| Summary: | Vegetation patches are frequently found in river channels. Besides being able to affect aquatic environments biologically and environmentally, they also impose a greater resistance to the channel flow. Although there have been many studies established on the drag induced by a single cylinder and a group of uniformly distributed cylindrical rods, the understanding of the drag induced by a porous vegetation patch remains unclear. A few studies have been conducted on this subject but the results are still preliminary. Hence, this study aims to investigate the drag induced by vegetation patches of varying porosities and the flow structure around these patches.
Two series of experiments were conducted to measure the drag induced by a rigid and emergent vegetation patch. The first series involved the use of a load cell to directly measure the drag acting on several simulated vegetation models that were subjected to a flow. Additionally, the Particle Image Velocimetry (PIV) technique was used to observe the flow structure around the models. In contrast, the second series involved an indirect measurement of the drag force by using the analogy between the drag imposed by a vegetation patch in an open channel and the drag imposed by a settling vegetation model in a tank. The terminal velocity of each settling model was computed from the time logger measurements and was used to obtain the drag coefficient.
A key finding is that the results from the two series of experiments could be unified to establish a single relationship between the drag coefficient and the Reynolds number. The implications of the assumption of sheltering effect and the use of different types of velocities to define the average pore velocity were also explored and analysed. From the PIV results, it is found that the vegetation models with a higher porosity display an individual wake behaviour while the models with a low porosity exhibit a group wake behaviour. However, a prominent finding is that for a specific model (Model C025), the wake behaviour varies with the model’s orientation relative to the flow direction. It is also established that the porosity of a model can affect the flow structure around the model, such as the magnitude of the velocities near the model.
Further studies can be conducted on other patch densities that were not tested in the present study. In addition, the assumption of sheltering effect could be investigated in greater depth, such as determining the circumstances for sheltering effect to be present in a vegetation patch. The effects of using submerged and flexible vegetation patches on the drag induced as compared to the use of emergent and rigid vegetation patches could also be explored further. The results obtained could be beneficial for the modelling of flood routing in rivers with vegetation.
|
|---|