic magma rising episodes beneath La Palma within the upper mantle towards an ephemeral magma reservoir in the period 2007-2017. However, in the period 2017-2020, magma rising have produced seismic swarms that were accompanied also by the highest ³He/⁴He ratio measured at Dos Aguas (10.42 R_A); both geochemical and geophysical signals confirm an upward magma migration towards a subcrustal magma reservoir beneath La Palma island.

Padrón et al., (2015). Bull Volcanol 77:28. DOI 10.1007/s00445-015-0914-2

How to cite: Padrón, E., Pérez, N. M., Melián, G. V., Sumino, H., Alonso, M., Recio, G., Asensio-Ramos, M., Rodríguez, F., and D’Auria, L.: Temporal evolution of 3He/4He isotopic ratio at Dos Aguas cold mineral spring, La Palma, Canary Islands, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-14994, https://doi.org/10.5194/egusphere-egu21-14994, 2021.

Significant temporal variations in the chemical and isotopic composition of Taal fumarolic gas as well as in diffuse CO₂ emission from Taal Main Crater Lake (TMLC) have been observed across the ~12 years of geochemical monitoring (Arpa et al., 2013; Hernández et a., 2017), with significant high CO₂ degassing rates, typical of plume degassing volcanoes, measured in 2011 and 2017. In addition to these CO₂ surveys at the TCML, soil CO₂ efflux continuous monitoring was implemented at Taal volcano since 2016 and a clear increasing trend of the soil CO₂ efflux in 2017 was also observed. Increasing trends on the fumarolic CO₂/St, He/CO₂, CO/CO₂ and CO₂/CH₄ ratios were recorded during the period 2010-2011 whereas increasing SO₂/H₂S, H₂/CO₂ ratios were recorded during the period 2017-2018. A decreasing on the CO₂/CH₄ and CO₂/St ratios was observed for 2017-2018. These changes are attributed to an increased contribution of magmatic fluids to the hydrothermal system in both periods. Observed changes in H₂ and CO contents suggest increases in temperature and pressure in the upper parts of the hydrothermal system of Taal volcano. The ³He/⁴He ratios corrected (Rc/Ra), and δ¹³C of fumarolic gases also increased during the periods 2010-2011 and 2017-2018 before the eruption onset. During this study, diffuse CO₂ emission values measured at TMCL showed a wide range of values from >0.5 g m⁻² d⁻¹ up to 84,902 g m⁻² d⁻¹. The observed relatively high and anomalous diffuse CO₂ emission rate across the ~12 years reached values of 4,670 ± 159 t d^-1 on March 24, 2011, and 3,858 ± 584 t d^-1 on November 11, 2017. The average value of the soil CO₂ efflux data measured by the geochemical station showed oscillations around background values until 14 March, 2017. Since then at 22:00 hours, a sharp increase of soil CO₂ efflux from ~0.1 up to 1.1 kg m^-2 d^-1 was measured in 9 hours and continued to show a sustained increase in time up to 2.9 kg m^-2 d^-1 in 2 November, that represents the main long-term variation of the soil CO₂ emission time series. All the above variations might be produced by two episodes of magmatic intrusion which favored degassing of a gas-rich magma at depth. During the 2010-2011 the magmatic intrusion of volatile-rich magma might have occurred from the mid-crustal storage region at shallower depths producing important changes in pressure and temperature conditions, whereas a new injection of more degassed magma into the deepest zone of the hydrothermal system occurring in 2017-2018 might have favored the accumulation of gases in the subsurface, promoting conditions leading to a phreatic eruption. These geochemical observations are most simply explained by magma recharge to the system, and represent the earliest warning precursor signals to the January 2020 eruptive activity.

Arpa, M.C., et al., 2013. Bull. Volcanol. 75, 747. https://doi.org/10.1007/s00445-013-0747-9.

Hernández, P.A., et al.,  2017. Geol. Soc. Lond. Spec. Publ. 437:131–152. https://doi.org/10.1144/SP437.17.

How to cite: Hernández, P. A., Melian, G., Asensio-Ramos, M., Padron, E., Sumino, H., Perez, N. M., Padilla, G., Barrancos, J., Baldago, M. C., Rodriguez, F., Alonso, M., Amonte, C., Arcilla, C., and Lagmay, M.: Geochemical evidence of volcanic plumbing system processes from fumarolic gases and diffuse CO2 degassing of Taal volcano, Philippines, prior to the January 2020 eruption , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15073, https://doi.org/10.5194/egusphere-egu21-15073, 2021.

The Cape Verde islands are located about 800 km west of Senegal, at 14°-17° latitude and 21°-25° longitude. The archipelago consists of a volcanic chain of 10 major islands and eight minor islands The only currently active volcano in the Cape Verde archipelago is Pico do Fogo, which is located on the island of Fogo. Rising to 2829 m a.s.l., it is the most active volcano of the Cabo Verde Island. We report the results of the geochemical monitoring of the fumarolic discharges at the Pico do Fogo volcano in Cape Verde from 2007 to 2016. During this period Pico do Fogo experienced a volcanic eruption (November 23, 2014) that lasted 77 days. Two fumaroles were sampled, a low (F1~100ºC) and a medium (F2~300ºC) temperature. The variations observed in the δ¹⁸O and δ²H in F1 and F2 suggest different fluid source contributions and/or fractionation processes. Although no significant changes were observed in the outlet fumarole temperatures, two clear increases were observed in the vapor fraction of fumarolic discharges during the periods 2008-2009 and 2013-2014. Also, two sharp peaks were observed in CO₂/CH₄ ratios at both fumaroles, in November 2008 and November 2013, coinciding with significant increases in the emission rate of diffuse CO₂ and He, and heat flow measured in the crater of Pico do Fogo volcano. This confirms that gases with a strong magmatic component rose towards the surface within the Pico do Fogo system during 2008 and 2013. Further, F2 showed two CO₂/St peaks, the first in late 2010 and the second after eruption onset, suggesting the occurrence of magmatic pulses into the volcanic system. Time series of He/CO₂, H₂/CO₂ and CO/CO₂ ratios are low in 2008-2009, and high in 2013-2014 period, supporting the hypothesis of fluid input from a deeper magmatic source. Regarding to the isotopic composition, increases in ³He/⁴He (R/R_A)_cor are observed in both fumaroles; F1 showed a peak in 2010 from a minima in 2009 during the first magmatic reactivation onset and another in late 2013, while F2 displayed a slower rise to its maximum in late 2013. The high ³He/⁴He ratios in both fumaroles are close to the magmatic end-member, indicating that He is predominantly of upper mantle origin. This work supports that monitoring of the chemical and isotopic composition of the fumaroles of the Pico do Fogo volcano is a very important tool to understand the processes that take place in the magmatic-hydrothermal system and to be able to predict future episodes of volcanic unrest and to mitigate volcanic risk.

