Istituto Nazionale di Geofisica e Vulcanologia (INGV), Sezione Roma1, Roma, Italia

On September 18, 2023, an earthquake with a magnitude of Ml=4.8 (Mw=4.9) occurred a few kilometers SW of Marradi (FI) at a depth of about 8 kilometers. The computed TDMT solution of the mainshock suggests a normal fault oriented NW-SE (Scognamiglio et al., 2006). The earthquake, preceded by a foreshock with a magnitude of Ml=3.3 (Mw=3.4), triggered a seismic sequence characterized, in the first two months, by approximately 700 aftershocks localized by the staff on duty in the Seismic Monitoring Room of the INGV in Rome, including 6 events with a magnitude of Ml≥3.0 occurred in the first two days. The sequence occurred in a high seismic hazard region. The two closest historical earthquakes occurred in the Mugello area about 30 km SW of Marradi (and about 25 km north of Florence): one, whose magnitude (Mw) is estimated to be about 6.0, occurred on June 13, 1542 while the other with estimated magnitude (Mw) of about 6.4 occurred on June 29, 1919. The second one is among the strongest (most significant) Italian earthquakes of the 20th century, and also one of the strongest known to date with its epicentre in the northern Apennines. The affected area was that of Mugello, with extensive damage both in the province of Florence and on the Romagna side of the Apennines. The analysts of the BSI (Italian Seismic Bulletin) reviewed the initial three days of the sequence, paying special attention to the hours directly following the mainshock. The BSI work mainly consists in revising the picking of P and S phases and assigning them appropriate weights, retrieving previously unused phases, and evaluating the maximum amplitudes necessary for calculating the value of Ml. The latter is a critical aspect of the initial phases of a seismic sequence; in fact, the occurrence of events is very close in time, making it challenging to estimate the maximum amplitudes and, consequently, the magnitude automatically. Through this analysis, they have identified an earthquake with a magnitude of Ml=3.4, occurring approximately one minute after the mainshock and overlooked during the surveillance service. Furthermore, a comprehensive effort to recover smaller seismic events not initially analyzed in the Seismic Monitoring Room resulted in the localization of 498 earthquakes, nearly a 30% increase within the first three days. In Figure 1(a-d), hypocentral parameters and time readings of the 352 earthquakes detected in the Seismic Monitoring Room have been compared with those of the same events revised by the BSI. It is evident as both the horizontal and vertical errors, as well the seismic gap associated with the location decrease for the dataset analyzed from the BSI, while the number of P and S phases increases for the same dataset. Subsequently, the events revised from BSI were initially relocated by applying the NonLinLoc code (Lomax et al., 2000) using a 1D regional velocity model from Pastori et al. (2019). Following this, a double-difference technique (Waldhauser and Schaff, 2008) was applied to improve the geometries of the activated structures (see Figure 2). Concomitantly, an analysis using the template matching technique was applied, for the period from September first to October 10th, to identify events overlooked by the Earthworm system in the Seismic Monitoring Room. The results are shown in Figure 3. It was found that the number of detections increased by approximately 60%. Furthermore, the figures 3c and 3d put in evidence as the larger number of new detected earthquakes is characterized by lower magnitude. This result highlights the value of integrating this type of analysis to complement the efforts of the Italian Seismic Bulletin (BSI).

