Case study = undisturbed sampling, cyclic testing and numerical modelling of a low plasticity silt

A potential development site in Central Otago, New Zealand, is situated on alluvial sediments that comprise very soft to soft low plasticity silts. Liquefaction prediction methods using CPT, SPT and conventional laboratory testing indicate that liquefaction would be expected to occur in moderate earthquakes. Given the sloping nature of the site this indicated potentially significant lateral spreading displacements, which is inconsistent with the observed landforms in the vicinity. Therefore, a more detailed assessment of liquefaction was carried out using advanced undisturbed sampling, cyclic triaxial testing and slope stability analysis using a dynamic effective stress finite element model. The purpose of the paper is to demonstrate the applicability of these advanced techniques to geotechnical engineering practice in New Zealand. The cyclic laboratory testing showed a cyclic mobility type soil response. Dynamic effective stress finite element modelling, using parameters derived from the cyclic laboratory testing, was able to demonstrate that a catastrophic flow failure type response was not expected at the site. This suggests that while moderate horizontal and vertical displacements may occur in large earthquakes, they are expected to build up gradually throughout the earthquake rather than occur suddenly during a large scale slope failure. Deep ground improvements to ‘hold the slope back’ were not considered necessary and robust shallow foundations with shallow ground improvement should provide acceptable building performance