How to cite: Melián, G. V., Hernández, P. A., Asensio-Ramos, M., Pérez, N. M., Padrón, E., Alonso, M., Padilla, G. D., Barrancos, J., Sortino, F., Sumino, H., Rodríguez, F., Amonte, C., Silva, S., Cardoso, N., and Pereira, J. M.: Insights from fumarole gas geochemistry on the recent volcanic unrest of Pico do Fogo, Cape Verde, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-14763, https://doi.org/10.5194/egusphere-egu21-14763, 2021.

The Eastern Carpathians are characterized by intense gas emissions starting from the Neogene to Quaternary volcanic structures, especially the youngest dormant volcano, Ciomadul, but occurring also far away from these, in the Cretaceous flysch units. This is the most intensive degassing area from Romania. The gas emissions appear in different forms: dry gas, named mofettes and bubbling gas when they are accompanied by groundwater. The major components of these gas emissions are: CO₂, CH₄, N₂ and sometimes H₂S. Recent studies reveal a magmatic contribution up to 60% in these emissions (Vaselli et al., 2002, Kis et al., 2019). Gases are also present dissolved in groundwater and transported to the surface by CO₂-rich springs. Besides these visible emissions, the gases come to the surface as diffuse degassing from the soil. We started a systematic geochemical investigation of the gas emissions in the volcano-tectonic environment of the southern part of the Eastern Carpathians, together with a 5-year monitoring of the gas emissions. Our primary aims are to constrain the flux of CO₂, the origin of the different gas species, their interaction, and their relationship with the geodynamic background. Our findings could be integrated to the global carbon estimations, currently missing from the worldwide evaluations and could help the establishment of a long-term monitoring system of the gases in the area.

This work was supported by a grant of the Romanian Ministry of Education and Research, CNCS - UEFISCDI, project number PN-III-P1-1.1-TE-2019-1908, within PNCDI III and the project GTC 32144 supported by Babes-Bolyai University, Romania.

How to cite: Kis, B.-M., Palcsu, L., Zsigmond, A.-R., Tamas, D. M., Szollosi, I., Szalay, R., and Harangi, S.: Geochemistry of gas emissions in the volcano-tectonic environment of the Eastern Carpathians, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-13506, https://doi.org/10.5194/egusphere-egu21-13506, 2021.

Late Miocene to Pleistocene volcanism within the Vardar zone (N. Macedonia) covers a large area, has a wide range in composition and it is largely connected to the tectonic evolution of the South Balkan extensional system, the northern part of the Aegean extensional regime. A recent study indicated an increasing rate of mantle metasomatism towards the younger centers in the region [1]. During the last stage of activity, ultrapotassic (UK) centers that formed between ca. 3.2 and 1.5 Ma originated from the lithospheric mantle beneath the region [2]. Although there are no reported mantle xenoliths from these centers, the erupted mafic rocks contain abundant olivine as phenocrysts [3]. Noble gas isotopic characteristics of fluid inclusions in olivine can reveal important information about the origin of the fluid and the metasomatic state of the lithospheric mantle. We analyzed for the first time the noble gas composition of fluid inclusions of olivine phenocrysts from the Mlado Nagoričane volcanic center, the northernmost member of the UK centers with an eruption age of 1.8 ± 0.1 Ma [2]. The R/R_A ratios give a range of 3.1-4.5 with ⁴He/²⁰Ne values of 11.7-14.6. These R/R_A values are lower than the MORB and the averaged subcontinental lithospheric values, and considering the negligible amount of atmospheric contribution, imply a more metasomatized character for the underlying lithospheric mantle beneath the region. Mantle-derived noble gases were detected in a recent geochemical study on the thermal springs and gas exhalations in the region, with up to 20% of mantle contribution calculated based on their noble gas composition using the MORB R/R_A value [4]. These new Mlado Nagoričane fluid inclusion noble gas values indicate that the mantle contribution in the recent gas emissions in the region could be higher than what was thought.

This research was supported by the European Union and the State of Hungary, financed by the European Regional and Development Fund in the project of GINOP-2.3.2-15-2016-00009 ‘ICER’ project

[1] Molnár et al. 2020 – EGU2020-13101.

[2] Yanev et al., 2008 – Mineralogy and Petrology, 94(1-2), 45-60.

[3] Yanev et al., 2008 – Geochemistry, Mineralogy and Petrology, Sofia, 46, 35-67.

[4] Temovski et al. 2020 – EGU2020-2763.

How to cite: Molnár, K., Temovski, M., and Palcsu, L.: First noble gas results from fluid inclusions of the Late Miocene-Pleistocene Macedonian volcanics, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-9440, https://doi.org/10.5194/egusphere-egu21-9440, 2021.