Bare twenty years into the XXI century – and what a treat. Damaging earthquakes with regional impact, climate extremes disrupting weather cycles, water shortages in high-income regions, scarcer (and costlier) energy and mineral resources, rising population. Add a slice of global geopolitical instabilities – even where one would never expect to report them from. And, well, why not: a novel pathogen, so little yet so commanding that the world is still vying with it. Natural hazards and anthropogenic factors interact in multiple ways and across various scales, close or afar, in time and space. They interweave a web of complexities that can appear deceitful, capricious, or otherwise overwhelming to the citizens of contemporary societies – even in statistically affluent and educated ones. There comes the role of geosciences, from paleontology to high-atmosphere physics, from energy to oceanography, from the solid to the not-so-solid earth. There comes their transformative, instrumental task – as new and as pressing as ever. Geosciences are not (and will not) what they used to be, bound as they are to glean lessons learned from the past to provide insight into the future. Geoscientists were once thought to study ancient rocks, fiddle with very slow-moving tectonic plates, and bantering about invisible earth’s features, too large, or too deep, or too far away to even imagine for us earthlings. But this is no longer the case – and maybe never has been. At the core of geosciences’ interests lies Nature, for what it is – with all its grand size, seemingly slow processes that unveil sudden effects, complex interactions among forces and bodies across distances and time. These prove to be paramount tools to probe a world perceived as inscrutable, increasingly richer in risks and poorer in resources. Therefore, tools of yesterday’s intellectual quests prove instrumental to decipher tomorrow’s societal issues, such as: - The long records of natural events (hazards) - Far-flung origins (our solar system and the universe) - Far-reaching effects (feedback, periodicity, and recurrence times) - Need to forecast (or at least account for) the irregular behaviors of modern phenomena (not always known or detectable by current means). The knowledge of compounded risks of natural origin provides an outlook on where and what to call for enduring communities. This applies also to risks resulting from interaction among natural events and anthropogenic components. Since natural phenomena embed complexities due to multiple variables and intrinsic feedback, interaction among natural and non-natural ones brings novel issues, requiring a remarkably broad outlook – global and beyond. The natural consequence is then to envision natural risks against population distribution, spatial extents of natural resources, size, and time window of induced effects. Picking a selection of examples, this talk thus tries to put into perspective: 1. Hazards stemming from multiple, at times unpredictable sources; 2. The precious role of geosciences to decipher them – and to forecast them; 3. The complexity of natural hazards, the flexibility of human planning; 4. Modern issues challenging societies and economies – today, tomorrow, and thereafter.

We present the results of an ongoing research for assessing tsunami risk perception in southern Italy. The study is motivated by the need of addressing a sound communication strategy for tsunami risk reduction, related to the activities of the Tsunami Alert Centre (CAT) of INGV, operating within the framework of the Italian civil protection system. The area of the second step of this study includes five regions of Italy (Basilicata, Calabria, Molise, Puglia, Sicily), facing on Adriatic, Ionian and Tyrrhenian seas, located in one of the most hazardous areas of the Mediterranean. In all the area the memory of relevant tsunamis is loose, since the last destructive event dates back to 1908 (due to the Messina-Reggio Calabria M~7 earthquake). The main goal of this study is to verify how people’s perception of tsunami risk compares with the hazard assessed by scientific data, and which are the main factors controlling people’s knowledge and awareness. We analysed a sample of more than 1,600 interviewees representing about 4 million people living in the coastal municipalities of the five considered regions. Results show that risk perception appears to be generally low, with significant differences among different areas, likely due to the the time elapsed since the last events. The survey results for the first two investigated regions (Calabria and Puglia, see Cerase et al., NHESS, 2019) showed that people’s perception and understanding of tsunamis are affected by media accounts of the mega-tsunamis of Sumatra 2004 and Japan 2011. At the same time, the risk posed by small tsunamis is basically underrated or neglected, posing some critical questions for risk mitigation strategies, particularly in touristic areas. Furthermore, the survey’s results show that for lay people the word ‘tsunami’ has a different meaning with respect to the Italian traditional word ‘maremoto’, implying that the same physical phenomenon would be understood in two different ways by younger, educated people and elders with low education level. In addition, people have high expectations from authorities, CPAs, research institutions about warnings. Moreover, living in different coastal areas appears to have a significant influence on the way tsunami hazard is perceived: Interviewees of Tyrrhenian Calabria are more likely to associate tsunami risk to volcanoes with respect to those living in the Ionian coastal areas, coherently with the presence of Aeolian volcanic islands and feared submarine volcanoes in the Tyrrhenian. A somehow unexpected result is that TV emerges as the most relevant source of knowledge for 90% of the sample. Some categories declared to prefer getting early warnings through broadcast media and sirens rather than receiving by SMS or apps, suggesting the need for redundancy and modulation of EW messages. We will present an update of the survey which is presently ongoing, related to the five regions. These results could help in addressing risk communication and mitigation policies.

