The 1933 Long Beach, California, Earthquake: Ground Motions and Rupture Scenario

Date: 4/26/2019

Time: 06:00 PM

Room: Grand Ballroom

We present a first-ever synoptic characterization of the distribution of ground motions from the 11 March 1933 Mw6.5 Long Beach, California, earthquake, using available macroseismic and limited instrumental data. The detailed shaking intensity pattern supports the conventional association of the earthquake with the Newport-Inglewood fault; it further reveals a concentration of especially severe damage centered in the town of Compton, where several credible accounts suggest vertical ground motions exceeding 1g, and intensities as high as Modified Mercalli Intensity 9. We use the broadband simulation approach of Graves and Pitarka (2010) to develop a rupture scenario for the earthquake, informed by the detailed damage distribution. Overall, the predicted shaking matches the observed intensities well using a 25-km long by 12-km wide rupture model with a hypocenter at 11 km depth near the southern end of the rupture, and a high-slip-patch near the northern end of the rupture. The modeling predicts high peak ground velocities in some near-field regions where liquefaction was documented but observed shaking intensities were relatively low, suggesting that high-frequency (≈2-8 Hz) energy was de-amplified by nonlinear site response. The concentration of damage near Compton can be explained by a combination of local site amplification, source-controlled directivity, and three-dimensional basin effects whereby energy was channeled towards the central Los Angeles Basin. In contrast with other inferred high-strain regions, it appears that high frequency energy was not de-amplified in this area. The overall success of the modeling in matching observed intensities provides a measure of confidence that the same modeling approach can be used to predict shaking from future earthquakes, with the caveat that, where the response is strongly nonlinear, ground motions may be lower than predicted using a simple Vs30-based model for non-linear site response, as used in our modeling.

Presenting Author: Susan E. Hough

Authors