Journal of Hydraulic Engineering

The durability of superhydrophobic surfaces in fully immersed conditions is a major obstacle to their application in engineering applications. We perform an experimental study to measure the friction factor $f_d$ as a function of time for a new superhydrophobic surface that is capable of recovering the Cassie-Baxter wetting state. Values of $f_d$ were obtained by measuring the pressure drop and volume flux of a turbulent water flow in a 1.5 m long duct containing one superhydrophobic wall. The Reynolds number of the flow was approximately $4.5\times {10}^4$ for all experiments. Reductions in $f_d$ were 29%–36% relative to a hydraulically smooth surface. The Cassie-Baxter state could be recovered by blowing air through the porous surface for 10 min. The durability of the drag-reduction, as quantified by the relaxation time $T$ in which the surface loses its superhydrophobic characteristics, were measured to be between 10 and 60 min depending on the initial head. The relaxation time $T$ was highly dependent on the pressure difference across the surface. In contrast to models based on Darcy flow through a porous medium, the study indicates that there seems to be a critical pressure difference beyond which the Cassie-Baxter state cannot be sustained for the material under consideration.

The sediment transport in the seabed boundary layer due to wave-induced streaming has been investigated for nonzero angle between the current and the waves. The mean sediment transport is a result of interaction between streaming due to wave skewness, the Longuet-Higgins streaming, and the wave–current interaction. For colinear waves, the mean sediment transport is directed along the wave propagation direction, with the largest transport taking place beneath the waves propagating in the direction of the current. Mean sediment transport decreases with increasing angle between the current and the waves. For a particular angle, the mean sediment transport beneath second-order Stokes waves is largest and decreases in the following order: linear propagating waves followed by horizontally uniform Stokes forcing and smallest for horizontally uniform linear forcing. The direction of the mean sediment transport is rotated toward the current away from the direction in which the wave propagates. The magnitude of rotation is smallest for second-order Stokes waves and increases in the following order; linear propagating waves followed by horizontally uniform Stokes forcing, and largest for horizontally uniform linear forcing.

A hydraulic jump in a stilling basin at the base of spillway is typically characterized by preaerated inflow conditions. This paper presents an experimental study of high-Froude-number hydraulic jumps with different preaeration levels. Two characteristic lengths of the jump, namely the jump length and roller length, which are defined in various studies but sometimes not clearly distinguished, are obtained from the same flows by means of free-surface and air-water flow velocity measurements, respectively. The effects of changes in preaeration level are found to be opposite on the jump and roller lengths. Enhanced preaeration allows for a shorter length of the hydraulic jump, and the jump roller is elongated to achieve the same energy-dissipation rate in a shorter distance. The results imply a predominant effect of energy-dissipation enhancement over reductions associated with the extra air–water mixing and bubble–turbulence interplay. Tailwater wave characteristics are presented, showing similar frequencies for the far-field wave fluctuations and upstream jump toe oscillations on the two ends of the jump.

Fluid transient phenomena involving pressure wave propagation have often been studied and solved with the method of characteristics. Only recently has the finite-volume method (FVM) been proposed and implemented to solve the transient fluid flows for a one-dimensional water-hammer–based analysis. The use of the FVM permits the introduction of new solution algorithms and, at the same time, deals with more general conditions, including multiphase flow and cavitation. The research presented in this paper investigates improvements to the solution methods for one-dimensional flow simulation with compressibility and multiphase liquid-gas flows induced by cavitation in which the gas phase consists of two distinct components: noncondensible gas and vapor. The effects of the second phase and the compressibility play an essential role in the density and, consequently, the speed of sound variation in the flow, and accounting for these provide a more accurate prediction of pressure wave propagation. The simulations carried out were second-order accurate in time and space by using the monotonic upwind scheme for conservative laws (MUSCL). The total variation diminishing (TVD) strategy was also implemented for stability reasons. To consider the second phase, a variation of the discrete gas and vapor cavity model was used. In conclusion, a comparison with experimental data, similar algorithm approaches, and the classical method of characteristics indicate a more effective approach for the simulation of pressure-wave propagation for compressible conditions.

