The SEISMOFAULTS project: first surveys and preliminary results for the Ionian Sea area, southern Italy

The SEISMOFAULTS project (www.seismofaults.it) was set up in 2016 with the general plan of exploring the seismicity of marine areas using deep seafloor observatories. The activity of the first two years (Seismofaults 2017 and 2018) consisted of the installation of a geophysical-geochemical temporary monitoring network over the Ionian Sea floor. Eleven ocean-bottom seismometers with hydrophones (OBS/H) and two seafloor geochemical-geophysical multiparametric observatories were deployed to: (1) identify seismically active faults; (2) identify potential geochemical precursors of earthquakes; and (3) understand possible cause–effect relationships between earthquakes and submarine slides. Furthermore, five gravity cores were collected from the Ionian Sea bottom and ~4082 km of geophysical acquisition, including multibeam and single channel seismic reflection data, were acquired for a total of 4970 km 2 high-resolution multibeam bathymetry. Using Niskin bottles, four water column samples were collected: two corresponding at the location of the two multiparametric observatories (i.e., along presumably-active fault zones), one corresponding at a recently discovered mud volcano, and one located above a presumably-active fault zone away from the other three sites. Preliminary results show: (1) a significant improvement in the quality and quantity of seismological records; (2) endogenous venting from presumably active faults; (3) active geofluid venting from a recently-discovered mud volcano; and (4) the correct use of most submarine devices. Preliminary results from the SEISMOFAULTS project show and confirm the potential of multidisciplinary marine studies, particularly in geologically active areas like southern Italy and the Mediterranean Sea. anomalously. earthquake, in Rn concentration recorded et In northern Iceland to earthquake [Barberio et al., Barbieri et 2020].


Introduction
Although the focus of scientific research is rapidly expanding towards and beyond the limits of our solar system, the surface and shallow layers of our planet, particularly the (deep) marine areas, are still largely unexplored. The marine areas often host active faults, volcanoes, and other structures that are potentially dangerous or able to have a strong impact on human life and activities [Buck et al., 2005;Smith et al., 2008;Trippetta et al., 2019].
These events have been studied by many geologists and geophysicists for the great force they had; however, their cause and origin (zone and generation mechanism) is still greatly debated. For instance, it is not yet known whether the tsunamis were caused directly by the earthquake-related seabed displacement or indirectly by seismically triggered submarine landslides [Valensise and Pantosti, 1992;Billi et al., 2008Billi et al., , 2010Tappin et al., 2008;Argnani et al., 2009;Favalli et al., 2009;Casalbore et al., 2012;Ridente et al., 2014;Polonia et al., 2016b;Schambach et al., 2020]. Andrea Billi et al. The lack of an adequate network of seismometers over the bottom of the Ionian Sea and of a continuous monitoring of other geophysical and geochemical data prevents a full comprehension of the tectonic, seismological, and geomorphological processes that are active in the western Ionian Sea. In recent years, however, the acquisition of multibeam and multi-channel seismic reflection data [Minelli and Faccenna, 2010;Polonia et al., 2011;Gutscher et al., 2016Gutscher et al., , 2017 and gravity cores taken from the seabed , together with the observations from a permanent multiparameter station [Monna et al., 2005;Sgroi et al., 2007] have suggested a new tectonic framework of the Ionian Sea. This framework has led to significant improvements in our knowledge of the hazards [Polonia et al., 2012;Sgroi et al., 2014]. Faults that may have caused one or multiple deadly earthquakes and tsunamis have been recently recognised and mapped [Polonia et al., 2016a[Polonia et al., , 2016b; Figure 1], such as the Ionian and Alfeo-Etna fault systems (Figure 1).
In the case of the Ionian Sea, it is therefore necessary to: (1) instrumentally-observe these faults ( Figure 1) and establish whether they are seismically-active; (2) understand whether the seismic activity along these structures could be predicted by possible precursory phenomena, such as geofluid venting from fault zones and/or mud volcanoes; and (3) understand whether submarine slides along the Sicilian-Calabrian margins can be caused by seismic events of low-to-intermediate magnitude. Our main objective is to highlight the potential of this type of multidisciplinary research.

