Wenxin Huai

Bed-load transport capacity in vegetated open-channel flow is critical for ecological river management. However, accurate prediction of bed-load transport under submerged vegetation conditions remains challenging due to difficulties in determining bed shear stress. Drawing on the phenomenological theory of turbulence, this study derived an equation for bed shear stress for flows with submerged vegetation and integrated it with the bed-load transport formula originally developed for unvegetated flows to establish a predictive model applicable to both rigid and flexible submerged vegetation. The model involves the deflection height of flexible vegetation, which has conventionally been calculated via iteration and thus impairs model practicality. This study innovatively employs genetic programing to develop an explicit predictor for deflected height, whose integration into the bed-load transport framework markedly enhances the model's applicability. The proposed bed-load transport model elucidates distinct response regimes. Under a constant energy slope, the bed-load transport rate increases with submergence ratio but decreases with vegetation density, whereas these trends are reversed under constant flow velocity. Additionally, under identical hydraulic conditions, bed-load transport rate is lower in flexible vegetation than in rigid vegetation, indicating that flexible vegetation features better performance in controlling bed erosion. The model thus provides a robust tool for sediment management in river restoration projects.

Floating vegetation significantly alters the hydrodynamic structure of rivers and wetlands, yet its influence on air–water surface gas transfer remains poorly understood. Conventional prediction models focus on flow without vegetation and rely predominantly on bulk flow velocity and largely neglect additional turbulence caused by aquatic vegetation, leading to limited use in ecological water bodies. This study investigates surface gas transfer rates in open channel flows with floating vegetation through flume experiments conducted under varying cross-sectional averaged velocities, vegetation densities, and vegetation heights, and provides several new fluid physics insights. First, velocity shear at the vegetation–gap interface induces Kelvin–Helmholtz instability, generating large-scale coherent vortices, and constitutes the primary mechanism driving surface element renewal. Second, the surface gas transfer rate does not change regularly with any individual factor among vegetation density, vegetation height, and bulk flow velocity. Instead, variations in flow and vegetation conditions primarily modulate the turbulent velocity scale within the mixing layer, which exhibits a strong positive correlation with the surface gas transfer rate. Based on these findings, a modified surface renewal model is developed for flow with floating vegetation by incorporating the turbulent diffusion coefficient and characteristic scale of vortices in the mixing layer in place of the molecular diffusion coefficient and cross-sectional averaged velocity. The proposed model achieves a coefficient of determination of R2 = 0.94 across all experimental cases, demonstrating high accuracy. These findings provide mechanistic insights into evaluating greenhouse gas emissions and reaeration processes in natural aquatic systems and wetlands dominated by floating vegetation worldwide.

Accurately determining flexible vegetation resistance remains challenging due to vegetation bending and resistance reconfiguration under water flow. This study conducted laboratory experiments to explore the bending and resistance reconfiguration for single submerged flexible vegetation under water flow, and corresponding factors influencing resistance have been analyzed and quantified. Four diameters and five heights of vegetation were tested to observe the changes in bending posture of vegetation under a wide velocity range (0.05–1.85 m/s), with the resistance including drag force and skin friction measured. Results indicate that resistance changes from quadratic law to linear relationship with velocity as both the Cauchy number and vegetation bending angle increase, attributed to streamlined structure and resistance reconfiguration. The skin friction acted on flexible vegetation became significant under both low and high flow velocities, and thus should be considered in related analysis. Furthermore, the calculated drag coefficient accounting for micro‐element bending angle presents a trend of first decreasing and then increasing with the Reynolds number and Cauchy number. The drag coefficient factor and effective length combined with the Cauchy number were used to compare drag force between flexible and rigid vegetation and explain the resistance reconfiguration effects. Finally, a new empirical formula integrating the Reynolds number, Cauchy number and bending angle was established. The new formula outperforms traditional formulas based solely on the Reynolds number in predicting the drag coefficient for both high and low ranges. This study provides valuable references for flexible vegetation‐flow interaction mechanisms and improves resistance calculations in vegetated open channels.