Seismic attenuation is generally thought to be a constant, or a simple monotonic function of frequency, and generally not a function of time. Examples of exceptions include attenuation enhancement due to shallow earthquake-induced damage, and fluctuations due to fluid diffusion. In reality, seismic attenuation fluctuates continuously in time at all frequencies, and the presence of cracks, their density and connectivity, as well as the presence and saturation of fluids, play a central role in defining such behavior. Due to multiple mechanisms, the crack density within a fault’s damage zone varies throughout the seismic cycle. Moreover, non-tectonic stress loads, seasonal or tidal, can change the crack density of crustal rocks, and leave detectable signatures on seismic attenuation. A strong signature can also be left on the crustal attenuation by a stress transfer from a nearby fault. Here we show that attenuation time histories from the San Andreas Fault (SAF) at Parkfield are affected by seasonal loading cycles, as well as by 1.5–3 year periodic variations of creep rates, consistent with published results that documented a broad spectral peak, between 1.5 and 4 years, of the spectra calculated over the activity of repeating earthquakes, and over InSAR time series. After the Parkfield mainshock, the modulation of seismic attenuation is clearly correlated to tidal forces. Opposite attenuation trends are seen on the two sides of the fault up to the M6.5 2003 San Simeon earthquake, when attenuation changed discontinuously, in the same directions of the relative trends. Attenuation increased steadily for over one year on the SW side of the SAF, until the San Simeon earthquake, whereas it decreased steadily on the NE side of the SAF, roughly for the 6 months prior to the event. Random fluctuations are observed up to the 2004 M6 Parkfield mainshock, when rebounds in opposite directions are observed, in which attenuation decreased on the SW side, and increased on the NE side. Another example of changes of attenuation with time is given for the Geysers geothermal field, where a large data set of earthquake recordings from a dense temporal deployment are analyzed and results are given in terms of 1/Q(f,t).

Studies of risk perception examine the judgments that people make when they are asked to characterize and evaluate hazardous activities and technologies (Slovic, 1987). Risk perception is a cognitive process involved in several daily activities that orients people’s behaviors on the impact of uncertain events. This process is both individual and collective; it selects and interprets signals related to direct observations or information received from others. Some researches highlighted that in many cases there is a difference between subjective risk perception and objective evaluation (Slovic, 1992, 2000). To understand risk perception it is necessary to consider a number of social, psychological and cultural ambits, as well as interaction among them (Wachinger & Renn, 2010). In our opinion, in agreement with the constructivist approach, seismic risk perception does not depend on the actual value of the seismic risk but there are others factors influencing it. In the specific case of seismic risk, the role of people perception is very important, especially in the absence of clear communication strategies. The clarity of the language used by mass media, scientists and decision-makers in communicating seismic risk to people is essential for a proper knowledge and awareness. This paper presents the results of the research on seismic risk perception in Italy, started by INGV researchers on funding by the DPC 2012-2015. In 2013, we built a questionnaire to assess seismic risk perception, with a particular attention to compare hazard assessed by scientific data and methodology, then translated into a national law. The Seismic Risk Perception Questionnaire (SRPQ) is designed by semantic differential method, using opposite terms on a Likert scale to seven points. The questionnaire allows to obtain scores for five risk indicators: Hazard, Exposure, Vulnerability, People and Community, Earthquake Phenomenon. In 2015, a CATI survey was conducted on a statistical sample of the Italian population (N=4,012), its results represent a benchmark for the seismic risk perception in Italy. At the same time, from 2013 to today, we have gathered on the web more than 9,800 responses to the SRP-Q (www.terremototest.it) and these data allow us to make comparisons for year and for different reference periods, also on the basis of the seismic phenomena that have occurred. The questionnaire is given by web on a random sample since 2013, January 21th, to April 2018, collecting over 9,800 answers. In this contribution we compare and discuss data collected during the last six years (2013-2018) trying to investigate if and how the occurrence of strong and damaging earthquakes (in particular those occurred in Central Italy in 2016) has changed the seismic risk perception in some regions of Italy. We hope that the analysis of risk perception data is able to produce useful indications to design seismic risk reduction activities and address risk communication strategies

