Critical Appraisal of Common Simplified Assumptions in Seismic Stability Analyses of Gravity Dams

Abstract

In practical seismic stability analyses of gravity dams, it is common to (1) oversimplify irregular dam-rock interfaces, (2) neglect the effects of cohesion and tensile strength, and (3) ignore the effects of vertical ground accelerations. In this paper, a critical appraisal of such simplifying assumptions is proposed. For this purpose, an efficient procedure for dam seismic stability analysis not requiring the above-mentioned assumptions is first proposed. The developed technique is applied to investigate the seismic stability of an existing gravity dam with a stepped dam-rock foundation. The response of the dam is studied under the effects of two earthquakes differing by their frequency contents. The resulting stresses at the dam-rock interface, sliding safety factors, crack lengths, and residual displacements are examined. Detailed discussion of the effects of dam-rock geometry, mechanical properties at the dam-rock interface, and the vertical seismic component is presented. It is primarily shown that a simplified dam-rock geometry should be used with caution because it may lead to inaccurate results for the dam sliding safety factor and residual displacement. The results also emphasize the importance of cohesion, tensile strength, and friction in dam seismic stability assessment. The sensitivity of stress distributions at dam-rock interface to ground accelerations is illustrated. The sliding safety factor is found to be practically insensitive to the vertical seismic component, whereas dam residual displacements are more affected.