Incorporating river bed level variability into flood risk assessment: framework and application in the Netherlands

Abstract

The current flood risk assessment framework in the Netherlands aims at ensuring a nation-wide protection level with a long-term vision. It assumes that future river bed level changes will not pose negative effects on flood safety and it adopts a fixed-bed assumption. However, the river bed level shows a strong spatial and temporal variability, where bed aggradation trends could lead to higher water levels in the future. Recent measurements in the Dutch Rhine have shown both bed degradation and aggradation in different river sections along the lower river domain. This raises questions to what implications does future river bed level changes have on flood safety in the Netherlands and whether the current fixed-bed assumption has adverse connotations to it.

The objective of this MSc thesis is to study the impact of river bed level changes on flood safety in the Netherlands. To achieve this, a framework for incorporating large-scale bed level changes into flood risk assessments is developed and applied under the current Dutch context. The implications on the long-term and short-term flood safety are analyzed separately, as fundamentally different approaches and uncertainties relate to the flood safety standards derivation and for the periodic safety assessment in the Netherlands.

In the long-term, I propose to incorporate morphodynamic modelling scenarios into the flood safety standards derivation process to account for future bed level changes and their long-term uncertainties. Modelling scenarios should include future river discharge, sea level rise, river maintenance and the evolution of river bifurcations, as they represent main drivers in the long-term river bed response in the Netherlands. Based on existing long-term morphodynamic modelling studies of the Rhine River, it is estimated that the extreme water levels decrease over the upper Waal, but increase over the lower Waal by 0.15 m and 0.30 m in 2050 and 2100, respectively. These scenarios consider that river dredging is not maintained after 2025.

In the short term, I propose to incorporate the analysis of the recent river bed level behaviour into the safety assessment process to account for its uncertainty in the near future. Based on yearly bed level measurements in the Waal from 2005 to 2020, decreasing and increasing water level trends have been identified in the upper and lower Waal, respectively, with increments in the order of 0.05 m. This is small in absolute terms, but significant compared to the accuracy level of the hydraulic loads used in the safety assessment process. As a consequence, these increments can have important economical implications on dike reinforcement programs.

In summary, this thesis provides a framework incorporating large-scale river bed level changes into the current flood risk assessment framework in the Netherlands. Furthermore, it quantifies the impacts of bed level variability on flood safety in the Waal River. In general, the fixed-bed assumption in the current flood safety framework is a conservative, or safe, assumption for the hydraulic loads’ calculation in the upper Waal. However, it may underestimate the future hydraulic loads in the lower Waal and impose adverse conditions on flood safety.