PSI - Issue 78

Marco Fasan et al. / Procedia Structural Integrity 78 (2026) 831–838

835

Fig. 2. Examples of simulated slip distribution on the fault plane. L= fault length while W= fault width in km. White contour lines indicate the rupture isochrones (seconds) while color scheme refers to the amount of slip as percentage of the maximum value. 3. Results and discussion Several ground motion intensity measures were considered. In this work, for the sake of brevity, only the distributions of Peak Ground Acceleration (PGA – Fig. 3) and Pseudo Spectral Acceleration (PSA – Fig. 4) at the first mode of vibration of the structure (mean between x and y horizontal directions as evaluated by in situ dynamic identification, about 0.5s) are reported. Pictures show the simulated values for each earthquake scenarios (sce01 – sce24). The data are grouped into three horizontal blocks, separated by vertical dashed lines, each representing a different fault strike angle: i.e. 285°, 70° and 170°. Colors indicate the fault dip: red dip 30°and dip 66°. Marker shapes represent the rake angle: circles rake 190°, triangles rake 350°. Black circles mark the median value for each scenario, with vertical black lines showing the 16th and 84th percentiles. Results are reported for three horizontal components: North-South, East-West and RotD100 (Boore, 2010). For each scenario the obtained variability is due to the uncertainty due to the slip distribution. The uncertainty due to the source parameters can be identified by the variability between scenarios. The results do not reveal a clear or consistent pattern across the scenarios. While the variability among different cases is somewhat reduced when considering the rotated ground motion component, it remains significant. This highlights the importance of accounting for “epistemic” uncertainties in complex settings by systematically varying the possible key parameters and combining them across multiple scenarios. Such an approach allows for a more comprehensive exploration of the possible outcomes. Naturally, this uncertainty could be narrowed through future studies that improve our understanding of the underlying processes and reduce “ epistemic ” gaps.