| Abstract: | Newmark's sliding block analysis for evaluating earthquake-induced permanent displacements of earth dams and slopes did not consider the effects of elastic dynamic response. Makdisi and Seed extended the analysis to include such effects, using the simpfying assumption that the computation of dynamic response and plastic slip can be decoupled. This paper examines the error introduced by this assumption, using three idealized lumped-mass models for a dam. Both sinusoidal and synthetic earthquake motions are employed. When the predominant frequency of the input motion lies in the proximity of the fundamental frequency of the dam, the slip based upon the decoupling assumption exceeds the exact value. This error decreases as the threshold acceleration required to initiate slip and the damping in the soil increase. For practical application, the error should seldom exceed 20 per cent. |
