In the Netherlands, robust dike and dam design is a major concern in the context of flood defence. Due to heterogeneity of the subsoil on which these structures are founded, the validity range of in situ tests decreases drastically. Consequently, large uncertainties regarding spatial variation of soil stratification and soil layer parameters are incorporated in the cross-sectional reliability requirements, resulting in conservative designs. This thesis presents a Machine Learning application, which, by learning locally measured information and analysing high spatial resolution surface settlement data, can provide insights into spatial variation of soil stratification. Through the analysis of these insights, the uncertainties regarding spatial variability in cross-sectional reliability requirements can be reduced, which leads to less conservatism in dike and dam construction.

Building with Nature (BwN) is a relatively novel way of thinking for flood protection. Unlike traditional flood protection solutions that are poised to bring forces of nature to a halt, the BwN’s philosophy is to use natural dynamic processes for engineering advantage as a protection against natural disasters and events. Yet, the BwN design philosophy faces obstacles in terms of technical efficacy and evaluating added benefits when compared to traditional flood defence solutions.

This research sets out to understand how these obstacles faced by the BwN design philosophy play a role in the consideration of a BwN flood defence solution during project development. This is done by relating the assessment of the added benefits to five discussion fields, namely Costs, Function, Policy, Responsibility and Support. In doing so it can be determined which discussion fields play an important part in embracing a BwN design based on its added benefits while considering its technical uncertainties. This research limits itself to a case study of the Houtribdijk reinforcement project where both traditional and BwN flood defence solutions were employed. Furthermore, this research limits itself to the project phase where flood defence options are still considered.

This research found that although BwN flood defence project face technical uncertainties and discussions on efficacy, there is willingness to embrace a BwN solution when involved parties can agree on shared responsibilities. The curators of a flood defence project were found to be least willing to accept a technically uncertain design as they are accountable should it fail. The curators’ concern regarding their responsibilities can be mitigated and they can be more open to alternative solution if they are ensured of proper research and shared accountability when other values are destroyed in favour of flood protection. It was also found that the effect on stakeholders in financing and taking share of the responsibilities also have a positive effect on embracement of a BwN solution when the discussions on efficacy have not yet been settled.

The Netherlands is a low-lying delta and therefore prone to flooding. After the flood in 1953, the Delta Committee was established which commenced the Delta Works to enhance water safety for the society and economy. The Maeslant Barrier and Eastern Scheldt Barrier are the largest and most technologically advanced flood defences which are part of the Delta Works. These storm surge barriers shorten the effective coastline to reduce the risk of flooding within the Rhine-Meuse Delta and the Eastern Scheldt. However, the climate is changing and the risk of accelerated sea level rise increases. The problem is that the current storm surge barriers are not designed to withstand significant sea level rise. Sea level rise influences the hydraulic loads on the barriers which consequently increase the failure probability of these flood defences. International research argues that future sea level rise is much more uncertain than previously thought because of the instability of the West-Antarctic ice sheet. Therefore, it is critical to identify the degree of uncertainty of climate change and sea level rise to develop effective strategies for the storm surge barriers. This study uses a risk analysis to analyze the critical risk factors which could affect the performance and remaining lifespan of the Maeslant Barrier and the Eastern Scheldt Barrier. When sea level rise exceeds the tipping point of the storm surge barrier, the barrier fails to deliver sufficient water safety and adjustments are required to comply with the strategies of the Delta Programme. Therefore, this research has developed feasible and affordable solutions to increase the performance and remaining technical lifetime in case of accelerated sea level rise.

The transition towards a new risk approach for the Dutch national safety assessment of primary flood defences has been taken as an opportunity to improve the dealings with uncertainties. The probabilistic models for the new safety assessment (WBI2017) not only deal with the natural variability, but also with so-called epistemological uncertainties. One class of epistemological uncertainties is the model uncertainty in the hydrodynamic models used. The quantification of the water level uncertainty depends on the dominant hydraulic processes in each water system and is chosen to be independent of the return period. However, water level frequency lines derived including model uncertainty sometimes conflict with the physics. A comparative analysis is carried out to assess the performance of Hydra-NL w.r.t. observations and to analyse if physical processes that play an important role in each fresh water system are represented correctly after the inclusion of model uncertainty with the WBI2017 method. This study showed that the estimated exceedance probability of actual water levels in the tidal river area and lake area are often overestimated by the Hydra-model even without model uncertainty. When model uncertainties are included the overestimations become even larger. Furthermore, the inclusion of model uncertainty sometimes gives an incorrect representation of the underlying physical processes. The effect of the model uncertainty gets larger when the water level frequency line becomes flatter. This is e.g. the case at locations where the water levels are influenced by the closure of the Europoortkering and upstream of the flood channel near Veessen-Wapenveld. It can be argued that water level uncertainties are likely to decrease in these situations, because a reservoir is more predictable than a flowing river and the operation of the flood channel results in an enlarged conveyance that is less sensitive than an average river profile. The hypothesis that the inclusion of model uncertainty according to WBI2017 results in an incorrect representation of the underlying physical processes is further analysed for several case studies in the upper river area and the lake area. Varying river schematisations and two river interventions (flood channel and retention area) are modelled to compare the WBI2017 method with a more physics-based approach. In this physics-based approach the hydraulic roughness of the main channel and/or floodplains is incorporated as additional stochastic variable for the derivation of water level frequency lines. It is shown that the water level uncertainty becomes smaller for wide rivers where the water level frequency line becomes flatter. The cases where river interventions are present the water level uncertainty is bounded, because of the environment/geometry that influences the water levels. According to the physics-based method the water level uncertainties are location and discharge dependent, which are both not integrated in the WBI2017 method. For the lake area the focus is on locations where high water levels are predominantly determined by the wind that is causing a set-up on the lake. By considering an empirical parameter of the "capped Wu" formula as additional stochastic variable the uncertainty of the drag coefficient is modelled for the physics-based method. The physics-based method demonstrates that the water level uncertainty is larger for higher decimate heights and vice versa. It can be concluded that it is not valid to choose one uniform standard deviation of the water level for all wind-dominated locations, as it is done for WBI2017. It is recommended to include the model uncertainty by considering a model parameter in the hydrodynamic model for the upper river area and lake area as additional stochastic variable in the Hydra-models, to model uncertainty in the water level. In other water systems it becomes more complex, because several equally important sources of uncertainty are present. Still, the most important model uncertainty sources could considered stochastically, but computational effort would become the limiting factor