GNGTS 2013 - Atti del 32° Convegno Nazionale

2. Level 1 and Level 2 hazard analyses produced similar results, with larger differences (of about 0.1g), between 0.1 and 0.2s, at the site of Soncino. At Casaglia, Level 1 and Level 2 UHSs are very close to those obtained following the non-ergodic approach (Level 3) 3. Except for Mirandola, where the Level 3 approach produces hazard values that are significantly higher than those resulting fromthe other approaches (due to the combination of positive site terms and large single station sigmas), the Level 4 assessment led to the highest estimates. However, although the figure may deceive the eye of the reader due to the different GMPEs used, the Level 4 UHSs are on average 1.2-to-1.5 times greater than the other spectra; at Casaglia, the large peak at 0.1s, which is particularly evident when the Level 4 simplified approach is applied, is due to the very large uncertainty in the soil amplification). Apart from the approach used, part of the differences in the results may be explained by the propagation of the uncertainty in soil amplification, which is not present within the framework of the Level 1 and 2 hazard calculations. Finally, concerning the Level 3 GSH assessment, one should note that, although the repeatable components of ground motion variability at a single site were identified and removed from the total aleatory variability, they were not transferred to the quantification of the epistemic uncertainty affecting the GMPE site correction term (e.g., through logic tree branches), with possible implications on final hazard estimates. 4. Except for the site of Soncino, where soil nonlinearity should be negligible, the Level 4 approaches take into account the nonlinear behaviour of soils which, subject to the condition that the real response of the sites investigated is accurately represented by the 1D response computed by the software at hand, is reflected in the UHSs of both Mirandola and Casaglia with evident differential amplifications at different spectral periods. Note that the soil nonlinearity resulting from 1D analysis, which was not observed from the experimental spectral ratios, may be related to “extreme” cases (here the word “extreme” is used to indicate outcomes having low probability of occurrence) resulting from the combination of “extreme” random soil samples and accelerograms having PGA > 0.25g or larger. 5. Although Level 3 and Level 4 lead to potentially more accurate estimates, they cannot be simply applied to large-scale GSH assessment as they require a large amount of site-specific data. However, they can be extended to small areas surrounding sites with sufficient recorded data, subject to geological analysis; both approaches could be applied to target soil profiles representative of different lithological conditions in order to extend results to larger areas (e.g., regional hazard assessment) 6. Although they cannot produce rigorous probabilistic results, Levels 0 and 1 appear to be the most efficient for large-scale hazard mapping. Acknowledgements. The process of collection of geological, geotechnical and geophysical data was strongly supported by the Region of Emilia Romagna that provided data without a formal participation to the project. References Al Atik L., Abrahamson N., Bommer J. J., Scherbaum F., Cotton F. and Kuehn N.; 2010: The variability of ground-motion prediction models and its components . Seismological research letters, 81 , pp. 794-801. Ambraseys N. N., Simpson K. A. and Bommer J. J.; 1996: Prediction of horizontal response spectra in Europe . Earthquake Engineering and Structural Dynamics, 25 , pp. 371-400. Anderson J. G. and Brune J.; 1999: Probabilistic seismic hazard analysis without the ergodic assumption . Seismological Research Letters, 70 , pp. 19-28. Barani S., De Ferrari R. and Ferretti G.; 2013: Influence of Soil modeling Uncertainties on Site Response . Earthquake Spectra, 29 , pp. 705-73. Barani S., De Ferrari R., Ferretti G. and Eva C.; 2008: Influence of Soil Property Uncertainties on Ground-motion Amplification . Proceedings of the 31st Assembly of the European Seismological Commission - ESC08, Hersonissos, Greece, pp. 49-58. Barani S., Spallarossa D. and Bazzurro P.; 2009: Disaggregation of probabilistic ground-motion hazard in Italy . Bulletin of the Seismological Society of America, 99 , pp. 2638-2661. 21 GNGTS 2013 S essione 2.1