Seismic Response of the Temple of Pythian Apollo in Rhodes Island and Recommendations for Its Restoration

In this paper, the response of a restored part of the ancient Temple of Pythian Apollo under strong seismic motions is examined. The monument is located on the ancient Acropolis of Rhodes and, in its present condition, consists of a single free-standing column and a colonnade composed of three columns and an entablature. Only a few drums contain ancient parts found in the excavations, while the rest is newer material, placed during the Italian restoration of 1937-38. The examined structure corresponds to the monument after a probable restoration scenario, which accounts for the replacement of all complements that were installed during the Italian restoration with new parts from natural or artificial stone. The performed investigation concerns the response of the structure under six strong seismic excitations, chosen to be compatible with the tectonic environment and past seismic events, scaled to two levels of ground acceleration, corresponding to return periods of 500 years and 1000 years. The structural analyses were performed with the finite element software ABAQUS. The monument was modelled using three-dimensional deformable finite elements, while special attention was paid to the modelling of the mortar that connects structural members, as well as the interface between the drums, where rocking and sliding are allowed. The results of the analyses lead to various conclusions about the response of the structure under the considered seismic scenarios, including the danger of collapse of the monument, the residual displacements induced by the earthquakes, the stress levels that develop in the structure and across the interfaces and the expected damage to the structural members. Based on these results, conclusions are drawn regarding the adequacy or not of the strength capacity of the proposed new parts (natural and artificial stones and connecting mortar), while the possibility of damage to ancient parts during earthquakes is also examined. The main conclusion is that, if the structure remained undamaged, it could survive five out of six of the examined earthquakes, except of the stand-alone column which collapses in half of them. However, this is not the expected situation, since damage will occur to several new and ancient parts during a strong seismic event, increasing significantly the vulnerability of the structure and the risk of partial or total collapse.