La sequenza sismica Amatrice-Visso-Norcia (AVN.s.s nel seguito) include il terremoto più forte avvenuto negli ultimi 30 anni in Italia. La sequenza sismica è iniziata il 24 agosto 2016 con due terremoti di Mw 6.0 e Mw 5.4 che hanno provocato ingenti danni e 294 morti; questi eventi sono stati seguiti da migliaia di aftershocks. Altri due terremoti forti sono avvenuti il 26 ottobre, Mw 5.4 e Mw 5.9. Il 30 ottobre Il più forte terremoto della sequenza (Mw 6.5) ha colpito l'Italia centrale, senza causare vittime ma oltre 40.000 sfollati. Altri quattro terremoti moderati (Mw>5.0) sono avvenuti il 18 gennaio 2017. A inizio ottobre 2018, l'INGV aveva già localizzato più di 90000 terremoti della AVN.s.s. Il Bollettino Sismico Italiano (BSI) ha deciso di revisionare attentamente i terremoti che hanno seguito immediatamente i tre principali eventi del 24 agosto, del 26 e del 30 ottobre 2016 poiché le prime decine di ore dopo un forte terremoto sono le più critiche per le operazioni di monitoraggio sismico svolte presso la Sala Operativa dell'Istituto Nazionale di Geofisica e Vulcanologia (INGV) di Roma. Infatti, l'occorrenza di innumerevoli terremoti complica le normali operazioni di sorveglianza sismica. Gli analisti sismologi in servizio presso la Sala Operativa (Marchetti et al., 2016) non possono revisionare in tempo quasi reale tutti i terremoti localizzati in modo automatico dal sistema di acquisizione e analisi dati Earthworm. Sebbene il personale in servizio di sorveglianza sia stato raddoppiato immediatamente dopo l’inizio della sequenza, presso la Sala Operativa è stato possibile revisionare solo gli eventi maggiori, al di sopra della soglia di risentimento (ML  2.5). Gli analisti del BSI hanno analizzato tutte le forme d’onda disponibili per localizzare gli eventi avvenuti tra il 24 e il 26 agosto, tra il 26 e il 27 ottobre e tra il 30 ottobre e il 1 novembre 2016., Per la prima volta l'analisi ha riguardato anche le registrazioni di 9 stazioni temporanee che non entravano automaticamente nelle localizzazioni di sala, in aggiunta alle stazioni sismiche permanenti e temporanee (12) dell'INGV (Moretti et al., 2016) utilizzate per il monitoraggio in tempo reale della sequenza. Sono stati analizzati manualmente, oltre agli eventi precedentemente revisionati dai turnisti in Sala Operativa, anche tutte le localizzazioni automatiche realizzate dal sistema Earthworm. Per ogni intervallo analizzato, la revisione ha permesso di aumentare il numero di terremoti localizzati di un fattore due o tre e di ridurre la magnitudo di completezza (Mc) di 0.5 per gli eventi successivi alle scosse di Amatrice (Mc 2.2) e Norcia (Mc 2.9) e di 0.2 per gli eventi successivi alla scossa di Visso (Mc 2.2). E’ importante sottolineare l’utilità di un catalogo come questo, basato su fasi riviste e che include eventi in un ampio range di magnitudo, come riferimento per: (i) istruire algoritmi di picking automatico e/o validare cataloghi dei traveltimes letti automaticamente, (ii) migliorare la completezza di cataloghi mediante l'applicazione di tecniche di matched filter basate sulla cross-correlazione di registrazioni di terremoti templates con dati in continuo; (iii) determinare localizzazioni ipocentrali di alta precisione mediante l'applicazione di tecniche di localizzazione relativa DD che dipendono criticamente dall'accuratezza delle localizzazioni assolute iniziali. Gli early aftershocks sono stati localizzati utilizzando il codice NonLinLoc ed un modello di velocità 1-D locale. Il catalogo finale include circa 10,500 early aftershocks, di magnitudo locale compresa tra 0.3 e 5.4, caratterizzati da ottimi parametri di localizzazione. Le localizzazioni di questi aftershocks avvenuti nei primi giorni dopo gli eventi principali, insieme ad una attenta revisione degli ipocentri degli eventi con magnitudo maggiore di 5.4, forniscono indizi molto importanti sui sistemi di faglie immediatamente attivati dall’occorrenza dei terremoti principali (Improta et al., in preparazione), aggiungendo importanti dettagli a quanto è già stato mostrato in diversi lavori che localizzano gli eventi della AVN.s.s utilizzando i tempi di percorso delle fasi P ed S letti in tempo quasi reale dagli analisti sismologi in servizio presso la Sala Operativa INGV (Chiaraluce et al., 2016) o mediante sistemi di picking automatico (Chiarabba et al., 2018). In particolare, l'analisi dell'evoluzione spazio-temporale della sismicità, anche di bassa magnitudo, fornisce nuove indicazioni per capire le relazioni esistenti tra i sistemi di faglie normali Quaternarie, anche caratterizzate da episodi di fagliazione superficiale (Villani et al., 2018), e faglie inverse pre-esistenti e per investigare la geometria e i meccanismi fisici della sorgente dei terremoti più forti della AVN.s.s. (Scognamiglio et al., 2018; Cheloni et al., 2017).

