The effects of spatial variability of soil properties on the behaviour of saturated soil deposits subjected to seismic excitation are analysed, and their implications for geotechnical design are discussed. A Monte Carlo simulation methodalogy, combining digital generation of non-Gaussian stochastic vector fields with dynamic, nonlinear finite element analyses, is used for that purpose. The proposed procedure is applied to assess the soil liquefaction potential, which is found to be considerably affected by the inhereat spatial variability of soil properties. Parametric studies, involving the degree of soil liquefaction and the frequency content of the seismic input, are performed. Finally, a characteristic percentile value of soil strength, which will predict a response equivalent to that provided by the more expensive Monte Carlo simulations, is proposed for use in deterministic design.
All Science Journal Classification (ASJC) codes
- Geotechnical Engineering and Engineering Geology
- Earth and Planetary Sciences (miscellaneous)
- In-situ testing
- Numerical modelling and analysis
- Shear strength
- Statis-tical analysis