Comparing river routing concepts in distributed hydrological modeling

Distributed hydrological models are valuable tools for operational and strategic water management planning. These models include a representation of vertical processes such as interception, transpiration and infiltration and require a lateral component to route the water downstream along the river network. The kinematic wave is a commonly used approach for lateral flow in distributed hydrological models, which assumes that topography mainly controls the water flow. While this applies in steep terrain, the assumptions of the kinematic wave do not apply in flatter landscapes. The Wflow framework is a free and open-source distributed hydrological modeling platform developed by Deltares (van Verseveld et al., 2022). The Wflow framework has been extensively tested in catchments around the world using the SBM vertical concept in combination with the kinematic wave for lateral river, overland flow and subsurface flow routing. Recently, the local inertial approximation, which only neglects the convective acceleration term in the Saint-Venant equations, was implemented in the Wflow framework as an alternative lateral routing concept to accurately represent river routing processes in flatter areas. In addition, the numerical scheme proposed by de Almeida et al. (2012) was implemented for the simulation of 2D overland flow. Using the HydroMT (Hydro Model Tools, https://github.com/Deltares/hydromt) Python package, we set-up wflow_sbm models for the Rhine and the Meuse basins and compare alternative concepts for river (and overland flow) routing, including kinematic wave, local inertial 1D and local inertial 1D2D. The results show significant differences in the shape and magnitude of modeled peak flows and the importance of floodplain storage and floodwave attenuation processes, as the local inertial 1D2D simulations were closest to streamflow observations. With this study, we stress the importance of including relevant routing processes (floodwave attenuation through overbank flow in the floodplains) as opposed to a more extensive calibration which would compensate for these lacking processes.

 

References:

de Almeida, G. A. M., P. Bates, J. E. Freer, and M. Souvignet, 2012, Improving the stability of a simple formulation of the shallow water equations for 2-D flood modeling, Water Resour. Res., 48, W05528, https://doi.org/10.1029/2011WR011570.

van Verseveld, W. J., Weerts, A. H., Visser, M., Buitink, J., Imhoff, R. O., Boisgontier, H., Bouaziz, L., Eilander, D., Hegnauer, M., ten Velden, C., and Russell, B.: Wflow_sbm v0.6.1, a spatially distributed hydrologic model: from global data to local applications, Geosci. Model Dev. Discuss. [preprint], https://doi.org/10.5194/gmd-2022-182, in review, 2022.