PSI - Issue 44

Annalisa Rosti et al. / Procedia Structural Integrity 44 (2023) 91–98 Annalisa Rosti et al. / Structural Integrity Procedia 00 (2022) 000–000

93

3

Fig. 1. Spatial distribution of the considered masonry churches and geographical location of the epicentres of the reference main shocks of the seismic sequence (http://terremoti.ingv.it/). 3. Seismic input definition The peak ground acceleration (PGA) was estimated from the latest INGV ShakeMaps (Michelini et al. 2020) and adopted for locally characterizing the ground motion severity. To suitably account for the earthquake sequence (Fig. 1), a single value of PGA was associated with each post-earthquake survey form, by taking the maximum value evaluated at each church location before the post-earthquake field inspection. 4. Classification of the observed seismic damage Post-earthquake surveys were conducted using the A-DC survey form (PCM-DPC MiBAC 2006), which defines twenty-eight possible collapse mechanisms (Fig. 2), referred to different macroelements constituting the church (Doglioni et al. 1994). For each damage mechanism with potential activation, the post-earthquake survey form allows for reporting the observed level of damage, based on a graduated scale from 0 (no damage) to 5 (collapse), in line with the EMS-98 (Grünthal et al. 1998). A suitable damage metric has thus to be established for passing from damage evaluation at the level of individual collapse mechanisms/macroelements to the global scale (i.e. church). In this context, alternative damage metrics, either based on the mean damage level (e.g. Cescatti et al. 2020; Canuti et al. 2021; Salzano et al. 2022) or maximum level of damage (e.g. Carbone 2021), may be employed, also depending on the purpose of the study. Approaches based on the mean level of damage are suitable for arranging short-term countermeasures, provisional and restoration interventions. On the other hand, metrics based on the maximum damage could be more appropriate for large-scale impact scenarios in terms of unusability (e.g. da Porto et al. 2021), which is generally driven by the peaks of damage (e.g. Rosti et al. 2018). Alternative damage metrics, accounting for the geometrical extent of individual macroelements, can be also found in the literature (e.g. Lagomarsino et al. 2019). In this study, global damage levels were defined based on the maximum observed seismic damage, considering damage information collected during the first post-earthquake survey. In case of churches with multiple inspections, damage levels were also evaluated based on damage data related to the last post-earthquake survey. Fig. 3 shows the spatial distribution of the identified masonry churches. Colors identify the associated damage level, based on damage data gathered during the first (Fig. 3 a) and last (Fig. 3 b) post-earthquake survey. Damage progression due to the cumulated effect of repeated earthquake shaking involves the increase of the frequency of occurrence of higher damage levels from the first to the last post-earthquake inspection. However, also considering that churches with multiple surveys represent a limited fraction (<10%) of the adopted dataset, increment of damage driven by repeated ground shaking is generally not very evident.