04.07. Tectonophysics

Two major deformation belts occur in the portion of the Adriatic Sea offshore the Gargano Promontory. The NE-SW - trending Tremiti Deformation Belt, located north of the Gargano Promontory, originated during the Plio- Quaternary, while the E-W-trending South Gargano Deformation Belt, located south of the Gargano Promontory, formed in a time span from Eocene to Early Pliocene. These deformation belts may have originated by tectonic inversion of Mesozoic extensional faults. This inversion tectonics, of Tertiary age, can be related to the evolution of the fold-and thrust belts surrounding the Adriatic Sea. The whole of the study area is, at present, seismically active and represents a preferential site of deformation.

Subduction zones represent a tectonic region where intense deformations and complex dynamic processes are expected. Although several progress have been made in understanding the structure and the geodynamic evolution of the subduction zones, the active interaction among the subducting slab and the surrounding mantle material remains still debated. The Southern Italy Subduction System is part of the complex tectonic boundary between the Africa-Eurasia macroplates and has been inherited from several phases of fragmentation of the Western Mediterranean subduction zone. It is widely accept that the geodynamic setting of the Southern Italy Subduction System results from the southeast retrograde motion of the northwestward subducting Western Mediterranean slab (i.e. Gueguen et al., 1998; Carminati et al., 1998; Faccenna et al., 2005 and refrences therein). The retrograde motion of the slab was responsible for the creation of the backarc extensional Tyrrhenian Sea and the building of the Southern Apennines and Calabrian arcuate orogenic belts. At present, only the portion of subduction beneath the Calabrian Arc, in the Ionian area, may be active, while a young slab window develops at the Southern Apennines (Lucente et al., 2006). The purpose of this study is to characterize the seismic structure beneath the Southern Italy in order to better define the geometry of the Ionian slab and of the surrounding mantle flows. We therefore analyzed the anisotropic and attenuation properties beneath the study region. Seismic anisotropy is found to be a ubiquitous properties of the Earth due to the mantle deformation and, thus, it is represent a powerful tool to constrain the anisotropic behavior of the upper mantle and of the subducting plate. In particular, the observed anisotropy can help to understand the mantle and the slab deformation and the dynamic processes occurring in the upper-mantle wedge above the sinking oceanic slab and in the mantle below the slab. In this study we present a large collection of shear wave splitting measurements in the Calabrian Arc - Tyrrhenian basin Subduction System. The data analyzed consist of several teleseisms and subduction zone local deep earthquakes (Baccheschi et al., 2007, 2008). We used the method described by Silver and Chan (1991), assuming that shear waves pass through a medium with homogeneous anisotropy and with an horizontal fast axis. We analyzed SKS phases from earthquakes with magnitude greater than 6.0 and epicentral distance Æ° ranging from 87° to 112°. In addition, to obtain the best signal to noise ratio, all teleseisms are band-pass filtered between 0.03-0.3 Hz. The pattern of SKS fast directions, with delay times up to 3.0 s, reveals the existence of a strong seismic anisotropy in the sub-slab mantle region. We observe both trench-parallel and trench-perpendicular fast directions. Fast axes are oriented NE-SW along the Calabrian Arc, parallel to the strike of the subduction. To the N they rotate to NNW-SSE following the curvature of the slab. Fast directions are almost perpendicular to