Analysis of simplified time of liquefaction triggering methods by laboratory tests, physical modelling and numerical analysis

The damage resulting from earthquakes can result from the combination of seismic excitation and/or due to a build-up of excess pore pressure in the soil (liquefaction). These two effects are related since the reduction of soil stiffness due to a decrease in effective stress, modifies the seismic response of the soil deposit. Therefore, the expected level and type of damage is dependent on the amount of seismic energy reaching the ground surface before liquefaction. The development and validation of simplified liquefaction assessment methods to provide reasonable estimates of the build-up of excess pore pressure is essential for improving estimates of the level of seismic demand (ground shaking and permanent ground deformation) that may be experienced by a building. This paper presents two methods, one based on equivalent cyclic stress loading, and another based on the cumulative strain energy, which are used to predict the evolution of the pore pressure build up throughout time. The centrifuge tests performed in ISMGEO (Italy) during the LIQUEFACT project (www.liquefact.eu) were used as a benchmark to evaluate the predictive performance of the methods. Additionally, a series of one dimensional soil column effective stress and total stress analyses and single soil element simulations were run. Available laboratory tests were used to calibrate the parameters of the simplified methods, as well as input parameters for the numerical simulations. The results showed that both simplified methods had considerable bias. A direct comparison of the effective stress analyses, a set of effective stress analyses with limited drainage, and the centrifuge results, highlighted that the centrifuge experiments exhibited significant pore water flow that was not captured in the simplified models. Comparisons between the irregular loading in the one dimensional analyses compared to the uniform loading in the element tests highlighted shortfalls in the conversion from irregular to equivalent uniform loading. Comparisons between stress demands from total stress, effective stress and the simplified methods illustrated the limitations of relying on the total stress acceleration to estimate demands on a soil in a liquefying deposit.