Uncertainty in Coastal Structure Reliability

Large improvements in hydrodynamic, statistical and structural modelling accuracy have resulted in more disparity in study accuracy. Some studies utilize almost no modelling and rely on expert judgement while others utilize highly accurate modelling. This has produced a demand for uncertainty estimates in order to understand the impact of low fidelity modelling. For the present study, high fidelity hazard, statistical and structural models are utilized. Monte Carlo simulation is used in a Level III reliability analysis to estimate the impact of uncertainty in reliability estimates of rubble mound damage for 3 structures with varied hazard and structure characteristics. Varying epistemic uncertainties illustrate the impact of hazard model accuracy. The results show that input epistemic uncertainties in the range of typical studies produce fairly small variations in reliability as long as the uncertainties are unbiased. Wave height and water level uncertainty of similar magnitudes produce similar variations in reliability. Biased uncertainty produces significant variations in reliability. Nearshore wave transformation is shown to produce significant improvements in reliability estimates.