Geological setting
Convergence and contraction along the Africa-Eurasia plate boundary in the Mediterranean are partly accommodated by the Calabrian Arc subduction system, where the Ionian crust and lithosphere subduct toward NW beneath the Calabrian accretionary prism and the Tyrrhenian lithosphere ( Figure 1). The accretionary prism, which is partly on-shore (Calabria and Peloritani Mts.) and mostly off-shore (Ionian Sea), developed through deformation of the thick sedimentary sequence lying on the African subducting plate. In particular, shortening, was taken up both along the outer deformation front and in the inner portions of the accretionary wedge [Amodio-Morelli et al., 1976;Cernobori et al., 1996;Doglioni et al., 1999;Minelli and Faccenna, 2010;Polonia et al., 2011;Gallais et al., 2012].
Three main morpho-structural domains have been defined within the Calabrian accretionary prism in the Ionian Sea area [Polonia et al., 2011]. Moving from SE to NW, these domains are (  processes as testified by mantle sourced diapirs aligned along such faults . A main NWstriking deformation zone (Ionian Fault system) delimits the western and eastern lobes of the accretionary prism. This deformation zone elongates between the Messina Straits region and the Ionian abyssal plain, cutting through the entire subduction complex.
The western and eastern lobes of the Calabrian prism (WL and EL, respectively) are very different from a structural point of view. The western lobe (WL; offshore Sicily) is characterised by a very low (about 1.5°) tapered salt-bearing accretionary prism delimited towards the Italian peninsula by a slope terrace hosting a Messinian thrust-top basin. This flat region, located between the inner and outer wedges, occurs where the basal thrust involves deeper layers down to the basement with formation of out-of-sequence thrusts (splay faults) and duplexes. At the front of central Calabria, the eastern lobe (EL) shows a more elevated accretionary prism that is between 1000 and 1500 m shallower than in the WL. Moreover, the EL is characterized by steeper topographic slopes and higher deformation rates than the WL. Variations in the structural styles between the EL and WL coincide with differences in the depth of the basal thrust (that is about 4 km shallower in the WL) and with the occurrence of thrust faults affecting the basement in the EL [Polonia et al., 2011[Polonia et al., , 2012.
A set of NNW-SSE trending fault systems characterizes the WL, inducing across-strike margin segmentation [Hirn et al., 1997;Bianca et al., 1999;Nicolich et al., 2000;Argnani and Bonazzi, 2005;Chamot-Rooke et al., 2005;Del Ben et al., 2008;Rosenbaum et al., 2008;Polonia et al., 2011Polonia et al., , 2012Gallais et al., 2013;Scarfì et al., 2016]. The major fault in this region is an active transtensive fault elongating between the Alfeo seamount and the area offshore of the Etna volcano. This set of transtensional faults possibly re-activates Messinian-Pliocene thrust faults, causing a vertical offset of the accretionary prism. The transtensional set occurs along a preexisting Mesozoic structure and is accompanied by a basin floored by a 700-m-thick relatively-undeformed sedimentary infilling [Polonia et al., 2011[Polonia et al., , 2012[Polonia et al., , 2016a. For its geographical location, the main NNW-SSE trending fault in the WL was named Alfeo-Etna Fault (AEF) system after Polonia et al. [2016a]. The transtensional faults, segmenting the continental margin offshore of Sicily (Figure 1), are likely the most seismically-active faults in the region [Monaco and Tortorici, 2000;Presti et al., 2013;Totaro et al., 2013;Polonia et al., 2016a;Barreca et al., 2018;Presti, 2020] and hence, they were the main target of the Seismofaults 2017 and 2018 surveys. The multidisciplinary approach of these surveys is described below.