Volcano monitoring is commonly performed through acquisition and interpretation of real-time signals able to track changes in the magmatic system and the eventual migration of magma toward the surface. Petrological monitoring, in particular,  focus on magma history in terms of depth of storage zones, transport pathways, mechanisms of differentiation, and timescales of involved processes with aim to extrapolate information about the trigger of magma ascent and the eruptive behaviour, and its possible variation over the course of an eruption.

In the present study, conducted in the framework of the EUROVOLC project, we developed a questionnaire that aims to survey the most common petrological monitoring procedures performed by volcano monitoring institutions, in order to identify prevailing techniques and most critical issues, and to rate the suitability of specific investigations in terms of costs versus benefit. The final goal is to identify essential and mandatory petrologic techniques to accomplish for an efficient petrological monitoring during ongoing eruptions, so that can be assessed the minimum logistic requirements (e.g., facilities, infrastructures, operators) and can be defined operative best practices protocols to achieve petrologic results in a timeframe short enough to be well of use for monitoring purposes.

The surveyed information, which resulted from a sample of eighteen interviewed institutions that deal with monitoring of active volcanoes with a variety of eruptive behaviour, provide insights about the whole steps of petrologic monitoring including sampling, sample preparations and analyses, data interpretation and dissemination. The survey reveals that efforts have been made to organize petrological monitoring with standardized procedures similarly to the other monitoring disciplines. For example, some institutions suffer lack of dedicated staff that can be operative with short forewarning. The objects of petrological investigation include all the types of volcanic products from lava to pyroclastic and there are attempts to deal with fixed sampling schedule. Moreover, there is consciousness that the capability to acquire and to interpret the most valuable analytical results at in situ institutions provide a quick image of ongoing eruptive processes and improve the interaction with other disciplines. Therefore, concerning the analytical procedures, which is the core of petrological monitoring, an important results is the cross correlation between the analyses that are easy to acquire (in terms of resources, equipment and time availability) and their effective role for the petrological monitoring.

The expectations include an augmented perception of the benefits that petrologic monitoring brings in the comprehension of eruptive processes. Filling the gap of the primary needs required to accomplish the identified best practices within a short timeframe is a compelling need to lead advancement of the volcano monitoring science.

How to cite: Re, G., Corsaro, R. A., D'Oriano, C., and Pompilio, M.: A review of petrological monitoring procedures: suggestion of best practice protocols for eruption monitoring, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-5758, https://doi.org/10.5194/egusphere-egu21-5758, 2021.

One of the main volcano-structural and geomorphological feature in Tenerife (2,034 km²) is the triple rift system, formed by aligned of hundreds of monogenetic eruptive products of shield basaltic volcanism. At the intersection of this triple rift system rises the Teide-Pico Viejo volcanic complex. These volcanic rifts are considered as active volcanic edifices. The North East volcanic Rift Zone (NERZ, 210 km²) form a main NE-SW structure. The North West volcanic Rift Zone (NWRZ, 72 km²) is oriented in NW-SE direction and the North South volcanic Rift Zone (NSRZ, 325 km²) comprises a more scattered area on the south of these monogenetic cones. The most recent eruptive activity of Tenerife has taken place in these rift systems. NERZ host the fissural eruption of Arafo-Fasnia-Siete Fuentes (1704-1705). NWRZ host two historical eruptions: Arenas Negras in 1706 and Chinyero in 1909. Recently the eruption of Boca Cangrejo (1492) has been added to the historical register through ¹⁴C dating. NSRZ does not host historical volcanism, although it is recent, up to 10,000 years old.

In order to provide a multidisciplinary approach to monitor potential volcanic activity changes at the NERZ, NWRZ and NSRZ, diffuse CO₂ emission surveys have been undertaken since 2000, in general in a yearly basis, but with a higher frequency when seismic swarms have occurred in and around NWRZ volcano. Each study area for NERZ, NWRZ and NSRZ comprises hundreds of sampling sites homogenously distributed. Soil CO₂ efflux measurements at each sampling site were conducted at the surface environment by means of a portable non-dispersive infrared spectrophotometer (NDIR) LICOR Li820 following the accumulation chamber method. To quantify the CO₂ emission rate from the NERZ, NWRZ and NSRZ a sequential Gaussian simulation (sGs) was used as interpolation method.

The diffuse CO₂ emission rate for the NERZ ranged from 532 up to 2823 t d^-1 between 2001 and 2020, with the highest value measured in 2020. In the case of NWRZ, the diffuse CO₂ emission rate ranged from 52 up to 867 t d^-1 between 2000 and 2020, with the highest value measured in one of the surveys of 2005. Finally, and for NSRZ, the diffuse CO₂ emission rate ranged from 78 up to 819 t d^-1 between 2002 and 2020, with the highest value measured in 2019. The temporal evolution of diffuse CO₂ emission at the NERZ, NWRZ and NSRZ shows a nice and clear relationship with the volcanic seismicity in and around Tenerife Island, which started to take place from the end of 2016. The good temporal correlation between the volcanic seismicity and the increase trend observed in the time series of diffuse CO₂ emission rates at NERZ, NWRZ and NSRZ is also coincident with the observed increase of diffuse CO₂ emission rate at the summit crater of Teide. This work demonstrates the importance of performing soil CO₂ efflux surveys at active rift systems in volcanic oceanic islands as an effective geochemical monitoring tool.

How to cite: Rodríguez, F., Padrón, E., Melián, G., Asensio-Ramos, M., Alonso, M., Amonte, C., Martín, A., Hernández, P. A., Pérez, N. M., Barrancos, J., Padilla, G., and D'Auria, L.: Monitoring of diffuse CO2 degassing at NERZ, NWRZ and NSRZ volcanic systems of Tenerife, Canary Islands, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15427, https://doi.org/10.5194/egusphere-egu21-15427, 2021.

