Abstract Flow‐vegetation interaction affects fluid flow hydraulics and associated material transport in river corridors. Concomitant changes in pressure within the flow field due to the presence of vegetation may act as a driver for the formation of hyporheic flow across the sediment‐water interface. This potentially important process, however, has yet to be studied. In order to investigate vegetation‐induced hyporheic exchange, a series of numerical models of interlinked surface‐subsurface flow modified by plant stems was conducted. Periodically staggered plant stem arrays on a flat sediment bed were considered within a coupled multiphysics computational fluid dynamics approach. Plants were idealized as rigid cylinders and arranged in different streamwise and spanwise spacing distances. Each vegetation array was then subjected to a broad range of flow Reynolds Numbers ( Re ). The results showed that hyporheic flow occurs in all conditions with the presence of vegetation. The vegetation‐induced hyporheic flux is found to be a function of Re via a power law. The flux increases with interstem space until the space reaches the distance that rigid stems no longer affect the flow structures in the vicinity of each other. Larger intervegetation distances lead to a larger hyporheic zone. A direct comparison with bedform‐induced hyporheic flow showed that vegetation can induce higher hyporheic flux through relatively shallower exchange zones. The results of all the simulations were synthesized into predictive models for hyporheic flux, bulk residence time and exchange depth based on drag coefficient, vegetation density, and Reynolds Number.
Physics-based models (PBMs), such as shallow water equations (SWEs) solvers, have been widely used in flood simulation and river hydraulics analysis. However, they are usually computationally expensive and unsuitable for parameter optimizations that need many runs. An alternative is the machine learning (ML) method, which can be used to construct computationally efficient surrogates for PBMs that can approximate their input-output dynamics. Among many ML techniques, convolutional neural network (CNN) is a prevalent method for image-to-image regressions on structured or regular meshes (e.g., mapping from the boundary conditions to flow solutions of SWEs). However, CNN-based methods have significant limitations because of their raster-image nature. Such methods cannot precisely capture the boundary geometry of obstacles and near-field flow features, which are of paramount importance to fluid dynamics. We introduced an efficient, accurate, and flexible neural network (NN) surrogate model [which is based on deep learning and can make point-to-point (p2p) predictions on unstructured meshes] called NN-p2p. The new method was evaluated and compared against CNN-based methods. NN-p2p improves the accuracy of the near-field flow prediction with a mean relative error of 0.56% for the velocity magnitude around piers with unseen length/width ratios. It also respects conservation laws more strictly than the CNN-based models and performs reasonably well for spatial extrapolation. The surrogate reduces computing time by almost 3-orders of magnitude in comparison with its corresponding PBM. Moreover, as a demonstration of the NN-p2p model's practical applicability, we calculated drag coefficient using NN-p2p for piers of varying length-to-width ratios and obtained a novel linear relationship between the drag coefficient and the logarithmic transformation of the pier's geometry.
Understanding the complex flow and sediment transport on vegetated slopes is important for ecological restoration and conservation projects. This study quantifies the erodibility of sand infill through densely vegetated engineered turf on steep slopes. Flume testing was conducted on four different sand infill materials. The initially lain bed material had artificially high mobility due to the infill application method. Grains were elevated by the vegetation and protruded into the flow. Then, the bed material gradation during subsequent flows became progressively coarser. Two regimes were identified. Poorly sorted infill soils underwent noticeable changes to gradation and had decreasing mobility with increasing shear stress. Conversely, well-sorted soils had minimal changes to gradation and resulted in the expected trend of increasing sediment flux with increasing shear stress. Existing predictive formulas performed poorly, in particular for the soils with evolving gradation. An updated formulation to predict sediment flux is proposed based on a reduction to the effective bed shear stress and dimensionless parameters relating to the flow, sediment, and vegetation characteristics. The proposed modification results in greatly improved predictions for both sediment flux magnitude and trend.
Abstract Hyporheic exchange induced by riverbed topography and roughness provides important ecosystem functions. We investigated how cobble clusters embedded in a finer‐grained sand bed affect the near‐bed channel flow and the exchange of surface and subsurface water. We tested how the spacing and embeddedness of cobbles altered hyporheic exchange through a three‐dimensional fully coupled surface‐subsurface model. The 3D modeling framework captured the full physics of the exchange process locating the lateral position of upwelling zones on the side of the cobble. As the cobbles protrude more into the channel, eddies appear downstream, with smaller vortices where cobbles are closer together. In our simulations, hyporheic exchange increased with the spacing and protrusion ratio. The travel time in the hyporheic zone also increased with the protrusion ratio because the hyporheic flow paths are longer when the cobble is more protruded. With different spacings at the highest protrusion ratio, the average slope of early time breakthrough curve was steeper for large spacing settings: the travel time increases as the spacing decreases.
Abstract Bedform‐driven hyporheic exchange is conditioned by the head gradients at the sediment‐water interface. Local exchange phenomena between the surface and the subsurface are often driven by the dynamic forces related to the velocity distributions around sediment waves. In open channels, the static forces are represented by the water depth, yet most computational fluid dynamics models use a rigid‐lid approximation. We investigated whether and when the deformation of the river’s free‐surface influences bedform‐driven hyporheic exchange. This was done through simulations of coupled open channel and hyporheic flows with the air‐water interface modeled either as a free‐surface or a rigid‐lid across increasing subcritical Froude numbers from ∼0.05 to ∼0.95. The normalized hyporheic flux was higher when considering free‐surface deformation across most of the range of Froude numbers considered. When the Froude number was larger than 0.6 and smaller than 0.85, both hydrostatic and nonhydrostatic‐driven fluxes increased significantly compared to the rigid‐lid approach and the total flux was about five times that predicted with the rigid‐lid. These results indicate that bedform‐driven hyporheic fluxes are underestimated in most studies that use a rigid‐lid assumption in subcritical flows.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
'/%3e%3cuse xlink:href='%23a' transform='rotate(144 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(216 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(288 450 300)'/%3e%3c/svg%3e)