Analyzing the Seismic Integrity of the San Francisco-Oakland Bay Bridge: An MASW Case Study

On October 17, 1989, the magnitude 6.9 Loma Prieta earthquake caused a 15-meter section of the upper deck of the San Francisco-Oakland Bay Bridge to collapse onto the lower deck at Pier E9. This structural failure highlighted the extreme vulnerability of infrastructure built upon the soft, unconsolidated sediments characteristic of the San Francisco Bay. In the decades following the disaster, the California Department of Transportation (Caltrans) initiated a detailed seismic safety program that shifted from traditional borehole-based geotechnical assessments to sophisticated geophysical imaging, primarily focusing on surface wave propagation.

The geotechnical investigation of the Bay Bridge relied heavily on Multichannel Analysis of Surface Waves (MASW) and Spectral Analysis of Surface Waves (SASW). These non-destructive testing (NDT) techniques were deployed to map the shear-wave velocity (Vs) of the subsurface, a critical parameter for predicting soil-structure interaction during seismic events. By analyzing the dispersion characteristics of Rayleigh waves, engineers were able to characterize the varying thicknesses of the Young Bay Mud and the underlying dense sands and clays, providing a two-dimensional profile of the soil's stiffness that traditional point-based drilling could not achieve.

At a glance

• Location:San Francisco-Oakland Bay Bridge, connecting San Francisco and Alameda Counties.
• Primary Hazard:Soil liquefaction and seismic amplification within Holocene-age Young Bay Mud.
• Primary Methodology:Multichannel Analysis of Surface Waves (MASW) and Spectral Analysis of Surface Waves (SASW).
• Key Parameter:Shear-wave velocity (Vs), utilized to determine the small-strain shear modulus of the soil.
• Outcome:Redesign of the Eastern Span as a self-anchored suspension (SAS) bridge with foundations anchored into the Franciscan Complex bedrock or deep Pleistocene sediments.

Background

The original San Francisco-Oakland Bay Bridge, completed in 1936, was designed before modern understanding of plate tectonics and seismic wave amplification. The eastern span, in particular, was a truss bridge supported by caissons and timber piles driven into the soft mud of the bay. Geologically, the area is defined by a complex stratigraphy: the top layer consists of highly plastic, soft silty clay known as Young Bay Mud, which can range from 10 to 40 meters in thickness. Below this lie the Merritt Sand and the Posey Formation, followed by the stiffer Old Bay Mud and eventually the Franciscan Complex bedrock.

During the 1989 earthquake, the soft mud acted as a filter that amplified low-frequency seismic waves while simultaneously losing shear strength, a process known as soil liquefaction. The resulting differential movement between the bridge piers led to the failure of the bolts securing the deck trusses. This event necessitated a total re-evaluation of the bridge’s seismic integrity, moving beyond simple boring logs to an integrated geophysical approach that could account for the lateral variability of the seafloor sediments.

The Transition to Surface Wave Analysis

Historically, geotechnical engineers relied on the Standard Penetration Test (SPT) and the Cone Penetration Test (CPT) to assess soil density and strength. While these methods provide high-resolution data at a specific point, they are invasive, expensive, and fail to capture the continuous horizontal variations in soil stiffness. In the context of a massive infrastructure project like the Bay Bridge retrofit, the need for a non-destructive, cost-effective method to profile large areas led to the adoption of SASW and later MASW.

The Mechanics of Rayleigh Wave Dispersion

Surface wave analysis is based on the dispersive nature of Rayleigh waves in a layered or heterogeneous medium. In such environments, different wavelengths (and thus different frequencies) of the Rayleigh wave travel at different velocities. High-frequency waves, which have shorter wavelengths, propagate through the shallower layers, while low-frequency waves with longer wavelengths penetrate deeper into the earth.

By measuring the phase velocity of these waves across a range of frequencies, researchers can construct a dispersion curve. This curve is the raw data used in the inversion process to calculate the vertical shear-wave velocity profile of the site. In the Bay Bridge surveys, geophones were placed in linear arrays along the shoreline and on the bridge footings to capture both ambient microtremors and controlled-source signals, such as those generated by a heavy impactor or a seismic vibrator.

Comparison of Historical Boring Logs and Vs Profiles

One of the primary objectives of the Caltrans seismic surveys was to reconcile historical boring logs from the 1930s with modern shear-wave velocity data. The following table illustrates the contrast between these two data acquisition methods in the context of the Bay Bridge eastern span.

The 2D Vs profiles derived from MASW revealed that the transition between the Young Bay Mud and the stiffer underlying strata was not a flat plane, as previously assumed in many legacy engineering models. Instead, it was an undulating surface with paleo-channels that could focus seismic energy, creating localized zones of intense ground motion.

MASW Application in Foundation Evaluation

The application of MASW for the San Francisco-Oakland Bay Bridge was not limited to soil mapping; it extended to the non-destructive evaluation of the bridge foundations themselves. By analyzing how surface waves interact with engineered interfaces, such as the contact between a concrete pier and the surrounding soil, engineers could detect voids or zones of poor consolidation.

‘The precision of inversion algorithms allows for the inference of material properties like porosity and elastic moduli from observed wave velocities, providing a diagnostic tool for infrastructure health that was previously unavailable.’

In the development of the new Eastern Span, which features a single 160-meter-tall pylon, MASW was used to verify the integrity of the large-diameter cast-in-steel-shell (CISS) piles. These piles were driven deep into the Posey Formation. The spectral analysis of waves reflected from the pile tips and the shaft-soil interface allowed for the empirical verification of the piles' load-bearing capacity without the need for destructive testing.

Seismic Modeling and Inversion Challenges

While MASW provided a wealth of data, the process of inversion—converting frequency-velocity data into a depth-velocity model—presents significant mathematical challenges. The heterogeneous nature of the Bay’s sediments means that the inversion is often non-unique, meaning several different soil models could potentially produce the same dispersion curve. To mitigate this, Caltrans engineers utilized a multi-objective optimization approach, constraining the geophysical models with the existing physical data from borehole samples.

The study of acoustic wave propagation within these heterogeneous solid-state media required the precise calibration of accelerometers to ensure that subtle microtremors were not lost in the urban noise of the San Francisco metropolitan area. Through the meticulous interpretation of these wavefields, the project team was able to map the depth to bedrock with a high degree of confidence, ensuring that the new bridge foundations were anchored in material capable of resisting the lateral forces of a maximum credible earthquake (MCE) on the nearby Hayward or San Andreas faults.

Conclusion

The case study of the San Francisco-Oakland Bay Bridge represents a landmark in the practical application of surface wave characteristics. By moving beyond the limitations of historical geotechnical methods and embracing the empirical study of Rayleigh and Love waves, engineers were able to construct a detailed picture of the subsurface risks. This discipline, which sits at the intersection of seismology and civil engineering, remains essential for the non-destructive testing of critical infrastructure and the detection of subsurface anomalies that threaten the stability of the built environment.