The fumarolic field of Pisciarelli is the most active vent of the Campi Flegrei caldera, a volcano in the metropolitan area of Naples (Italy) in a current state of unrest. Recent studies have identified a clear escalation of degassing activity at Pisciarelli since 2012, raising concern on a possible acceleration of the unrest. The absence of sulfur dioxide prevents UV spectroscopy from determining the volcanic gas flux, and researchers have tried alternative techniques for measuring CO₂ and H₂S fluxes. Here we report observations of CO₂, H₂S, and H₂O concentrations in the plume of Pisciarelli derived on December 2019 and October 2020 with a hexacopter drone equipped with miniaturized diffusive gas sensors. The drone flew at a constant altitude (~50 m above ground level), transecting the gas plume multiple times. This technique allowed us to calculate the CO₂, H₂S, and H₂O gas fluxes by combining the georeferenced gas concentrations with the plume vertical rising speed derived from thermal and visible camera footages. Similar to previous gas composition and flux measurements, our results suggest that gas-sensors-equipped drones are a cost-effective technique for monitoring gas fluxes on active volcanoes, where UV spectroscopy cannot be used, and that can be made from safe distances.

How to cite: Tamburello, G., Marotta, E., Belviso, P., Avvisati, G., Ricci, T., Rouwet, D., Avino, R., and Caliro, S.: Gas-sensors-equipped drone measurements of volcanic plume gas composition and flux at Pisciarelli, Campi Flegrei, Italy, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15492, https://doi.org/10.5194/egusphere-egu21-15492, 2021.

In seismic active areas, the primary composition of natural gas emissions can be modified upon migration to the surface and storage in crustal reservoirs as the result of secondary chemical processes at shallow levels that can change the pristine composition of the fluids creating misunderstanding in the evaluation of the contributions due to different sources. Noble gases are among the most powerful indicators of such natural processes.  In particular, Helium (hereafter He) is a reliable geochemical tracer for discriminating the crustal and mantle components in the outgassing gases due to the different origin of its two isotopes (³He has a primordial origin, whereas ⁴He is continuously produced by radioactive α-decay of ^235,238U and ²³²Th).  Therefore, the ³He/⁴He ratio is considered one of the most efficient geochemical tracers, whose variations can be directly ascribed to magmatic/crustal dynamics and therefore it is of primary importance in volcanic and seismic forecasting. In this study, we report chemical and isotopic (helium and carbon) data of gases and water emitted from three areas characterized by a high seismic hazard and located within the southern Apennines seismogenic belt. Through two fieldwork campaigns in 2019-2020, about 15 sites were inspected. Carbon dioxide is the main component in most of investigated sites (> 90 vol.%), except for Pozzo Tramutola, that is CH₄-dominated. He and N₂ concentrations are significantly variable (from 6 to 260 ppm and from 0.22 to 12.78 vol%, respectively). In agreement to previous investigations (Italiano et al., 2001; Caracausi and Paternoster, 2015), the sites in the Matese area are characterised by typical metamorphic [MOU1] N₂ values and low content of He and Ar and seem to be the result of mixing processes between crustal and/or metamorphic and atmospheric or ASW end-member. The sampled fluids have ³He/⁴He ratios from 0.02 to 2.92 Ra with corresponding He/Ne ratios in the range of 0.353-508.10. These ⁴He/²⁰Ne ratios are much higher than the same ratio in the atmosphere (He/Ne=0.318; Ozima-Podosek, 2002) supporting that atmospheric He component in the sampled fluids is negligible for most sites. In general, we recognized that ³He/⁴He ratios indicate mixing between radiogenic and mantle end-members and Mefite site has highest mantle values that are close to the ratio at Mt Vesuvio and Pleghreian volcanic systems (< 60 from the study area). The ⁴⁰Ar/³⁶Ar ratios show a small range from values close to atmosphere up to ⁴⁰Ar/³⁶Ar = 325. We also investigated the carbon species and their isotopes. To investigate the genetic origins of the methane we have used web-based machine learning tool that determines the origin of natural gases (Snodgrass-Milkov, 2020) and the results shown that methane is mainly thermogenic even if we also recognized an abiotic component in a few of sites. This study will provide data for the reconstruction of a basic model for interpreting the relationships between outgassing and tectonics, and further for interpreting possible seismic-induced variation

Ozima & Podosek. 2002. Noble Gas Geochemistry.

Tsunogai & Wakita. 1995. Science

Snodgrass and Milkov, 2020. Comput. Geosci

How to cite: Buttitta, D. and Paternoster, M.: Noble gases and carbon isotopes in natural gas samples from seismic active areas of southern Apennines (Italy), EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15484, https://doi.org/10.5194/egusphere-egu21-15484, 2021.

The oceanic active volcanic island of Tenerife (2034 km²) is the largest of the Canarian archipelago. There are more than 1,000 galleries (horizontal drillings) in the island, which are used for groundwater exploitation and allow reaching the aquifer at different depths and elevations. During a two-year period (July 2016 to July 2018), a hydrogeochemical study was carried out in two galleries on Tenerife (Fuente del Valle and San Fernando) for volcanic monitoring purposes with weekly sampling. Physicochemical parameter of water, such us temperature (ºC), pH and electrical conductivity (E.C., µS·cm^-1), were measured in-situ at each sampling point and chemical/isotopic composition of the water determined later in the laboratory.

Temperature values showed mean values of 28.1 ºC and 19.0 ºC for Fuente del Valle and San Fernando galleries, respectively. According to the average pH values, which were 6.30 for Fuente del Valle and 7.13 for San Fernando, and based on the chemical composition, both galleries are sodium-bicarbonate (Na-HCO₃) type. E.C. values in both galleries presented high ranges, with mean values of 975 and 1648 µS·cm^-1 for Fuente del Valle and San Fernando, respectively. The total alkalinity mean value of groundwater from Fuente del Valle gallery was 11.3 mEq·L^-1 HCO₃^-, while that from San Fernando was 17.3 mEq·L^-1 HCO₃^-. The SO₄^2-/Cl molar ratio was 0.59 and 3.4 for the samples from Fuente del Valle and San Fernando galleries, respectively.

