Earthquake-induced diffuse landslide can cause considerable harm. This study used discrete element modeling (DEM) to investigate the diffuse failure of soil slopes triggered by earthquakes. Initially, the DEM was validated through demonstrating satisfactory agreement with the experimental results of a shaking table test. Subsequently, the time and most realistic location of diffuse failure occurrence was quantified using discrete second- order work. Diffuse failure was associated with abrupt fluctuation of the second-order work and a local burst of kinetic energy. Within the context of second-order analysis, the frictional resistance of the slope was analyzed. Both the in-phase and the out-of-phase evolutions of frictional resistance and seismic loading were captured. The distinctive in-phase response was identified as an important feature of diffuse failure. Moreover, the predominant influence of particle rolling behavior on frictional resistance development during diffuse failure progression was highlighted. The magnitude and anisotropy of the contact force are profoundly influenced by the input vibration amplitude, resulting in variations in frictional resistance, which suggests that stress-induced anisotropy plays a crucial role in diffuse failure. The results of this study shed light on the grain-scale failure mechanisms of earthquake-induced landslides, and serve as the basis for elucidating the failure behavior of seismic-induced landslides.

To address the limited comprehension of the dynamic response characteristics of soil-rock mixture (SRM) slopes, three sets of large-scale shaking table model tests of SRM slope with different rock contents were designed and conducted based on the similarity principle. The differences in dynamic response of SRM slope with different rock contents were systematically compared and analyzed. The research results indicate that the acceleration response of SRM slopes under earthquake action conforms to the free surface effect, that is, the acceleration amplification effect of the slope is significantly stronger near the top of the slope than within the slope. However, the dynamic response of SRM slopes with different rock contents under sine wave excitation of different frequencies is significantly different, this is due to the differences in the dynamic properties of slope structures with different rock contents. Under seismic action, the dynamic earth pressure of SRM slopes with different rock contents increases continuously from the shallow surface to the interior of the slope, but due to the different degrees of deformation and damage of the slope body, the overall dynamic soil pressure response of slopes with different rock contents is different. Moreover, during the entire seismic wave grading loading process, the sudden changes in dynamic soil pressure at different parts of the slope can serve as the basis for dynamic failure of the slope. As the rock content rises, the overall deformation of the slope under seismic action decreases gradually. For instance, a slope with 20% rock content exhibits continuous sliding from shallow to deep layers, while slopes with 40% and 60% rock content have relatively small deformation. A slope with 40% rock content only experience sliding of surface rock and soil, and a slope with 60% rock content only experience peeling of shallow surface soil. This indicates that higher rock content reinforces the stability of the SRM slopes.