9422238 Graves Objective The objective of the proposed work is to understand and quantify the effects that the San Fernando and Los Angeles sedimentary basins had on the long-period (1-10 sec) ground motion response during the Northridge earthquake, and to formulate recommendations for the inclusion of basin effects investigations in future seismic hazard analyses. Significance The Northridge earthquake and its aftershocks have produced the most comprehensive set of strong motion data ever recorded for a single earthquake sequence. These data allow for a wide range of research involving the analysis and modeling of earthquake source, path and site effects within the Los Angeles region. One important area of concern is to understand the effect that the deep sedimentary basins which underlie the San Fernando and Los Angeles regions had on the amplification and focusing of long-period (1-10 sec) seismic energy during this earthquake. Previous work has shown that wave propagation effects related to the geologic structure of the Los Angeles and San Fernando basins have a significant impact on the observed ground motions, particularly for periods longer than about 1 sec. and in many cases, these basin-induced effects dominate the ground motions in the period range 1-10 sec. During the Northridge earthquake, localized areas of intense ground shaking and associated damage were observed that cannot be explained by conventional attenuation relations (e.g., Santa Monica, I-10 freeway collapse). It is important to understand the cause of these phenomena and address their implications for engineering design, particularly for large structures (e.g., tall buildings, freeway bridges, base-isolated structures), which are most sensitive to long-period energy. Method and Anticipated Results This investigation involves the analysis of ground motion recordings of the Northridge earthquake and its aftershocks in order to systematically map out the observed basin effects within the San Fernando and Los Angeles regions. A key element of this analysis is the numerical modeling of these data using a 3D elastic finite-difference technique. The finite-difference approach is attractive because it allows flexibility in incorporating strongly heterogeneous velocity models, as well as using a finite, variable-slip source model. The goal of this project is to: (1) produce long-period (1-10 sec) time histories calculated on a fine spatial mesh throughout the San Fernando and Los Angeles regions; (2) display the results in map format, which will illustrate the spatial distribution of generated ground motions (e.g., wave field timeslices, distribution of peak velocity and duration of shaking) and the correlation of these motions with the subsurface geology; and (3) compare these results with those obtained using existing empirical attenuation relations, as well as the results calculated for 1D velocity models, in order to quantify the effects of the basin response for use in future seismic hazard analyses. ***
