Breach Forming Scenarios at Concrete Faced Rock-fill Dams

Abstract

Rock-filled dams with reinforced concrete facing, as the proposed study case of Bolboci Dam on the River Ialomiţa, Dâmboviţa County, centre Romania, are characterized by a very high operation safety. On the international level, there is no record of failed concrete faced rock-fill dams. However, during their usage there have been several incidents mainly due to the unsatisfactory behaviour of the reinforced concrete masks. Specifically, in an ICOLD statistic that presented 17 cases of damages to rock-fill dams with reinforced concrete facing, 14 were due to the unsatisfactory behaviour of the reinforced mask. The other 3 cases were due to piping phenomena or percolation through the dam foundation or the lake’s shore. The failure scenarios of this dam type, Bolboci Dam including, can be as follows: the retaining structure over-spilling during floods due to an insufficient or poor operation of top and/or bottom dischargers - the overflow gradually erodes the body of the dam until the breach occurs; concentrated infiltrations with a high gradient through the dam body as a result of a serious malfunction in the concrete facing sealing, with material carving in, leading to breaches in the dam body; piping in the dam foundation / side banks that could finally lead to slippage and finally dam failure (improbable due to the existing geotechnical conditions); overpassing the safety guard of the dam’s crown leading to the dam over-discharging as a result of irreversible excessive subsidence due to exceptional incidents. There have been studies on three statistical methods on how characteristical breaches occur: MLM Empirical Method, Washington State and Froehlich. The Froehlich method has been endorsed and modified, because Bolboci Dam is a well built, well executed and a well employed dam. The failure hypotheses of the dam were done with applications using MathCAD software, their algorithm being based on the finite differences method. It must be specified that the statistical approach of dam failure is a trap, due to the fact that it is well know they are a work of one of a kind engineering art. This fact shows the difficulty of the elaboration of the failure scenarios and the need of a sensitivity study for the comparison of the scenarios. The results show that the failure of rock-filled dams with reinforced concrete facing is highly improbable and leads to small flows with a sluggish evolution of the breaches.