In the years between 2010 and 2015 in the Apennines-Calabrian arc boundary, in the Pollino massif, a long seismic sequence took place. The area is subject to Northeast- Southwest extension, which results in a complex system of normal faults striking Northwest-Southeast, nearly parallel to the Apenninic mountain range. The seismic sequence includes more than 6000 earthquakes in the Pollino region, the maximum magnitude recorded is Ml=5.0 and it happened in October 25th 2012 after about two years of ongoing activity; the peculiar temporal evolution of the seismic sequence allows us to catalogue it as a swarm. Here we describe the main seismological characteristics of this seismic sequence and characterise the fracture field of the region. We analyse thousands of seismograms, deriving accurate locations crust velocity model and anisotropic parameters in the crust. These parameters yield clues and insights that may help understanding the physical mechanisms behind the seismic swarm. Since the late 60s-early 70s era seismologists started developing theories that included variations of the elastic properties of the Earth crust and the state of stress and its evolution prior to the occurrence of a large earthquake. Among the others the theory of the dilatancy: when a rock is subject to stress, the rock grains are shifted generating microcracks, thus the rock itself increases its volume. Inside the fractured rock, fluid saturation and pore pressure play an important role in earthquake nucleation, by modulating the effective stress. Thus, measuring the variations of wave speed and of anisotropic parameter in time can be highly informative on how the stress leading to a major fault failure builds up. We systematically look at seismic-wave propagation properties to possibly reveal short-term variations in the elastic properties of the Earth crust. In active fault areas, tectonic stress variation influences fracture field orientation and fluid migration processes, whose evolution over time can be monitored through the measurement of the anisotropic parameters. We analysed waveforms recorded at permanent and temporary stations hold by the Istituto Nazionale di Geofisica e Vulcanologia.

Il fenomeno sismico può influenzare in modo determinante alcune società. La Calabria è stata colpita nel passato da terremoti distruttivi che ne hanno segnato la storia: nella mappa di pericolosità sismica, la Calabria è una delle aree dove sono attesi gli scuotimenti più forti. A questa situazione si contrappone una distribuzione di sismicità strumentale che presenta uno scarso numero di terremoti con magnitudo significative. I risultati ottenuti dall’indagine sulla percezione del rischio sismico, che ha analizzato 182 questionari relativi alla Calabria, somministrati su web nel periodo gennaio 2013 - luglio 2017, indicano una bassa consapevolezza, sia del fenomeno sismico che della pericolosità sismica e della vulnerabilità del territorio. Il presente lavoro ha l’obiettivo di evidenziare il divario tra pericolosità del territorio e carenze informative della popolazione, e l’importanza fondamentale del dialogo tra Ricerca e società in un trasferimento complesso ma imprescindibile ai fini di una corretta informazione del rischio alla base della prevenzione sismica. Il fenomeno sismico può influenzare in modo determinante alcune società. La Calabria è stata colpita nel passato da terremoti distruttivi che ne hanno segnato la storia: nella mappa di pericolosità sismica, è infatti una delle aree dove sono attesi gli scuotimenti più forti. A questa situazione si contrappone una distribuzione di sismicità strumentale che presenta uno scarso numero di terremoti con magnitudo significative. I risultati ottenuti dall’indagine sulla percezione del rischio sismico indicano una bassa consapevolezza, sia del fenomeno sismico che della pericolosità sismica e della vulnerabilità del territorio. Il presente lavoro ha l’obiettivo di evidenziare il divario tra pericolosità del territorio e carenze informative della popolazione, e l’importanza fondamentale del dialogo tra Ricerca e società in un trasferimento complesso ma imprescindibile ai fini di una corretta informazione del rischio alla base della prevenzione sismica.