The δ¹⁸O and δD isotopic analyses showed a meteoric origin of groundwaters, with mean values of -6.2‰ and -26‰ vs. VSMOW for Fuente del Valle and -6.2‰ and -21 ‰ vs. VSMOW for San Fernando. The isotopic data showed a strong interaction with endogenous gases such as CO₂, H₂S, H₂, etc. Regarding the isotopic composition of total dissolved carbon species, expressed as δ¹³C_TDIC, average values of -0.17‰ and 0.26‰ were obtained for Fuente del Valle and San Fernando galleries, respectively. These results show an endogenous origin CO₂ signature, heavier for Fuente del Valle gallery groundwater compared to that of San Fernando.

Groundwater physicochemical parameters exhibited stable values throughout the study period, while significant temporal variations were observed in the total alkalinity, SO₄^2-/Cl^- molar ratio, δ¹⁸O and δD. Changes in isotopic ratios coincided with variations observed in the alkalinity and the SO₄^2-/Cl^- molar ratio. On October 2, 2016, a seismic swarm of long-period events was recorded on Tenerife followed by a general increase of the seismic activity in and around the island. A correlation was observed between some hydrogeochemical parameters in the groundwaters of the galleries, related to observed changes of the seismic activity. This study demonstrates the suitability of monitoring the chemical and isotopic composition of groundwater from Fuente del Valle and San Fernando galleries, as they are sensitive to changes in volcanic activity on Tenerife island. The study of groundwaters associated to a volcanic system can provide information about the magmatic gas input in the aquifer, modelling how the groundwaters flow through the edifice, and offer important geochemical information that could herald a future eruption.

How to cite: Amonte, C., Asensio-Ramos, M., Melián, G. V., Pérez, N. M., Padrón, E., Hernández, P. A., Rodríguez, F., D'Auria, L., and López, D.: Hydrogeochemical temporal variations related to changes of seismic activity at Tenerife, Canary Islands, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15098, https://doi.org/10.5194/egusphere-egu21-15098, 2021.

In this study the geochemical composition of the fluids belonging to the geothermic reservoir of Casaglia is presented. The site is located few kilometers northward of Ferrara, probably the only city in Italy whose heating system is fed by the geothermal heat near the top of the Dorsale Ferrarese, a structural anticline raising the Mesozoic limestones up to few hundred meters below the surface. Measurements of the chemical and isotopic composition of the gas phase (e.g., CO₂ and noble gas) were carried out, together with a full characterization of the physico-chemical parameters and the chemistry of the water phase.

Fluids derive from a well at a depth of about 322+15meters and the temperature of the emerging water is of 78,6 °C, pH of 6.29 and Eh of -470 mV. Salinity is up to 115.6 mS/cm with a TDS varying between 71024 mg/L and 73718 mg/L. The hydrochemical facies is identified as clorurato-alkaline and the Cl/Br ratio suggest mixing with fossil brines. dD and d¹⁸O vary from 4.70 to 5.02 and from -12.0 to -12.2 respectively. The volatile phase is mainly composed of N₂ (24.9-40.5 %),CH₄ (21.1-29.5 %) and CO₂ (37.1-18.6 %), with d¹³C(CO₂), d¹³C(CH₄) and dD(CH₄) varying from -4.4 to -3.7 ‰, from -41.7 to 41.2 ‰ and from -152 to -171 ‰, respectively.  The He amounts are extraordinary high (up to 3956 ppm)  with a ³He/⁴He of 0.02Ra unequivocally pointing to a crustal origin (e.g., Caracausi & Sulli, 2019). The ⁴⁰Ar/³⁶Ar ratios span the range 300-374, being very close to the same ratio in atmosphere.

Such high He concentration cannot be explained by a simple steady-state crustal degassing, taking into account the Th and U contents of the sedimentary cover and the metamorphic basement (Coltorti et al. 2011) which lead also to consider that the thermal state of the Casaglia reservoir involve the entire crustal thickness and not only the Mesozoic carbonate succession that hosts the reservoir itself.

It is inferred that under an active tectonic regime, as it is that where Casaglia is located, the formation of micro-fracturation, due to the field of stress generated by the local seismicity, increases the He release from the rocks and can contribute to the observed He excess in the geothermal reservoirs (e.g., Buttitta et al., 2020). In this respect, the fault system of Dorsale Ferrarese contributes to generate a preferential pathway for rising fluids with consequent mixing phenomena and provides a reasonable explanation about the presence of this high He content in the reservoir.

References:

Buttitta D. et al. (2020). Continental degassing of helium in an active tectonic setting (northern Italy): the role of seismicity. Scientific Reports, 10(1), 1–13.

Caracausi A. & Sulli A. (2019). Outgassing of Mantle Volatiles in Compressional Tectonic Regime Away From Volcanism: The Role of Continental Delamination. Geochemistry, Geophysics, Geosystems, 20(4), 2007–2020.

Coltorti M. et al. 2011. U and Th content in the Central Apennines continental crust: a contribution to the determination of the geo-neutrinos flux at LNGS. Geoch. Cosmoch. Acta 75, 2271-2294.

How to cite: Balzan, S., Caracausi, A., Ferretti, G., Saroni, A., Martinelli, G., Italiano, F., and Coltorti, M.: Geochemical composition of Casaglia geothermal fluids and its relationships with the tectonic regime (Emilia-Romagna Region, Italy) , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-11088, https://doi.org/10.5194/egusphere-egu21-11088, 2021.

