Interpretation of subway-induced soil stress-wave signals for detecting coupled cavity-boulder defects

Abstract

Underground cavities and isolated stone clusters constitute critical geotechnical hazards that threaten the stability of surface infrastructure and subsurface engineering structures. Conventional geophysical prospecting techniques for detecting subsurface anomalies are often constrained by high operational costs, limited spatial resolution, or low detection efficiency. To overcome these limitations, this study investigates the feasibility of utilizing subway-induced soil stress waves as an alternative geophysical excitation source for detecting concealed geological defects, particularly within complex soil media containing alternating distributions of cavities and isolated stones. A laboratory-scale physical modeling experiment was designed to replicate the propagation behavior of soil-borne stress waves under varying combinations of cavity-stone configurations. A scanning laser Doppler vibrometer (LDV) was employed to record full-field wavefield evolution, enabling high-resolution visualization of wave propagation patterns and energy attenuation characteristics. Experimental observations reveal that changes in waveform amplitude, energy distribution, and spatial imaging signatures can be leveraged to preliminarily localize the horizontal extent and approximate dimensions of both cavities and stone clusters. By applying the Akaike information criterion to extract the first-arrival times of stress wave signals, this study quantitatively evaluates the influence of subsurface anomalies on wave travel-time perturbations, thereby enabling effective differentiation between cavities and isolated stones. Furthermore, frequency-domain analyses demonstrate that the presence of isolated stones substantially enhances the high-frequency content of the recorded signals in contrast to cavity-affected conditions. LDV-based visualization of stress wave propagation provides an intuitive representation of scattering, reflection, and mode conversion behaviors induced by buried defects. Overall, the findings underscore the promising potential of subway-induced soil stress waves as a non-invasive, cost-effective, and long-term monitoring approach for urban subsurface defect detection. This work provides a theoretical foundation and experimental evidence supporting the application of stress-wave-based geophysical techniques in urban geological hazard assessment and tunnel engineering.