Effects of spatial variability of soil properties on the seismic response of an embankment dam

• Seismic response of an embankment dam is studied using the probabilistic approach.
• Cross-correlated random fields are used to define the soil characteristics and spatial variability.
• 2D advanced numerical analyses are carried out using stochastic models of an embankment dam.
• Seismic demand hazard curves are derived using the PEER-PBEE framework.

In this paper a numerical procedure for a two-dimensional seismic analysis of an embankment dam is presented taking into account the spatial variability of soil properties. The approach integrates advanced numerical modelling and random field theory into a probabilistic framework. Geotechnical parameters are modelled as anisotropic random fields with different autocorrelation distances in the vertical and horizontal directions. The correlation between cohesion and friction angle values is taken into account with Gaussian cross-correlated random fields using a specific marginal distribution, autocorrelation functions and cross-correlation coefficients. Based on this stochastic representation of the soil domain, fully nonlinear time-history analyses are carried out for several earthquake records of varying intensities. The results of the analyses are used to explicitly derive fragility functions that include the influence of spatial variability of soil parameters and the uncertainty of ground motion.