Evaluation of site effects by means of 3D numerical modeling of the Palatine Hill, Roman Forum, and Coliseum archaeological area

In this study we perform 3D nonlinear analyses of seismic site response of the Central Archaeological Area of Rome, which includes the Palatine Hill, Roman Forum, Circus Maximus, and Coliseum. The geological bedrock of the study area is constituted by a Pliocene marine sandy-clayey unit (MonteVaticano Formation, MVA). At top of this unit a continental Quaternary succession is superimposed. Previous studies available for this area (Pagliaroli et al. 2014a; Mancini et al. 2014; Moscatelli et al. 2014) enabled to define a detailed three-dimensional reconstruction of the subsoil conditions, characterized by complex surficial and buried morphology, lateral heterogeneities and dynamic properties of involved material, natural as well as anthropogenic. The area of Rome is affected by earthquakes from different seismogenic districts: i) the central Apennine mountain chain (D = 90–130km and M = 6.7–7.0); ii) the Colli Albani volcanic district (D = 20km and M=5.5); iii) Rome area itself, which is characterized by rare, shallow, low-magnitude events (M < 5). Both natural and artificial signals have been considered to define the input motion for the numerical modeling of the site response of the whole archeological area. This was accomplished by means of the finite differences code FLAC3D. To evaluate the seismic hazard and, consequently, to assess possible priorities for seismic retrofitting of the monuments, contour maps of Housner intensity amplification ratio FH (defined as the ratio between Housner intensity at the top of the model and the corresponding input at the bedrock outcrop), are carried out. To cover the entire range of natural periods pertaining to the monuments in the examined area, FH was evaluated over three ranges of period: 0.1–0.5s, 0.5–1.0s, and 1.0–2.0s. Numerical results shown that: 1) within the range of periods 0.1–0.5s, high values of FH = 2.2–2.6 occur both in correspondence of narrow valleys filled with soft alluvial deposits and at top of Palatine Hill; 2) within the range of periods 0.5–1.0s, high values of FH occur in correspondence of the deepest valleys; 3) within the range of periods 1.0–2.0s, low values of FH occur except in correspondence of the deepest valleys.Results show a good agreement with the previous 2D numerical modeling and with the microzonation maps (Pagliaroli et al 2014a, b), even if interesting differences show up highlighting the usefulness of 3D modeling in such complex settings. Such results are significantly relevant for the monumental and archaeological heritage of this area, as it is highly vulnerable due to its old age and state of conservation.