GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 1.1 21 A widely used practice invokes the criterion of “segment seismic moment conservation” proposed by Field et al. (1999), where the T mean can be obtained by estimating the M max , provided that three-dimensional geometry and slip rate of a seismogenic structure are known. Peruzza et al. (2010) extended this approach by introducing the estimated T mean and α via errors propagation which occur in estimating maximum magnitude and slip-rate. Applying this methodology, Peruzza et al. (2011) demonstrated that the probability of occurrence of an event with M > 6 for the Paganica fault before the April 6, 2009 earthquake, considering an exposure time of 5 years, was the highest of central Apennines (~3.5%). Actually the EP code uses as input information for each seismogenic source the following parameters: 1) fault name, 2) kinematics, 3) length along strike, 4) width along dip, 5) minimum and maximum slip-rate, 6) observed characteristic/maximum magnitude (optional), 7) standard deviation of the observed characteristic/maximum magnitude (optional), 8) elapsed time since the last characteristic/maximum earthquake (optional). In detail the code uses different empirical and analytical relationships between the geometry of each input source and the characteristics of the expected earthquake, in order to quantify several values of M max and associated T mean . The EP code, therefore, formally propagates the errors of magnitude and slip-rate obtaining, for each seismogenic source, the most likely value of recurrence interval and the associated error. Finally, it uses the selected values to calculate the hazard rates, for a given exposure time, following a BPT probability density function (time- dependent) and a Poissonian distribution. Fault parameters and earthquake scaling relationships. ��� �������� ��� ����������� �� The analysis and integration of different types of data such as tectonics, active faulting and long-term seismicity have produced a first seismotectonic model of the Etna region including information on segmentation, kinematics and seismic behaviour (Azzaro, 2004). Later, geometry and slip-rates of active faults have been constrained by geological/geomorphological field investigations (Azzaro et al. , 2012a), while geodetic data modelling provided information on the extension at depth of faults as well as slip-rates and kinematics in the short-term (Azzaro et al. , 2013a). Finally, the magnitude of the historical earthquakes has been calibrated by means of ���new ad-hoc relationships in terms of M l and M w (Azzaro et al ., 2011). In short, most of the input parameters needed for the EP code are available. A scheme of the faults considered in our analysis is shown in Fig. 1, while the values of input parameters are reported in Tab. 1. Tab. 1 - Fault and seismic parameters used in the analysis. Abbreviations: FF = Fiandaca fault; STF = S. Tecla fault; SVF = S. Venerina fault; MF = Moscarello fault; SLF = S. Leonardello fault; PF2 = Pernicana fault, central segment; kinematics 8 = extensional volcanic context. Fault Kinematics Length Width Min slip-rate Max slip-rate M max σ Elapsed time (km) (km) (mm/yr) (mm/yr) (observed) M max (yrs) FF 8 7.7 3.5 0.9 1.1 4.8 0.36 120 STF 8 7.6 5 4.2 4.4 5.3 0.36 100 SVF 8 5.6 5 0.9 1.1 4.8 0.36 135 MF 8 8 3.5 1.4 2.7 5.1 0.36 149 SLF 8 4 3.5 2.5 2.7 4.4 0.36 113 PF2 8 4.5 5.7 3.3 5.2 4.3 0.3 4