GNGTS 2015 - Atti del 34° Convegno Nazionale

GNGTS 2015 S essione 2.1 19 The results pertain to the fields of the probabilistic seismic hazard analysis related to the tuning and application of validation methods, the implementation of innovative methodologies to include site effects, the exploitation of detailed seismotectonic models, and the development and acquisition of new tools and data useful for seismic hazard. Finally, the project has promoted some activities related to induced seismicity and seismic risk perception, with potential significant implications regarding the seismic hazard reduction strategies. The focus of the project was the validation exercise of Probabilistic Seismic Hazard Assessment (PSHA), started in S2-2012 project and continued within the Task 6 in 2014-15. The scoring test completed during the S2-2012 Project highlighted the need of having: i) high quality site response characterizations for the accelerometric stations, ii) a careful check on completeness of recordings, not fully accomplished by the information available in 2013 in the ITACA archive (http://itaca.mi.ingv.it ), and iii) reinforced understanding of some methodological aspects of scoring. These issues have been addressed in this project and the validation exercise strongly benefitted from the results obtained by the Task 2, Task 3 and Task 4. In particular, the adoption of site amplification inferred from the experimental data analysis carried out within the Task 2, instead of the simplified NTC8 coefficients, has significantly changed the previous ranking results, highlighting the importance to have realistic amplification factors to correct the ground motion observed on soil categories different from rock condition. Furthermore, significant methodological improvements have been obtained taking into account the ground motion spatial correlation on seismic hazard estimate validation. This fact influences the validation test, too. In general the ground motion spatial correlation increases the variance of the variable used in the statistical test, and it makes more acceptable the ground motion observations. Under Task 2, an extensive campaign of site characterization has been carried out, in synergy with the project GEORAN coordinated by CNR-IGAG and funded by DPC, in order to establish and apply the best practice for site characterization, following the standards in data acquisition and post-processing of the seismic microzonation defined at national level. Among the 77 RAN stations, active since 1975 or at least continuously operating for 25 years, 46 sites have been investigated by different research units to provide detailed site-specific characterizations. All the collected data have been included (or, have being progressively included) in the ITACA database, increasing of about 50% the number of stations for which good geological and geophysical information is available. Notably, about 1/2 of the studied stations had changed their previously soil categories, assigned only on the basis of surface geology. Innovative methodological approaches in PSHA have been proposed within Task 4 and Task 5, helping to clarify certain aspects still debated in the context of seismic hazard. The Task 4 implemented and tested different methods (with increasing complexities) to introduce the local seismic response in PSHA at three sites (Mirandola, Casaglia and Soncino) that have been characterized in the context of the previous project. The PSHA for generic sites (Level 1), estimated considering the appropriate values of the site parameters (Vs30 or soil categories) in the Ground Motion Prediction Equations (GMPEs), performs quite well when compared to a more sophisticated approach (Level 4). Major problems can rise for sites that can have non-linear soil behaviour, or in case of shallow alluvium soil condition. The PSHA for generic site was then performed at the 77 RAN stations, exploiting the results obtained by Task 2. Thirteen different GMPEs were applied with the aim of examining the influence of different GMPEs on the final hazard and evaluating improvements in ground motion prediction by the comparison with observations at the single recording station. Task 5 acquired new data and tested an alternative approach with respect to the standard model (MP04) for the Calabria region, reaching important results such as the development of a new seismotectonic model that allows the evaluation of earthquake probabilities for the largest