Journal of the Virtual Explorer

Pauselli and Federico (2002) produced a regional scale study of the rheology of the NA crust, based on the study of the thermal field along the CROP03 transect, aimed to locate the brittle/ductile transition (as defined by Brace and Kohlstedt, 1980). According to this study, in the westernmost part of the region (TD) the brittle/ductile transition occurs at a depth of 10 to 12 km; moving farther east, it deepens from 12 km (beneath the TZ) to about 25 km (in the AD). In the easternmost part of the profile the rheological behaviour of the lithosphere is more complex, with rocks being brittle at a depth less than 25 km but also at depth greater than 34 km. A more complete study, performed by Pauselli et al. (2010), in which a high-pressure brittle fracture mechanisms (Zang et al., 2007) is included in the thermo-rheological model across the NA, shows important lateral variations of both total lithosphere strength and of the strength distribution with depth. The TD is characterised by a “crème brulée” structure, where only one load-bearing layer (the top half of the upper crust, 6 to 8 km thick) contributes to most of the lithosphere strength. In the AD the lithosphere is much stronger and a “jelly sandwich” structure is present, with both lower crust and upper mantle contributing to the lithosphere strength.

The different rheological profiles characterising the TD and the AD are reflected by the different distribution of the earthquake’s hypocenters (Chiarabba et al., 2005; De Luca et al., 2009).

The TD is characterised by scarce, shallow-seated (mostly < 10 km deep), low-magnitude seismicity, concentrated in the shallower part of the upper crust: the reduced thickness of the seismogenic layer can be easily referred to the weak and thinned crust of the Tuscan region and to the geothermal and post-magmatic processes active in the area (e.g. Liotta and Ranalli, 1999). In the AD the distribution of the seismicity is more complex. Both instrumental and historical seismicity (Gruppo di lavoro CPTI, 2004) including the larger (5 < M < 7) extensional earthquakes of the region, are concentrated along a NW-SE trending belt, narrower in the northern part (about 20 km wide), wider in the southern part. Across this belt, grossly corresponding to the TZ and to the westernmost part of the AD, crustal seismicity deepens from west to east, down to a depth of 25 km, locally highlighting two separate clusters, located in the upper and lower crust respectively. The eastern part of the AD is characterised by a diffuse microseismicity, mostly located in the shallower part of the crust (< 15 km deep) and by few moderate events (M < 5.5). The AD is also characterised by less frequent, but regularly recorded intermediate seismicity, occurring along a west-dipping structure down to about 70 km. This sub-crustal seismicity is generally interpreted as connected to the descending Adriatic continental slab.

Summarising, the available data highlight a good correspondence between the distribution of the shallow seismicity and the rheology of the upper crust, which is thicker and stronger in the AD with respect to the TD. For a full comprehension of these relationships, better earthquake locations are needed, along with more detailed rheological profiles, keeping in count e.g. the mechanical variations within the upper crust, where different rock-types (soft sediments, sedimentary rocks, crystalline rocks) are superposed. For example, below the axial ridge of the Apennines the distribution of the seismicity suggests that the brittle crust is partitioned into three different layers (Mirabella et al., 2008): a low-velocity horizon, corresponding to the shallower part of the basement, decouples the overlying sedimentary cover (where moderate seismicity is produced) from the underlying crystalline basement (where only microseismicity occurs).

High-pressured fluids can also play an important role in earthquake triggering (Miller et al., 2004), in a region where a huge CO₂ flux is observed (Chiodini et al., 2004).

Finally, we have to consider that the seismicity is not only a function of the rheological properties of the lithosphere, but also reflects the active tectonic structures and processes.

The focal mechanisms of the earthquakes reflect the contemporaneous activity of extension in the hinterland and compression in the foreland area. This pattern was early described since the 80's (e.g. Lavecchia et al., 1984; 1994; Frepoli and Amato, 1997) and is now supported by a large collection of focal mechanisms, whose quality and quantity have continuously increased with time (e.g. Pondrelli et al., 2006, De Luca et al., 2009).

The axial ridge of the Apennine belt is characterized by moderate to large normal-fault earthquakes (5<M<7), mostly occurring on NW-trending adjacent segments (Amato et al., 1998, Chiaraluce et al., 2003). The contiguity of these segments delineates an elongated extensional belt that obliquely crosses the arc-shaped, pre-existing compressional structures and is mostly confined in the upper 6-8 km of depth. This extensional belt longitudinally crosses the whole NA and it is clearly marked by the alignment of several intra-mountain basins, which, from NW to SE, are labeled the Lunigiana-Garfagnana, Mugello, Casentino, Sansepolcro, Gubbio, Colfiorito and Norcia basins (fig. 1). This is part of a much longer alignment, affecting the Central (L’Aquila, Fucino) and the Southern (Irpinia, Val D’Agri) Apennines as well.

In Northern Umbria, the seismogenic normal faults splay out from a major low-angle NE-dipping detachment (Alto Tiberina Fault, see also chapter 3), whose presence and activity is supported by both active and passive seismic data (Chiaraluce et al., 2007 and references therein). This fault represents the most recent, still active extensional master fault, driving the extension of the NA hinterland.

A belt of shallow (D < 15 km), contractional or traspressional earthquakes occurs in the eastern part of the AD, in the Po Plain-Adriatic foreland, from the southern edge of the Po Plain (e.g. Reggio Emilia, Forlì), down to the Adriatic coast, from Pesaro to Ancona to Porto S. Giorgio. This seismicity could mark the position of the currently active compressional front of the NA, even if the activity of the compressional front in the AD is presently debated (e.g. Di Bucci and Mazzoli, 2002; Scrocca, 2006).

In an intermediate position, between the presently active extensional and compressional belt, some earthquakes are registered at a depth of 15-25 km, showing transpressional or compressional kinematics. In this region, also some significant historical earthquakes occurred (e.g. Cagli, 1791). These events could represent the expression of the deeper part of the Adriatic crust, even if the oblique kinematics also supports the hypothesis that these events are nucleated along transfer faults, segmenting the major thrusts.

The Italy stress map (Montone et al., 2004), merging the seismological data with other geological information (i.e. borehole breakout data and structural data of active faults), confirms and reinforces this setting, showing two coupled, NW-SE trending, nearly parallel belts of extension and compression.

In the last few years GPS data provided new effective constraints to the kinematic framework of the region (D’Agostino et al., 2009; Hreinsdóttir and Bennett, 2009). At present these data indicate active extension across the main ridge of the Apennines, with an average rate of about 2.5-3 mm/yr, in good agreement with the seismicity data and the other stress filed indicators. The kinematics of the compressional belt are still undefined, and different hypothesis have been made about the present-day tectonic setting of the Adriatic region (e.g. D'Agostino et al., 2008) and about the activity of the Po Plain-Adriatic thrust (Scrocca, 2006).

It is certain that geodetic data will improve continuously in the next years, giving a further contribute to the definition of the present-day tectonic setting.