Identifying the effects of parameter uncertainty on the reliability of modeling the stability of overhanging, multi-layered, river banks

grainedmaterial. In such banks, fluvial erosion of the lower, non-cohesive, layer typically occurs at a muchhigher rate than erosion of the upper part of the bank. Consequently, such banks normally develop acantilevered bank profile, with bank retreat of the upper part of the bank taking place predominantly by thefailure of these cantilevers. To predict the undesirable impacts of this type of bank retreat, a number of bankstability models have been presented in the literature. These models typically express bank stability bydefining a factor of safety as the ratio of resisting and driving forces acting on the incipient failure block. Theseforces are affected by a range of controlling factors that include such aspects as the overhanging blockgeometry, and the geotechnical properties of the bank materials. In this paper, we introduce a new bankstability relation (for shear-type cantilever failures) that considers the hydrological status of cantileveredriverbanks, while beam-type failures are analyzed using a previously proposed relation. We employ thesestability models to evaluate the effects of parameter uncertainty on the reliability of riverbank stabilitymodeling of overhanging banks. This is achieved by employing a simple model of overhanging failure withrespect to shear and beam failure mechanisms in a series of sensitivity tests and Monte Carlo analyses toidentify, for each model parameter, the range of values that induce significant changes in the simulated factorof safety. The results show that care is required in parameterising (i) the geometrical shape of theoverhanging-block and (ii) the bank material cohesion and unit weight, as predictions of bank stability aresensitive to variations of these factors