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.

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.

The Storm Water Management Model 5.1 (SWMM) is a widely adopted dynamic hydrologic and hydraulic model often used to estimate runoff quantity and quality in urban drainage systems. SWMM’s unsteady flow algorithm, EXTRAN, is based on a link-node solution that enables it to represent typical stormwater inflows well. Yet, for rapid inflow conditions associated with more extreme inflows, predictions yielded by SWMM underestimate surges and sometimes underrepresent sudden changes sewer flow conditions. Recent research showed the benefits of introducing artificial spatial discretization (ASD) models in SWMM to represent rapid inflows in sewers. However, with the recent addition of the Preissmann slot algorithm in the SWMM formulation, a systematic evaluation of the performance of this pressurization algorithm in complex and highly dynamic inflow scenarios is still missing. The present study applied the conditions presented in the Storm Water Management Model Quality Assurance Report and compared the use of either link-node or ASD along the original EXTRAN and the new Preissmann slot algorithm in modeling results. The performance of each of the selected modeling alternatives was evaluated in terms of continuity errors and numerical stability. The findings obtained in this study indicate modeling result improvements with an adequate selection of temporal and spatial discretization.

Air expulsion from an end-of-pipe orifice in a rapidly filling horizontal pipe is investigated experimentally and analytically in order to more completely characterize the system’s transient response. In particular, images of air–water patterns, air-volume variations, orifice flow regimes, and measured pressure histories are synchronized to elucidate the process of air expulsion. Air expulsion typically undergoes an early stage involving pressurization, expansion, and release of a portion of the initial air, events that generally occur even before the advancing water column reaches the pipe end. The next stage depends strongly on the orifice size. For a small discharge orifice, an oscillation of the residual air occurs with the discharge orifice being intermittently choked by water; by contrast, larger discharge orifices rapidly and completely expelled the air, often leading to high water-hammer pressures. Three distinct patterns of pressure oscillation are typically observed. With small orifices, the cushioning effect of the initial air tends to dominate, whereas slightly larger orifices lead to a more complex process of expulsion and more persistent and larger pressure oscillations. Even larger orifices often lead to severe water-hammer pressures. Thus, smaller orifices tend to result in smaller pressure fluctuations. As expected, both the initial-air volume and the inlet pressure significantly influence the transient response. A derived analytical model accurately captures the measured pressure oscillations during the intermittent release of residual air, including the water hammer that can arise due to suddenly arresting the liquid water column.

Acoustic Doppler velocimeters (ADVs) are used extensively in various field and laboratory studies of hydraulic engineering. However, their accuracy in predicting statistics of turbulence quantities has been questioned. Two fundamental limitations of this type of velocimeter are Doppler noise and the damping of fluctuations due to the temporal averaging performed by the instrument. An important factor that may affect both error sources is the sampling frequency of the ADV. An experimental investigation of the effect of the ADV sampling frequency on the measurement of both the mean and the high-order statistics of the flow in a turbulent jet was conducted. The experiments were carried out in the self-similar zone of an axisymmetric nonbuoyant jet at a Reynolds number of 10,000 released into quiescent water. Measurements of the mean and RMS velocities, spectra, and Reynolds shear stresses at different sampling frequencies are presented. Results were compared with those of other measurement techniques and interpreted using a novel analytical model quantifying the noise and the damping effect on the basis of their nonvanishing statistical correlation in the postaveraging domain, as well as the ratio of the flow’s integral timescale to the sampling interval. The damping effect at high sampling frequencies was eliminated using a hypothesis of proportionality of a relative change in the correlation coefficient to a change in the noise variance, provided that the integral timescale is adequately larger than the sampling interval. The proposed precision-enhancement technique (referred to herein as denoising and reverse-damping transformation) was shown to improve the accuracy of velocity variances. The results and model offer an opportunity to improve the precision of ADV measurements in turbulent flows.

In this paper, the mixing and dilution characteristics of vertical buoyant jets discharged from multiport diffusers are studied numerically. The performances of four different turbulence closures are investigated, including the standard $k\text{-}\epsilon$, renormalization group (RNG) $k\text{-}\epsilon$, standard $k\text{-}\omega$, and $k\text{-}\omega$ shear stress transport (SST) models, and the simulated results are compared to available experimental measurements. The comparisons demonstrate that a fully three-dimensional (3D) numerical model can be a reliable tool for the study of multiple vertical buoyant jets. Different fit measurement methods are employed to evaluate the turbulence models, and the results show that the predictions by the RNG $k\text{-}\epsilon$ model are the most accurate. The validated model is utilized to carry out additional computations with different port spacings and densimetric Froude numbers ($F_r$), and the simulated centerline and transverse concentrations at various cross sections are extracted and analyzed. The study reveals how varying port spacing and $F_r$ affect multiple vertical buoyant jets. A new empirical formula that considers the port-spacing effect is proposed to describe the concentration decay along the jet centerline. The study also provides clues about merging points, well-mixed locations, and jet spread under the influences of port spacing.

There has been little study on the microscopic bubble entrainment and diffusion process on high-speed self-aerated flows, although the problem under investigation is theoretically important and has important engineering applications. This study presents an experimental investigation on visual processes of free surface air entrainment and single-bubble diffusion in supercritical open-channel flows. The typical surface deformation, single air bubble rising, and penetration are recorded using a high-speed camera system. Results show that for a single-bubble formation process, surface entrapment development and bubble entrainment through a deformation evolution underneath the free surface are the two main features. The shape variation of local surface deformation with time follows an identical power law for different bubble size generations. The entrained bubble size depends on both the size scale and the shape of the entrapped free surface. As the single bubble moves downstream, its longitudinal velocity is approximately the same as that of the water flow surrounding it, while its vertical velocity for rising and penetration increases with the increase in water flow velocity. An empirical-linear relationship for the bubble rising and penetration velocity with water flow velocity is obtained. This study demonstrates that microscopic bubble movement can improve the self-aeration prediction in open-channel flow and advance our understanding of the macroscopic and microscopic air–water properties in hydraulic engineering.

Pipe leakage and partial blockage are the two common types of pipe defects in water supply systems. In practical systems, these defects, which may cause great waste of water and energy resources and can induce potential reduction of system operation capacity, are usually formed simultaneously. This paper develops an effective transient frequency response (TFR)–based method to characterize and detect simultaneously these two types of pipe defects. To this end, an analytical expression is first derived to characterize the simultaneous effect of leakage and partial blockage on TFR in water pipelines. The obtained results are then validated through laboratory experimental tests for the detection of multiple pipe defects. After validation, extensive numerical applications that cover typical ranges of test conditions are adopted to further investigate the applicability and accuracy of the developed TFR-based method. The application results demonstrate high accuracy and wide applicability of this TFR-based method to detect simultaneously potential leakage and partial blockage in water supply pipelines. The advantages and limitations of this TFR-based method are discussed in the final section for practical applications in urban water supply systems.

The morphology of the Scheldt estuary creates challenges for its different functions. Large shoals and adjacent channels with sharp bends result in challenging conditions for navigation. In order to provide estuarine managers with adequate answers, different research tools are necessary: numerical models, physical scale models, monitoring, and system understanding and expertise. This paper describes the validation of two numerical models, focusing on the Hansweert area, which is characterized by complex flow fields and important dredging works. In 2016, velocity measurements were executed over two spring-neap cycles using Acoustic Doppler Profiler devices. These velocity measurements are used for additional validation of two existing numerical models, describing the entire Scheldt estuary. Both models have shown good performance in large-scale tidal hydrodynamics, but uncertainty increases for local flow patterns, especially in shallow and intertidal areas. Adapting the local friction field improved significantly the representation of the measured complex flow patterns (e.g., large eddies) in the study area without impacting the overall model performance. This model allows both the improvement of our system understanding, e.g., formation of the large eddy, as well as the generation of flow patterns that are used in a ship simulator, where river pilots train to prepare for the challenging flow conditions in the field.

In this note, a two-dimensional depth-averaged numerical model (MassFlow2D) previously developed to simulate subaerial and subaqueous muddy debris flows is further improved. The retarding effects of ambient fluid, including shear resistance, dynamic pressure, and added mass, on debris-flow movement is considered. The critical depth is reformulated by considering the role of the earth pressure gradient of debris flow. The improved model is then validated against laboratory-measured datasets from two flume experiments and a slump test. The numerical results agree well with the laboratory datasets. Afterward, the robustness of the model is demonstrated by a schematized large-scale submarine landslide. The improved model is also applied to a more realistic field scenario with real bathymetry originating from the Northern South China Sea where historical submarine landslides occurred. The results show that resistances from ambient water have significant implications on the dynamics of failed sediments. It is observed that the earth pressure gradient of debris flow plays a major role in the stage of creep deformation.

Energy of debris flows in steep mountainous channels is mostly consumed by erosion, collision, and hydraulic jumps due to abrupt topographic changes. In this study, flume experiments were performed to study the energy dissipation of debris flows over stepped gradients and erodible beds. A power law relationship between the depth-averaged velocity and flow depth was used to calculate the kinetic energy of debris flows. The experimental data indicate that the energy loss of entrainment per unit volume of bed sediment has an approximately linear relationship with transition angle and porosity. The net resistance coefficient also demonstrates more energy dissipation when debris flows move downward on slopes with a greater gradient change.

This study experimentally investigated the axis switching of free-falling water jets issued from an elliptical nozzle with an aspect ratio of 1.6. High-speed photography was used to capture the flow behaviors. The axis-switching wavelength of the elliptical jets was found growing with the falling distance while its amplitude was decaying. The variation of the jet cross-sectional area could be estimated based on free-falling velocity. An increasing axis-switching frequency with the falling distance was observed in the experiments and was estimated by introducing a correction factor for the gravity effect into Rayleigh’s equation. Finally, an estimation method for the axis switching of elliptical water jets was developed and the predictions compared well with the experimental measurements.