Enhancing karst cave detection in Hangzhou using stationary-phase traffic noise imaging with seismic interferometry

In urban environments, pervasive noise, especially traffic-induced noise, presents significant challenges for subsurface geological surveys. However, recent advancements in seismic interferometry have demonstrated that such noise can be repurposed as a valuable resource for near-surface imaging. We have developed an efficient approach that utilizes traffic-induced noise data from stationary-phase zones to retrieve high-quality surface waves for enhanced karst cave detection. The noise data were collected using six densely spaced linear arrays, with approximately 5-m receiver interval and 10-m inter-array distance, deployed in an abandoned parking lot in the town of Yunqi, Hangzhou city, Zhejiang Province, China. Spectral analysis reveals that the traffic-induced noise exhibits dominant energy around 10 Hz, within a frequency range of 5 to 20 Hz. A beamforming algorithm was applied to determine the azimuthal distribution of noise sources, which facilitated the analysis of wavefield propagation direction and guided the preselection of a high-quality one-hour window of continuous noise recordings. This study further highlights the advantage of cross-coherence over cross-correlation in suppressing random perturbations caused by interfering noise. By selecting noise segments from this one-hour window whose sources fall within the stationary-phase zone and applying cross-coherence to them, high-quality Rayleigh waves are successfully retrieved, enabling the clear extraction of dispersion curves from virtual shot gathers. These dispersion curves are subsequently inverted using hunger games search optimization algorithm, yielding 2D shear (S)-wave velocity maps which show good agreement with borehole logging data. These results with sensitivity analyses provide a detailed interpretation of the karst cave distribution and bedrock structure, demonstrating the effectiveness of traffic-induced noise for near-surface geological imaging. This case study underscores the potential of passive surface-wave methods in complex urban environments and illustrates how everyday traffic-induced noise can be transformed into a powerful tool for geological exploration.