First surveys: Seismofaults 2017 and 2018
The first cruise of the SEISMOFAULTS project, i.e. the Seismofaults 2017 scientific cruise, lasted 14 days and took place in May 2017. The main purpose of the cruise was the deployment of eight OBS/H and two multiparametric geochemical-geophysical observatories at specific locations on the Ionian seafloor, as shown in Figure 2 (Table 1). Five gravity cores were also collected from the sea bottom ( Figure 2 and Table 1) and ~4082 km of geophysical data, including high-resolution multibeam bathymetry and single channel seismic reflection profiles, were acquired in three working areas (Areas I, II, and III; Figure 2), resulting in a total of 4970 km 2 highresolution multibeam bathymetry .
Conductivity-temperature-depth (CTD) casts were performed along the seawater column to obtain temperature, conductivity, salinity, and sound velocity profiles that are necessary to calibrate the multibeam measurements. Four water column samples were collected with Niskin bottles: two at the two multiparametric observatories sites, one on a recently discovered mud volcano, and one located along a fault zone far away from the other three sites ( Figure 2 and Table 1). CTD casts were also performed at the sites of water sampling. The SEISMOFAULTS project in Ionian Sea

High-resolution multibeam bathymetry
High-resolution bathymetric surveys (Seismofaults 2017) were carried out using a multibeam Teledyne Reson SeaBat 7160 (41-47 kHz) echosounder with a footprint size of 1° × 1°. Positioning was obtained by a differential GNSS system (accuracy ±0.5 m), while sound velocity profiles were estimated from multiple CTD casts (Seabird 911plus) to ray trace the acoustic wave along the water column. Multibeam data were processed onboard using

Single-channel seismic profiles
High resolution seismic stratigraphy is based on chirp seismic profiles. These profiles were acquired using a frequency-modulated source (Seismofaults 2017) operating in the frequency range of 2-7 kHz (Benthos Chirp III) and

BB -OBS/H (broad-band ocean bottom seismometer with hydrophone)
During to version A to simplify operational procedures and better assure data quality. Deep-sea-rated syntactic foam blocks are now used to provide the required buoyancy and titanium grade 5 for the frame and vessel that hold the instrument and batteries. These top-quality materials have become essential in undersea deep operations.
Furthermore, the OBS/H (version B) can now be equipped with a broad-band seismometer plus a hydrophone or a differential pressure gauge (DPG).

Andrea Billi et al.
8 the seismometer must be anchored as firmly as possible to the seabed to produce meaningful waveforms.
Furthermore, it must be shielded against marine currents and other noise sources. This new OBS/H has a special architecture that combines the burn-wire method for best coupling to the seabed with a telescopic shield for maximum protection against currents.  [Dahm et al., 2002;Sgroi et al., 2006;2007;Favali et al., 2013].

9
The SEISMOFAULTS project in Ionian Sea

Multiparametric observatories
We used two new 'stand-alone' multidisciplinary seafloor observatories developed in the mainframe of the scientific and technological activities of the EMSO -ERIC (European Multidisciplinary Seafloor and water-column Observatory, European Research Infrastructure Consortium http://emso.eu/). Photographs and the construction scheme of the multiparametric observatories are shown in Figure 5, with the main parameters and characteristics being reported in Table 3. The multiparametric observatories are built to operate in extreme submarine environments down to a depth of 4000 m.
In recent times, innovations in technology have increased the maximum number of synchronous acquisitions, data rates and resolution, and allowed for autonomous remote data synchronisation and reduced power To achieve this goal, the observatories were deployed along previously identified active tectonic structures [Argnani and Bonazzi, 2005;Argnani et al., 2009;Polonia et al., 2012Polonia et al., , 2016aGutscher et al., 2015Gutscher et al., , 2017 (Figure 2).

