One of the factors that has a significant effect on the tsunami inundation process in coastal areas with soft soil is ground subsidence caused by liquefaction, which must be taken into account when analyzing tsunami inundation.

Focusing on mixed sandy soil containing fine contents, which is the type of soil found in many coastal areas, we examined methods for predicting the amount of ground subsidence caused by liquefaction.

We conducted undrained cyclic torsional shear tests while varying the uniformity coefficient of the sandy base material and fine contents (content rate(Fc) and type), and observed which parameters had a high correlation to volumetric strain following liquefaction. We chose two different types of mixed silica sand with different uniformity coefficients (silica sand 1: uniformity coefficient of 3.891, silica sand 2: uniformity coefficient of 2.286) as base materials, and used sandy soil blended with DL clay, which is non-plastic silt, and kaolin, which is a type of plastic clay, as test specimens.

We applied the following three parameters and measured their correlation coefficients to volumetric strain: 1) Void ratio, 2) Skeletal void ratio, and 3) Density deviation ratio [(ρdmax-ρd) / (ρdmax-ρdmin), ρd: The dry density of sandy base material, ρdmax: The maximum dry density of sandy base material, ρdmin: The minimum dry density of sandy base material]. The comparisons showed that the correlation coefficient would become higher in the following order: 3) Density deviation ratio > 2) Skeletal void ratio > 1) Void ratio. The correlation between density deviation ratio and volumetric strain is illustrated in the figure below. In the case where pure sand was used as base material, it can be seen that there is a consistent high correlation, unaffected by the variance caused by the different uniformity coefficient of the sandy base material. In the other case where the base materials contained fine particle fractions, a positive correlation is observed, while the degree of data scattering is relatively large.

In the cases of the mixed sandy soil, volumetric strains tend to be larger than the results of pure sand. To improve prediction accuracy, it is necessary to further analyze the effects on volumetric strain following liquefaction, focusing on how the physical properties of fine contents act in such process.