Back water effects in the Groenlose Slinge resulting from bend flow
In the 20th century most Dutch brooks were channelized to increase discharge capacity and improve shipping conditions. This large intervention has had destructive consequences on many lowland ecosystems. A common procedure has been to re-meander these streams to reduce discharge capacity and increase ecological conditions. A disadvantage of this measure is that 1D hydrodynamic models underestimate backwater effects resulting from friction in channel bends and thus underestimate the occurrence of flooding. The goal of this study was to compare simulated water levels of a 1D model to a 2D model to determine if 2D models might be more reliable than 1D models and to improve our understanding about backwater effects resulting from stream bends.
In this research a casestudy of a re-meanderd brook, the Groenlose Slinge, has been modelled using a 1D-model (Sobek1D) and a 2D depth averaged model (Delft3D-2DH). This location was chosen because of a sharp bend which is expected to have caused flooding. Data measured at the location of the casestudy was provided by the Dutch water board Rijn & IJssel. Cross-sections were used to schematize the bathymetry and measured time series of discharge and water levels were used to determine boundary conditions to perform scenario analysis.
The resulting water levels along the waterline of the meandering stream, simulated by both models, were analysed and compared. The maximum water level difference simulated for the highest discharge scenario was 11 centimeters. This difference is not considered troublesome, as the scenario had a recurrence time of 1 event in 10 years. As the time series data was measured a relatively short period without high discharge events, the simulations could not be validated properly. Future research could be done using theoretical models to determine a relationship between stream bend characteristics and backwater effects.