Angela Saraò

In regions dominated by low-magnitude seismicity, obtaining reliable focal mechanisms (FMs) at the catalog scale remains challenging, particularly when waveform-based inversions are limited by simplified velocity models and network geometry. Here, we present an automated workflow for polarity-based FM determination that combines deep-learning first-motion classification using the convolutional first-motion model with probabilistic inversion through SKHASH, accounting for uncertainties in source location, velocity structure, and polarity assignment. Once event locations and P-wave arrivals are available, the workflow requires no manual polarity review and provides quantitative quality metrics for solution robustness. The workflow is validated against a reference focal-mechanism catalog of 159 earthquakes with manually reviewed polarities, used here as a validation dataset. Automated polarity classifications agree with manual picks in 92% of cases, and 73% of the resulting FMs differ by less than 30° in Kagan angle from the reference solutions, demonstrating reliable performance for small-magnitude events. We apply the workflow to approximately 1500 earthquakes recorded in northeastern Italy between 2024 and 2025 (1.0≤Md≤4.6), yielding 122 quality-controlled FMs, consistent with the established seismotectonic framework of the southeastern Alps, and producing the first automated FM catalog for this region. Nearly 75% of the retained solutions correspond to earthquakes with Md≤2.5, significantly densifying the available FM dataset at small magnitudes. An additional application to the 2025 Raveo earthquake sequence confirms the consistency of the automated solutions with independently derived moment tensor and manual focal mechanism results.

We present a workflow for the automatic detection of seismic events in volcanic areas, leveraging deep learning (DL) tools for the picking and association of seismic phases. Specifically, we employ PhaseNet and EQTransformer via the SeisBench open-source platform to perform automated picking of P- and S-wave arrivals from continuous seismic waveforms. To evaluate performance, we systematically test several pretrained models available within SeisBench using seismic data from Mount Etna (Italy), encompassing both volcano-tectonic (VT) and long-period (LP) events recorded from January 2019 to June 2020. Each model configuration is quantitatively assessed through statistical analyses, comparing results against existing seismic catalogs to identify the most effective approach for the Etna region. Our findings show that models trained on volcano-specific datasets significantly outperform those based on tectonic-only data. To further classify VT and LP signals, we apply a frequency-based algorithm using a modified frequency index, which effectively distinguishes VT, LP, low spectrum, mixed, and unclassified type events based on their spectral characteristics. This classification demonstrates strong agreement with the reviewed catalogs. Overall, the proposed workflow performs well in the complex setting of Mount Etna, highlighting the potential of pretrained DL models for semi-automated volcanic monitoring. At the same time, our results underscore the need for site-specific model validation and expert oversight to ensure reliable outcomes.

We present a comprehensive catalog of independent moment magnitude (Mw) estimates for northeastern Italy and its neighboring regions using data from the Istituto Nazionale di Oceanografia e di Geofisica Sperimentale (OX) network. The catalog, spanning from 2016 to 2023, covers the geographical area between 10.0°–14.5° north latitude and 44.5°–47.0° east longitude. In this region, we identified 11,685 low-to-moderate magnitude earthquakes (−0.70≤ML≤4.35) suitable for independent Mw computation. To enhance accuracy, we incorporated new homogeneous estimates of local magnitude (ML). A newly developed routine enables rapid near-real-time Mw estimation and permits regular, rolling seismic catalog updates. We also established an empirical relationship between Mw and ML specific to our study region. Our study provides a detailed and accurate Mw catalog of independent estimates that significantly improves critical applications such as seismic hazard assessment, and risk mitigation in this seismically active region. Regular updates will be made as new data become available, ensuring ongoing accuracy and relevance. The catalog is accessible online to the scientific community, fostering collaboration and further research, thereby enhancing our understanding of the geodynamics and seismic hazards in northeastern Italy and its surrounding areas.

Understanding what are the main characteristics of seismic risk communicationpractice and research is essential to depict best practices and gaps that can provideinsights for future improvements. Towards this task, and focussing on the Europeanframework, a scoping review based on the analysis of scholarly literature databases,was conducted. It reveals that, over the last 20 years, seismic risk communication hasbeen a research topic of increasing interest, trying to keep up with current riskcommunication trends and yet mostly under-researched. Recommendations frominternational disaster risk reduction frameworks show up also through the increasinginterest on the communication of seismic risk in Europe. However, it appears to bepracticed in an uneven way in the different European countries and not necessarilylinked to the level of hazard.It mostly occurs in the pre-crisis phase of the disaster lifecycle when risk awarenessand capacity to cope with hazards can be effectively built.An increasingly proactive, with an emphasis on a bottom-up strategy that relies onyouths to build the resilience of future generations is another key issue of thecommunication of seismic risk in the last 2 decades.Social media have had an increasing impact to provide timely and actionableinformation in times of crisis and to engage citizens, in the pre-crisis and post-disasterphase.Our data highlights that the future agenda for the communication of seismic riskshould be set on building trust with the public, tailoring communication to its needs.Actions are even more necessary to curb the spread of fake news and its negativeimpact on disaster management and build the communication practices on atheoretical background

