Hanging wall effects on cross-fault slope failures: Shaking table experiment insights

Abstract

Earthquake-triggered landslides are prone to occur on the hanging wall of faults, yet the failure mechanism of co-seismic landslides affected by reverse faulting remains poorly understood. In this study, we explore the dynamic response and failure mechanism of cross-fault slopes by conducting reverse fault physical modeling on large-scale shaking table model testing. A novel movable model box with a sliding bottom plate and an air cushion is used to simulate the reverse faulting of the horizontal layered slope models with fault dip angles of 30° and 50°. We analyze the effect of different reverse fault angles on the dynamic response and failure patterns, using various seismic waves, Hilbert-Huang transform (HHT), and particle image velocimetry (PIV). The results indicate that the dip angle of the reverse fault dislocation is crucial in influencing the dynamic response of the cross-fault slope. The 30° model is more sensitive to frequency changes and is prone to resonance at 24 Hz, while the 50° model produces stronger dynamic response to high amplitude seismic waves. Reverse fault dislocation amplifies the dynamic response and Hilbert energy at the hanging wall, with a larger amplification coefficient observed at a 30° dip angle. Slope models with different dip angles of the reverse fault produce distinct Hilbert energy distributions, resulting in two typical failure patterns. A “tension-shear” failure pattern, characterized by a shallow sliding surface, occurs in the 50° dip angle model, while a “tension-ejection” failure pattern with a vertical tensile sliding surface occurs in the 30° dip angle model. Our results provide important insights into the behavior of cross-fault slopes during seismic events, and provide guidance for better understanding and managing hazards associated with cross-fault slopes.