Water Balance of the Maurikse Wetering

Abstract

The Maurikse Wetering is a tributary of the Line river in the centre of the Netherlands. As the catchment is situated between and in close proximity to the larger Nederrijn and Waal rivers, intercatchment groundwater flow (IGF; groundwater flow crossing topographic divides) might have a significant influence on the water balance of the Maurikse Wetering catchment. As the IGF cannot be measured directly and due to the complex nature of the IGF, the IGF is considered to be one of the hardest fluxes to quantify in conceptual hydrological modelling. The objective of this thesis is to express the IGF as a function of easy-to-measure variables.
The most common method used to estimate the net IGF to a catchment in conceptual hydrology is to equate the IGF to the missing water in the water balance of the catchment. A different approach is analyzed in this thesis. To quantify the IGF, the Maurikse Wetering catchment and the surrounding area is modelled in the groundwater model MORIA. The direction of the groundwater flow shows that the groundwater flow in the region is heavily influenced by the Nederrijn and Waal rivers. A multiple linear regression (MLR) analysis was performed to study the influence of variables affecting the water balance of the Maurikse Wetering, e.g. the water level in the Nederrijn, the groundwater level in the Maurikse Wetering catchment and the precipitation, on the IGF. The resulting relations reliably model the IGF. The fact that multiple easy-to-measure factors have a relation to the IGF, shows that there are alternative methods to equating the IGF to the missing water in the water balance. This provides a basis for the usage of IGF relations in predictive modelling.
To analyse the effect the IGF has on conceptual hydrological models, the catchment is modelled in WALRUS. The model is calibrated, modelled and validated both without IGF data and with the IGF as modelled by the derived relations. Including the IGF in WALRUS shows an improvement in modelling the variation in groundwater level over smaller time steps compared to the WALRUS model without IGF, the latter only showing a seasonal change in groundwater level. During validation, the model with the IGF relation retains a higher efficiency in modelling the average groundwater level in the Maurikse Wetering than the model without an IGF flux. The results with and without the incorporation of IGF into the model show that the IGF can contribute to significant improvements for conceptual hydrological models.