Earthquake-induced sand liquefaction poses a significant threat to societal infrastructure, economic stability, and the environment. The liquefaction phenomenon is governed by a multitude of factors. While existing research on the influence of gravelly soils on sand liquefaction has primarily focused on parameters such as gravel content and the sand-to-gravel ratio, employing dynamic triaxial testing methods, significant limitations persist. The large particle size characteristic of gravelly soils is not adequately represented in small-scale dynamic triaxial specimens, which compromises the accurate characterization of their dynamic properties and overlooks the structural influence of gravelly soil layers within a stratigraphic profile.To investigate the impact of gravel layers on site liquefaction potential, this study draws upon field observations of varying liquefaction manifestations in New Zealand following seismic events. It employs an integrated methodology of physical modeling and numerical simulation to analyze the mechanisms through which factors such as the thickness and spatial distribution of gravelly layers influence sand liquefaction.

Seven distinct experimental models were designed for this investigation: one comprising pure sand, three with varying thicknesses of gravel interlayers, and three with varying thicknesses of overlying gravel layers. A comprehensive analysis of the effects of gravel interlayer thickness and overlying gravel layer thickness on the liquefaction characteristics of saturated sand was conducted using shaking table tests and numerical simulations performed with the finite difference code FLAC3D.

The results showed that the existence of gravel interlayer will increase the overall permeability of the liquefied site, increase the equivalent permeability coefficient, and reduce the liquefaction potential of the liquefied soil layer above the gravel interlayer. The size of the excess pore pressure ratio at the deep position is negatively correlated with the thickness of the gravel layer. With the increase of the thickness of the gravel layer, the excess pore pressure ratio of the underlying liquefied soil layer is smaller. This is due to the existence of the gravel interlayer. The initial overlying effective stress increases and the excess pore pressure ratio decreases. The greater the thickness of the gravel interlayer, the lower the excess pore pressure of the underlying liquefied soil layer. Because the permeability of gravel interlayer is much higher than that of liquefied soil layer, the seepage rate of pore water in the underlying liquefied soil layer is increased. The greater the thickness of gravel interlayer, the closer the monitoring point is to the interlayer, the more obvious the effect is.

In the time history curve of the excess pore pressure ratio of the overlying gravel layer with different thicknesses, it can be seen that the peak value of the excess pore pressure ratio decreases with the increase of the thickness of the gravel layer above the liquefied soil layer. Because the weight of gravel is higher than that of liquefied soil layer, the initial overburden effective stress of liquefied soil layer is increased, and the excess pore pressure ratio is effectively reduced. The greater the thickness of gravel layer, the greater the effective stress provided, and the lower the excess pore pressure ratio. At the same time, the thickness of the gravel layer increases, the equivalent permeability coefficient increases, and the peak value of the excess pore pressure ratio of the liquefied soil layer is lower.

Using the equivalent permeability coefficient of each model calculated by empirical value, it can be seen that the existence of gravel layer can improve the equivalent permeability of the whole liquefaction site, thus effectively reducing the liquefaction potential of the liquefaction site.

In general, this paper systematically analyzes the influence of gravel interlayer thickness and overlying gravel layer thickness on the liquefaction characteristics of saturated sand by combining physical test and numerical simulation, and studies the relationship between the liquefaction state and the equivalent permeability coefficient of the gravel layer soil model.