Tools, methodologies and discussions about single-site, multi-site and sequence-based probabilistic seismic hazard analysis.

In the framework of performance-based earthquake engineering (Cornell and Krawinkler, 2000), actions the structures must withstand are based on probabilistic seismic hazard analysis (PSHA). Classical formulation of PSHA goes back to the second half of the twentieth century (Cornell, 1968), but its implementation can still be demanding for engineers dealing with practical applications. Moreover, in the last years, a number of developments of PSHA have been introduced; e.g., vector-valued and advanced ground motion intensity measure (IM) hazard, the inclusion of the effect of aftershocks in single-site hazard assessment, and multi-site analysis requiring the characterization of random fields of spatially cross-correlated IMs. Although several software to carry out PSHA have been available since quite some time (see Danciu et al., 2010), generally, they do not feature a user-friendly interface and do not embed most of the recent methodologies relevant from the earthquake engineering perspective. These are the main motivations behind the development of a practice-oriented software, namely REgionAl, Single-SitE and Scenario-based Seismic hazard analysis (REASSESS V2.0). The tool, which has been developed within the activities of the AXA-DiSt 2014-2017 research program of AXA-Matrix Risk Consultants, Milan, Italy and Dipartimento di Strutture per l’Ingegneria e l’Architettura, is one main results of the thesis and has been used to develop all the other studies introduced in the following. In the most advanced countries, where PSHA is adopted for the definition of the design seismic actions, the code typically provides them in the form of hazard maps for different pseudo-spectral accelerations and return periods. In other words, for each of the sites the design spectrum is derived from the uniform hazard spectrum in which all the ordinates have the same return period of exceedance (the return period is usually a function of the design limit-state). PSHA has also been often questioned (e.g., Castanos and Lomnitz, 2002; Stein et al., 2003; Reiter, 2004; Musson et al., 2005; Wang, 2012). The ongoing debate on the adequacy of PSHA is (often) feed by the actually-observed seismic actions on structures. For example, when an earthquake occurs at a site, researchers typically compare the design spectrum with the recorded counterpart (e.g., Masi and Chiauzzi, 2009). In this sense, several studies show that the cases where the design spectra are exceeded are not rare (see, for example, Crowley et al., 2009). Another relevant issue concerning PSHA is that it is often implemented neglecting the effect of aftershocks, that can also be strong (e.g., Masi et al., 2011), in order to describe the occurrence of earthquakes according to the homogeneous Poisson process; as a consequence, seismic codes at the state-of-the-art worldwide implicitly assume that the effect of aftershocks is negligible and do not consider that failure of structures can be due to an aftershock rather than by a mainshock (i.e., the event of highest magnitude within a sequence). Finally, in the recent years, PSHA estimates have been confirmed or disproved through hazard validation studies performing formal tests against observed ground motions at multiple sites over the years (e.g., Schorlemmer et al., 2007; Albarello and D’Amico, 2008); the nature and form of these studies implies that results they provide are sensitive to the adopted hypothesis of spatial dependence/independence between ground motions at the sites and, also important, require a careful evaluation of the involved data. For these reasons, the thesis proposes a study which, not questioning PSHA (which is a rational method to quantify the seismic threat for a site), recalls some of the recent advances in the seismic hazard assessment to deepen the above-introduced issues. To do so, the whole discussion is addressed with reference to the case-study of Italy and adopting the same source model used to develop the national seismic hazard. In particular, PSHA is herein studied under three non-conventional points of view, by means of which: • it is demonstrated that the exceedance of design spectrum in the epicentral areas of earthquakes of even moderate magnitude is well expected, identifying the seismic scenarios for which such exceedance is more probable, and quantifying the expected amount of the exceedance when such exceedance occurs; • profiting of sequence-based PSHA (SPSHA) introduced by Iervolino et al. (2014), it is quantitatively shown that the hazard increase due to aftershocks for structural design is not very high, even if it is not negligible. It is also illustrated that the contribution of aftershocks to hazard for a site can strongly vary with return period and that, given the return period, it is different from site to site; • profiting of multi-site PSHA (MSPSHA; Giorgio and Iervolino, 2016), it is demonstrated that hazard validation studies via observed exceedances at multiple sites over the years should always consider the spatial dependence existing between ground motions at the sites generated by a common earthquake, to avoid erroneous conclusions about the inadequateness of PSHA.