UC Berkeley

Removal of individual blocks of rock is one of the principal mechanisms by which scour can occur, and prediction of block erodibility can be complicated due to the inherent variability associated with the rock mass as well as flow conditions in the vicinity of the block(s) in question. In order address the stochastic nature of the problem, we present a methodology for system reliability assessment of the probability of scour of 3D rock blocks subject to hydraulic loads within a block theory framework. Monte Carlo simulations are used to determine overall block failure probability, and to identify the most likely failure mode. A first-order reliability method (FORM) is then used to determine sensitivity to the different variables and hence the relative level of importance of the physical parameters with respect to the dominant failure mode. An example problem is used to illustrate the value of this information in focusing site investigations and analyses on the most important variables as well as in guiding decisions regarding scour mitigation strategies.