Sign In
Skip Navigation LinksHome > CGS > California Strong Motion Instrumentation Program > DOCS > Reports > Datautil > California Geological Survey - CSMIP 93-02

CSMIP 93-02

"Investigation of the Response of Puddingstone Dam in the Whittier Narrows Earthquake of October 1, 1987"

by J. Bray, R. Seed and R. Boulanger

December 1993, 60 pp

Click on the link below for the full text:

PDF (Adobe PDF (SIZE 764 KB)) - Requires Adobe Acrobat Reader or other PDF viewer.


The Whittier Narrows Earthquake of October 1, 1987 (ML ≈ 5.9) shook Puddingstone Dam, a primarily cohesive, homogeneous section, compacted earth dam which had previously been instrumented as part of the California Strong Motion Instrumentation Program (CSMIP). The resulting maximum (transverse, horizontal) crest acceleration was 0.19 g, and maximum accelerations recorded at abutment stations were on the order of 0.04 to 0.08 g. The resulting recorded accelerograms provided a valuable opportunity to investigate and evaluate the accuracy and reliability of conventional geotechnical procedures for evaluation of dynamic response characteristics of earth and rockfill dams. Presented in this report are the results of dynamic analysis studies of the response of Puddingstone Dam to the 1987 Whittier Narrows Earthquake performed using both (a) simple one-dimensional columnar analyses, and (b) two-dimensional (plane strain) dynamic finite element analyses. Nonlinear, strain-dependent dynamic shear moduli and damping characteristics were modelled using the "equivalent linear" method. Nonlinear modulus degradation and damping relationships for the compacted sandy silty clay which comprises a majority of the embankment were modelled based on the relationships proposed by Sun et al. (1988) for clays of low plasticity. The results of the two-dimensional finite element analyses were found to be in good agreement with the observed (recorded) field response, providing good support for these modelling and analysis techniques. The simpler, one-dimensional columnar analyses were found to significantly underestimate the crest response, as a result of their inability to model geometric effects or topographic amplification, but were also found to produce reasonably good agreement between calculated and recorded response at the center of the downstream face.