the strike of subduction in front of the slab (Aeolian Islands) and behind the slab (Straits of Messina). In the Apulian domain we observe trench-perpendicular fast directions, oriented N-S and ENEWSW. The pattern of SKS splitting measurements parallel to the strike of the slab suggests that the anisotropy is closely controlled by subduction and by the rollback motion of the slab. These two processes would be responsible for activating mantle flow below and around the slab itself. The pattern of SKS splitting in the Apulian domain seems to be not a direct results of the rollback motion of the slab and may be explained as frozen-in lithospheric anisotropy or as asthenospheric flow deflected by the structure of the Adriatic microplate. In order to obtain a detailed image of the anisotropic structure beneath the Southern Italy Subduction System we also used the direct S waves from earthquake located within the descending Ionian plate. The particular geometry of the Tyrrhenian subduction zone relative to the distribution of the land areas and, consequently, locations of the seismic stations provide an opportunity to collect unique data. In fact, the main massif Calabria is an uplifted fore-arc that lies well trenchward of the volcanic arc. In addition, the slab dips at high angle (about 70°) below Calabria and the lateral extension of the slab is limited and bounded at its edges by the Southern Apennines and Sicily.Seismic stations are distributed in Calabria, in the Southern Apennines and in Sicily and only few are in the Aeolian volcanic arc. This allows most recorded rays to travel through and along the subducted slab. This is not frequently observed worldwide since in most subduction zones, as in Japan, land corresponds to the volcanic arc and trenchward of this the forearc is submerged. This enabled us to sample rays that propagate up the slab and allowed us to separate the different sources of the anisotropy: the subducting lithosphere, the mantle wedge above it and the overriding plate. We analyzed several deep earthquakes, with depth greater tha 150 km, that occurred within the descending slab; S splitting parameters show a complex pattern of anisotropy with variable fast directions across the subduction zone and delay times ranging from 0.1 sec to 2.2 sec. Measurements at single stations are quite variable excluding the overriding plate as main source of anisotropy. The S wave splitting parameters also show frequency-dependent behaviour that we attribute to the presence of small-scale anisotropic heterogeneities. Comparison of the S splitting measurements to the Pwave velocity anomaly at 100-200 km depth shows that where the rays primarily sample the slab the delay times are small. In contrast, where the S rays sample the mantle wedge, the delay times are quite high. This dt pattern depicts the slab as a weakly anisotropic region and suggests that the main source of anisotropy in the subduction zone is the surrounding asthenosphere (Baccheschi et al., submitted to JGR). We also determined the attenuation structure of the slab and of the surrounding regions by the inversion of high quality S-waves t* from slab earthquakes. We obtained high resolution Qs model down to 300 km depth. The results indicate low values of Qs (Qs values down to 200) corresponding to crustal layers (down to 25 km depth), while the slab is characterized by higher but heterogeneous Qs structure (Qs values up to 1100). At 100 km depth the high Qs body is well reconstructed beneath the Calabrian arc and at 200 km depth it is extended offshore the Southern Tyrrhenian Basin beneath the Aeolian Islands. These preliminary attenuation results allowed us to better define the geometry and the boundary of the Ionian slab and distinguish between anisotropy in the slab and in the mantle wedge.