Although earthquakes are a threat in many countries and considerable resources have been invested in safety regulations, communities at risk often lack awareness and preparedness. Risk communication is a key tool for building resilient communities, raising awareness, and increasing preparedness. Over the past 2 decades, seismic risk communication has evolved significantly. This has led to a reorientation from a predominantly “one-way”, top-down communication model to the promotion of new models in which people, their needs, and their participation in disaster risk management are central elements. The 2015–2030 Sendai Framework recommendations, recent disaster experiences and research have highlighted that new models can improve communication effectiveness. In this paper, we critically explore this transition by conducting a scoping review (n=109 publications) of seismic risk communication in Europe. We analyse the approaches, messages, tools, and channels used for seismic risk communication and how they have changed over time. The results reveal that the stated goals of seismic risk communication are, in decreasing order, to share information, raise awareness, change behaviours/beliefs, and increase preparedness. Pupils, students, and citizens are the primary recipients of communication activities. Over the years, two trends have emerged. First, “two-way”, transdisciplinary and bottom-up communication models prevailed over the “one-way” model. Second, communication aimed more at promoting proactive behaviours than just informing the public. Face-to-face, hands-on activities, and serious games are key tools to engage with the public. The results also reveal the emerging role of social media to target different audiences/social groups. Strikingly, only one-fifth of the analysed publications explicitly build on or tests risk communication theories. Future research could focus on comparing practices across countries and risks (e.g., earthquakes and floods) and on innovating communication theories and methodologies, especially by incorporating the role of information technologies and social media.

Rupture processes of large earthquakes have been studied by seismic waveform analysis and directivity effects have also been observed in moderate and small earthquakes. This effect leads to azimuthal and spectral variations in ground motion that can be used to estimate the fault plane orientation or a predominant rupture propagation direction in a particular region or during a seismic sequence. For moderate-to-strong events, directivity at low frequencies can result in potentially destructive pulses with large ground motions, while at high frequencies and for small-to-moderate events, the most pronounced effect is the shift in corner frequencies that results in high-frequency energy arriving in short time intervals. It is therefore of the utmost importance to estimate the directivity effects in engineering applications and seismological studies of earthquake sources. While some methods appear to work well for high magnitude earthquakes, determining directivity and source parameters for small to moderate magnitude earthquakes remains a challenge. One of the most common methods to estimate the directivity from moderate to small earthquakes relies on measuring the duration of the source pulse (the apparent source time function) at each location and then modeling it using a line source. Some approaches rely on the deconvolution of waveforms by an empirical Green’s function (eGf), to overcome the problems associated with the presence of path and site effects.A promising approach for estimating the rupture directivity effect and associated source properties is based on the calculation of the second seismic moments. In this study we apply the method based on the calculation of the second seismic moments to estimate the rupture process and source parameters to study a Mw 4.7 earthquake that occurred in central Italy during the 2016 - 2017 seismic sequence recorded by the RAN (Rete Accelerometrica Nazionale) and RSN (Rete Sismica Nazionale) italian networks.We first used synthetic apparent source time functions calculated from a geometric source model obtained from a real event to test the robustness of the method. Then, we applied the second-seismic moment method and the approach based on high-frequency S wave amplitude variations versus source azimuths analysis with an empirical Green's function deconvolution approach and compare the results.

Accurate quantification of seismic activity in volcanic regions is an important asset for improving hazard and risk assessment. This is especially true for densely populated areas, as in the case of Etna volcano (Southern Italy). There, the volcanic hazard is amplified by the seismic risk of active faults, especially on the eastern flank of the volcano. In such a context, it is common to rely on moment magnitude (Mw) to characterize seismicity and monitor the energy released during an eruption. In this study, we calculate the moment-based magnitude (M_W) for selected seismic data sets, using different approaches in distinct magnitude ranges to cover the widest possible range of magnitude that characterizes Etna's seismicity. Specifically, we computed the M_W from a dataset of moment tensor solutions of earthquakes that occurred in the magnitude range 3.4≤ M_L ≤ 4.8 during 2005-2020; we created a dataset of seismic moment and associated M_W for earthquakes 1.0 ≤ M_L < 3.4 obtained by analyzing source spectra; we fine-tuned two relationships, for shallow and deep earthquakes, to obtain M_W from response spectra. Finally, we calibrated a specific relationship between M_W and M_L for the Etna area earthquakes in the range 1.0 ≤ M_L ≤ 4.8. All the empirical relationships obtained in this study can be applied in real-time analysis of the seismicity to provide fast and robust information on the released seismic energy.

To ensure environmental and public safety, critical facilities require rigorous seismic hazard analysis to define seismic input for their design. We consider the case of the Trans Adriatic Pipeline (TAP), which is a pipeline that transports natural gas from the Caspian Sea to southern Italy, crossing active faults and areas characterized by high seismicity levels. For this pipeline, we develop a Probabilistic Seismic Hazard Assessment (PSHA) for the broader area, and, for the selected critical sites, we perform deterministic seismic hazard assessment (DSHA), by calculating shaking scenarios that account for the physics of the source, propagation, and site effects. This paper presents a DSHA for a compressor station located at Fier, along the Albanian coastal region. Considering the location of the most hazardous faults in the study site, revealed by the PSHA disaggregation, we model the ground motion for two different scenarios to simulate the worst-case scenario for this compressor station. We compute broadband waveforms for receivers on soft soils by applying specific transfer functions estimated from the available geotechnical data for the Fier area. The simulations reproduce the variability observed in the ground motion recorded in the near-earthquake source. The vertical ground motion is strong for receivers placed above the rupture areas and should not be ignored in seismic designs; furthermore, our vertical simulations reproduce the displacement and the static offset of the ground motion highlighted in recent studies. This observation confirms the importance of the DSHA analysis in defining the expected pipeline damage functions and permanent soil deformations.