GNGTS 2022 - Atti del 40° Convegno Nazionale

GNGTS 2022 Sessione 2.2 331 Fig. 2b and 2c show the empirically-derived fragility curves for irregular mid/high-rise masonry and post-1981 RC building typologies as a function of PGA, the latter being estimated using both the PBS (solid lines) and the ShakeMap (dashed lines). The two sets of curves (PBS Vs ShakeMap) turn out to be consistent, without any systematic bias. PBS-derived fragility curves tend to be less conservative than ShakeMap-derived ones for masonry buildings (especially for the MH class), while a reverse trend is found for RC buildings. Fig. 2 - (a) Comparison between the map of PGA-Hmax in the detailed study area from PBS (left) and from the ShakeMap (right). (b) Fragility curves of irregular mid/high-rise masonry building typologies as a function of PGA estimated from PBS (solid lines) and ShakeMap (dashed lines). (c) As in (b) but for low, mid and high-rise seismically designed (post 1981) RC buildings. Finally, the two sets of fragility models, from PBS and ShakeMap, are coupled with the corresponding ground shaking scenarios to check the consistency of the predicted damage levels and of their spatial distribution in the L’Aquila municipality area, with respect to the observed ones. In Fig. 3a, the spatial distribution of prescribed damage levels (i.e. DS1, DS3 and DS4) is shown, as obtained from PBS and ShakeMap. Results are displayed in terms of expected exceedance probabilities evaluated at the building level. In addition to this comparison at spatial scale, Fig. 3b illustrates the comparison between the observed and predicted global