← Back to Sessions

Relating Peak and Cumulative Ground Motions for Earthquakes in the San Francisco Bay Area

Description: 

We examine ~5000 ground motion records from 39 Mw3.5–6 earthquakes in the San Francisco (SF) Bay Area with the goal of understanding site conditions leading to amplification in either the form of sharp velocity pulses or longer duration resonant motion. Most estimates of seismic hazard are influenced by predicted peak ground velocity or acceleration (PGV or PGA). These quantities are, in general, affected by higher frequency motions (>1Hz), and thus may not be good indicators of the kind of shaking that threatens long period structures, such as tall buildings or bridges. Also, the peak motions do not inform us about the duration or total dynamic load to a structure, which may be weakened after several successive pulses of significant amplitude. This is important for structures around the SF Bay shoreline, where sites in very low velocity bay mud experience long duration ringing at ~1 Hz (Hirakawa and Aagaard, 2022). This type of resonant motion is likely the cause of the collapse of the Cypress Viaduct during the 1989 Loma Prieta earthquake (Rogers, 1991).

We compute PGV and cumulative absolute displacement (CAD) for the SF Bay Area ground motion records. CAD, the time integral of the absolute horizontal velocity, inherently increases with PGV since it involves integration of the largest velocity pulse but also incorporates additional motions. Sites that experience long duration ringing will have larger CAD than average. We consider each ground motion record in normalized PGV-CAD space and rotate the dataset into principal components. We refer to the first principal component as the “primary intensity” which accounts for the expected increase in CAD with PGV. We refer to the second principal component as the “excess shaking”, which may be thought of as extra (or deficit) cumulative motion apart from what would be typical for a given PGV. We investigate site-specific trends in these two quantities and consider possible relations with subsurface geology. The results of this study will expand our definition of site response beyond changes to shaking amplitude, with direct application to the SF Bay Area.

Session: How Well Can We Predict Broadband Site-Specific Ground Motion and Its Spatial Variability So Far? [Poster Session]

Type: Poster

Date: 5/1/2024

Presentation Time: 08:00 AM (local time)

Presenting Author: Evan

Student Presenter: No

Invited Presentation: 

Authors