GNGTS 2017 - 36° Convegno Nazionale

292 GNGTS 2017 S essione 2.1 releasing this, with regard both to updated input elements and to the strategies to be followed, according to the most advanced procedures adopted worldwide and in order to obtain a shared and largely accepted probabilistic seismic hazard assessment. This new reference hazard model is in agreement with the classical probabilistic approach originally proposed by Cornell, using the Openquake software. The model presented in this study was developed with the aim to be applied as a branch of the logic tree to be used for the new Italian seismic hazard map (MPS16), close to the its conclusion Such seismic hazard map is going to substitute the MPS04, the national seismic hazard map still in force. We have developed for the PSHA in Italy a conceptual, modular approach to generate a smoothed stochastic earthquake annual rate model appropriate for long-term forecasts, considering both Historical and Instrumental seismicity. Our model extends the classical kernel- smoothing method, also to include geologic information: the moment rate contribution released annually on mapped faults with associated deformation rates. Building on the smoothing concept, we combine geological data with seismological data. The smoothed seismicity model takes into account the completeness magnitude of the seismic catalog that changes both in time and space, through the Eq. 1: Ri ≈ j =1Nexp ( dist i,j ) 222·12 π 2·10bmj-mminTc (1) where: is the spatial rate in each cell of the grid, N is the total number of events in the catalog, dist i,j is the distance between the centre of i -th cell and the j -th event, σ is the smoothing distance; b is the b -value of the Gutenberg-Richter law, is the magnitude of completeness relative to the j -th event, is the minimum magnitude of completeness and is the time length (in year) of the magnitude of completeness . Subsequently we merge the smoothed seismicity model with the one based on the mapped fault slip rate. At the end of the process we have got in an unique source model the seismic rates coming from three different databases in order to cover all the Italian territory: the database of Italian Seismogenic Sources (DISS 3.2.1, 2016, Fig. 1), the Historical seismic catalog (CPTI15) and the Instrumental catalog (1981-2016). The latter catalog was used because of its lower completeness magnitude M w 3.0. Regarding the magnitude distribution, we use for all catalogs a Tapered Gutenberg-Richter approach with a b -value equal to 1 and uniform for the whole Italian territory. The annual rate value for the final model, obtained for each geographical cell of 0.1° x 0.1° and for each magnitude bin starting from M w 4.45, is determined in the following way: if the cell falls within one of the seismic sources, we adopt the respective value of the smoothed seismicity rate averaged with the seismicity rate estimated from fault slip rate; if instead the cells fall Fig. 1 - Annual rates computed for the individual sources (DISS 3.2.1, 2016).