Partially spanning vegetated leaky barriers (VLBs) composed of large vegetative debris have become an effective solution for channel rehabilitation, stream stabilization, erosion control, and habitat restoration. Understanding the fluid mechanics of flow around VLBs is crucial for optimizing their design, accurately predicting their hydrodynamic effects, and evaluating their ecological functionality. This study employs an experimentally validated large eddy simulation approach to investigate the flow structures generated by VLBs with varying blockage ratios (0.1–0.5). Results demonstrate that, compared to a solid barrier, VLBs can suppress flow separation, attenuate turbulence, and reduce local scour. Through internal gap flows, VLBs inhibit the development of recirculation zones and lower turbulence intensity, reducing the recirculation zone length by 91% and decreasing the domain-averaged turbulent kinetic energy by approximately 64% relative to solid barriers. The power spectral density of lateral velocity fluctuations in the horizontal shear layer is substantially lower in VLBs. The cross-sectionally averaged turbulent kinetic energy decays more rapidly along the streamwise direction in VLBs. By distributing flow through their permeable structure, VLBs homogenize bed shear stress distribution and lower peak stress values. This work advances the understanding of fluid–barrier dynamics controlling leaky barrier efficacy, thereby informing the design of multi-functional structures that balance hydraulic engineering requirements with ecological restoration targets.

Understanding the drift dynamics of fish eggs provides a useful basis for estimating early fish resource abundance, yet existing studies rarely account simultaneously for submerged vegetation, egg water hardening, and deposition–resuspension dynamics. We developed an Euler–Lagrange framework that couples a vegetated flow field with an egg random displacement model, an egg settling velocity model, and time-dependent egg water hardening to simulate grass carp egg transport in submerged vegetated channels. The model reproduced a 3-stage evolution of the suspension fraction, with an early decline, a subsequent recovery, and final stabilization. Ignoring the postfertilization decrease in settling velocity reduced the predicted steady suspension fraction by about 10%, whereas neglecting deposition and resuspension overestimated the centroid drift distance by about 116.8 m after 4 h. Within the investigated parameter range of a submergence ratio of 2.0 to 4.8 and a vegetation density of 0.012 to 0.084, higher vegetation density lowered suspension fraction and weakened sustained downstream transport, whereas larger submergence ratio increased suspension fraction, raised the minimum suspension level, and promoted longer centroid drift distance. These findings clarify how submerged vegetation regulates egg retention and transport and provide a basis for improving early-stage egg resource estimation and evaluating ecological sustainability in vegetated rivers.

Leaky barriers serve as a soft-engineering approach for river management and have become indispensable for river restoration and bank stabilization. In this study, we idealized multiple full-span leaky barriers as horizontally arranged cylindrical arrays with uniform diameters (d) and equal spacing (0.5d). Flow field visualization was conducted using tracer techniques, and particle image velocimetry was employed to quantify the effects of blockage ratio on wake structures behind these barriers. Results demonstrate that flow deflection induced by barrier obstruction generates a primary jet through basal gaps. The maximum velocity of the primary jet (u¯max) increases with the blockage ratio (ψ), reaching 2.82–4.43 times the average velocity of the incoming flow (U0). The primary jet retains initial velocity at x/d≤0.5, decaying linearly within 0.5<x/d<2.4. Within 2.4<x/d<8, the decay rate diminishes due to jet merging and weakened adverse pressure gradient. Multiple secondary jets emerge within inter-cylinder gaps and are deflected by the low-pressure zones downstream of cylinders. These jets coalesce before merging with the primary jet. Coherent vortex bands with elevated vorticity develop near wake shear layers, significantly enhancing turbulent kinetic energy in these regions. Jet interactions modulate velocity decay patterns in both primary and secondary jets. This work reveals fluid-barrier interaction mechanisms, advances hydrodynamic understanding of leaky barriers, and evaluates their ecohydraulic implications for riverine ecosystems.

Numerous hydrodynamic processes—such as friction loss and bridge pier scour—depend on identifying dimensionless parameters that quantify system behavior. The Buckingham theorem offers a principled approach to reducing the number of variables by identifying essential dimensionless groups that govern these processes. While machine learning (ML) algorithms are powerful tools for uncovering complex relationships among variables, they often face limitations in interpretability and physical consistency. In this study, we argue that feature engineering—particularly dimensionality reduction—should precede ML model development to enhance the discovery of physically meaningful hydraulic relationships. We demonstrate this approach using Principal Component Analysis (PCA) applied to two canonical problems: estimating the turbulent friction factor and predicting equilibrium scour depth. Specifically, we integrate PCA with genetic programming and artificial neural networks to derive predictive equations. Our results show that PCA preprocessing improves prediction accuracy, reduces training time, and yields simpler, more interpretable equations. As a practical contribution, we present two concise predictive expressions: one for the friction factor in vegetated open‐channel flows and another for scour depth in turbulent channel flows. These findings highlight the value of PCA‐enhanced ML frameworks for advancing physically informed modeling in hydraulic and sediment transport systems.