Determination of site amplification in the Los Angeles urban area from inversion of strong-motion records

Abstract

AbstractThe amplification of strong ground motion at sites in the greater Los Angeles, California, region is determined using the generalized-inverse method of Andrews (1986). Site-amplification estimates are determined at 281 strong-motion sites that provided horizontal-component accelerograms from the 1971 San Fernando, 1987 Whittier Narrows, 1991 Sierra Madre, or 1994 Northridge mainshocks. The estimates are determined relative to the spectral level recorded at a single reference site. In a second inversion, a source-site interaction term is added to Andrews's (1986) model to quantify the effect selected mainshock records have on site-amplification estimates. The source-site interaction term is applied to the San Fernando Valley sites' records of the Northridge earthquake and to three Los Angeles basin sites' records of the Whittier Narrows mainshock.Site-amplification spectra are averaged within two frequency bands: the intermediate-frequency band (IFB) from 0.5 to 1.5 Hz and the high-frequency band (HFB) from 2 to 6 Hz. Results are displayed on maps of surficial geology. Average spectral levels are correlated with average shear-wave velocity in the uppermost 30 m, a geotechnical parameter used by the National Earthquake Hazards Reduction Program (NEHRP) to characterize site amplification. Statistically significant correlation is found in both frequency bands. Average spectral amplification levels for NEHRP class B, C, and D sites in the LA urban area are determined. These averages display an expected increase in spectral amplification with category, with a more pronounced variation in the IFB than in the HFB. Considerable overlap in the one-standard-deviation range of the C and D site levels is found in both IFB and HFB, suggesting that site-specific spectral amplification is influenced by more than just near-surface shear-wave velocity. Average site-amplification levels are compared with those obtained from Northridge aftershock records at 28 collocated sites and are found to be in reasonably good agreement in both frequency bands. Some interesting outliers, that is, sites for which the mainshock- and aftershock-determined amplifications differ significantly, are identified as sites that experienced ground failure during the Northridge mainshock.