Earthquakes occur regularly in the vicinity of La Réunion's two main hotspot volcanoes, Piton des Neiges and Piton de la Fournaise. While earthquakes at Piton de la Fournaise volcano are clearly linked to its volcanic activity, the seismicity beneath Piton des Neiges is not well understood. However, except during eruptive periods, we often record more seismic events at Piton des Neiges than at Piton de la Fournaise. This study aims to better capture this seismicity to understand the nature and causes of the activity beneath Piton des Neiges, a volcano that has been dormant for 27,000 years. We improve previously available seismicity catalog by using template matching, double relocation and focal mechanism determination. Results indicate that the seismicity is primarily concentrated on a northeast-dipping reverse fault located in the oceanic crust beneath the volcanic edifice. We also identify secondary seismicity clusters with the same orientation in the vicinity of the main fault. Seismicity occurs continuously since the installation of the first seismological stations in the vicinity of Piton des Neiges in 1999. Although occasional periods of increased swarm-like activity are observed in 2011 and 2018, they do not correlate with markers of deep magma transfers that are often observed prior to the eruptions of the Piton de la Fournaise. These variations in seismic activity are limited to the main reverse fault and might be associated with periods of creeping activity. Our findings suggest that the seismic activity beneath Piton des Neiges is likely caused by regional tectonic stress and edifice loading on pre-existing faults, rather than from deep magma transfers. This conclusion is supported by the presence of various reverse faults with similar orientation and the lack of correlation between seismicity fluctuations and deep magmatic activity.

How to cite: Firode, L., Duputel, Z., Ferrazzini, V., and Lengliné, O.: Seismicity under a dormant volcano: unveiling active crustal faulting beneath Piton des Neiges, La Réunion, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4933, https://doi.org/10.5194/egusphere-egu24-4933, 2024.

Geodetic observations indicate a seaward displacement of the eastern flank of Piton de la Fournaise volcano on La Réunion Island. Previous studies have suggested that the this displacement could be the result of a sheared sill. However, the location of the sill inferred from InSAR data is currently inconsistent with the distribution of earthquakes observed at greater depth. However, other results have attributed the current distribution of seismic activity beneath the eastern flank to the combined effect of overpressure in the magma chamber and loading of the volcanic edifice. The aim of the our study is to investigate the relationship between these two phenomena: the presence of active seismicity and the displacement of the eastern flank of Piton de la Fournaise. With this purpose, we enhance the characterization of seismic activity using template matching, double-difference relocation and focal mechanisms determination. We then explore the link between the spatio-temporal evolution of the seismicity and the displacement of the eastern flank. Additionally, we evaluate the impact of the velocity model to determine if we can reconcile hypocenter locations with the destabilization structure inferred from InSAR.

How to cite: Lengliné, O., Firode, L., Duputel, Z., and Ferrazzini, V.: Investigating the relationship between the presence of active seismicity and the displacement of the eastern flank of Piton de la Fournaise, La Réunion. , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5624, https://doi.org/10.5194/egusphere-egu24-5624, 2024.

Repeating earthquakes have overlapping rupture patches, similar focal mechanism and magnitudes. They are often detected based on highly similar waveforms using template matching techniques, which help to reconstruct complex sequences and swarms. Here, we investigate earthquakes with highly anti-correlated waveforms. Such poorly known observation implies the occurrence of reversed seismogenic processes at close hypocentral locations. We introduce the terms true and quasi anti-repeating earthquakes to denote cases affecting the same rupture patch or neighboring patches, respectively. We report about a number of observations of anti-repeating earthquakes in different environments, such as volcano, induced and intermediate-depth seismicity, and then review conceptual models to explain them. Some of these observations occurred during seismicity unrests, in the form of seismic sequences and swarms. Both true and quasi anti-repeating earthquakes are indicators for stress perturbation transients or local stress heterogeneities, often controlled by fluid migration processes. Therefore, their analysis may help the identification and tracking of fluids in the subsurface.

How to cite: Cesca, S., Niemz, P., Dahm, T., and Ide, S.: The role of anti-repeating earthquakes in seismic sequences and swarms, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-3876, https://doi.org/10.5194/egusphere-egu24-3876, 2024.

