GNGTS 2017 - 36° Convegno Nazionale

286 GNGTS 2017 S essione 2.1 Comparison of occurrence rates from NDSHA maps and observed intensities A. Magrin, A. Peresan CRS-OGS, National Institute of Oceanography and Experimental Geophysics, Udine, Italy Neo-Deterministic Seismic Hazard Assessment (NDSHA) (Panza et al. 2001, 2012) means scenario-based methods for seismic hazard analysis, where realistic and duly validated synthetic time series, accounting for source, propagation, and site effects, are used to construct ground motion scenarios. NDSHA, in its standard form, defines the hazard as the envelope ground shaking at the site, namely the maximum estimate computed from a large set of possible scenario earthquakes. Thus, the standard NDSHA maps provide rather robust and conservative hazard estimates, which do not require any assumption about the probabilistic model of earthquakes occurrence (e.g. Panza, 2017). Some specific applications, however, may benefit from temporal information about the computed ground shaking, including a gross estimate of its average occurrence time. The standard NDSHA procedure has been recently expanded and provides the option to take into account the information about earthquake occurrence rates (Magrin et al. , 2017), whenever this information is required and for the areas, if any, where reliable rate estimates are available. The frequency-magnitude relations for earthquakes in the Italian region used in Magrin et al. (2017) are obtained according to the multi-scale seismicity model (Molchan et al. , 1997), which accounts for the extent of the investigated territory. In the first step of the procedure an occurrence rate estimate is associated to each of the modeled sources. The synthetic seismograms are then computed at the different sites of interest following the standard NDSHA procedure (Panza et al. , 2001, 2012). The rate of occurrence of the seismic source is also associated to the related seismograms; therefore the occurrence rate of the maximum ground motion obtained by standard NDSHA can be estimated. The introduction of occurrence rates in NDSHA also allows for the generation of ground shaking maps associated with a specified occurrence time, which permit a straightforward comparison between the NDSHA and the PSHA maps in terms of rates. Here we report about a preliminary cross-checking between observations and model ground motion rates. Since the instrumental data cover just a short time window, we cannot compare the estimated rates of ground motion with the rates of recorded ground motion. Therefore the only possible verification is against the rates of observed macroseismic intensities. The use of intensity data poses two different problems: the comparison of ground motion parameter values with intensity values and the estimate of completeness of the dataset at each site. Following Nekrasova et al. (2014), we choose to convert design ground acceleration (DGA) values in intensity values in MCS scale (I MCS ), using the relation of Panza et al. (2001). To minimize completeness problems, we consider only I MCS ≥VII and the major cities of each Province; in such a way we consider sites with a generally long history and quite uniformly distributed over the Italian territory. We consider the macroseismic data from DBMI15 (Locati et al. , 2016), the latest release of the Italian Macroseismic Database. The database includes intensity data for the Italian region in the time window 1000-2014. In the comparative analysis we associate each selected site to the nearest point on the grid used for the NDSHA computation and we estimate the occurrence rate of intensity from DBMI15 at each site. For each site we estimate also the rate of occurrence of DGA associated with intensity values from VII to the maximum. From this preliminary verification test we can formulate some preliminary conclusions: • the conservative choices in the procedure lead to a conservative estimation of occurrence rates; • occurrence rates of earthquake used in the model could be overestimated, particularly in northern Italy;