Response spectrum shapes implied by earthquakes in Turkey: comparisons with design spectra

Abstract

Design spectrum shape in many recent standards and codes is defined by spectral accelerations for two periods, T_s,(~0.2s) and T_{1,(~1 s)}, modified by multiplicative factors that account for the site class and strength of the ground shaking. This article draws attention to apparent discord between spectra from actual recordings at a number of stations of the national strong motion network of Turkey with the design spectra for the same location and site characteristics as given by the national seismic hazard map. We find that, for deep basins with Site Class D or E profiles, the design spectrum seems not to recognize consistently the constant velocity and longer period demand. If this conjecture is true, it may foreshadow unsafe designs for the building stock in Turkey in similar environments. Many more earthquake recordings than are currently at hand are needed to verify the applicability of using the two-point design spectrum that is embedded in the regulation. This is not a unique problem only to the Turkey Building Earthquake Regulation (TBER 2018). The recently recorded strong ground motion records from the 06 February 2023, M 7.7 and M 7.5 Kahramanmaraş, earthquakes support the conjecture of this study and provide strong evidence for a need to revise the corner periods of the design spectrum in the Regulation.