In this chapter, we present scripts and programs that accompany this book. Five MATLAB scripts regard simple examples related to supervised learning, that is, linear discrimination, the perceptron, support vector machines, and hidden Markov models. Seven scripts are devoted to unsupervised learning, such as K-means and fuzzy clustering, agglomerative clustering, density-based clustering, and clustering of patterns where features are correlated. These scripts provide a starting point for the reader, who can adjust and modify the codes with respect to proper needs. Besides, we provide sources and executables of programs that can be readily applied to larger and more complex datasets. These programs regard supervised learning using multilayerperceptron and support vector machines. KKAnalysis is a toolbox for unsupervised learning and offers various options of clustering and the use of self-organizing maps. The programs offer graphical user interfaces (GUI) to facilitate their use and create both graphical and alphanumeric output that can be used in further processing steps. The programs come along with real-world datasets that are also discussed in the example applications presented in various chapters of the book. Other propaedeutic material can be found in a folder called “miscellaneous.”

We describe the main features of the developed software tool, namely PlatE-Motion 2.0 (PEM2), which allows inferring the Euler pole parameters by inverting the observed velocities at a set of sites located on a rigid block (inverse problem). PEM2 allows also calculating the expected velocity value for any point located on the Earth providing an Euler pole (direct problem). PEM2 is the updated version of a previous software tool initially developed for easy-to-use file exchange with the GAMIT/GLOBK software package. The software tool is developed in Matlab® framework and, as the previous version, includes a set of MATLAB functions (m-files), GUIs (fig-files), map data files (mat-files) and user’s manual as well as some example input files. New changes in PEM2 include (1) some bugs fixed, (2) improvements in the code, (3) improvements in statistical analysis, (4) new input/output file formats. In addition, PEM2 can be now run under the majority of operating systems. The tool is open source and freely available for the scientific community.

The Database of Individual Seismogenic Sources (DISS) was conceived at the end of the 1990s by a group of scientists at Istituto Nazionale di Geofisica e Vulcanologia. The database was designed to host data about seismogenic source models intended to serve as geological input for ground-shaking SHA applications and was continuously updated since then. In 2005 there was a big turn in this process as we launched a new version of the database (DISS 3) which augmented the database with two innovative categories. The first, now named “Composite Seismogenic Source”, was intended to overcome the inherent difficulties in identifying fault segment boundaries. The second, named “Debated Seismogenic Source”, was devised to host tectonic information about active faults that have been proposed in the literature as potential seismogenic sources but are not fully parameterized or are considered to be not reliable or have been deprecated by subsequent work. In 2005 the database was first made available to the public through a specifically designed web-based GIS application. This new database is now being widely used in various branches of ground-shaking SHA practice and tsunami hazard. The main strength of this database is that it stores fault parameters in a native 3D and flexible conceptual model. Lately, we also developed strategies to make it testable with independent data under a number of different tectonic and seismic hypotheses. During the years, DISS brought together a large amount of published and original data on Italian seismogenic sources having a potential for a magnitude 5.5+ earthquake and is now being extended to the rest of the Euro-Mediterranean area. We present highlights on the identification and characterization of new seismogenic sources in three key-areas in Italy, namely Lombardia/Veneto (Southern Alps), Adriatic Sea, and Abruzzo/Molise (central Apennines). These new sources describe youthful structures of the Alpine south-verging contractional system, the external fold-and-thrust system in the Adriatic offshore, and the extensional domain of the inner central Apennines.

L’Appennino Settentrionale è una catena montuosa NE-vergente ed è il risultato dell’affioramento del prisma di accrezione originato in seguito alla subduzione della litosfera adriatica sotto il mar Tirreno ed ancora in atto (Faccenna et al., 2001). Dall’Oligocene ad oggi, l’Appennino Settentrionale è stato interessato da due fasi deformative: inizialmente compressiva con la formazione di thrusts e più recentemente distensiva (Elter et al., 1975). Attualmente è caratterizzato da un regime crostale distensivo con una velocità stimata circa 2.5 mm/anno (Hunstad et al., 2003). Gli effetti e le conseguenze di questi episodi deformativi sono ben visibili attraverso un’analisi geologica e geofisica. L’area in studio è posta in corrispondenza della transizione tra la successione Toscana ed il settore Tirrenico del dominio Umbro-Marchigiano, quindi, una zona particolarmente dibattuta da un punto di vista geodinamico, a causa della presenza di diverse tipologie crostali, flusso di calore e anomalie gravimetriche.

During the summer of 2010 we carried out a survey to acquire a multidisciplinary dataset within the Gulf of Sant'Eufemia (SE Tyrrhenian sea, Italy), with the aim of studying the active tectonics affecting the region, including that potentially responsible for key, elusive earthquakes such as the to-date unexplained 8 September 1905 (Mw 7 - 7.5) earthquake. The data here analysed highlight the presence of several tectonic and morphologic features characterizing the investigated area. We have recognized the Angitola Channel, a deep and wide canyon showing a straight trend in its coastward segment, and a meandering trend in the seaward segment. Based on morpho-structural elements, we maintain that the Angitola Channel could be tectonically controlled. Moreover, several gravitational instabilities as slumps and collapses affect the flanks of the morpho-structural high, detected offshore Capo Vaticano. Very high resolution seismic data have unveiled the presence of numerous fluid escape features and several mud volcanoes straddling the sector from the coastline to seaward.