Volcanoes are inherently unstable structures that spread and frequently experience mass wasting events (such as slope failure, rockfalls, and debris flows). Hydrothermal alteration, common to many volcanoes, is often invoked as a mechanism that contributes significantly to volcano instability. We present here a study that combines laboratory deformation experiments, geophysical data, and large-scale numerical modelling to better understand the influence of hydrothermal alteration on volcano stability. La Soufrière de Guadeloupe (France) is a hazardous andesitic volcano that hosts a large hydrothermal system and therefore represents an ideal natural laboratory for our study. Uniaxial and triaxial deformation experiments were performed on samples prepared from 17 variably-altered (alteration minerals include quartz, cristobalite, tridymite, hematite, pyrite, alunite, natro-alunite, gypsum, kaolinite, and talc) blocks collected from La Soufrière de Guadeloupe. Our uniaxial compressive strength experiments show that strength and Young’s modulus decrease as a function of increasing porosity and increasing alteration. Triaxial deformation experiments show that cohesion decreases as a function of increasing alteration, but that the angle of internal friction does not change systematically. We first combined recent muon tomography data with our laboratory data to create a 3D strength map of La Soufrière de Guadeloupe. The low-strength zone beneath the southern flank of the volcano exposed by our 3D strength map is coincident with the hydrothermal system. We then assigned laboratory-scale and upscaled mechanical properties (e.g., Young’s modulus, cohesion, and angle of internal friction) to zones identified by a recent electrical survey of the dome of La Soufrière de Guadeloupe. Numerical modelling (using the software LaMEM) was then performed on a cross-section of the volcano informed by the recent electrical data, and on a cross-section in which we artificially increased the size of the hydrothermally altered zone. Our modelling shows (1) the importance of using upscaled values in large-scale models and (2) that hydrothermal alteration significantly increases the surface velocity and strain rate of the volcanic slope. We therefore conclude, using models informed by experimental data, that hydrothermal alteration decreases volcano stability and thus expedites volcano spreading and increases the likelihood of mass wasting events and associated volcanic hazards. Hydrothermal alteration, and its evolution, should therefore be monitored at active volcanoes worldwide.

How to cite: Heap, M., Baumann, T., Rosas-Carbajal, M., Komorowski, J.-C., Gilg, H. A., Villeneuve, M., Moretti, R., Baud, P., Carbillet, L., Harnett, C., and Reuschlé, T.: The influence of hydrothermal alteration on volcano stability: a case study of La Soufrière de Guadeloupe (France), EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-151, https://doi.org/10.5194/egusphere-egu21-151, 2020.

Despite the well-established interest of Synthetic Aperture Radar data for volcanoes study and monitoring, their integration to operational monitoring activities in volcanoes observatories remains limited so far. We here describe the effort in progress to integrate in near real time the information derived from Sentinel-1 satellites into the monitoring devices at BBPTKG in charge of Merapi volcano survey as well as the use of Sentinel-1 data during the recent period of  unrest. Merapi (7°32.5’ S and 110°26.5’ E) located in the densely populated Province of Yogyakarta in Central Java is one of the most active volcanoes in Indonesia. The eruptive history of Merapi is characterized by two eruptive styles: 1) recurrent effusive growth of viscous lava domes, with gravitational collapses producing pyroclastic flows known as « Merapi-type nuées ardentes » (VEI 2); 2) more exceptional explosive eruptions of relatively large size (VEI 3-4), associated with column collapse pyroclastic flows reaching distances larger than 15 km from the summit. The eruptive periodicity is 4 to 5 years for the effusive events and 50 to 100 years for the explosive ones. The last explosive events (VEI 3-4) occurred in November 2010 and was followed by a period of limited activity. In August 2018, a new dome was observed inside the summit crater, thus marking the start of a new phase of effusive activity. A new period of unrest then started in mid-October 2020, characterized by an increase in seismic activity as well as large and localized displacements in the summit area. Magma finally reached the surface on 4  January 2021. Deformation is currently recorded by EDM and tiltmeters together with a network of 10 permanent GNSS stations. GNSS data are automatically processed and inverted for a pressure source at depth. Both displacement time series as well as spatial probability distribution are directly available through WebObs (Beauducel et al., Frontiers, 2020), an integrated web-based system for monitoring. Sentinel-1 data are acquired over the volcano every 12 days on descending track 76 and every 6 days on ascending track 127. Since mid 2017, Sentinel-1 data are automatically downloaded on a local server at BPPTKG. Interferograms and coherence images are then produced using the NSBAS processing chain (Doin et al, 2012) and automatically integrated to WebObs to enable detection of potential rapid and significant changes in signal. Mean velocity maps are also produced as well as time series of surface displacement at given location enabling direct comparison with GNSS measurements. The descending InSAR time series shows a strong displacement away from the satellite in a 1.5 km wide area located on the north-eastern part  of the crater. This signal became significant in September 2020. It is consistent with field measurements recorded and allows to map the affected area. In mid-November 2020, Sentinel-1 data thus provided the first information on the spatial extent of the ongoing surface displacements, which was useful for crisis management.

How to cite: Pinel, V., Beauducel, F., Putra, R., Sulistiyani, S., Nandaka, G. M. A., Nurnaning, A., Budi Santoso, A., Humaida, H., Doin, M.-P., Thollard, F., and Laurent, C.: Monitoring of Merapi volcano, Indonesia based on Sentinel-1 data, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-10392, https://doi.org/10.5194/egusphere-egu21-10392, 2021.

Sudden steam-driven eruptions at tourist volcanoes were the cause of 63 deaths at Mt Ontake (Japan) in 2014, and 22 deaths at Whakaari (New Zealand) in 2019. Warning systems that can anticipate these eruptions could provide crucial hours for evacuation or sheltering but these require reliable forecasting. Recently, machine learning has been used to extract eruption precursors from observational data and train forecasting models. However, a weakness of this data-driven approach is its reliance on long observational records that span multiple eruptions. As many volcano datasets may only record one or no eruptions, there is a need to extend these techniques to data-poor locales.

