Lateral spreading of gently- sloping deposits of liquefiable sand is a cause of much damage in earthquakes, reportedly more than any other form of liquefaction - induced ground failure. Estimation of the forces induced on structures such as pile foundations within these deposits requires a knowledge of both the magnitude of spreading occurring and the distribution of displacement through the liquefied layer. This paper concerns itself with measuring the displacement profile through a laterally spreading slope by means of coloured sand marker lines within centrifuge models. A series of dynamic centrifuge model tests have been carried out on gently slopings and beds containing instrumentation for measuring pore pressures, acceleration sand displacements. It can be concluded that the surface displacements relatively easily measured in the field by such techniques as aerial photogrammetry do not necessarily indicate the magnitude of lateral spreading occurring at depth. Surface layers above the phreatic surface may rotate, causing increased displacement with depth. Alternately they may shear on the boundary with the underlying liquefied layer by virtue of support received from the toe, giving a displacement discontinuity.
A large volume of case study data is available on lateral spreading towards these free boundaries from previous earthquakes. Dynamic centrifuge modelling was carried out using sloping laminar boxes at RPI by Abdoun (1997), and finite element modelling at Cornell by Meyersohn (1994). Much of the data from field studies of the extent of lateral spreading is based on aerial photogrammetry or satellite imaging, comparing photographs from before and after the earth quake and measuring the movement of objects dragged with the lateral flow, such as man holes and some buildings, relative to certain fixed points. Unfortunately this technique can only give surface displacements for the soil.