In January-February 2023 a seismic sequence took place in the Central Mediterranean Sea,  ~90 km south of the island of Malta, and ~200 km from the coast of Sicily. The seismicity started in Mid January, in a region that experienced only sparse seismicity in the past. In the following days several M4+ events occurred. The largest earthquake, with moment magnitude Mw 5.3, took place a couple of weeks after the unrest onset, on January 30. The seismicity continued for several weeks, before fading down. Later seismicity was observed and still going on. This recent, unusual seismicity offers a unique opportunity to investigate seismogenic processes in this region with unprecedented detail. However, analyzing seismic sequences in offshore environments presents significant challenges due to the absence of optimal seismic monitoring conditions. These limitations compromise the effectiveness of conventional data analysis techniques, hindering the characterization of offshore seismic sequences. We tackled these limitations through the adoption of advanced, waveform-based seismic data analysis techniques that allow to investigate offshore seismic sequences, with the aim to provide insights into their origin. We combine full-waveform based detection and template matching methods to enhance the detection of events, advanced location techniques based on Distance Geometry Solvers (DGS), and probabilistic waveform-based methods for seismic source characterization. We combine the seismic source analysis for the 8 largest earthquakes in the sequence, with magnitude exceeding ML 4.5, with waveform-based and statistical analysis of the seismicity. About 500 events are identified. Their locations map a narrow lineament extending ~NW-SE. Full moment tensors for the largest events identify normal faulting mechanisms with a similar orientation, and shallow centroids of ~5 km depth. This result, combined with a waveform similarity analysis, suggests a predominant mechanism for the entire sequence. Using different seismicity indicators we classify the 2023 sequence as a seismic swarm. Indeed, the largest events in the sequence occur weeks after the unrest onset. Compared to previous seismicity, the sequence was outstanding in terms of maximum magnitude, seismicity rate and moment rate. While normal faulting earthquakes are not unusual in the Central Mediterranean, they differ from the few focal mechanisms previously proposed for the swarm focal region, which has important implications, considering that normal faulting earthquakes at shallow depth pose a tsunami hazard in the region.

How to cite: Grigoli, F., Cesca, S., Bocchini, G. M., D'Amico, S., and Agius, M.: High-Resolution Analysis of the 2023 Seismic Swarm Offshore Malta, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9984, https://doi.org/10.5194/egusphere-egu24-9984, 2024.

In this project I am trying to understand the mechanics of the 2007 Marianas earthquake-swarm region and surrounding vicinity; located in the asthenosphere and lithosphere of the Marianas microplate above the subducting Pacific plate. The earthquake swarm exhibited seismicity for almost 2 years, with events occurring from depths of 300 km and up to the surface, right underneath the Marianas arc currently-active basaltic volcanoes.
My approach is to estimate the magnitudes of mechanical variables (forces per unit area, forces per unit volume) that control the evolution of this system. These forces are evaluated for the swarm region, where the ambient rock is heavily fractured and intruded by migrating magmas, thus altered and weakened. From continuum-mechanics momentum equation, the specific terms under consideration are the elastic, viscous and inertial forces. These are evaluated as characteristic magnitudes, following approximated scaling forms of their analytical expressions. Additionally and alternatively, I am carrying out separate assessments of the viscous and elastic forces using visco-elastic (VE) theory (Kelvin-Voigt and Maxwell VE models).
For comparison, I am also carrying out the corresponding estimates for Japan’s Mt. Yake (1998) volcanic earthquake swarm, and also, for a typical continental crustal seismogenic environment.  
Importantly, the deformation of the system is very different when-and-where earthquakes are occurring, than at other times and locations. For this reason, my assessment distinguishes  two regimes: (i) short length-scales and short duration Co-Seismic, and (ii) long-length-scale and long duration A-/Inter-Seismic background. It is for these two time frames that I estimate the stress-strain and forces per unit volume. Additionally, associated energies and energy density-rates are also estimated, and finally the energy budget of the co-seismic and inter-seismic frameworks are studied.

How to cite: orellana-rovirosa, F. and phipps morgan, J.: Understanding the Marianas’ 2007 volcanic earthquake swarm: A perspective from fundamental quantities, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13642, https://doi.org/10.5194/egusphere-egu24-13642, 2024.