Transfer machine learning is one approach for generalising lessons learned at data-rich volcanoes and applying them to data-poor ones. Here, we tackle two problems: (1) generalising time series features between seismic stations at Whakaari to address recording gaps, and (2) training a forecasting model for Mt Ruapehu augmented using data from Whakaari. This required that we standardise data records at different stations for direct comparisons, devise an interpolation scheme to fill in missing eruption data, and combine volcano-specific feature matrices prior to model training.

We trained a forecast model for Whakaari using tremor data from three eruptions recorded at one seismic station (WSRZ) and augmented by data from two other eruptions recorded at a second station (WIZ). First, the training data from both stations were standardised to a unit normal distribution in log space. Then, linear interpolation in feature space was used to infer missing eruption features at WSRZ. Under pseudo-prospective testing, the augmented model had similar forecasting skill to one trained using all five eruptions recorded at a single station (WIZ). However, extending this approach to Ruapehu, we saw reduced performance indicating that more work is needed in standardisation and feature selection.

How to cite: Dempsey, D., Cronin, S., Kempa-Liehr, A., and Letourneur, M.: Broadening volcanic eruption forecasting using transfer machine learning, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-970, https://doi.org/10.5194/egusphere-egu21-970, 2021.

Geysers are characterized by regular eruptions of hot water fountains. Their internal system consists of a heat source at depth, an often complex crack system and a conduit linking it to the surface. The conduit and crack system is filled with water, steam and gases similar to a volcano. Bubble traps are sometimes and rarely mapped and alternative heat-driven models for geyser eruptions exist.

Using a multidisciplinary, dense and close network of video cameras, seismometers, water pressure sensors and a tiltmeter we studied pool geyser Strokkur in June 2018. These multidisciplinary observations and particle-motion based tremor locations enabled us to derive a schematic cross section describing the driving mechanisms and the fluid dynamic processes within the bubble trap, crack system and conduit. We imaged a bubble trap at 23.7+-4.4 m depth, 13 to 23 m west of the conduit. We divide the eruptive cycle into eruption, refilling of the conduit, gas accumulation in the bubble trap and a trail of bubbles from the bubble trap into the conduit where they collapse at depth and have gained novel insights in understanding the gas accumulation, migration and collapse in such hot geyser systems in different phases of the eruptive cycle.

The dataset of this experiment can be accessed here:

- Eibl, E. P. S., Müller, D., Allahbakhshi, M., Walter, T. R., Jousset, P., Hersir, G. P., Dahm, T., (2020) ' Multidisciplinary dataset at the Strokkur Geyser, Iceland, allows to study internal processes and to image a bubble trap.' GFZ Data Services. DOI: 10.5880/GFZ.2.1.2020.007

- Eibl, E. P. S.; Walter, T.; Jousset, P.; Dahm, T.; Allahbakhshi, M.; Müller, D.; Hersir, G.P. (2020): 1 year seismological experiment at Strokkur in 2017/18. GFZ Data Services. Other/Seismic Network. DOI:10.14470/2Y7562610816

How to cite: Eibl, E. P. S., Müller, D., Walter, T. R., Allahbakhshi, M., Jousset, P., Hersir, G. P., and Dahm, T.: Mapping the Bubble Trap Feeding Eruptions of Strokkur Geyser, Iceland, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-5089, https://doi.org/10.5194/egusphere-egu21-5089, 2021.

Monitoring the activity of a volcanic unrest in an archipelago is always a challenging task. Difficulties are even greater if we are also dealing with monogenetic volcanism, without a defined magma chamber, where each unrest can be related to a different magma intrusion, following different ascending paths towards an eruptive vent that can arise both on land or at sea. Moreover, if the repose time between eruptions is long, the historical eruptive record contains very few eruptions, and hence few data that allow an in-depth characterization of the dynamics of the volcanism in the area. 

This year marks the tenth anniversary of the beginning of the last eruption in the Canary Islands (submarine eruption of Tagoro volcano, 2011-2012). In this work we review the main difficulties, concerns and uncertainties that arose in the monitoring of this phenomenon. Some of these problems were solved during the crisis, throughout a multiparametric monitoring and the collaboration of different institutions; others would not be a major problem today, thanks to recent technological advances. On the other hand, there are still some unsolved monitoring difficulties when studying an event similar  to the one which lead to Tagoro volcano ten years ago. Part of the complexity is inherent to the spatial distribution of the islands in the archipelago and the limitations on the knowledge of the volcanic phenomenon. It is in these last challenges where the key to improve the volcano monitoring in oceanic islands is. 

How to cite: del Fresno, C., Felpeto, A., Domínguez Cerdeña, I., García-Cañada, L., Meletlidis, S., González-Alonso, E., Torres, P. A., Luengo-Oroz, N., Sainz-Maza, S., López-Díaz, R., Moure, D., and Casas, B.: The challenge of monitoring volcanic unrest processes in small oceanic islands: the case of Tagoro volcano (Canary Islands)  , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-11315, https://doi.org/10.5194/egusphere-egu21-11315, 2021.

Permutation Entropy (PE) has been suggested to be a promising tool for the prediction of volcanic eruptions. It is a robust yet simple tool to quantify the complexity of time series. The application has been used in the biomedical and econophysics fields and recently was adopted to find precursors of volcano eruptions and to identify tremor episodes. However, in the different eruption cases, the temporal variation of PE was found behaving in different ways. For example, a gradual drop of PE was observed few days prior to the 1996 Gjalp eruption while it remained high prior to the 2012 Copahua eruption. Our final aim is to quantify what features in the PE can be interpreted as eruption precursors and whether this is applicable to different eruptions from the same or different volcanoes. In calculating the PE, the determination of two key inputs, namely the delay time and the embedding dimension, is crucial as PE depends strongly on those parameters. Here we present several tests on different types of synthetic signals with different signal to noise ratios to determine the most suitable input parameters. We found that when the delay time is much shorter than or equal to the dominant period of the signal, the value of PE will be strongly influenced by the noise. Thus, the value of the delay time should be chosen in between. Furthermore, the embedding dimension should not be smaller than 5 to be able to identify the characteristic of the underlying signal. Finally, we show the application of this method to the seismic data during the dike formation and the effusive eruption at Holuhraun, Iceland, in 2014-2015.