Seawater column geochemistry
Three vertical casts by Rosette and Niskin bottles were carried out near the deployment area of the two multiparametric observatories (deployed at depths of 1088 m and 1689 m; Figure 2 and Table 1). Samples were compared with the local air-saturated seawater (ASSW). Vials were filled with 120 ml of sea water collected at different depths and crimped to avoid air contamination. Chemical analyses were done on the gas phase obtained after the attainment of the equilibrium (at constant temperature) between the water sample and a known volume of host high-purity gas (argon) injected into the bottle used for sampling [see Sugisaki andTaki, 1987 andCapasso andInguaggiato, 1998 for details].
The analytical determinations were done by an Agylent 7800B gas chromatograph characterized by a double detector (TCD-FID) using argon as carrier gas. Helium isotope analyses were done on the gas fractions obtained by using the same procedure described above for the gas chromatography. Next, following the procedures proposed by Sano and Wakita [1988], the sample was purified. Water samples of 100 ml were stored in PVC bottles for total alkalinity titration. A volume of 50 ml for each sample was filtered by a 0.45 μm filter and acidified by HNO 3 0.1 N for cations (Ca, Mg, Na, and K) determination, whereas the non-acidified samples were collected for anions (Cl, F, and SO 4 ) determination. pH and EC were measured by electronic instruments calibrated in situ using buffer solutions.
Chemical analyses of the major constituents were done in the laboratory on filtered (0.45 mm) and acidified (100 mL HNO 3 Suprapur) water samples (Na, K, Mg and Ca) and on untreated samples (F,Cl,Br,NO 3 , and SO 4 ) using ion-chromatography (Dionex ICS-1100). The HCO 3 content was measured by standard titration procedures using hydrochloric acid.

11
The SEISMOFAULTS project in Ionian Sea

Rationale
In the following sections, we briefly report some preliminary results from the Seismofaults 2017 and 2018 surveys to verify some of our hypotheses and to explore the potentialities of the scientific approach followed in the SEISMOFAULTS project.

High-resolution multibeam bathymetry and single-channel seismic profiles
The study area is located between the southern sector of the Calabrian margin and the eastern side of the Sicily escarpment ( Figs. 1 and 2).

