We present the IMAGMASEIS project, a large-N seismic experiment carried out on La Palma (Canary Islands, Spain) between 2023 and 2024, aimed at high-resolution imaging of the crustal and upper mantle structure using passive seismic methods. The project involved the deployment of 235 temporary broadband and short-period seismic stations, supplementing 21 permanent stations, thus creating the densest seismic network ever installed on the island. The main goal is to characterise the magmatic plumbing system beneath Cumbre Vieja volcano, identify magma accumulation zones, and investigate structural changes related to the 2021 Tajogaite eruption. We describe the experimental design, network configuration, instrumentation, deployment strategies, and challenges encountered, including difficult terrain and logistical constraints. Preliminary results demonstrate the potential of the dataset for ambient noise tomography, receiver function analysis, and local earthquake studies. IMAGMASEIS provides a valuable resource for understanding volcanic and tectonic processes in oceanic island settings and serves as a model for cost-effective, high-density seismic deployments in similar environments.

Anticipating volcanic eruptions remains a challenge despite significant scientific advancements, leading to substantial human and economic losses. Traditional approaches, like volcano alert levels, provide current volcanic states but do not always include eruption forecasts. Machine learning (ML) emerges as a promising tool for eruption forecasting, offering data-driven insights. We propose an ML pipeline using volcano-seismic data, integrating precursor extraction, classification modeling, and decision-making for eruption alerts. Testing on six Copahue volcano eruptions demonstrates our model's ability to identify precursors and issue advanced warnings pseudoprospectively. Our model provides alerts 5-75 hr before eruptions and achieving a high true negative rate, indicating robust discriminatory power. Integrating short- and long-term data reveals seismic sensitivity, emphasizing the need for comprehensive volcanic monitoring. Our approach showcases ML's potential to enhance eruption forecasting and risk mitigation. In addition, we analyze long-term geodetic data (Interferometric Synthetic Aperture Radar and Global Navigation Satellite System) to assess Copahue volcano deformation trends, in which we notice an absence of noteworthy deformation in the signals associated with the six small eruptions, aligning with their small magnitude.

This study focused on seismic event detection in a volcano using machine learning by leveraging the advantages of software/hardware co-design for a system on a chip (SoC) based on field-programmable gate array (FPGA) devices. A case study was conducted on the Copahue Volcano, an active stratovolcano located on the border between Argentina and Chile. Volcanic seismic event processing and detection were integrated into a PYNQ-based implementation by using a low-end SoC-FPGA device. We also provide insights into integrating an SoC-FPGA into the acquisition node, which can be valuable in scenarios where stations are deployed solely for data collection and holds the potential for the development of an early alert system.

Abstract Understanding seismic tremor wavefields can shed light on the complex functioning of a volcanic system and, thus, improve volcano monitoring systems. Usually, several seismic stations are required to detect, characterize, and locate volcanic tremors, which can be difficult in remote areas or low-income countries. In these cases, alternative techniques have to be used. Here, we apply a data-reduction approach based on the analysis of three-component seismic data from two co-located stations operating in different times to detect and analyze long-duration tremors. We characterize the spectral content and the polarization of 355 long-duration tremors recorded by a seismic sensor located 9.5 km SE from the active vent of Copahue volcano in the period 2012–2016 and 2018–2019. We classified them as narrow- (NB) and broad-band (BB) tremors according to their spectral content. Several parameters describe the characteristic peaks composing each NB episode: polarization degree, rectilinearity, horizontal azimuth, vertical incidence. Moreover, we propose two coefficients $$C_P$$ C P and $$C_L$$ C L for describing to what extent the wavefield is polarized. For BB episodes, we extend these attributes and express them as a function of frequency. We compare the occurrence of NB and BB episodes with the volcanic activity (including the level of the crater lake, deformation, temperature, and explosive activity) to get insights into their mechanisms. This comparison suggests that the wavefield of NB tremors becomes more linearly polarized during eruptive episodes, but does not provide any specific relationship between the tremor frequency and volcanic activity. On the other hand, BB tremors show a seasonal behavior that would be related to the activity of the shallow hydrothermal system. Graphical Abstract

The characterization of dominant frequencies is essential for tracking significant temporal variations during and between tremor episodes. In this work, we proposed a method for quantifying the dominant frequencies and their attributes. We applied it to characterize three tremor episodes recorded several hours before ash emissions at Copahue volcano during June-August 2020. The method consisted in (i) extracting dominant peaks (relative maximum in successive PSD) and their polarization attributes: polarization degree, rectilinearity, and polarization angles; and (ii) extracting dominant frequencies (i.e., relative maximums in the PDF of dominant peaks) and their polarization attributes. We applied the method at three stations located at 4.5, 9.5, and 11 km from the crater vent to investigate how the characteristics for dominant frequencies change among the stations. We found linearly polarized peaks in the three stations whose azimuth coincides with crater direction. The station closest to the crater (NAN5) recorded linearly polarized frequencies above 3.0 Hz. The time evolution of dominant peaks suggests that different sources compose the seismic tremor. Our results represent a forward step in the understanding of eruptive tremors of Copahue, being a clear example that their implementation at other volcanoes can improve the monitoring tasks.

Cerro Domuyo, in northwestern Neuque?n province of Argentina, shows notable geothermal activity, although it is located at a considerable distance from the actual volcanic arc. Many studies have been developed in this area with the aim of investigating its geothermal field. Despite these efforts, the characterization of the dynamic activity in the area is still poorly known. This work shows the results of a network of seismological stations, which registered volcano-tectonic, long-period and hybrid events. A large number of volcano-tectonic events concentrated mostly in the Cerro Domuyo geothermal area were detected. These events can be divided into two groups, those that occurred at shallow depths below the geothermal area and those at greater depths below the high region of Cerro Domuyo. Shallow events were located around the Manchana Covunco fault and many of them were particularly clustered at its intersection with the Humazo fault. Seismicity in the area shows that these faults are continually active due to fluid movement, as was evidenced by the hydrothermal manifestation of the Humazo in 2003. Moreover, new studies provide evidence that the Cerro Domuyo is experiencing an important inflation caused by a magmatic body. Deep volcano-tectonic events are evidencing this activity. Considering the large distance between the actual volcanic arc and the study region, and the high density of shallow volcanotectonic events, it is highly likely that this magmatic body is increasing the geothermal activity. Additionally, the aeromagnetic anomaly over Cerro Domuyo is showing a thin magnetic crust of less than 6 km, suggesting a clear link between the geophysical results and the inflation in Cerro Domuyo.

Fil: Hantusch, Marcia. Universidad Nacional de Rio Negro. Instituto de Investigacion en Paleobiologia y Geologia. Rio Negro, Argentina.

Improving the ability to detect and characterize long-duration volcanic tremor is crucial to understand the long-term dynamics and unrest of volcanic systems. We have applied data reduction methods (permutation entropy and polarization degree, among others) to characterize the seismic wave field near Copahue volcano (Southern Andes) between June 2012 and January 2013, when phreatomagmatic episodes occurred. During the selected period, a total of 52 long-duration events with energy above the background occurred. Among them, 32 were classified as volcanic tremors and the remaining as noise bursts. Characterizing each event by averaging its reduced parameters, allowed us to study the range of variability of the different events types. We found that, compared to noise burst, tremors have lower permutation entropies and higher dominant polarization degrees. This characterization is a suitable tool for detecting long-duration volcanic tremors in the ambient seismic wave field, even if the SNR is low.