Armenia, SE Turkey and NW Iran are located in the central part of the Arabian lithosphere collision zone, a region, which experiences N–S shortening and E–W extension accompanied by intense faulting, strong earthquakes and active volcanism. The Gegham volcanic ridge (GVR) is located in the center part of the Neogene-Quaternary volcanic belt formed within the territory of the Armenian Highland. The duration of volcanism within the Gegham Ridge spans from the Late Miocene to the Holocene. The GVR in central Armenia represents one of the densest clusters of individual monogenetic volcanoes in the world.

The study area of the Gegham Volcanic Ridge is located between the system of the Gegham Ridge Fault (GF1) and the Gavaraghet Fault (GF2).

The faults in the axial part of the Gegham ridge (GF1) represent structures of fracturing related to eruptions of numerous Quaternary volcanoes. The Gavaraghet Fault (GF2) related to a few historical and recent earthquakes is the most active one in the Gegham Fault system.

During the period of 2014-2018, earthquake swarms were recorded in the area of the Gegham Volcanic Ridge. Further studies of the relationship between tectonic and volcanic processes will contribute to the assessment of the volcanic hazard for this area. Beyond its importance for natural hazard assessment, the volcanically young Gegham Ridge also holds great potential as a source of geothermal energy.

Seismic activity in the study area was recorded using data from the advanced and dense seismic network installed recently. This network equipped by full-broad band (permanent and temporary) stations has been developed since 2012 and covers both the study region and adjacent areas.

Seismic activity of this area is manifested in the form of small-size earthquakes. Earthquake epicenters tend to concentrate in volcanic vents, as well as near active faults areas. The depths of earthquakes are up to 25 km.

In this study, source mechanisms of earthquakes were investigated using digital waveform data recorded by seismic stations of the Armenian seismic networks (included the 3 temporary seismic stations). The focal mechanisms of a set of earthquakes that occurred within 2022-2023 were constructed with high reliability, based on the polarity of the first motion of the P-wave.

All data were relocated using local and regional seismic network data with the aim to reduce main parameter uncertainties in the catalogue, used in the current study. 

How to cite: Sargsyan, L., Sahakyan, E., Demirchyan, H., Navasardyan, G., Gevorgyan, M., and Meliksetian, K.: Studies of recent seismo-volcanic activity in the Gegham volcanic ridge (Armenia) , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-14803, https://doi.org/10.5194/egusphere-egu24-14803, 2024.

The use of template matching to detect previously missed small earthquakes is widespread in seismology, due to its power in searching for similar signals. The newly detected earthquakes improve understanding of the geology, seismo-tectonics and seismogenesis of the area explored. The usefulness of template matching has sparked the development of many software tools (e.g. QuakeMatch; Toledo et al., 2024) that allow seismologists to easily apply them to their area of interest. 

Like every detection technique, the performance of template matching shows a tradeoff between sensitivity and false detection rate that is dependent on the choice of the detection threshold. The value of the correlation coefficient between two earthquake signals is highly dependent on several properties such as the distance of the earthquake from the station, the noise level at the station, the magnitude of the earthquake, the focal mechanism of the earthquake, etc, making the selection of a specific correlation coefficient threshold hard. To detect a larger number of earthquakes, researchers often use a lower correlation coefficient detection threshold and manually inspect the detected events to classify them as true events (Toledo et al., 2024). This is, however, a tedious task, especially when using a large number of templates. To reduce the human workload, which can be especially important during evolving earthquake sequences, we employ a Convolutional Neural Network (CNN) to discriminate between earthquakes and noise using the template matching detections as input. We use the data from several microearthquake natural and induced Swiss sequences (Simon et al., 2024, in prep.), to train and test the developed CNN model. Our CNN model uses single-station 3-component waveforms of any length and outputs an earthquake detection score. We demonstrate that the developed CNN can be used to significantly reduce the human workload with high accuracy, allowing the use of low correlation coefficient value thresholds for template matching detections. Furthermore, we show the implementation of the method inside the QuakeMatch software (Toledo et al., 2024).

How to cite: Jozinović, D., Toledo, T., Simon, V., and Kraft, T.: Improving template matching detections using a Convolutional Neural Network, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13232, https://doi.org/10.5194/egusphere-egu24-13232, 2024.

