GNGTS 2014 - Atti del 33° Convegno Nazionale

24 GNGTS 2014 S essione 1.1 with respect to the worldwide used relationship (Wells and Coppersmith, 1994) here given as a function of rupture area, subsurface rupture length and surface rupture length (all referred to normal faults). It is important to highlight that the Wells and Coppersmith (1994) relationships are extrapolated outside its definition ranges and applied to volcano-tectonic environments. These considerations suggested us introducing the Villamor et al. (2001) relationship in the area-based computations of EP code, but leaving the Wells and Coppersmith (1994) ones. The EP code, in addition to the M max s calculated by the above defined empirical scaling relationships, defines for each fault other two expected M max : one from the general formula of magnitude as a function of the scalar seismic moment (M0 in Fig. 3), starting from a constant strain drop value (here 2 X 10 -5 ); and the other (MAS in Fig. 3) by using the aspect ratio relationships derived by Peruzza and Pace (2002) on a slightly modified Wells and Coppersmith (1994) data set. Preliminary results and conclusions . ��� ������ �� ��� �� ���� ������� �� ��� ���� ���� The output of the EP code applied to the Etna case is reported in Tab. 2, where the most likely values of characteristic expected magnitude ( M char ) with the associated standard deviation σ, the corresponding mean recurrence times ( T mean ) and the aperiodicity factor α , are indicated for each fault. The obtained α values suggest fault behaviours potentially modelled by a time-depended approach. Fig. 3 reports the calculated M max values for four faults, following all the different approaches with the associated uncertainty and, if available, the observed (historical) M max . The upper curve (‘SUM’ in Fig. 3) represents the summation of the M max ’s, treated as probability density functions, in order to evaluate a reference “mean” value (‘Ref’ in Fig. 3). This representation is useful to evaluate whether the observed M max value is in agreement or not with the values of maximum rupture calculated ( M char ) from the geometry/kinematic. It is important to stress that the M max values calculated by the different methods are comparable, also with the observed historical earthquakes except the case of PF2 and partially the SVF. As regards PF2, the partial discrepancy between the M max calculated by the proposed Etna scaling relationship and the others values, can be ascribed to the “strange” fault geometry showing a length along dip ( W ) larger than the one along strike ( L ). The results of this work, actually in progress, suggest that a geological approach based on geometric-kinematic parameters to estimate the expected seismicity rates can be adopted with success on the volcanic context of Etna. A comparison between our results with scalar moment rates estimated from seismic and geodetic data will provide important constraints on the fault parameters and validate the goodness of the methodology. Tab. 2 - Output of EP code for the studied faults. The characteristic magnitude ( Mchar ) is calculated according to the Etna SRL-M relationship shown in Fig. 2. Fault abbreviations as in Tab. 1. Fault M char (calculated) σ M char T mean (yrs) α FF 5.2 0.3 264 0.54 STF 5.4 0.2 67 0.47 SVF 5.1 0.3 182 0.52 MF 5.3 0.2 138 0.59 SLF 4.8 0.3 45 0.52 PF2 4.9 0.4 27 0.65