The main purpose of this study is to obtain the damage scenario for residen
tial buildings in the occurrence of a destructive earthquake (M = 7+) in th
e city area of Catania, Eastern Sicily, and to illustrate the comparative p
erformance of two alternative methods used for this purpose. The methods ar
e representative of two different approaches to estimating the seismic vuln
erability of structures, i.e., an empirical approach based on statistical s
core assignments (widely used in Italy and other countries) and a more rece
nt, mechanical approach that uses displacement limit states associated with
well-defined thresholds of structural damage. A special concern for seismi
c vulnerability in Catania is caused by the fact that earthquake design nor
ms were enforced in its municipal area only since 1981. We emphasise some t
ypical problems encountered in earthquake scenario work, such as the diffic
ulty of assembling a reliable building inventory, and the uncertainties inh
erent in the vulnerability assessments through different probabilistic assu
mptions. Different criteria for the representation of damage are applied an
d discussed. It is shown that the main scenarios obtained by the two method
s are in reasonable agreement, provided a suitable percentile level for dam
age is chosen in the statistical score assignment approach.