Estimation of response spectra and simulation of nonstationary earthquake ground motions

Abstract Italian strong-motion data were used to study the attenuation of response spectra and to simulate artificial accelerograms as a function of magnitude, distance, and site geology. The database has already been utilized for the study of the attenuation of peak ground acceleration (PGA) and velocity and consists of 95 accelerograms from 17 earthquakes of magnitudes ranging from 4.6 to 6.8. Using multiple regressions, we developed empirical predictive equations for the vertical and horizontal components of response spectra corresponding to 14 frequencies ranging from 0.25 to 25 Hz. Predictive equations, aimed at the ground-motion simulation, were also estimated for time-dependent frequency parameters, strong ground motion duration, and Arias intensity. The shape of the predicted spectra is strongly dependent on magnitude and nearly independent of distance. Alluvium sites show an amplification effect, with respect to stiff sites, in different frequency ranges according to the thickness of the soil deposit. The vertical/horizontal spectral ratio in far field varies, with magnitude and frequency, from 0.35 to 0.85. The resulting response spectra are compared with the predictions of some recent attenuation relationships and with those proposed by the Eurocode EC8. The simulation of nonstationary strong ground motions is achieved through an empirical method where time and frequency features of the motion are represented through the physical spectrum, extending the spectral moments theory to the nonstationary case. The simulated time histories fit the recorded accelerograms in terms of several ground-motion amplitude measures, such as peak acceleration, peak velocity, Fourier spectra, and response spectra. The principal advantage of the proposed method consists in correlating the simulation parameters with earthquake magnitude, source distance, and soil conditions.