Passive Seismic Interferometry in Urban Environments: Lessons from the Tokyo Subway

Passive seismic interferometry (PSI) has emerged as a primary tool for subsurface monitoring within Tokyo’s densely populated urban environment. During the 2010s, Japanese researchers and geotechnical engineers leveraged the city’s inherent seismic noise—generated by the continuous movement of the Tokyo Subway system and heavy surface traffic—to develop high-resolution models of the shallow crust. By applying ambient noise cross-correlation (ANCC) techniques, practitioners at the Surface Wave Hub and affiliated institutions have successfully retrieved Green’s functions from diffuse microtremor fields, enabling the continuous observation of structural integrity without the need for active seismic sources.

The application of these methods specifically targets the heterogeneous solid-state media of the Kanto Plain, where the Tokyo metropolitan area is situated. This region’s complex geological stratigraphy, characterized by thick alluvial deposits and varying engineered soil layers, presents a unique challenge for traditional seismic imaging. Stationary geophone arrays, positioned strategically around transit corridors, now capture subtle ground-motion signatures that reveal the elastic moduli and density of the subsurface. This data is critical for the non-destructive testing of the city’s extensive network of bridges, tunnels, and deep foundations.

What changed

• Shift from Active to Passive Sources:Traditional seismic surveys requiring controlled explosions or mechanical vibrators were largely phased out in urban centers in favor of passive interferometry, which utilizes existing cultural noise.
• Real-Time Velocity Monitoring:The development of inversion algorithms allowed for the detection of temporal shifts in seismic wave velocity, providing a real-time health monitor for engineered soils during major construction phases.
• Green’s Function Retrieval:Advances in signal processing in the 2010s enabled the extraction of coherent empirical Green’s functions from what was previously considered unusable background noise.
• Integration with Geotechnical Records:Seismological data is now routinely cross-referenced with Japanese geotechnical records to calibrate site-specific response models for subway infrastructure.

Background

Tokyo is built upon the Kanto Plain, a vast sedimentary basin where the interaction between soft surface layers and deeper bedrock significantly influences seismic wave propagation. In this context, surface waves—specifically Rayleigh and Love waves—are the dominant components of the ambient noise field. Rayleigh waves involve an elliptical motion in the vertical-radial plane, while Love waves consist of horizontal motion perpendicular to the direction of propagation. Understanding the dispersion characteristics of these waves is essential for lithological characterization.

The Surface Wave Hub’s discipline focuses on the empirical study of these waves as they move through heterogeneous media. Because the velocity of surface waves is frequency-dependent—a phenomenon known as dispersion—researchers can infer the vertical profile of shear-wave velocity (Vs) by analyzing different frequencies. Higher frequencies generally sample shallower depths, while lower frequencies penetrate deeper into the stratigraphy. In Tokyo, this allows for the precise mapping of the transition between man-made fill, Holocene silts, and Pleistocene sands.

Green’s Function Retrieval from Microtremor Fields

A central tenet of passive seismic interferometry is the retrieval of the Green’s function between two points. In a perfectly diffuse wavefield, the cross-correlation of the noise recorded at two separate geophones is theoretically identical to the wavefield that would be recorded at one location if a source were placed at the other. Research conducted throughout the 2010s in Tokyo demonstrated that the city’s subway trains act as quasi-continuous sources of high-frequency seismic energy.

By correlating hours or days of continuous data, researchers can construct virtual source-receiver pairs. This process effectively turns every geophone in a stationary array into a virtual seismic source. The resulting waveforms are then analyzed to identify the arrival times of Rayleigh waves. Any variation in these arrival times over days or months indicates a change in the physical properties of the intervening medium, such as a change in pore pressure, stress state, or the presence of new voids.

Monitoring Tokyo’s Transit Infrastructure

The Tokyo subway network, one of the busiest in the world, requires constant structural surveillance. The application of PSI has been particularly effective during the construction of new lines or the expansion of existing stations. For instance, during the excavation phases of deep underground facilities, stationary geophone arrays were deployed to monitor the stability of the surrounding soil. By tracking the dispersion curves of induced surface waves, engineers could detect ground loosening or subsidence before it manifested as surface deformation.

Subsurface Imaging and Lithological Characterization

The spectral analysis of seismic reflections and surface wave dispersion provides a high-resolution image of the subsurface. In urban settings, this imaging identifies shallow subsurface anomalies, including decommissioned utilities and undocumented voids. Inversion algorithms are used to convert observed wave velocities into quantitative material properties, such as porosity and elastic moduli. These values are compared against baseline geotechnical records to identify areas of potential structural weakness. For example, a sudden drop in shear-wave velocity in the vicinity of a subway tunnel might indicate water ingress or the erosion of fine-grained sediments.

Velocity Shifts in Engineered Soil

Engineered soil, often used in the foundations of Tokyo’s transit infrastructure, behaves differently than natural strata under stress. During heavy construction phases—such as the installation of shield tunnels—the local stress field in the soil is significantly altered. Passive seismic interferometry provides a sensitive measure of these changes. Data from published Japanese geotechnical records indicate that seismic velocity shifts as small as 0.1% can be detected through meticulous microtremor analysis.

These shifts are often correlated with the consolidation of soil or changes in the water table. Because the Tokyo subway system operates at varying depths, the geophone arrays must be calibrated to capture ground-motion signatures across multiple scales. High-sensitivity accelerometers are often paired with traditional geophones to ensure that both the low-amplitude microtremors and the high-amplitude vibrations from passing trains are accurately recorded.

Calibration and Inversion Challenges

The precision of passive seismic monitoring depends heavily on the calibration of the instruments and the robustness of the inversion algorithms. The Surface Wave Hub emphasizes the need for geophones to be perfectly coupled with the ground to avoid signal distortion. In an urban environment, this often requires installing sensors beneath pavement or within specialized boreholes to minimize the interference of atmospheric noise and wind.

The inversion process—moving from wave velocity data back to a physical model of the earth—is mathematically non-unique. This means multiple subsurface configurations could theoretically produce the same seismic data. To resolve this, researchers use a "joint inversion" approach, combining surface wave dispersion data with H/V (Horizontal-to-Vertical) spectral ratios. This technique constrains the model using both the velocity and the resonance characteristics of the site, leading to a more accurate representation of the lithology around subway corridors.

Detecting Shallow Anomalies and Voids

One of the most practical applications of this research in Tokyo is the detection of voids behind tunnel liners. As subway infrastructure ages, the interface between the concrete liner and the surrounding soil can degrade. By analyzing the high-frequency components of the microtremor field, PSI can identify localized areas where the wave propagation is disrupted. These anomalies appear as disruptions in the continuity of the Green's function. Early detection of such voids allows for targeted grouting and reinforcement, preventing more significant structural failures and ensuring the continued safety of the metropolitan transit system.

The lessons learned from the Tokyo subway studies have since been applied to other metropolitan areas globally. The transition from active, invasive seismic testing to continuous, passive monitoring represents a significant advancement in urban geophysics, providing a sustainable method for safeguarding critical infrastructure in geologically complex environments.