Hydraulic jumps, frequently present in energy dissipation structures, have been thoroughly investigated over the past $70\mathrm{years}$. However, studies have focused primarily on the more common and more stable cases of large Froude number jumps. Investigations on low Froude number jumps, usually called transition jumps, are scarce, although they are used in a significant number of run-of-the-river, high-discharge, and low-head dams. This paper addresses the characterization of hydrodynamic pressures of low Froude number jumps. A two-dimensional experimental facility was used for that purpose. Results were compared with those from published studies covering transition jumps and stabilized jumps, to identify differences and similarities between them. The analysis considered common dimensionless statistical parameters, duly nondimensionalized. It was found that, for most of these parameters, low Froude number jumps behave differently from stabilized jumps. This can be attributed mainly to the hydraulic jump’s oscillating jet formed under the surface macroturbulent roller, as well as to the partially developed boundary layer of the incoming flow in low-head spillways.

Irrigation reservoirs are widely used in parts of the Mississippi River alluvial floodplain to supplement limited groundwater resources. Constructed using local soils that are often low in clay content, the earthen embankments of these reservoirs are susceptible to wave erosion. In this study, laboratory experiments were performed to quantify wave-induced erosion and retreat of cohesive embankments. Erosion rates were measured for a model erodible embankment in a laboratory wave tank for a range of wave characteristics, and relationships were established between volume of embankment erosion, edge retreat rate, and incident wave height. An analytical model was developed to predict the embankment erosion and retreat rate. The model was calibrated using laboratory data and then compared with field measurements of erosion and retreat in an irrigation reservoir. The results show agreement between the field measurements and model predictions of bank retreat.

In the literature, a classical free-surface inception point empirical relationship for smooth chutes serves as a basis for the development of a free-surface inception point relationship for stepped chute applications. This study expands research on the free-surface inception point for stepped chutes with multiple sections with different step heights along the chute bottom surface. Data show that existing free-surface inception point relationships for stepped chutes do not always accurately predict where the free-surface inception point will occur when assuming a single step height in stepped chutes designed with multiple sections with different step heights. Accounting for the different step height sections and assuming a constant rate of boundary layer development for each step height allows the error for predicting the free-surface inception point location to be reduced to $\pm 10\%$ for stepped chutes with two different step height sections. The research results contribute to improving knowledge on the developing free-surface inception point in stepped chutes, and it provides additional validation data for modeling these flows in computational fluid dynamics models.

It can be difficult to predict sediment transport in sand-bedded ephemeral streams because of rapidly changing conditions that may not be sustained long enough to reach equilibrium. This situation often leads to the inheritance of bed topography from previous flows, which may result in under- or overpredictions of transport rate due to bed forms and flow resistance that are not in equilibrium with the flow conditions. To address this difficulty, a series of lab experiments was used to examine changes in sediment transport rate and bed topography after rapid increases in flow depth and discharge. It was found that sand transport rates reached equilibrium conditions in as little as 1–15 min after the step-up in flow rate and depth. Transport rate equilibrium was likely reached in advance of centerline bed-form amplitude, although bed forms also responded quickly to step-up flows, and the response was much more rapid than in previous experiments that examined the effects of decreases in flow rate and depth.

There are numerous situations in nature and industrial flow in which particles are transported by a vortex-flowing fluid, such as near hydraulic structures such as bridge piers and abutments, spur dikes, vanes, intakes, root wads, and settling tanks. To explore vortex transport of particles, an experimental study and empirical model were performed in which particles were introduced into a vortex contained within a test section. The vortex was induced by tangential jets located in the test section just above the point of sediment introduction. Quantities varied in the experiments were particle diameter and vertical axial velocity. Dimensional analysis was performed and resulted in a relationship of normalized particle discharge as a function of particle Froude number. It was found that at low particle Froude numbers, there was no particle transport due to the absence of sufficient turbulence to entrain the particles. Also, above a critical value, particle transport increased quickly due to sufficient entraining turbulence and relatively high vertical velocity with little particle collision. However, at intermediate values, the transport decreased due to heavy particle–particle collisions. At high values, the transport increased exponentially again.