How to cite: Sudibyo, M., Eibl, E., and Hainzl, S.: Testing the application of Permutation Entropy to characterize the precursory phase of volcanic eruptions, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-12299, https://doi.org/10.5194/egusphere-egu21-12299, 2021.

How well are our volcanoes monitored? When and why should we review and enhance the monitoring setup for volcano surveillance? These questions are often raised at Volcano Observatories or at those Institutions in charge of monitoring volcanoes and their associated hazards. The Icelandic Meteorological Office (IMO) is responsible for monitoring natural hazards in Iceland, including volcanoes and volcanic eruptions. IMO operates an extended multidisciplinary monitoring network which comprises seismometers, cGPS, gas sensors, MultiGAS and DOASes, hydrological stations, strainmeters and tiltmeters, infrasound networks and webcams, with the aim of detecting in a timely manner potential unrest at any of the 32 active volcanoes in the country. Limited resources and funding opportunities often pose limitations on how extensive (in terms of number of sensors and their variety) a volcano monitoring network can be. Therefore, the Volcano Observatories are often required to decide how to prioritize the monitoring needs and find a balance in sensitivity, reliability, and efficacy of the network.  

In this contribution, we will present the results of the analysis performed at the IMO to rank the Icelandic active volcanoes by their threat and, consequently, to prioritize their monitoring needs. Some criteria (based on eruption frequency, potential hazards, infrastructure exposure and current status) are defined as guidelines and they are used to drive decisions regarding when and how to alter the monitoring setup. The specific case of Hekla volcano is used here to evaluate the validity of such criteria and to perform an analysis of the current capability of issuing a timely warning for one of the most dangerous volcanoes in Iceland. 

How to cite: Barsotti, S., Parks, M., Melissa, P., Jónsdóttir, K., Vogfjorð, K., and Ófeigsson, B.: Ranking Icelandic volcanoes by threat and prioritizing their monitoring requirements , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15542, https://doi.org/10.5194/egusphere-egu21-15542, 2021.

Ground deformation is frequently one of the first detectable precursors alerting scientists to changes in behavior or the onset of unrest at active volcanoes. GNSS, InSAR, strain and tilt measurements are routinely used by volcano observatories for monitoring pre-eruptive, co-eruptive and post-eruptive deformation. In addition to monitoring signals related to magma migration, deformation observations are used as an input into geodetic modeling to determine the location and rate of magma accumulation and help define the structure of magma plumbing systems beneath active volcanoes.

This presentation will provide an update of how geodetic observations are being used in conjunction with seismicity and gas measurements, for the near-real time monitoring of key Icelandic volcanoes; to determine their current status, identify the onset and likely cause of unrest, locate magmatic intrusions, determine magma volumes and supply rates and assess the likelihood of eruption. An overview of the current status of the following active volcanoes will be provided: Hekla, Bárðarbunga and Grímsvötn, along with an update of the recent volcano-tectonic unrest on the Reykjanes Peninsula.

Hekla is one of the most active and dangerous volcanoes in Iceland with approximately 18 eruptions since 1104. Over the past few decades, Hekla erupted at almost regular ~10 year intervals, with the last four eruptions occurring in 1970, 1980–1981, 1991 and 2000. Previous geodetic studies have suggested magma storage at depths of 12-25 km directly beneath the volcanic edifice. However, recent InSAR analysis has detected a localized inflation signal to the west of the volcano. A regional borehole strain meter network has proven instrumental for real-time eruption forecasting at Hekla.

In the Bárðarbunga volcanic system, the six-month long effusive 2014-2015 Holuhraun eruption was accompanied by gradual caldera collapse of up to 65 m and was preceded by a two-week period of 48 km long lateral dyke propagation with extensive seismicity and deformation. Geodetic observations indicate that Bárðarbunga began to slowly inflate in July 2015. This may be explained by a combination of renewed magma inflow and viscoelastic readjustment of the volcano.

The Grímsvötn subglacial volcano is the most frequently erupting volcano in Iceland, with eruptions in 1998, 2004 and 2011. A GPS station shows a prominent inflation cycle prior to eruptions. Observations during the 2011 eruption suggest a pressure drop at a surprisingly shallow level (about 2 km depth) during the eruption, in a similar location as in previous eruptions. Deformation at this volcano has now surpassed that observed prior to historic eruptions and its aviation color code is currently elevated to yellow.

In December 2019, the Reykjanes Peninsula entered a phase of volcano-tectonic unrest characterized by seismic swarms, followed in late January 2020 by inflation detected in near-real time by GNSS and InSAR observations. At the time of writing (mid-January 2021) there have been three magmatic intrusions in the vicinity of Svartsengi, an intrusion beneath Krýsuvík and indications of magma migration at depth along the entirety of the Peninsula.

How to cite: Parks, M., Ófeigsson, B., Geirsson, H., Drouin, V., Sigmundsson, F., Hooper, A., Hreinsdóttir, S., Li, S., Ducrocq, C., Sturkell, E., Vogfjord, K., Hjartardóttir, Á. R., Fridriksdóttir, H. M., Þrastarson, R., Barsotti, S., Pfeffer, M., and Roberts, M.: The application of geodetic observations for near-real time monitoring of Icelandic volcanoes, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-14271, https://doi.org/10.5194/egusphere-egu21-14271, 2021.