OBS/H: Results from previous experiments and an example from the SEISMOFAULTS dataset
A permanent seafloor observatory has been deployed in the Ionian Sea off-shore Catania to better identify active seismogenic structures (Submarine Network 1 -SN1) ]. In the Southern Tyrrhenian Sea, the TYDE (TYrrhenian Deep sea Experiment -TYDE) [Dahm et al., 2002] involved the temporary deployment of several OBS/H modules. The data recorded by these marine seismic stations helped to improve the location accuracy of offshore earthquakes and the identification of active tectonic structures 2007]. The same data were also used to compute 1D and 3D seismic velocity models [Monna and Dahm, 2009;Monna et al., 2013] and to evaluate the activity of seamounts and emerged volcanoes, such as Stromboli and Etna [Sgroi et al., 2009;Beranzoli et al., 2015] (Figs. 7 and 8). The OBS/H modules also recorded 275 "new" low-magnitude events (not recorded by LS) with a minimum of seven P-and S-phase arrival times that were located thanks to the OBS network.   Sgroi et al. [2007], who analysed the crustal and subcrustal seismicity recorded by the seismometer hosted within the SN1. Figure 8 shows the single-station location of earthquakes recorded by the SN1 (blue triangles).
Eight events were recorded by the SN1 and inland stations and were located using the integration of travel times (Table 4).
These travel times were used to compute an apparent mean velocity value used in the inversion of P-and Sarrival times versus the hypocentral distances of the same events. Using the computed apparent velocity value, t st p intervals, a V p /V s ratio of 1.73, and the results of polarisation and particle-motion analyses, Sgroi et al. [2007] estimated the location of a dataset of 213 earthquakes recorded only by the SN1 (Figure 8). This seismicity is linked to the main tectonic structures that are active in the Ionian Sea area, namely the external Calabrian Arc accretionary prism associated with the subduction process, the tectonic structures related to the Malta Escarpment, and the seismogenic structures that are active in the deep Ionian Basin [Polonia et al., 2016a. SN1 recorded many low magnitude earthquakes that were not recorded by land stations. This is certainly due to the low signal-to-noise ratio due to the volcanic tremor associated to some explosive phases that accompanied the 2002-2003 Mt. Etna eruption [Sgroi et al., 2007]. High amplitudes of volcanic tremor obscured low magnitude earthquakes on land seismograms. On the other hand, the same increase in the background noise due to the volcanic tremor was also visible on SN-1 seismograms, but it did not obscure the signal of interest [Sgroi et al., 2007;Sgroi et al., 2019].  Table 4. Comparison [modified from Sgroi et al., 2007] between locations using only the data of the land-based network and integrated ones obtained including the SN1 arrival times. Hypocentral coordinates, gap, rms, horizontal and vertical errors are reported for both locations.
Using the data recorded by the marine seismometers (OBS/H) during the SEISMOFAULTS project, we expect to improve the location accuracy of earthquakes in the Ionian Sea by decreasing the azimuthal gap and decreasing the error of the hypocentral parameters. For instance, in Table 4, the location parameters are shown for eight events (sketched with blue circles in Figure 8 of this work) located with only land stations vs. the integrated locations with SN1 arrival times. Note that a decrease of 52° in azimuthal GAP is obtained for the earthquake occurred on 2 October 2014 at 13:17pm. In Figure 9, we show all the waveforms from the SEISMOFAULTS marine seismometers (i.e. seven OBS/H) and the waveforms from several ISN land stations for the 11 April 2018 M L 3.7 earthquake, which was located offshore of southern Calabria by the ISN (Lat. 37.7313, Lon. 15.9405, depth 42.4 km; http://cnt.rm.ingv.it/). In general, despite marine seismograms having more complex waveforms due to multiples coming from the water surface and sediment layers, various seismic phases are clearly detectable by applying an adequate digital filter on the data.
Thanks to the seismological data from the OBS/H deployed during the SEISMOFAULTS project, we can confidently identify and locate a 'new' low-magnitude seismicity not recorded by the ISN, thus contributing to the identification of new seismically active faults or fault segments in the Ionian domain.

Seawater column geochemical data
The locations of the sites for the seawater column geochemical analyses are listed in Table 5 and given in WGS84 geographical coordinates.

17
The SEISMOFAULTS project in Ionian Sea   Table 6. Geochemical-geophysical data from the analyses of the seawater columns. To enhance differences in geochemical composition of the samples from the three sites, analytical results are graphically presented in Figures 10(a) and 10(b). The sampled waters show a pH between 8.10 and 8.37 and conducibility between 46.2 and 46.6 mS/cm. The chemical composition of the water samples is described in terms of major ion contents. The results of the chemical analyses of the gas extracted from the sampled seawaters point to the existence of a dissolved gas phase that is different from the atmosphere. Figure 10(a) shows the contemporary presence of atmospheric components (represented by O 2 and N 2 ), as well as CO 2 typically originating from an endogenic source.
The analytical results are plotted on the CO 2 -O 2 -N 2 triangular diagram (Figure 10a), where typical atmospheric components (oxygen and nitrogen) are plotted besides an endogenic gas (CO 2 ). The air-saturated seawater (ASSW) end-member is also shown.
All dissolved gases have an O 2 /N 2 ratio slightly lower than that of ASSW due to a relative decrease in oxygen concentration. The increase in CO 2 content is quite evident. Figure 10b shows typical endogenic components (CO 2 and CH 4 ) versus the atmospheric component (here represented by O 2 ) and the injection of CO 2 in several sites. The concentrations of dissolved CO 2 and CH 4 are higher than those of ASSW in contrast to a lower O 2 concentration, suggesting a contribution of endogenic gases to the expected equilibrium with the atmospheric gases. Figure 10b shows the occurrence of the dissolution processes responsible for CO 2 loss and the enhancement of the less soluble species, such as CH 4 .

