Seismic liquefaction and lateral spreading assessment for a pipeline river crossing using simplified and advanced methods

抄録

For assessing seismic liquefaction triggering and its consequence of lateral spreading and permanent ground displacement estimation, there are various methods of evaluation, with varying degrees of sophistication and complexity. At one end of the spectrum, simplified, empirical or semi-empirical based approaches are available in the literature to assess the seismic liquefaction triggering and estimation of lateral spreading, primarily used as screening level tools. At the other end of the spectrum, advanced seismic numerical modelling using sophisticated soil constitutive models can be utilized to evaluate the seismic performance of liquefiable soils and an estimation of the permanent ground displacements due to liquefaction. This paper focuses on the seismic liquefaction and lateral spreading displacement assessment for a proposed pipeline river crossing using both simplified and advanced numerical methods.

The proposed pipeline crosses a major river using horizontal directional drilling (HDD) method, which will go through potentially liquefiable fluvial sand deposits and soft clays. The estimation of lateral spreading and permanent ground deformation along the proposed HDD path was required to ultimately assess the integrity of the pipeline under the design seismic load. At the initial phase of the study, simplified liquefaction triggering assessment using Idriss and Boulanger (2008) method was performed and then estimation of lateral spreading was made using semi-empirical correlations in the literature such as Youd et al (2002). Then refinements to the simplified liquefaction triggering assessments were made by performing one dimensional site response analysis to have a better estimate of seismic shear stresses. In the last phase of the study, detailed seismic numerical model of the crossing was analyzed using FLAC software. The seismic performance of the potentially liquefiable unit was assessed using the advanced soil constitutive model of UBCSAND and for non-liquefiable units UBCHYST constitutive model was used. Furthermore, in the FLAC model of the pipeline crossing, the spatial variability in the liquefiable sand unit was also accounted for, by having a distribution of Standard Penetration Test (SPT) measurements, using regression analyses, honoring field measurement values. The initial phase of the study indicated widespread liquefaction and large permanent ground deformation which required both ground improvement and thickening the pipe wall thickness as mitigation measures. However, through the more rigorous advanced FLAC analysis, it was illustrated that the spatial extent of the liquefaction was more constraint, and the predicted permanent ground deformation was also smaller, leading to only requiring thickening the pipe wall thickness.