The “alternanza scuola lavoro” (interchange school/work) has been recently introduced by Minister of Education, University and Research (Law 107/15 - MIUR) in the Italian high school as a methodology for implementing the second cycle teaching. We have implemented the method in the Colli Albani, an area located 20 Km southeast Rome, the site of a quiescent volcanic district, a place very popular since prehistoric times. Educating secondary school students to planet sustainability and involving them in a deepest knowledge of the Earth is a challenge today. The reason is not only in the curricula: a recent study in four European countries, including Italy, has shown that none of the countries involved provides a special course to educate students on earthquakes and volcano hazards and risks (Bernhardsdottir et al. 2012). In our present and past experiences we have noticed a poor knowledge of the territory they inhabited. Very often, in accordance with that study, a total ignorance of the natural risks that in certain areas of our country can become serious. Our Laboratory then is getting year by year specialized in searching new methodologies to attract and motivate students. In this particular case, we have used three methods with students of a Classical Lyceum: gamification, science narrative and museology to transfer the knowledge of a territory interested by volcanism for thousands of years. We divided the students into three groups, each group for a different methodology. Apart from an area of common activities, the students of the three groups have worked on their own project with the same goals: obtaining a deeper knowledge of the area they inhabit and become aware of the natural risks. We present some preliminary results of the three groups activities obtained also by comparing the three methods efficaciousness.

Amplification at rock sites in areas of high topographic relief has been increasingly observed in the last years, with unexpected level of damage after strong earthquakes. In regions affected by recent tectonic activity, topographic irregularities can include fault damage zones. In such conditions, seismic waves can be locally amplified as a double effect of wave focusing along the topography and /or the presence of fractures/joints or locally weakened rocks. The role of topography vs. geological complexities in controlling the ground motion amplification at rock sites is a newly debated issue in the seismological community. The most crucial questions regard what is the real contribution of the topography shape and fracturing, and how to parameterize such effects for their inclusion in the seismic design codes. In this framework, the EMERSITO INGV task force installed 7 seismic stations across the San Giovanni fault, after the Amatrice mainshock of the 2016 sequence in Central Italy. This active normal fault is located in the area of the Montereale intermountain basin (Abruzzi region, Italy) and bounds the southwestern slope of Mt. Mozzano, a roughly 2D-shaped, up to 1450 m high pronounced topography. Moreover, this fault has been recently studied by several authors who performed detailed geological and geophysical surveys. Our stations recorded more than 100 earthquakes with magnitude ranging from 2.5 to 3.9 as well as a 4.4 M earthquake with hypocenter in Capitignano district, few kilometres far. We have analyzed in detail the recorded signals calculating the traditional spectral ratios at single station (HVSRs) and using the reference site (SSRs) using both ambient noise and earthquakes. In order to obtain a robust estimate of the site amplification effect at each station, we have investigated the influence of backazimuth and epicentral distance. We have also applied the time-domain covariance matrix analysis and the frequency domain polarization analysis. We have found that, in spite of the complexity of the seismic data, the observed polarization pattern is generally oriented orthogonal to the ridge elongation, as well as to the fault strike, suggesting the existence of a high angle relation between ground motion polarization and fracture systems