A suitable characterisation of seismic input at the sites where damage is reported is a key aspect of the study of empirical seismic fragility. The availability of a shakemap, specifically derived for the event of interest from recorded ground motions, allows considering site conditions, spatial variability and possible directivity effects. Whenever this is not available, alternative solutions must be considered, such as for example the use of ground motion prediction equations (GMPEs) or the generation of conditional spatially correlated ground motion random fields (CSCRFs). Each of these approaches bears several uncertainties, which obviously have an effect on the derived empirical damage probability matrices. This paper explores issues and procedures related with the ground motion characterisation and assesses their impact on empirical damage distributions derived from a comprehensive post-earthquake damage database, compiled after the 2009 L'Aquila (Italy) seismic event. Several GMPEs are qualitatively and quantitatively ranked with respect to the available accelerometric data and the damage probability matrices obtained with the best performing GMPE are then compared with results obtained using more refined ground motion characterisations.
Influence of seismic input characterisation on empirical damage probability matrices for the 2009 L'Aquila event
Rosti A.;Rota M.;Penna A.
2020-01-01
Abstract
A suitable characterisation of seismic input at the sites where damage is reported is a key aspect of the study of empirical seismic fragility. The availability of a shakemap, specifically derived for the event of interest from recorded ground motions, allows considering site conditions, spatial variability and possible directivity effects. Whenever this is not available, alternative solutions must be considered, such as for example the use of ground motion prediction equations (GMPEs) or the generation of conditional spatially correlated ground motion random fields (CSCRFs). Each of these approaches bears several uncertainties, which obviously have an effect on the derived empirical damage probability matrices. This paper explores issues and procedures related with the ground motion characterisation and assesses their impact on empirical damage distributions derived from a comprehensive post-earthquake damage database, compiled after the 2009 L'Aquila (Italy) seismic event. Several GMPEs are qualitatively and quantitatively ranked with respect to the available accelerometric data and the damage probability matrices obtained with the best performing GMPE are then compared with results obtained using more refined ground motion characterisations.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
