GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 2.1 GNGTS 2024 Site-specific checks of probabilistic seismic hazard models with macroseismic historical records Roberto Paolucci Dipartimento di Ingegneria Civile e Ambientale, Politecnico di Milano, Milano, Italy Abstract: Probabilistic Seismic Hazard Assessment (PSHA) provides estimates of annual probability of exceedance of ground motion amplitude at a site. Validation or falsification of PSHA results from ground motion records suffer of two main limitations: (i) since a sufficiently large amount of data should be collected at that specific site to make statistical analysis meaningful, checks cannot be carried out at the site scale, but by integrating records from a more or less large portion of the territory where the PSHA is carried out and (ii) the time interval is not sufficiently extended to cover those that are of relevance for the PSHA applications. In a nutshell, testing with ground motion records the PSHA result at a specific site (say, the town of Florence) is not presently feasible, unless very short return periods are considered that are of no relevance from the practical viewpoint. A different perspective comes if, instead or in addition to ground motion records, use is made of the macroseismic historical records. If reference is made to the Italian context, likely representing that of several European countries, completeness of the catalogue of locally observed macroseismic effects (including lack of effects)) during past earthquakes at many historical sites in terms of moderate-to-large values of macroseismic intensity may extend back to at least several centuries. While several researchers argue that correlations between ground motion and macroseismic intensity may be relatively poor, so to prevent the use of the latter one for PSHA testing, such lack of correlation occurs also because the ground motion amplitude recorded at a given site may not be in itself representative of that of a wider urban area, as, instead, is the case of the macroseismic intensity. With no ambition to introduce novel advanced approaches, in this contribution we will present criteria and application examples with reference to a consistency check, with macroseismic historical records, of the hazard estimates at several Italian towns from two seismic hazard models.