Multichannel reflection seismic data were acquired south of the Salento peninsula, in an area where crustal seismicity has been recorded. Seismic profiles show the presence of small grabens bounded by extensional faults with NW-SE direction. These grabens are filled with Plio-Quaternary sediments and represent the prolongation of the grabens located onshore in the Salento peninsula. Outer arc extension due to flexuring of the Adriatic-Apulian lithosphere under the double load of the Hellenides and Apennines-Calabrian arc is thought to have originated these grabens. The Adriatic-Apulian continental lithosphere presents a very small radius of curvature and a decoupling between upper crust and mantle lithosphere is expected. Inner arc compression within the upper crust may be responsible for the seismicity recorded in the area.

Geothermobarometry on rocks from the Joshua Flat – Beer Creek Pluton and its aureole was combined with fl uid investigations and numerical modelling. The pluton is composed of distinct bodies of diorites, monzonites and granodiorites. The granodiorite intruded the still partially molten monzonite. This led to reheating of the pluton and its aureole. Hornblende thermobaro metry gives temperatures of ca. 700 °C and pressures of about 2 kbar. In the contact aureole three metamorphic zones were distinguished with decreasing distance from the pluton: the andalusite-cordierite, sillimanite, and cordierite-K-feldspar zones, respectively. Leucocratic veins together with reaction fabrics between biotite and K-feldspar document dehydration melting due to biotite breakdown in the innermost aureole. This is supported by Na-in-cordierite thermometry and infra-red spectroscopy measurements at cordierite from metapelites. The latter showed relatively low contents of channel volatiles indicating that fl uids did not play a major role during the prograde contact metamorphic evolution of the pluton’s thermal aureole and that heat transport in the aureole was conductive. Numerical modelling of the Tt path of the pluton/wall rock system resulted in temperatures, which are close to those determined by geothermometry when assuming reheating of the system by a second intrusion. Moreover, modelling suggests a conductive nature of heat fl ow around the pluton.

In November 2002 a submarine gas eruption started offshore 3 Km east of Panarea island (Aeolian Island) on top of a shallow rise of 2.3 km2 surrounded by islets forming a small archipelago. This event has posed new concern on a volcano generally considered extinct. Panarea island and its archipelago (~ 3.3 km2) are the emergent portion of submarine stratovolcano more than 2000 m high and 20 Km across; exhalative activity due to a shallow hydrothermal system is well known since historical times. To monitor and study ground deformation associated with anomalous gas emission, a local GPS network (PANAREA) was designed, set up and measured during time span December 2002 - October 2006. The network consists of nine sites (six constructed after 2002) located on Panarea and on the islets. GPS data analysis was performed combining episodic campaigns of Panarea and other local networks located in the Aeolian area, carried out between 1995 and 2006, and data of continuous European and Italian sites. The results show at Panarea volcano two distinct crustal domains characterized by different kinematics and styles of deformation. The merging of GPS and structural data suggest the relationship among gas vent distribution, submarine volcanological structures and ground deformations. The actual distribution of the estimated strain-rate is consistent with the structural setting.The general subsidence and shortening in the islets area can be interpreted as the response of the surface to the variation of the hydrothermal system reservoir which is progressively reducing its pressure after the gas eruption. A simple first order approach to the modelling of the hydrothermal system is the use of Okada sources.To evaluate the coupled thermo-hydro-mechanical processes going on in Panarea, a two-step model will be implemented. The model first involves the simulation of pore pressure and temperature changes due to fluid circulation. Then the mechanical response of the porous rock is calculated based on the linear theory of poro-elasticity.