GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 1.1 35 In order to estimate the geologic moment-rate for active faults cutting the eastern flank of Mt. Etna, we collected all fault geometric and kinematic information from published and unpublished studies (see Azzaro et al. , 2013 and references therein). More in detail, for each structure we collected the following parameters: length ( L ), strike ( θ ), dip ( φ ), down-dip width ( H ), and slip-rate. For this last parameters, when available we used both short- and long-term estimations. The considered faults have length typically in the ~5-8 km range and down-dip width in the 1.6-4.5 km range. Long-term slip-rate estimations range from 1 up to 5.2 mm/yr while short- term slip-rate values for the last decades range from 2 up to 28 mm/yr. Higher slip-rate values are observed along the Pernicana fault, especially on its easternmost segment (Palano et al. , 2006). Method and preliminary results. �� The methodology applied in the southern Apennines (Palano et al. , 2011) to estimate and compare the scalar seismic, geodetic and geological moment-rates has been used in �� the eastern sector of Mt. Etna. In particular, taking into account the spatial distribution of the active faults and their seismogenic thickness H s , as a first step we divided the volume into prismatic bodies. Then, by adopting the classical formulations below reported, for each prismatic body we estimated and compared the aforementioned moments- rate scalar values as follows. The seismic moment rate was calculated according to the Kostrov (1974) general formulation where N is the number of events occurred in the selected interval in the volume A· H s ( A and H s represent the surface area and the seismogenic thickness of the selected prismatic body), M seis is the seismic moment of the n-th earthquake from the N total earthquakes, and ΔT represents the considered time interval (in our case 8 years). To perform the computation we considered all earthquakes with magnitude M ≥ 1.5 - i.e. the value of the completeness magnitude - selected for the investigated area (see Fig. 1). Earthquake magnitudes were converted into scalar moments through the relationship proposed by Giampiccolo et al. (2007). Our preliminary estimations range in the interval 5.82 · 10 14 - 7.21 · 10 14 Nm/yr. The geodetic moment rate was estimated by taking into account the observed horizontal velocity field and associated covariance information; in a first step we computed the expected velocity values at the nodes of the selected area (i.e. at the top surface of each prismatic body) and as a second step we derived the average 2D strain-rate tensor at the center of each selected area. Then, by adopting the previous formulation, we computed the geodetic moment-rate using the Savage and Simpson (1997) formulation , where μ is the shear modulus of rocks within the focal volume, with assigned value of 10 GPa, H s is the seismogenic thickness, A is the surface area of the selected prismatic body and and are the principal horizontal geodetic strain-rates. We found that our preliminary geodetic moment-rate estimations range between 3.18·10 18 - 1.03·10 19 Nm/yr. Lastly, the geological moment rate was computed using the Brune (1968) formulation, where μ is the shear modulus of the rocks involved in faulting with assigned value of 10 GPa, L and H are length and down-dip width, respectively, of the dislocation along which faulting and u is the average displacement, which corresponds to the estimated geologic slip-rate (see Azzaro et al. , 2013 and references therein). Our preliminary estimation of the geological moment rate ranges in the interval 4.92·10 14 - 2.81·10 16 Nm/yr. As already observed by Azzaro et al. (2013) comparing the short- vs. long-term slip-rates of active faults in the Etna region, also the geodetic moment-rate estimations computed in this work are generally larger than the seismic and the geologic ones. In general, a number of reasons may account for the observed discrepancy among the estimated moment-rates, such as: i) uncertainties of the geologic slip-rate estimations; ii) poorly defined geometry of modeled faults, especially for down-dip width; iii) role of buried/hidden fault segments in accommodating deformation; iv) limited length of the instrumental earthquake catalogue. Despite this, it is well evident that a large amount of the deformation affecting the eastern flank occurs aseismically,