Graduation of Jakob Christiaanse

Continuous sea level rise and growing environmental awareness have led to increasing implementation of nature-based solutions to counter coastal erosion. An example in the Netherlands is the Sand Engine—a mega-scale nourishment designed to feed the Dutch coast over a period of 20 years. For such designs to work, a good understanding of the governing natural processes is paramount. The behaviour of the sandy coast, however, is subject to natural variability in future weather and uncertainties in the interactions between sand dynamics and hydrodynamic forcing. To predict the evolution of coastal systems, engineers often apply numerical models. Next to uncertainty due to variability in natural forcing, such models introduce a series of model-related uncertainties, which are often not consistently included in predictions. With limited knowledge on the magnitude of these uncertainties, the long-term strategy development and design of projects are impeded. Parameter uncertainty denotes our limited knowledge on the values of free model parameters and is an important source of model uncertainty. This study aims to quantify parameter uncertainty in process-based coastal area predictions by analysing uncertainty bounds for morphodynamic predictions of the Sand Engine.

A study period of 14 months was chosen, from August 2011 (directly after construction of the Sand Engine) until October 2012. Using advanced numerical acceleration techniques, a synthetic dataset of 1024 morphological Delft3D predictions was generated, each with an identical hindcast period but different model parameter settings. First, a sensitivity analysis (elementary effects method) was performed to find the most influential parameters on three morphological indicators: cumulative volume change, shoreline position, and bed level change. Based on the sensitivity analysis, five parameters were selected for the large dataset. Subsequently, parameter uncertainty was quantified by a Generalised Likelihood Uncertainty Estimation (GLUE). To increase the convergence speed of the samples, quasi-random Sobol sampling was applied. To validate the applied parameter ranges and assess model performance, the predictions were compared to observations. Although the GLUE method has been successfully applied to limited morphodynamic problems, it has not been used to date for complex 2D coastal predictions.

Using GLUE, posterior likelihood distributions of the five parameters were derived from the 1024 model runs, revealing optimal values for each parameter. These showed significant differences to the default values from the Delft3D user manual. Uncertainty bounds derived for the morphological indicators provided an observation-based value of the parameter uncertainty in the predictions. They showed that uncertainty in input parameters translates to significant output uncertainty. Extended in 2D, the uncertainty bounds provided spatial uncertainty maps, revealing areas where parameter variation resulted in high uncertainty in bed level changes. Uncertainty growth appeared to be correlated to periods of high morphological activity, especially in the initial response phase (first seven months). Finally, a simplified comparison of parameter uncertainty versus wave climate variability indicated that both uncertainty sources are in the same order of magnitude, hence form significant contributions to overall prediction uncertainty.

The presented results have an impact on two levels. First, they can be used to communicate and address uncertainty in predictions of coastal change. For example, the spatial uncertainty maps can let stakeholders understand the potential range of outcomes for a certain design. Secondly, the results, combined with the created dataset, can provide valuable information for future morphodynamic studies. This work contributes to addressing the need for stochastic simulations, which are expected to be increasingly used for many coastal engineering and management purposes in the years to come.