Bed-load sediment transport in flows with emergent vegetation

• Main Author: Wu, Haoliang
• Other Authors: Chiew Yee Meng
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Civil engineering::Water resources
• Online Access: https://hdl.handle.net/10356/152967
• Institution: Nanyang Technological University

Flows in regions with vegetation such as wetlands have received much attention in the past two decades because the ecological values of aquatic vegetation have been increasingly identified. In the perspective of hydrodynamics, current researches have mainly focused on evaluating the impact of vegetation on the flow conditions rather than the complete flow-wetland feedback system. Previous studies proposed methods for predicting drag and turbulence of vegetated flows over immobile beds. However, for mobile-bed condition, a model for evaluating the fluid and solid fluxes has not yet been established, mainly because of the complex vegetation-water-sediment interactions. To simplify the investigation of these interactions, this study considers only bed-load sediment transport in equilibrium through emergent vegetation. By analysing the present measurements and also those from other relevant studies, a model is proposed to evaluate the weak and moderate bed-load transport rates in flows with emergent vegetation. There are two major factors that affect bed-load transport in vegetated flows, namely the spatial variability and temporal variability (due to turbulence) of the flow. First, to investigate the influence of the spatial variability induced by vegetation, two series of laboratory experiments were conducted: (1) measurements of the bed-load transport rates through the whole vegetation patch (patch-scale) in uniform flows through different vegetation models, and (2) flume tests with marked bed surface to illustrate the spatial variability of bed-load sediment fluxes. During the first series of experiments, the vertical velocity profiles between the bed and mid-depth level were measured by using an Acoustic Doppler Velocimeter (ADV). The vertical velocity profiles, which are similar to the previously published results obtained in vegetated flows with fixed beds, have significant spatial variability in both the streamwise and lateral directions. Thus, if the flow velocity at a specific location is used to predict the patch-scale bed-load transport rate, the error would be substantial because the selection of the measurement location itself could have introduced errors to the flow velocity. This study uses bulk flow velocity to evaluate bed-load transport rates, but without omitting the need to consider the influence of spatial variability. This is achieved by conducting the second series of experiments, which are designed to characterize the spatial variability of sediment fluxes in the vegetated flows. The observed results clearly show the spatial distributions of scour holes, deposited dunes, and non-sediment-exchange areas, indicating that the spatial variability in sediment fluxes is a crucial factor in evaluating the bed-load transport rate. Finally, a bed-load model is proposed based on a simplified spatial distribution of sediment fluxes. With this approach, only the bed-load transport across the scour hole around a vegetation stem caused by the local turbulent flow needs to be evaluated. Despite that the area of investigation is restricted to a limited region around the stem, it is still difficult to evaluate the average bed-load transport rate through the stem-scale flows related to an individual scour hole. To predict the bed-load transport rates affected by the vegetation stems, two approaches are applied. The first approach is proposed based on a dimensionless mean flow intensity, which contains the spatially averaged turbulent intensity. The second approach involves a probabilistic bed-load model, which is obtained by using a modified Einstein’s bed-load formula to flows with vegetation. Undetermined coefficients in the two approaches are calibrated by fitting the measured experimental results. Finally, the validity of the predictive formula is verified using the compiled experimental data, including those from previous studies. In summary, by analysing the observed phenomena and experimental results through the foregoing methods, the spatial variability is considered while predicting the patch-scale bed-load transport rate, and the influence of the turbulence around each vegetation stem is parameterized based on a probabilistic theory. As a result, a novel bed-load transport model is established to evaluate bed-load transport rates through different vegetation patches using the averaged flow velocity and turbulent intensity.