Discussion and Conclusions
The preliminary results from the Seismofaults 2017 and 2018 surveys in the Ionian Sea are useful for several reasons:

19
The SEISMOFAULTS project in Ionian Sea 1. Recently, Polonia et al. [2011Polonia et al. [ , 2012Polonia et al. [ , 2016aPolonia et al. [ , 2017 have drawn a series of long strike-slip to transtensional or transpressional faults cutting across the Calabrian accretionary wedge along the NW-SE direction ( Figure 1); [see also Argnani and Bonazzi, 2005;Minelli and Faccenna, 2010;Gutscher et al., 2016Gutscher et al., , 2017Dellong et al., 2018]. Moreover, the accretionary wedge is still active over thrust faults that dip towards the northwest and that developed in part over thick deposits of Messinian salt [Gutscher et al., 2006;Minelli and Faccenna, 2010;Polonia et al., 2011;Bortoluzzi et al., 2017]. As mentioned in the introductory section, the location, physical extension, and seismic potential of these faults is poorly constrained. The ensamble of OBS/H deployed in SEIMOFAULTS is a promising tool to detect local seismicity and identify unknown or hypothesized seimogenic structures in the western Ionian Sea as shown by previous experiments (Figs. 7-9). In particular, the many low magnitude earthquakes that are potentially missed by inland seismic stations can be recorded and precisely located by marine stations. Moreover, the location accuracy of all earthquakes (i.e. both those recorded solely by marine stations and those recorded by land and marine stations) can be greatly improved by the presence of a marine network of seismic stations. Data from seafloor instruments is necessary to obtain more realistic crustal velocity models. In turn, the velocity models are the base for the calculation of precise hypocentral locations of earthquakes [e.g., Dahm et al., 2002;Sgroi et al., 2006Sgroi et al., , 2007Monna et al., 2013;Beranzoli et al., 2015]. Preliminary results indicate a good performance/response for identification of lowmedium magnitude earthquakes (Figure 9), thus from the data collected during the SEISMOFAULTS project, we expect to achieve a substantial improvement in earthquake location accuracy in the Ionian Sea area and therefore a substantial improvement in the knowledge of seismically active faults in this region.
2. Earthquake forecasting has been attempted using a wealth of different methods [e.g. Wyss, 1991]. Recent retrospective studies on geochemical anomalies as potential seismic precursors suggest that they can be a good candidate for earthquake forecasting [Inan et al., 2012]. To correctly understand the overall concept underpinning our attempt in the Ionian Sea, a few background studies must be considered: (a) In 1995, eight months before the M7.2 Kobe earthquake (Japan), the Cl and SO 4 concentrations in groundwaters started to increase significantly and anomalously. Nine days before the earthquake, a peak in Rn concentration was recorded [Igarashi et al., 1995]. (b) In 2002, anomalies in the Cu, Zn, Mn, and Cr concentrations in groundwaters were recorded one, two, five, and 10 weeks, respectively, before a M5.8 earthquake in northern Iceland [Claesson et al., 2004]. (c) In 2012, anomalous increases of Ca, Mg, K, and Cl concentrations in groundwaters together with decreases of Na and SO 4 concentrations started between 20 and 30 days before the M7.1 Van earthquake (Turkey) [Inan et al., 2012]. (d) In 2012, significant increases in the Na, Si, and Ca groundwater concentrations started four to six months before two M≥5.5 earthquakes in northern Iceland [Skelton et al., 2014]. (e) In 2016, concentrations of As, V, and Fe in groundwaters started to anomalously increase three to four months before the M6.0 Amatrice earthquake (Italy) [Barberio et al., 2017;Petitta et al., 2018;Boschetti et al., 2019;Barbieri et al., 2020]. The aforementioned works suggest that within the seismic cycle and during the preparatory phase for moderate-large earthquakes, the Earth's intermediate crust undergoes fracturing, and might allow deep fluids to ascend and contaminate shallow aquifers. The continuous geochemical and physical monitoring of groundwaters could therefore help scientists identify anomalies in temporal series that could potentially constitute reliable hydrogeochemical seismic precursors. For these reasons, we decided to use two multiparametric geochemical-geophysical observatories coupled with a dense network of OBS/H ( Figure 2). Moreover, we searched and surveyed venting structures such as mud volcanoes ( Figure 6). We expect that the two geochemical modules (i.e. located along potentially active faults) can record possible geochemical anomalies that are associated to the preparation phase of an earthquake or, more in general, to the seismic cycle of seismically-active faults. Our preliminary results on the geochemistry of the seawater column show seawater contamination by endogenous fluids (Figure 10) along faults in the Ionian Sea. Further studies are necessary to confirm how these observations might help in the detection of an earthquake preparation phase.
