Numerical Investigation of Straight Longitudinal Training Walls on the River Bank Protection

Introduction: River training is the stabilization of the channel in order to maintain the desired cross-section and alignment. Channel stabilization and restoration efforts have been increased dramatically with excessive cost since 1990. However, it is estimated that at least 50% of these projects fail and others may not perform as expected. Therefore, stream restoration today is more of an art than a science. Rivers have been trained for centuries by series of transverse groynes which are structures constructed at an angle to the flow in order to deflect the flowing water away from critical zones. This generally results in damages to their ecosystems as well as in undesirable long-term morphological developments. To maintain a navigable channel and improve flow conveyance by dividing the channel into main and side channels, engineers have recently proposed constructing longitudinal training walls as an alternative to the traditional transverse groynes. The effectiveness of longitudinal training walls in achieving these goals and their long-term effects on the river morphology have not been thoroughly investigated yet. In particular, studies that assess the bed and wall shear stresses of the parallel channels separated by the training walls are still lacking.
Methodology: In the present study, the performance of longitudinal training walls on bed and wall shear stresses was compared with the traditional transverse groynes. Three-dimensional Reynolds-averaged Navier-Stokes equations with the k-ɛ turbulence model were solved numerically by applying finite volume method using Flow-3D software. Results of the single groyne simulation were validated based on experimental data with a good agreement. The experimental channel had rectangular cross-section of width B=0.9144m and slope S0=10−4, groyne was a parallelepiped of length b=B/6=0.1524m and thickness of 3 mm. In addition, results of the series of groynes simulation were compared to the available numerical data, shown a good agreement. After validation, continuous longitudinal training walls were simulated with equivalent length to a series of groynes with uniform and non-uniform configuration in three transverse positions y/B=1/2, y/B=1/3 and y/B=1/6. Uniform configuration of groynes was included three groynes with D/b=6 and 23 spacing and non-uniform configuration was included nine groynes, first five groynes with D/b=1.5 and four other groynes with D/b=6 spacing. Also, non-continuous longitudinal training walls were simulated with the same transverse positions for D/b=23. The hydrodynamic behavior of the bed and wall shear stresses was investigated, since the starting point of longitudinal training walls affects the morphodynamic behavior of the flow, the starting position of the walls was fixed and only different transverse positions with different lengths were simulated.
Results and Discussion: Here we analyze the bed and wall shear stresses in open channel flow by training channel in a new way by subdividing their channel in parallel channels with specific functions with longitudinal training walls. The results showed that the longitudinal training walls system reduces the bed shear stress compared to the groyne system and as a result we will have less erosion in the bed. However, some small increases in wall shear stresses were seen in all simulations of longitudinal training walls. The closer the transverse position of the longitudinal training walls to the channel wall, the lower wall shear stresses. Furthermore, if the longitudinal training wall system is implemented in non-continuous form, due to the drop in flow energy at the beginning and end of each wall, the flow energy drop leads to a decrease in velocity and due to the direct relationship between velocity changes and shear stress, bed and wall shear stresses in side channel can be reduced further, that this reduction depends on the transverse position and, more importantly, whether or not this transverse position has created a mild flow conditions in the side channel. Therefore, this system by providing a mild flow conditions in the side channel which is shallower channel, has a favorable effect on aquatic habitat, ecosystem, and wall shear stress. Besides, by providing a deep navigable main channel, can increase the ability of transferring additional flood discharge.
Conclusion: In this study, twelve simulations were performed with different lengths for longitudinal training walls in different transverse positions with continuous and non-continuous forms. The results showed that this system, compared to the traditional groyne system, introduces less bed shear stress and by providing a mild flow conditions in the side channel, has the ability to control the wall shear stress dynamically. This system has a favorable effect on aquatic habitat and ecosystem, beside high ability to transfer additional flood discharge. So, longitudinal training walls can be used as an appropriate replacement for traditional transverse groynes.