Importance sampling for efficient modelling of hydraulic loads in the Rhine–Meuse delta

摘要: During flood events, breaching of defences along a river system can have significant reducing effect on downstream water levels and risks. This paper presents Monte Carlo based risk framework for policy decision making, which takes this retention into account. The is developed to estimate societal in terms potential numbers fatalities associated probabilities. It tested the Rhine–Meuse delta Netherlands, where floods be caused by high flows Rhine Meuse rivers and/or sea North Sea. Importance sampling applied procedure increase computational efficiency computations. focuses development testing efficient importance strategies framework. an strategy deltas more challenging than non-tidal only discharges are relevant, because relative influence discharge level differs from location location. As consequence, methods that accurate one may inefficient other locations or, worse, introduce errors computed design levels. Nevertheless, case study described required simulation time was reduced factor 100 after introduction method framework, while at same accuracy estimates were improved.