When the excess pore water pressure generated during an earthquake dissipates in saturated loose sand, it causes post-liquefaction reconsolidation that can potentially yield substantial damage to the structure. To build resilient infrastructure, it is paramount to estimate these settlements as well as introduce soil reinforcement techniques to mitigate associated risks. Although there are abundant studies on liquefaction triggering assessment, the study of post-liquefaction settlement and the effects of stone columns as soil reinforcement is a relatively less established field. Generally, simplified empirical methods are employed for settlement evaluations. However, they possess several limitations such as the influence of non-liquefiable layers, soil fabric, permeability, and so on. Numerical models can be utilized to capture these effects with proper validation. This study evaluates the performance of stone columns in reducing seismically induced post-liquefaction settlement utilizing the Finite Element Method (FEM) and constitutive relationship, PM4Sand model, as it has been extended to account for reconsolidation settlement. The ability of the numerical framework to capture reconsolidation settlement is validated by replicating a shake table test performed on Ottawa F-55 sand. Results are compared with a previous numerical study inspired by the same experiment. After validation, a generic numerical model is proposed, and the performance of the natural ground and the reinforced ground is compared. A parametric analysis using 12 different ground motions is performed to assess the effect of varying ground motion intensity on the post-liquefaction settlement. The analysis is also performed with the conventional PM4Sand model (without the extension for reconsolidation). Finally, simulations are performed with a footing load above the soil model. The results demonstrate that (a) the presence of stone columns reduces post-liquefaction settlement, and (b) conventional constitutive models can highly underpredict post-liquefaction settlement. Further research is required to assess the effects of (a) 3D, (b) variations in permeability, (c) parametric analysis of stone columns, and (d) densification of stone columns. / Master of Science / When subjected to an earthquake, loose saturated sand may undergo liquefaction and exhibit a reduction in shear strength due to a rise in excess pore water pressure and the corresponding reduction in effective stress. This leads to failures associated with settlements resulting from the gradual dissipation of excess pore pressures. This mechanism results in post-liquefaction settlement. Several authors have investigated the mechanism of the post-liquefaction behavior of sand and proposed methodologies to assess the deformation caused by seismic loads. They mainly conclude that the reconsolidation mechanism is characterized by a decrease in the overall soil stiffness and an increase in permeability. Among different methodologies to quantify this settlement, finite element numerical modeling is the most widely used. The primary task in performing such numerical simulation is to select the best constitutive model (i.e., stress-strain relationships) that can accurately capture post-liquefaction behavior. In this study, the capabilities and limitations of the most common constitutive models are reviewed. Moreover, the efficacy of stone columns is also assessed to mitigate the risk posed by liquefaction. Firstly, the numerical framework is validated against data from a shake table test experiment. Then, a numerical model is proposed and subjected to different seismic motions. The settlement of the ground with and without stone columns is assessed and compared for all motions. In addition, the efficacy of stone columns is also analyzed by simulating the model with a footing load. Thus, this study provides insights into the effectiveness of stone columns under different seismic motions.

Frequently, deep foundations extend through potentially liquefiable soils. When liquefaction occurs in cohesionless soils surrounding a deep foundation, the skin-friction in the liquefied layer is compromised. After cyclical forces suspend and pore pressures dissipate, effective stress rebuilds and the liquefied soil consolidates. When the settlement of the soil exceeds the downward movement of the foundation, downdrag develops. To investigate the loss and redevelopment of skin-friction, strain was measured on an instrumented micropile during a blast-induced liquefaction test in Mirabello, Italy. The soil profile where the micropile was installed consisted of clay to a depth of 6m underlain by a medium to dense sand. The 25cm diameter steel reinforced concrete micropile was bored to a depth of 17m. Pore pressure transducers were placed around the pile at various depths to observe excess pore pressure generation and dissipation. Soil strain was monitored with profilometers in a linear arrangement from the center of the 10m diameter ring of buried explosives out to a 12m radius. Immediately following the blast, liquefaction developed between 6m and 12m below ground. The liquefied layer settled 14cm (~2.4% volumetric strain) while the pile toe settled 1.24cm under elastic displacement. The static neutral plane in the pile occurred at a depth of 12m. From 6m to 12m below ground, the incremental skin-friction was 50% compared to pre-liquefaction measurements. The decrease in residual skin-friction is consistent with measurements observed by Dr. Kyle Rollins from previous full-scale tests in Vancouver, BC, Canada, Christchurch, New Zealand, and Turrel, Arkansas.

Frequently, deep foundations extend through potentially liquefiable soils. When liquefaction occurs in cohesionless soils surrounding a deep foundation, the skin-friction in the liquefied layer is compromised. After cyclical forces suspend and pore pressures dissipate, effective stress rebuilds and the liquefied soil consolidates. When the settlement of the soil exceeds the downward movement of the foundation, downdrag develops. To investigate the loss and redevelopment of skin-friction, strain was measured on an instrumented micropile during a blast-induced liquefaction test in Mirabello, Italy. The soil profile where the micropile was installed consisted of clay to a depth of 6m underlain by a medium to dense sand. The 25cm diameter steel reinforced concrete micropile was bored to a depth of 17m. Pore pressure transducers were placed around the pile at various depths to observe excess pore pressure generation and dissipation. Soil strain was monitored with profilometers in a linear arrangement from the center of the 10m diameter ring of buried explosives out to a 12m radius. Immediately following the blast, liquefaction developed between 6m and 12m below ground. The liquefied layer settled 14cm (~2.4% volumetric strain) while the pile toe settled 1.24cm under elastic displacement. The static neutral plane in the pile occurred at a depth of 12m. From 6m to 12m below ground, the incremental skin-friction was 50% compared to pre-liquefaction measurements. The decrease in residual skin-friction is consistent with measurements observed by Dr. Kyle Rollins from previous full-scale tests in Vancouver, BC, Canada, Christchurch, New Zealand, and Turrel, Arkansas.