Estimating the roughness of muddy beds
A study based on in-site measurements and numerical modeling
H.J. Speerstra (TU Delft - Civil Engineering & Geosciences)
Peter M.J. Herman – Mentor
D. S. Van Maren – Graduation committee member
I. Colosimo – Graduation committee member
TJ Zitman – Graduation committee member
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Bed roughness is an important parameter for the prediction of sediment transport as well as calculating flow conditions near the bed. Numerical models are extensively used to make these predictions. In these models it is fundamental to have proper values for bed roughness. The bed roughness is being estimated using a dataset from an ongoing pilot project in the Dutch western Wadden Sea. This dataset consists of 40 days of velocity measurements using ADV (8 Hz) and ADCP (1 Hz) instruments and concentration of SPM measurements using OBS instruments.
To estimate the bed roughness, the dataset is analyzed using four methods for calculating the bed shear stress. The logarithmic profile, turbulent kinetic energy, vertical turbulent kinetic energy and the Reynolds stress method.
The data has been processed and averaged per tidal phase, so statistical analysis can be applied to it. From this analysis, it is found that the concentration of SPM increases at 6 cm with increasing wind speed. As a result of this increasing of SPM the bed roughness also increases. This leads to the hypothesis that suspended sediment makes the bed rougher, and is not primarily governed by horizontal advection but also local resuspension.
A 1DV numerical model is used in which horizontal advection is excluded to test this hypothesis.
Simulations are performed with stationary boundary conditions, using combinations of water depth (0.2 to 2.8 m) and velocities (5 cm/s to 70 cm/s). These simulations are imposed with an initial homogeneous concentration. For every combination, the concentration is increased gradually until the concentration profile becomes L-shaped.
Besides simulations with stationary conditions, timeseries of water depth and velocity are used to simulate one tidal cycle.
All simulations performed with this numerical model do not take wind and waves into account and water-bed exchange is excluded.
From the simulations with stationary boundary conditions the roughness is calculated using the LP method. It is found that bed roughness increases with increasing initial homogeneous concentration.
From the simulation of one tidal cycle, it is found that the roughness increases towards the turn of the tide. After the turning of the tide, the concentration profile becomes L-shaped, and the roughness is decreased. The concentration profile becomes homegeneous again after a certain threshold of velocity and waterdepth and at the same time the roughness increases again. During a tidal cycle, it is possible to have a collapsed concentration profile, which indicates a lower bed roughness and thus a smoother bed.