Rock debris is considered as a highly stable material that can serve as foundation for structures. However, with the recent increase in earthquakes, even landfill grounds comprising rock debris need to be evaluated for potential effects of spatial variations in liquefaction strength. Existing evaluation methods may not be accurate as they often use the lower limit of liquefaction strength to ensure conservatism. In this study, three centrifugal model tests were performed to examine the influence of spatial variations in the physical properties of landfill grounds composed of rock debris, which support a cement-improved embankment. Case A represents a ground model devoid of weak components, whereas Cases B and C include weak elements with a volume ratio of 29.3%. These latter cases were established under varying conditions with respect to the positioning of weak components. It was ascertained that the displacements of the embankment in Cases B and C, where weak components were present, were marginally greater than those in Case A, under the application of analogous seismic motion. Upon comparing the outcomes of Cases B and C, the displacements of the embankment appeared consistent and independent of the configuration of the weak component.

The objective of the research is to investigate and compare the behaviour of impervious and pervious sheet piles to mitigate liquefaction-induced foundation settlement. A scaled model at a ratio of 1/5 was constructed based on a comprehensive large-scale shake table test. This model consisted of three soil layers with varying relative densities: 50% for the upper crust, 30% for the intermediate liquefiable layer, and 85% for the lower dense layers. A shallow foundation was then placed on the crust layer. Following model preparation, a sheet pile (both impervious and pervious) was installed using the press-in technique at 0.625 times the foundation length from the center of the foundation. To prepare the pervious sheet pile, holes sized 5 mm in diameter were made, followed by the layering of a non-woven geotextile on the outer surface of the sheet pile. This geotextile facilitated the exclusive flow of water while restricting soil passage. Additionally, two different sheet pile thicknesses were employed to assess the effect of thickness on the performance of pervious sheet piles. The experimental outcomes indicated impervious sheet piles performed better in reducing foundation settlement than pervious sheet piles, with a 26% improvement observed for thinner and a 13% improvement for thicker sheet piles.

The real response of a structure during earthquakes depends on the foundation input motion. The foundation input motion is usually considered as free field surface accelerations based on the site-specific ground response analysis. In this study, the effect of soil-structure interaction (SSI) on the foundation input motion and base accelerations was investigated using a typical 4-5 story residential building in liquefiable soils. The foundation input motion (FIM) was evaluated in finite element geotechnical software using a constitutive model representing the liquefaction behavior. The effect of the existence of the building was examined considering the massless boundary of the foundation (kinematic effect) and the building itself (inertial effect). The results demonstrated that the foundation input motion remarkably differs from free field surface motions in the existence of the building. The spectral base accelerations were obtained with higher damping and with a wider range of predominant periods in liquefiable soils when compared to the base acceleration spectra without SSI. The modal analysis was performed for the building in a structural finite element software incorporating the SSI. The structural 1^st mode natural period significantly increased in liquefiable soils. Therefore, the actual accelerations that the building is exposed to were estimated to be higher or lower than the computed accelerations without the effect of SSI, depending on the structural natural period and thickness of the liquefiable layer.

During previous earthquakes in Japan, underground structures, particularly reinforced box culverts (RBCs), have frequently suffered from loss of serviceability, due to joint openings among multiple units of box culverts. This issue has been mainly reported on liquefiable grounds and soft grounds, suggesting that little attention has been paid to the seismic performance of the entire road embankment system. From the current study, joint openings were found to be higher in undrained conditions. This was influenced by high liquefaction levels at the middle joint of the culvert hence generating high stresses. Also, longer cyclic stress paths were noted for the undrained conditions generating very high strains beneath the culvert as compared to the drained conditions. Ground outside the culvert region behaved the same; in terms of liquefaction, accelerations, stress paths and stress-strain relationships; irrespective of the drainage condition.

Rock debris is considered as a highly stable material that can serve as a foundation for structures. However, with the recent increase in earthquakes, landfill grounds comprising rock debris must be evaluated to determine the potential effects of spatial variations on liquefaction strength. Existing evaluation methods may not be accurate because they often use the lower limit of the liquefaction strength to ensure conservatism. In this study, numerical simulations of centrifugal model tests were conducted to analyze the effects of spatial variations on the physical properties of landfill ground composed of rock debris that supports a cement-improved embankment. The ground models used for the centrifugal model tests comprised 29.3% weak parts with a small liquefaction strength ratio, and the parameters of the constitutive model were set based on laboratory test results. Similar to the model test results, the numerical simulations revealed that the difference in the placement of the weak parts had little effect on the amount of embankment settlement. In addition, the weighted average ground physical properties based on the volume ratio of the weak parts approximately reproduced the same amount of embankment settlement in the tests.

This study aims to use the actual cases of liquefaction-induced building settlements during the Hualien earthquake on February 6, 2018, to examine the empirical prediction formula for liquefaction-induced building settlements developed in Taiwan for liquefaction risk assessment. The empirical prediction formula is established by collecting actual cases of liquefaction-induced building settlements during the 1999 Chi-Chi Earthquake and the 2016 Kaohsiung Meinong Earthquake events, and using various indicators obtained from the site liquefaction assessment for statistical regression. This study first reviewed the relevant literature to obtain the analysis information such as basic data of buildings affected by soil liquefaction in the Hualien City, the scope of liquefaction-induced settlements, and representative geological boreholes in the liquefaction area. Second, to use the SPT-N based simplified soil liquefaction evaluation methods to calculate the relevant indicators of soil liquefaction potential such as liquefaction potential index P_L, post-liquefaction ground settlements S_L of free field sites, and thickness of the surface non-liquefaction layer. The liquefaction indicators and basic parameters of shallow-founded buildings such as the number of floors and the width of buildings were input to the empirical prediction formula for liquefaction-induced building settlements to estimate the empirical values of liquefaction-induced building settlements. Finally, the estimations would be compared with the actual settlements of shallow-founded buildings during the 2018/02/06 Hualien earthquake to understand the applicability of the empirical prediction formula for the soil liquefaction risk assessment in the Hualien City area. The research results show that the calculated empirical values of liquefaction-induced building settlements are similar to the actual values of settlements, which implies soil liquefaction impacts significantly on the shallow-founded buildings in the Hualien city during 2018/02/06 Hualien earthquake. This empirical prediction formula for Taiwan liquefaction-induced building settlements could be preliminarily applied to the risk assessment of soil liquefaction in the Hualien City area.

Effective stress analyses of various soil-structure systems are performed, including six case histories of river dykes with induced settlements ranging from 2.7m to none (no damage), three case histories of caisson quay walls with induced lateral displacement of 3.7m to none (no damage), four sheet pile quay walls with induced lateral displacement of 1.8m to none (no damage), one breakwater with induced settlement of 2.0m and one pile-supported pier with induced lateral displacement of 1.5m. The analyses are performed with a SPT N-value based parameter calibration procedure of strain space multiple mechanism model of cocktail glass model type (Iai et al, 2011) allowing generation and dissipation of excess pore water pressure. The results of analyses are generally consistent with those of case histories with and without damage. Introduction of Sus (steady state strength) in the analysis shows better representation of degree of damage for river dykes, caisson walls, breakwaters, and pile-supported piers.