Selection and scaling of earthquake records in assessment of structures in low-to-moderate seismicity zones

The behaviour of a building during an earthquake depends on many factors including the characteristics of the ground motion, the configuration and dynamic characteristics of the structure, the hysteretic behaviour of the structural components, the building's non-structural parts, and the local soil behaviour. When conducting seismic assessment studies, controlling the first factor involves selecting and scaling appropriate earthquake records. It is a common practice to scale the records to fit a target acceleration response S_a determined from a pre-determined spectrum or attenuation law. In doing so, two key-problems must be faced: (i) the value adopted for the “effective” period T₁,which lengthens as the structure enters the non-linear range, and (ii) the reliability of the reference value adopted for S_a(T₁). In this work, the influence of selecting appropriate values for these two parameters—T₁ and S_a(T₁)—on the response of the building is investigated from the study of two structures, of 4 and 8 stories. The study does not focus on evaluating a particular seismic code; rather, it uses a European attenuation model as the target spectrum. Non-linear dynamic response analyses are carried out to investigate the Collapse Prevention State of these structures subjected to records representative of a Performance-based Earthquake Engineering level characterized by 2% exceedence in 50 years with 2475 years of return time. The records are selected from a database of European earthquakes, and the failure of the structures is evaluated by statistical means. From the results of the analyses, it is proposed that in seismic assessment studies the reference period for scaling should be 1.1 times the fundamental period of the structure, and at least 30 records should be used to guarantee reliable results.