Graduation of Noor ten Harmsen van der Beek

Mixing of salinity by ship traffic in canals

• Professor of graduation: Dr. ir. R.J. Labeur

• Supervisors of graduation: Prof. dr. J.D. Pietrzak (TU Delft), ir. O.M. Weiler (Deltares), ir. A.C. Bijlsma (Deltares), ir. A.J. van der Hout (TU Delft, Deltares

Salt intrusion poses a threat to the fresh water supply function of inland waterways. This threat is increasing due to climate change and deepening of channels. How far inland salt intrusion reaches is dependent on a balance of buoyancy forcing, water depth, discharge of the canal and the amount of mixing. Wind, bottom roughness, and sailing ships contribute to mixing. Currently little is known about the contribution of ship traffic to mixing. It is important to better understand the role of ship traffic in order to make more accurate models and take the right measures to protect the fresh water supply function of inland waterways. The aim of this study is to get insight in the amount of mixing ships generate and which processes are responsible for this. The final goal is to be able to implement mixing by ship traffic in large-scale numerical models.

The flow field in an unstratified canal has been well researched, however, less is known about the effects ships have in a stratified canal. Ships move water when sailing through a waterway. This induces three water movements: primary and secondary waves and the propeller jet flow. It is unknown how the return current will change for a stratified canal and how much shear due to the return current contributes to mixing. A ship sailing in a stratified canal generates secondary waves on the water surface as well as internal waves.  These waves could contribute to mixing by shear instabilities in the wave field or wave breaking on the side slopes. The high velocities in the propeller jet could entrain surrounding fluid and contribute to mixing.

A 3D non-hydrostatic finite element model is set up in order to study these processes in more detail.  A sailing ship in a stratified canal is modelled using FinLab. The propeller is neglected to simplify the model. It is expected to have a smaller role compared to the return current and internal waves. A parameter study is performed to observe the influence of several parameters (such as the canal blockage, internal Froude number, and side slope) on the flow pattern and the amount of mixing.

The results show that the interface goes down around the vessel and comes up again behind the vessel. Internal waves follow the vessel in a V-shape, comparable to a supercritical surface wave pattern. The increased velocities around the vessel cause shear at the ship hull and the bed. Internal wave instabilities and interaction with the slope cause additional mixing.  The amount of mixing is found to be in the order of magnitude of 1 percent over the modelled length with a slightly higher change of density at the top of the water column. The actual effect will be larger since the internal wave field is still present at the outflow boundary.

Processes around the vessel and the internal waves contribute about equally to the amount of mixing in the model domain. As the interface comes up behind the vessel, it is likely that the propeller jet can have a large impact on the density field directly behind the vessel. Due to the cumulative effect of the ship traffic, mixing by ship traffic is estimated to be large importance on the density distribution in a canal. More research is needed to include this effect in large-scale numerical models and find a good parameterisation. How mixing by ship traffic could finally be implemented will be dependent on the numerical model that is used, and the amount of detail that is needed.