Numerical Study About Effect of Distribution of Thickness and Strength of Liquefiable Layer on Ground Deformation

Abstract eng:
When the liquefaction of the ground occurs, the settlement is often caused due to the shear deformation during the earthquake and the consolidation after the earthquake. However, this settlement does not occur regularly but usually varies in the space. One of the reasons of this variety would exist in the liquefiable layer itself, such as the thickness of the layer and the liquefaction resistance. However, it is not easy to identify the reasons and to predict such an irregular settlement in the field. In this paper, a series of effective stress analyses were performed in order to clarify these effects on irregular settlement. A railway yard damaged in the 2011 off the Pacific coast of Tohoku earthquake was selected as the objective for conducting numerical simulations. In this field, the coal ash under the railway track would have been liquefied. The layer under the coal ash was alluvial clay layer of about 20 m in thickness with a small N-value of SPT. The coal ash produced by steam locomotives was presumably deposited in the field in order to keep the height of the railway tracks against the consolidation of the clay layer. The height of the railway tracks after the earthquake was measured and the surface layer survey was conducted precisely in order to clarify the structure of the liquefiable layer. From these surveys, the settlement of the railway tracks has good correlation with the distributions of the thickness of the liquefiable layer and the liquefaction strength. In the simulations, five cases of calculations were performed. In Cases 1 and 2, 1-dimensional dynamic response analyses were conducted in which the horizontally layered ground was supposed. The maximum and minimum thicknesses of the liquefiable layer were supposed in Cases 1 and 2 based on the survey. In Case 3, 2-dimensional model in consideration of the distributions of the thickness of the layers at the field was prepared. In Cases 4 and 5, the distributions of the liquefaction strength were considered in the same model as in Case 3. Among the series of simulations, the simulation in Case 3 in consideration of the distributions of the thickness of the liquefiable layer reproduced well the measured settlement. On the other hand, 1-dimensional analysis showed smaller settlement even if the maximum thickness of the liquefiable layer was considered. Though the tendencies for the excess pore water pressure to increase with increasing thickness of the liquefiable layer in Case 3 were almost the same as those obtained from 1-dimensional analyses in Cases 1 and 2, the time histories of the displacement were different. This would be caused by the increased response of acceleration in the space due to the distributed layer. Furthermore, the settlement due to the consolidation after the earthquake was different from Cases 1 and 2. The simulations of Cases 4 and 5 in consideration of the distributed liquefaction strength showed better coincidence with the measured settlement than Case 3. In these cases, the occurrence of the liquefaction changed partially according to the strength of liquefaction. Therefore, larger settlement was evaluated in the region where the liquefaction strength decreased.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket