Journal of Hydraulic Engineering

Forum papers are thought-provoking opinion pieces or essays founded in fact, sometimes containing speculation, on a civil engineering topic of general interest and relevance to the readership of the journal. The views expressed in this Forum article do not necessarily reflect the views of ASCE or the Editorial Board of the journal.

Streaky structures have been reported in open-channel flows by various investigations in the inner and outer layer. Although the ubiquity of streaks is well recognized, it remains unclear how the streaky structures in the inner and outer layers are related to each other. This paper employs a unified algorithm to identify streaky structures which can be applied in both inner and outer layers with exactly the same subjective threshold. The characteristic scales of streaky structures are quantified, including the width and the spanwise spacing. The results reveal that the spatial scales increase from the near-wall region to the outer layer. The continuous linear growth trend indicates a close relationship between the streaky structures in the inner and outer layers, which can be plausibly explained with the attached-eddy hypothesis.

Asymmetric turbulent channel flow, such as ice-covered river flow, is a century-old problem but still unsolved in hydraulics and fluid mechanics. This study finds exact solutions for its eddy (or turbulent) viscosity and mean velocity distributions, which are independent of any assumption without any fit parameter. Specifically, it first applies Guo’s quartic eddy viscosity and complete log-law from high-Reynolds-number pipe flow to symmetric turbulent channel flow. It then formulates a functional equation, involving both bottom and top plane shear velocities, to govern the eddy viscosity distribution in asymmetric channel flow. The analytic solution for the eddy viscosity then leads to a velocity distribution solution that includes four components: bottom shear velocity effect, top plane shear velocity effect, symmetric interaction between both about a critical point, and antisymmetric interaction between both. The velocity distribution solution agrees well with field data and so is applicable in ice-covered rivers. Laboratory data also confirm the velocity distribution structure, but a turbulent mixing intensity parameter depends on the Reynolds number. Therefore, future laboratory tests should focus on high-Reynolds-number flow.

In this paper, a three-dimensional (3D) finite-element model for flooding has been developed, which can be applied to different types of flooding. The theoretical methodology has been introduced in details. The new model has been tested with an idealized case and the Thacker test case, then validated by a tsunami-caused coastal flooding experiment. The advantages of the new flooding model developed in this paper over existing ones are (1) the use of unstructured meshes that makes it possible to obtain more detailed information about local flow structures close to isolated objects and long linear objects; (2) the 3D modeling ability that provides increased accuracy where vertical inertia is important; (3) the use of a wetting and drying (WD) method that allows the free-surface height to be treated with a high level of implicitness and stability, allowing relatively large time steps to be used; and (4) control of the conditioning of the resulting matrix equations in high-aspect ratio domains.

Uncontrolled air pocket release from water-filled shafts can lead to geysering in stormwater systems. Such occurrences are deleterious from public health and environmental standpoints and can cause property and structural damage. Causes, frequency, magnitude, and location of geysering events still are poorly understood and pose practical difficulties to designers regarding how to create dropshafts that are less likely to present this issue. This paper presents results from experimental and numerical investigations on air-related geysers that provide insight into the mechanisms of air release and the displacement of water in vertical shafts. A 302-mm schedule 40 clear PVC apparatus is constructed with the essential features of a stormwater tunnel, and is fitted with vertical shafts with diameters ranging from 0.10 to 0.20 m. Predetermined air pocket volumes are released in the horizontal pipe and eventually reach shafts, causing water displacement and often causing geysers. Kinematics of the air pocket release are measured along with pressures at selected points in the apparatus. These results are used in the calibration of a computational fluid dynamics (CFD) model, which compares well with the experimental measurements. The model subsequently is used in a larger geometry that allows the evaluation of air pocket release kinematics for a wider range of conditions. Findings of this work provide further details on the nature and strength of geysering events, and suggestions for future studies in this topic are also provided.

Eigenfrequency shift in a pipe with a blockage is investigated experimentally. The experimental test rig consists of a reservoir-pipe-valve (RPV) system that contains a single partial blockage. Blockages with different lengths are considered. The experiments confirm the existence of certain frequency bands where waves reflect strongly from the blockage and other bands where waves reflect weakly (Bragg-type resonance). These bands agree with the theoretically derived Bragg resonance condition. It is found that the values of the resonant frequencies are highly sensitive to the wave speed. For example, a 4% error in wave speed results in an error in the eigenfrequency estimates that increases almost linearly from approximately 4% for the first mode to 60% for the seventh mode. A direct and efficient approach is proposed for using the Bragg resonance condition to detect blockages in pipes.

In comparison to sand and gravel, cohesive soils, and clays in particular, present significant challenges when critical entrainment thresholds and erosion rates are to be determined for a particular basin or modeled in the laboratory. A wide variation in cohesive soil characteristics and behavior is usually encountered, even within relatively short reaches of river and creek beds. It is, therefore, evident that assessing the erodibility of such beds requires analysis of a large number of locations within reach, resulting in a costly and time-consuming undertaking. The objective of this paper is to introduce a practical procedure and test device designed to allow a fast and accurate determination of the critical shear stress of cohesive (and some noncohesive) soils, both in laboratory samples and in the field. The device, herein referred to as the erosionometer, is based on physical shearing of the soil surface. Initial calibration of the device showed a very good correlation with critical shear stress values for particle entrainment measured in a laboratory flume.

Results are presented for a novel experimental investigation of live-bed scour at a cylindrical pier in covered flow. The experimental campaign also comprehended a reference scour test in clear-water conditions. Results showed that the time to achieve a dynamic equilibrium was a decreasing function of the flow rate. The same was found for the mean depth around which the scour fluctuated after equilibrium. The bed elevation was simultaneously measured within the scour hole and upstream of it to investigate the bed-form control of the scour fluctuations. Taking advantage of a high temporal detail used for the measurements, several statistical tools (cumulative frequency distributions, autocorrelation functions and spectra) were employed to characterize the properties of bed form and scour fluctuations. Amplitudes of the scour fluctuations were similar to those of incoming bed forms. On the contrary, a difference in the temporal scales for bed forms and scour fluctuations was detected and explained recognizing a mechanism of bed-form coalescence in the scour hole. The presented results enabled a phenomenological depiction of the process to be obtained, and are significant for analysis of scour data and also for engineering purposes.

The classical Joukowsky formula for the head jump associated with a sudden flow stoppage is an engineering rule of thumb that helps estimate peak surge pressures in pipelines. However, the Joukowsky formula does not take account of friction. Although this is usually of little concern for short pipelines, the effects of pipe friction can cause significant deviations for long pipelines and for high frictional flows, where rarefaction, line packing, and attenuation phenomena can be significant. This paper describes some extensions of the classical Joukowsky formula that take frictional effects into account. By using suitable nondimensional variables based on the system parameters, a single universal formulation of the underlying water hammer equations is developed and some simple analytic approximations are derived. Both instantaneous and finite-duration stoppages of flow are considered. For an instantaneous stoppage, a first-order analysis shows that the head behind the surge front is uniform in space but continues to fall with time at a nondimensional rate of 0.5. A finite shutdown can usefully be thought of as being intermediate between two instantaneous shutdown cases, one occurring at the start of the finite shutdown period, and the other at the finish. The intuitive notion that a finite shutdown period has a moderating effect on the subsequent development of rarefaction is shown to be incorrect for long pipelines.