ABSTRACT
Failure of retaining walls has occurred many times during earthquakes, especially when the backfill was saturated. However, there is a general lack of data recorded during earthquakes in the field about seismic response of retaining walls. The results of some laboratory tests and numerical analyses have shown that fabric anisotropy of soil grains has a significant influence on the soil behavior. As a result, many researchers realized the importance of soil anisotropy. Meyerhof (1978) extended the theory of ultimate bearing capacity of shallow foundation on homogeneous isotropic soil to anisotropic cohesionless soils semi-analytically. Wang (2007) developed an analytical solution of lateral force induced by rectangular surcharge loads on a cross anisotropic backfill. Arthur (1972; 1975) and Oda (1972; 1977; 1979) studied the effects of anisotropy systematically using conventional lab tests and obtained some useful results. Peijun Guo (2008) used a new type of container to characterize strength variation of anisotropic sand in direct shear tests. Li & Dafalias (2000; 2002) incorporated fabric tensor concept into a constitutive model to simulate the anisotropic effect of granular materials. However, there are hardly any real physical data available to analyze the influence of anisotropy on the response of retaining walls during earthquakes. Thus, verification and validation of design principles and numerical simulations have to rely on physical data such as full scale model test or scaled centrifuge model test. Data from centrifuge tests have been used in recent years to study the mechanism of response and to verify the results of numerical simulations for seismic soil-structure interactions in a number of important projects.
