Liquefaction response of reclaimed soils from effective stress analysis

Abstract

One-dimensional (1D) dynamic effective stress site response analysis (ESA) is performed for profiles at the port of Wellington, New Zealand (CentrePort), which contains reclamation fills comprised of gravel-sand-silt (G-S-S) mixtures and hydraulic fills. The first phase of the study realistically simulates three recent earthquake case histories while considering modelling uncertainties by using the PM4Sand and the Stress-Density constitutive models. The results illustrate possible mechanisms explaining the severity of liquefaction manifestation and soil ejecta characteristics observed in G-S-S fills through careful engineering interpretation of the response. Challenges for 1D ESA to explain complex manifestation patterns affected by two-dimensional variability in fill composition and response characteristics are illustrated for the hydraulic fills. In the second phase of analyses, ESA-based response measures are proposed to quantify the severity of the liquefaction response for a range of input seismic demands. The response characteristics show very small scatter despite using a range of different input ground motions and two soil constitutive models. Results illustrate the capability of ESA to capture details of the liquefaction response such as the similar threshold seismic intensity for liquefaction triggering of the loosely deposited fills, different maximum response of the sites reflecting the differences in the thicknesses of the fills, and the evolution of the response from triggering to maximum reflecting differences in depositional characteristics.