Turbulence structure of open channel flows over permeable and impermeable beds: A comparative study

Abstract

The behavior of turbulent open channel flows over permeable surfaces is not well understood. In particular, it is not clear how the surface and the subsurface flow within the permeable bed interact and influence each other. In order to clarify this issue we carried out two sets of experiments, one involving velocity measurements in open channel flows over an impermeable bed composed of a single layer of spheres, and another one where velocities were measured over and within a permeable bed made of five such layers. Comparison of surface flow velocity statistics between the two sets of experiments confirmed that bed permeability can significantly affect flow resistance. It was also confirmed that even in the hydraulically rough regime, the friction factors for the permeable bed increase with increasing Reynolds number. Such an increase in flow resistance implies a different distribution of normal form-induced stress between the permeable and impermeable bed cases. Subsurface flow measurements performed within the permeable bed revealed that there is an intense transport of turbulent kinetic energy ͑TKE͒ occurring from the surface to the subsurface flow. We provide evidence that the transport of TKE toward the lower bed levels is driven mainly by pressure fluctuations, whereas TKE transport due to turbulent velocity fluctuations is limited to a thinner layer placed in the upper part of the bed. It was also confirmed that the turbulence imposed by the surface flow gradually dissipates while penetrating within the porous medium. Dissipation occurs faster for the small scales than for the large ones, which instead are persistent, although weak, even at the lowest bed levels.

Turbulent boundary layers over permeable walls: scaling and near-wall structure

“…This makes C. Manes, D. Poggi and L. Ridolfi the results of the present study directly comparable with the literature pertaining to aquatic and atmospheric canopy flows (Ghisalberti & Nepf 2002;Poggi et al 2004;Finnigan et al 2009), and also with the DNS of Breugem et al (2006), but to a lesser extent with flows over permeable granular walls (Manes et al 2009;Sarkar & Dey 2010;Detert, Nikora & Jirka 2010). The reason is that the latter class of porous media is characterized by large roughness elements, which makes the effects of roughness and permeability both significant.…”

“…In the literature, this problem has been overcome by following two approaches. The first, particularly used in the study of flows over granular beds, involves a comparative analysis between flow statistics measured over rough impermeable and rough permeable walls characterized by the same surface roughness texture (Zagni & Smith 1976;Manes et al 2009). For example Manes et al (2009) present a comparison between statistics of velocities measured in open channel flows over an impermeable wall made of one layer of beads and a permeable wall made of five of such layers.…”

“…The first, particularly used in the study of flows over granular beds, involves a comparative analysis between flow statistics measured over rough impermeable and rough permeable walls characterized by the same surface roughness texture (Zagni & Smith 1976;Manes et al 2009). For example Manes et al (2009) present a comparison between statistics of velocities measured in open channel flows over an impermeable wall made of one layer of beads and a permeable wall made of five of such layers. The rationale behind this approach lies in the concept that (i) a wall is considered as permeable (impermeable) if its thickness is significantly larger (comparable or smaller) than the characteristic size of its constitutive components; (ii) the comparison between flow statistics measured over the two beds highlights the effects of wall permeability since the surface roughness is kept the same.…”

First-Order Contaminant Removal in the Hyporheic Zone of Streams: Physical Insights from a Simple Analytical Model

“…Turbulent eddies in the water column of a stream, for example, can cause flow across the sediment-water interface in the presence or absence of bed forms. 33 Third, plants and animals colonizing the hyporheic zone can exert profound impacts on stream-sediment exchange, by forming mounds across which pumping occurs, inducing pore water flow within sediments, and structuring the permeability field with burrows and roots (reviewed in 34 ). Fourth, at a larger scale geomorphic features of a stream such as riffle-pool sequences, debris dams, meander bends, and regional groundwater flow all influence hyporheic exchange.…”

Free surface flow over square bars at intermediate relative submergence

“…Nikora (2009) extended the use of the double-averaging methodology to develop a theoretical expression that explicitly showed that the friction factor can be accounted for by six additive components, of which only three are present in two-dimensional uniform spatially-averaged flow without secondary currents: viscous stress, turbulent stress and forminduced stress. The contribution of the form-induced stress, ũw , has recently attracted substantial attention as it has been shown to play an important role in the overall streamwise momentum balance in the near-bed region (Aberle, Koll, & Dittrich, 2008;Dey & Das, 2012;Ferreira et al, 2010;GimenezCurto and Corniero, 1996;Gimenez-Curto and Corniero Lera, 2003;Manes, Pokrajac, Coceal, & McEwan, 2008;Manes, Pokrajac, & McEwan, 2007;Manes, Pokrajac, McEwan, & Nikora, 2009;Mignot, Barthelemy, & Hurther, 2009a,b;Nikora et al, 2007b;Sarkar & Dey, 2010). Indeed, Gimenez-Curto and Corniero (1996) and Gimenez-Curto and Corniero Lera (2003) suggested that at very low submergences the form-induced stress could become the dominant component in the streamwise momentum balance.…”