Modelling the time variance of the river bed roughness coefficient for improved simulation of water levels

ABSTRACT

Aquatic macrophytes create seasonally variable changes in river bed roughness. These changes in roughness influence the hydrodynamic behaviour of the river, for example, elevated water levels for the same discharge. In most hydrodynamic river models, the roughness is assumed constant in time, which may lead to high errors in the river water level results. When the river water level results are applied to water quality studies, biased estimates may be obtained of river flow velocities, dilution and other water quality processes. This article investigates how models implemented in two common hydrodynamic modelling software packages (MIKE11 and InfoWorks RS) can be adopted to take into account the time-varying river bed roughness due to plant growth. Albeit the use of a different approach by the two models, it is shown that both can account for that influence. After application to the Grote Nete river catchment in Belgium, far more accurate river water level results are obtained. Overall, the increase in accuracy of the water level simulations outweighs the increase in model complexity required to enable simulation of the vegetation changes. The model results can also be used to correct discharge estimates based on the measured water levels as shown for this case study.

Keywords:

• InfoWorks RS
• MIKE11
• plant growth
• rating curve
• roughness coefficient
• water level prediction

Acknowledgements

The Province of Antwerp and the Flemish Environment Agency (VMM) are gratefully acknowledged for providing the measurement data on cross-sections and structures along the rivers in the Grote Nete catchment; the Hydraulic Information Centre (HIC) for the provision of the water height measurements and the derived discharge estimates. Finally, DHI Water & Environment and Innovyze are respectively acknowledged for the provision of the license for the MIKE11 and InfoWorks RS software.

ORCID

Ingrid Keupers http://orcid.org/0000-0001-5844-5860

Thuy Nguyen Thanh http://orcid.org/0000-0003-0546-6603

Patrick Willems http://orcid.org/0000-0002-7085-2570

Additional information

Funding

This research has been supported by a Ph.D. fellowship grant of the Research Foundation – Flanders (FWO) to the first author and an IRO scholarship to the second author.