Dealing with Uncertainties in Long-term Predictions of a Coastal Nourishment

In coastal engineering, the wave- and tide-driven alongshore sand drift and the resulting shore erosion or accretion can be fairly well predicted with numerical 2DH models. Despite recent advances in process-based morphodynamic modeling, application of such models is inevitably associated with considerable uncertainties from various sources.

This paper describes a comprehensive approach identifying and quantifying individual sources of uncertainty and integrating these into a single probability density function (PDF) of aggregated model outcomes. The approach was developed for the tender of a Design, Build and Maintenance (DBM) contract to upgrade a sea dike at the Holland coast. The reinforcement comprises a sand beach and dune, constructed at the foreshore of an existing 6 km long hard sea defense: the Hondsbossche and Pettemer Zeewering” (referred to as HPZ from here onwards). The main challenge was to design the nourishment so that it meets the functional requirements whilst minimizing the maintenance costs over the 20 year post-construction period.