Premium
Influence of spectral acceleration correlation models on conditional mean spectra and probabilistic seismic hazard analysis
Author(s) -
RodríguezCastellanos Ali,
Ruiz Sonia E.,
Bojórquez Edén,
ReyesSalazar Alfredo
Publication year - 2021
Publication title -
earthquake engineering and structural dynamics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.218
H-Index - 127
eISSN - 1096-9845
pISSN - 0098-8847
DOI - 10.1002/eqe.3331
Subject(s) - seismic hazard , seismology , probabilistic logic , peak ground acceleration , correlation , vibration , spectral acceleration , intensity (physics) , conditional probability , conditional expectation , computation , geology , structural engineering , engineering , mathematics , ground motion , statistics , physics , algorithm , geometry , acoustics , quantum mechanics
Summary Innovative engineering applications are proposed to satisfy the current necessities of earthquake engineering, which are well established from a conceptual perspective, and frequently, the development of some missing inputs is only required to employ them, such as the case of the correlations between spectral values at different vibration periods. Accordingly, we computed the correlation between spectral accelerations at two vibration periods using ground motions from interplate and, alternatively, intraslab events, recorded in firm ground of Mexico City. Results show that the spreading of correlation values depends on the rupture mechanism. Then, based on hypothesis test analyses, we used four correlation models available in the literature to predict our correlation values. According to the findings, we proposed a predictive correlation mathematical model for interplate and, separately, for intraslab seismic events. We evaluated the influence of the predictive correlation models on the results corresponding to two earthquake engineering applications. The first refers to compute conditional mean spectra and the second to perform probabilistic seismic hazard analysis (PSHA) using I Np and, alternatively, Sa avg , intensity measures. We found that while the conditional mean spectra might be affected in the region of short vibration periods, the computations of the PSHA with improved intensity measures are not affected significantly.