3D ambient seismic noise tomography and monitoring of unstable rock slope

Unstable rock slopes pose significant risks to infrastructure and human safety. Understanding their internal structure and monitoring dynamic changes are crucial for assessing slope stability and developing early warning systems. Conventional methods primarily focus on geological and geomorphological features, limiting their ability to capture subsurface structural changes. The passive seismic method offers a cost-effective and non-invasive approach for investigating subsurface structures and detecting underground changes. In this study, we utilize both short-term array data and long-term single-station ambient noise data from the Brienz/Brinzauls rock slope, Switzerland. For short-term data, we evaluate wavefield diffuseness to select segments with diffuse wavefields for seismic interferometry processing. We then apply the Generalized Phase Shift S-Transform (GPST) to extract surface wave dispersion curves and reconstruct a 3D S-wave velocity model. This model delineates the slip surface and reveals the intrinsic relationship between the June 2023 Insel collapse and the internal S-wave velocity structure. For long-term monitoring, we track slope changes using velocity ratio variation (dv/v) derived from the Autocorrelation Function (ACF), along with rockfall events, daily average peak ground acceleration (PGA), and daily average peak ground displacement (PGD). Anomalies in these parameters indicate that internal slope failure occurred on October 28, 2018, preceding the significant acceleration of the rock slope. Case studies have shown that ambient noise seismic data can effectively image the internal structure of slopes, and special attention should be paid to low-velocity interlayers with large inclination angles. These applications provide valuable supplementary tools for monitoring and evaluating unstable rock slopes.