Flow characteristics of rectangular open channels with compound vegetation roughness

An idealized parallel flow caused by a lateral bed roughness difference due to the partial vegetation across a channel is investigated. Similar to the flow in a compound channel, there are mixing layers adjacent to the interface between the vegetation and the non-vegetation lanes, and a lateral momentum exchange occurs between the slow- moving water in the former lane and the fast-moving water in the latter lane. Under a uniform flow condition, the three-dimensional with different discharges and water depths are （3D） instantaneous velocities of two cases measured with a 16 MHz acoustic Doppler velocimeter （micro ADV）. The longitudinal variation of the streamwise velocity and the vertical variation of the Reynolds stress are analyzed. A quadrant analysis is carried out to investigate the outward and inward interaction, ejection, and sweep phenomenon caused by the vegetation variation across the channel. The results show that the flow characteristics in the vegetation lane are similar to those in an open channel fully covered with submerged vegetation, and the flow characteristics in the smooth non-vegetation lane are similar to those in a free open channel. For the cases studied here, the width of the mixing region is about 10~ of the channel width, and the mixing region is mainly on the non-vegetation half.

Hydraulic calculation of steady uniform flows in trapezoidal compound open channels

Hydraulic calculation of steady uniform flows in trapezoidal compound open channels is studied. Based on the force balance of water in each sub-section, the average velocities of the main channel, side slope, and floodplain are derived. The lateral momentum exchanges between the sub-sections are expressed by using the apparent shear stress. To verify the model, seven groups of UK Flood Channel Facility （UK-FCF） measured data with a relative water depth between the floodplain and the main channel varying from 0.057 to 0.4 are used for comparison. The result shows that the calculated velocity is larger than the measured data when the relative water depth is small, while it is less than or close to the measured value in the case of a larger relative water depth. The influence of the apparent shear stress on the calculation of velocity on the floodplain is not obvious, while it is much greater on the main channel. The three-stage model is compared with Liu’s two-stage model, showing that the former can give a better prediction for a three-stage trapezoidal compound channel. Finally, the apparent shear stress is calculated and compared with the measured data. The result shows that the chosen values of the momentum transfer coefficients are appropriate.