GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 2.1 37 connection between major earthquakes in the periAdriatic regions are given in a number of papers (e.g., Mantovani et al. , 2012a, 2012b, 2013, 2014; Viti et al. , 2012, 2013). The available data set (a seismic history of some hundreds of years) can be considered representative of the future behaviour of seismicity . This assumption is clearly unreliable, since the spatio-temporal distribution of main shocks depends on the progressive development of tectonic processes, whose temporal scale is much longer than the known seismic history. However, the earthquake catalogues available for the Italian region are reliable and satisfactorily complete for the last centuries at most (e.g., Gruppo di Lavoro MPS, 2004). This time interval is much shorter than the duration of the present seismotectonic setting, presumably started in the Middle Pleistocene (e.g., Mantovani et al. , 2009). The shortness of the available seismic history may crucially affect the results of probabilistic procedures, as pointed out by Swafford and Stein (2007). Seismic activity in the study area can be considered as the effect of a number of seismogenetic zones. This idea could be reasonable, but one must be aware that it makes the results of this approach very dependent from the proposed seismogenetic zoning. In this regard, it has been pointed out that the zones adopted for elaborating the present hazard map are scarcely compatible with the seismotectonic setting in the northernApennines. ��� ��� ������� ������� ��� For the Italian region, the detailed ZS4 model [80 zones: Meletti et al. (2000)] has been replaced by the much simpler ZS9 model [36 zones only: Meletti et al. (2008)]. This choice has favoured the statistical analysis, since in average each new, larger zone includes more earthquakes. On the other hand, the new seismic zoning is less realistic with respect to the seismotectonics context in the Italian aera. For instance, the Apennine chain is currently subdivided in few, very long belts, which heavily affect the resulting hazard estimates. However, the distribution of historical seismicity (Fig. 1) shows that the real seismic sources are not uniformly distributed within the adopted seismogenic zones (Mantovani et al. , 2011, 2012b, 2013). Thus, considering the scarce reliability of its basic assumptions, the results obtained by the Cornell’s approach can hardly be used to exclude the occurrence of earthquakes with intensity values I>VIII in the two Regions here considered during the next 50 years. The need of having a more realistic evaluation of seismic risk has led the Toscana and Emilia-Romagna Regions to promote further studies based on more reliable approaches. The results of such investigations (Mantovani et al. , 2011, 2012b, 2013) are presently utilized by the above Regions to manage the initiatives for seismic risk mitigation in the respective territories. In the next section we describe the results of an analogous study carried out for the Marche and Umbria Regions. A novel seismic hazard assessment for Umbria and Marche. The profound difference between the methodology here proposed and the Cornell’s approach is that we do not try to estimate the probability of future earthquakes (and related seismic shaking), since we are aware that such evaluation would be affected by unknown uncertainty. Thus, our primary interest is defining the highest values of intensity ever felt in each commune, by the analysis of the available macroseismic data set (DBMI11, Locati et al. , 2011), and tentatively recognizing the zones that, in spite of the fact that were not hit by strong events in the past, may be prone to major events. This last result has been tentatively achieved by exploiting the detailed reconstruction of the present tectonic setting we have achieved during a long and accurate analysis of a huge amount of observations in Earth Sciences (e.g., Mantovani et al. , 2009; Viti et al. , 2011). The geometries of the zones finally adopted are shown in Fig. 2. For each zone it is defined a maximum intensity (Imax), on the basis of the seismic history and of the tectonic setting and it is assumed that at any time an event of the proposed Imax can take place anywhere in that zone. Taking into account the implications of the seismogenic zones considered and the documented damages, a value of Imax is proposed for each commune of Umbria and Marche. The information acquired during the study is synthesized in provincial tables, of the type shown in Tab. 2. For each commune, the table reports the Imax value we propose on the basis of