In the realm of seismic and microseismic event detection and localization, our research marks a significant step forward by integrating machine learning with advanced waveform-stacking techniques for detecting, locating and characterising seismic events. This integration is crucial for unravelling the complex spatio-temporal patterns of seismicity sequences. Our study addresses the challenges posed by noise-dominant microseismic events, which are typically overlooked by conventional detection methods.

Building upon the foundational work on migration and stacking, we have developed an automated, data-driven method utilising a neural network trained in seismic phase arrival identification. This approach, underpinned by stacking and migration techniques, is enhanced by the incorporation of a spatial octree to precisely and efficiently localise seismic sources. These enhancement gives insights into complex seismic sequences, such as volcanic swarms and regional tectonic sequences.

The software framework facilitates extensive feature extraction, such as local and moment magnitudes, enabling the study of seismic events across various scales and tectonic settings. This is exemplified in our validation studies using data from the Eifel region, Germany, and the Reykjanes Peninsula, Iceland. These regions, known for their diverse seismic activities including tectonic earthquakes and fluid-induced swarm activity, provide a rich dataset for testing our method's efficacy in different geological contexts.

Our research contributes to the session's overarching goal of understanding the physical processes behind complex seismic sequences. By enhancing detection and localization capabilities, we aim to offer new perspectives and tools for the geophysical community to investigate the triggering mechanisms of these sequences with unprecedented resolution.

How to cite: Isken, M. P., Dahm, T., Heimann, S., Münchmeyer, J., Cesca, S., and Niemz, P.: Advancing Seismic Event Detection: Integrating Machine Learning with Waveform-Stacking Techniques, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18706, https://doi.org/10.5194/egusphere-egu24-18706, 2024.

The Pollino area, one of the largest seismic gaps in Italy, was struck between 2010 and 2014 by a long-lasting seismic sequence. More than 10,000 small-to-medium earthquakes occurred as a swarm-like sequence and, to a lesser extent, as  aftershocks following the two largest events: a M_L 4.3 on 28 May 2012 and a M_L 5.0 on 25 October 2012. A slow slip event began about three months before the strongest earthquake. 

Our study focuses on this complex sequence and the recent advancements obtained by our group in terms of crustal structure characterization and fault imaging. We integrate the most recent findings in terms of 3D scattering and absorption imaging, high sensitivity to fluid content, deformed fractured structures, and impermeable layers, with already achieved seismic and focal mechanism tomographic results and available geological information for the area. 

High absorption topping the western Pollino seismic volume appears pressurized between the low-Vp/Vs and low-scattering San Donato metamorphic core and a deep basement. This high absorption volume is also characterized, at the same depth, by an excess of fluid pressure, mapped by applying the focal mechanism tomography, where clusters of events of similar waveforms occurred. These events were caused by a slow slip event, similar to the transient deformation event, and favored by pore-pressure increases in fluid-saturated fault networks. 

This work was supported by the PRIN-2017 MATISSE project (no. 20177EPPN2), funded by the Ministry of Education and Research.

How to cite: Napolitano, F., Amoroso, O., De Siena, L., Gabrielli, S., and Capuano, P.: Elastic and inelastic properties of the Pollino (Italy) seismogenic volume, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8106, https://doi.org/10.5194/egusphere-egu24-8106, 2024.

The analysis of the precursory behavior has been growing in the last decades around the paradigm of earthquake prediction. Amongst the plethora of precursors, the b-value has been deemed as a proxy for the state of stress on a seismogenic source. On 6th February 2023 a mainshock of Mw 7.8 followed by another event of Mw 7.5 within a few hours of difference hit Kahramanmaraş region. Both events occurred on different fault zones of the East Anatolian Fracture Zone (EAFZ). While the first one struck the Pazarcik segment on the main segment of the EAFZ, the second event did it on the Sürgü-Misis fault zone at the west of the EAFZ. In a context of complex rupture behavior, a stress redistribution and transient stress caused by the first event are two of the likely triggering mechanisms of the second event. Our b-value analysis relies on the earthquake catalog from AFAD in the period from early 2015 till the previous moments of the first event on February 2023. To homogenize magnitude reported we excluded Md (few events) and convert Ml to Mw using an empirical relation. As the seismicity along the EAFZ is high, and to avoid statistical bias caused by earthquake clusters, we declustered the earthquake catalog. After assessing the magnitude of completeness to guarantee that the catalog gathers the minimal quality requirements, we examined the temporal evolution and the spatial distribution of b-value. Concerning temporal evolution, we detected that 40 days before the mainshock a sudden decrease of the b-value from 0.85 until its minimum 0.5 on the event’s day. With regard to the spatial distribution of b-value, mainshock’s epicenter is on an elongated region with general strike NE-SW and minimum b-value of 0.76.  

