A single surface slope effects on seismic response based on shaking table test and numerical simulation

Abstract

This paper presents the effects of a single surface slope on seismic ground responses and exhibits the peak horizontal acceleration (PGAh) distribution regulations in the slope cross-section under vertically impinging SV wave excitation. The seismic response of a homogeneous slope was investigated through a series of shaking table tests, and physical modelling was used to verify the numerical simulation. Subsequently, many numerical simulations were conducted considering slope geometry and input ground motion using Fast Lagrangian Analysis of Continua in Three Dimensions (FLAC3D). Two patterns of PGAh distribution are summarised for single surface slopes: one type is the low slope dynamic response, which shows that PGAh increases along the slope surface with the increase in the slope elevation, varies little with the increase in the distance from the slope surface to the inside in the horizontal direction. The other type is the high slope dynamic response, which shows that PGAh changes periodically and takes on a certain pattern along the slope surface with the increase in the slope elevation, and decreases gradually with the increase in the distance from the slope surface to the inside in the horizontal direction. Meanwhile, the critical height of the slope for different distribution patterns of PGAh varies with the slope material and the harder the rock material, the higher the critical slope height for the distribution patterns of PGAh. The critical height of Hcr which is approximately 0.17–0.20 times the incident wavelength (λ), divides the two types of dynamic response of the slope based on the dimensional analysis method and numerical simulations. That is, the chart of its dynamic response will have a high dynamic response when the slope height is greater than Hcr, and its dynamic response will present as a low slope dynamic response when the slope height is less than Hcr. The amplification coefficient at the crest increases with an increase in the dimensionless slope height (H/λ) and reaches the first peak value when H/λ reaches 0.17–0.20. The de-amplification at the crest occurred when H/λ was >0.4. The novel findings can support the input ground motion parameters for earthquake slope engineering design.