The dependence of peak horizontal acceleration on magnitude, distance, and site effects for small-magnitude earthquakes in California and Eastern North America

Abstract

Abstract One-hundred and ninety free-field accelerograms recorded on deep soil (> 10 m deep) were used to study the near-source scaling characteristics of peak horizontal acceleration for 91 earthquakes (2.5 ≦ M L ≦ 5.0) located primarily in California. Of the six attenuation relationships developed in this study, the one considered most reliable is given by the expression: ln P H A = - 2 . 5 0 1 + 0 . 6 2 3 M L − 1.0 ln [ R + 7 . 2 8 ] + ∈ where PHA is the mean of the two horizontal components of peak acceleration in g, M L is local magnitude, R is epicentral distance in kilometers, and ∈ is a random error term with mean of zero and standard deviation of 0.506. An analysis of residuals based on an additional 171 near-source accelerograms from 75 earthquakes indicated that accelerograms recorded in building basements sited on deep soil have 30 per cent lower accelerations, and that free-field accelerograms recorded on shallow soil (≦ 10 m deep) have 82 per cent higher accelerations than free-field accelerograms recorded on deep soil. Accelerations recorded on rock were found to be similar in amplitude to those recorded on deep soil. A separate analysis of site effects for the Imperial Valley and Oroville regions of California indicated that peak accelerations from individual sites can depart significantly from predictions based on broad site classifications. An analysis of residuals based on 27 selected strong-motion recordings from 19 earthquakes in Eastern North America indicated that near-source accelerations associated with frequencies less than about 25 Hz are consistent with predictions based on attenuation relationships derived for California, once site effects and differences in regional magnitude scales are taken into account.