Work supported by the Portuguese Fundação para a Ciência e a Tecnologia (FCT) I.P./MCTES through national funds, UIDB/04683/2020 (ICT) and UIDP/04683/2020 (ICT)

How to cite: Fontiela, J. and Seivane, H.: b-value variations preceding the mainshock of the 2023 Kahramanmaraş earthquakes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17086, https://doi.org/10.5194/egusphere-egu24-17086, 2024.

This study delves into the phenomenon of dynamic triggering of earthquakes in Yunnan, China, a region renowned for its abundant geothermal activity. Through an extensive analysis spanning from 2006 to 2021, we unveil the impact of 13 distant M>6 earthquakes on seismic clusters in the region, emphasizing the unique clustering of these seismic events at specific fault-related locations. Advanced methods, including the Epidemic-Type Aftershock Sequence (ETAS) model, were employed to identify the spatiotemporal patterns of seismic activity before and after these distant M>6 earthquakes.

Noteworthy observations highlight the preferential distribution of earthquake clusters at specific fault-related locations, such as fault ends, bends, intersections, and fault step-overs. Some earthquake clusters exhibit clear fluid diffusion processes, validated by increased water temperature in nearby wells. The applied ETAS model underscores a high proportion of forced seismic activity, elucidating the subtle relationship manifested as delayed triggering effects.

The results of our study emphasize the association of dynamic triggering with specific fault-related locations, emphasizing the potentially significant role of subsurface geothermal fluids in this process. This research deepens our understanding of seismic activity patterns in the Yunnan region, revealing the intricate interplay between distant M>6 earthquakes, fault dynamics, and geothermal fluid activity.

How to cite: Wang, Z., Lei, X., and Ma, S.: Dynamic Triggering of Earthquakes in Yunnan, China: Insights into the Influence of Distant M>6 Earthquakes and Geothermal Fluids, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2644, https://doi.org/10.5194/egusphere-egu24-2644, 2024.

The occurrence of fluid-induced, moderate-to-large earthquakes in several locations around the globe sparked interest in the relationships between fluids and seismicity over the last few years. Several studies suggest variations of the stress state of rocks, due to the increase or drop of the pore fluid pressure, can be a mechanism that can trigger earthquakes in the presence of fluid phases. In this scenario, the Val d’Agri represents a precious case study where the effect of fluids on seismic activity can be observed. In this region, wastewater reinjection reactivated the Costa Molina blind thrust in the eastern sector of the Val d’Agri, where present-day seismicity was almost absent. A few kilometers SW from this cluster, seasonal water loading from the artificial Pertusillo reservoir generates further seismic activity within the buried carbonatic platform. The formation and evolution of the faults generating seismicity are still a matter of debate, especially in the compressional/extensional tectonic setting that characterizes the southern Apennines geological history. Consequently, the distribution of the seismic potential in the region is largely unconstrained.

We built up a numerical, thermo-mechanical model to identify the principal mechanisms that generated the present-day tectonic setting observed in the Val d’Agri and the surrounding region, and to assess the seismic hazard characterizing this area. We suggest the presence of a major dècollement layer that decouples deformation between the sedimentary cover and the crystalline basement, represented by the Triassic Burano Formation. Our model quantifies the stress field and estimates Coulomb stress values in the Val d’Agri crust, allowing us to assess the potential of the rocks to generate earthquakes. We suggest that Coulomb stress values are positive in a large part of the crust, and therefore that fluid injection may be particularly effective for the reactivation of buried structures, especially within the carbonatic platform at a depth between 2 and 6 km.

How to cite: Lavecchia, A., Serlenga, V., Filippucci, M., Stabile, T. A., Prosser, G., and Tallarico, A.: Fault (re)activation and fluid-induced seismicity: an example from the Val d'Agri intermontane basin (southern Italy), EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12026, https://doi.org/10.5194/egusphere-egu24-12026, 2024.