phenomena including several submarine slides. The occurrence of strong earthquakes in this region, as well as its fast uplift and presence of a giant volcano (Mt. Etna) make gravitational slides a significant hazard source in the Ionian region. In particular, Billi et al. [2008Billi et al. [ , 2010 ascribed the occurrence of tsunamis that occurred after the 1908 earthquake and previous earthquakes, mostly to seismically triggered submarine slides [Schambach et al., 2020]. Seismic stations (both on land and in marine areas) are able to record, identify, and precisely locate (submarine) landslides [Kanamori and Given, 1982;Kawakatsu, 1995;Dreger et al., 1998;Ristau, 2008;Lin et al., 2010] as well as earthquakes. We therefore expect that data from the OBS/H devices deployed in the Ionian Sea will be useful for locating submarine mass movements and earthquakes and for identifying the possible temporal and spatial relationships between them.
5. The SEISMOFAULTS project is a significant test for the underwater devices used. The quality of the data is heavily dependent on several instrumental factors: (a) correct and reliable recording of all data; (b) duration of batteries that ensures full functionality, including data recording, clock use (for post-survey synchronisation), and release from ballast at the time of recovery; (c) complete impermeability and duration of materials; (d) resistance to water pressure (down to about 2800 m b.s.l. in the case of the Seismofaults 2017 campaign); (e) resistance to deep currents, and (f) resistance to fishing with trawl nets. All these aspects will be carefully evaluated and used to improve future scientific activities.
In conclusion, the SEISMOFAULTS project is a multidisciplinary geoscientific experiment, presently active in a densely populated region (Ionian Sea) that hosted some of the strongest earthquakes and tsunamis of the entire Mediterranean region along the Africa-Eurasia active plate boundary. Hence, this experiment should lead to a better understanding of seismogenic and tsunamogenic mechanisms, of which faults are seismically active, of which slope sectors are prone to gravity mass failure, and whether and where geochemical precursors of earthquakes can be identified and monitored. This experiment is engaging three main Italian scientific institutions for a total of about 35 scientists. It has so far required many technological facilities (scientific vessels, OBS/H, multiparametric observatories, and laboratories) and a connected large effort of management, coordination, and scientific planning to organize the marine campaigns and to coordinate all the scientists involved contemporaneously on the vessels and in the laboratories, as well as to organize the data management and processing. A first successful result of the experiment was the correct deployment of all marine devices and their safe recovery (except one OBS/H), the recovery of four gravity cores, the acquisition of 4970 km 2 of high-resolution bathymetry, the sampling of sea waters along four sea water columns, and, last but not least, the integration of a a significant number of geoscientists with different scientific and technological expertise to reach a common objective. The large set of acquired data and those that will be soon processed and released might give impulse to a wide range of multidisciplinary studies. At present, the main goal is to advance the knowledge of seismogenic/tsunamogenic faults in the Ionian Sea and to build an experience that may constitute a model or a reference for future similar studies in other seas. We hope that this experiment is the first step of a long-term research program over the Italian and European seas, in areas that are prone to earthquakes and other potentially hazardous geological phenomena.
Data and Sharing Resources. All data used for this article are reported in the associated figures and tables or the related sources are properly cited within the article.