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.

In this paper, a hybrid artificial viscosity–central-upwind (AV-CU) scheme is proposed for simulating recirculating turbulent shallow water flows by combining the artificial viscosity (AV) technique with the central-upwind (CU) scheme. Two-dimensional (2D) depth-averaged Reynolds-averaged Navier–Stokes (DA-RANS) equations are solved using the AV technique, whereas the CU scheme is employed to compute the $\kappa –ϵ$ model. The model is spatially and temporally second-order accurate. Scalable wall functions (ScWFs) are employed, thus becoming flexible in generating meshes without having to estimate the wall friction velocity at the initial time step as if the standard wall functions (StWFs) were used. The results benefit strongly from this hybrid approach being more accurate than the CU and Harten-Lax-van Leer-contact (HLLC) schemes—and $1.52\times$ cheaper than the HLLC scheme for modeling recirculating turbulent flows. As such, the proposed approach could become a promising method for practical engineering purposes to simulate turbulent shallow water flows more efficiently and accurately.

The protection of river banks in mountain rivers by riprap requires large blocks with weights typically exceeding 1 t. Thus, the blocks have to be placed individually. Such packed riprap has a relatively small spacing between the blocks. Consequently, the interlocking forces between the blocks are high, and the resistance of this protection to erosion is increased compared with traditional dumped riprap used for bank protection in low-grade rivers. Based on 98 systematical laboratory experiments, an adapted design method was developed for the case of riprap riverbank protection using large, individually placed blocks. Three different block sizes were tested by varying the longitudinal channel slope and bank inclination. For comparison, 34 tests were performed with dumped riprap. A design relationship is presented as a function of block size relative to the flow depth and a modified block Froude number considering the mean flow velocity. The additional resistance of riprap to erosion with large, individually placed blocks compared with dumped riprap is quantified. The proposed equations are compared with existing riprap sizing equations. The suggested method proved to be reliable in the assessment of the stability of packed riprap. Furthermore, it is shown that a second layer significantly delays riprap failure.

The rotational speed estimation of Francis and Kaplan turbines is of primary importance in the design of hydropower plants. However, most rotational speed equations were conceived for power outputs higher than 1 MW, i.e., for heads typically higher than 10 m for Kaplan turbines and 50 m for Francis turbines. Therefore, the accuracy of these equations is low when applied to reaction turbines with power outputs below 1 MW (mini-hydropower), which means the personal know-how of hydropower companies is generally used to select rotational speed. In light of this, a database composed of hundreds of full-scale Francis and Kaplan miniturbines was collected. Full-scale data were used in combination with the Buckingham $\pi$-theorem to determine a nondimensional equation for the rotational speed estimation of collected turbines. Flow rate and head were used as input variables. The dimensional analysis was also extended to high-head Kaplan and Francis turbines. A universal equation valid for both Francis and Kaplan turbines was obtained. Finally, the dimensionless terms obtained from the analysis were proposed as characteristic numbers for the selection of the hydraulic turbine type. The new equations were proposed as engineering tools for the design of hydroturbines.

The effect of receiving water depth on the dilution of a unidirectional diffuser discharging dense effluent in quiescent conditions is analyzed. Dilution measurements from previous studies are reviewed to determine the effect of shallowness on the mixing of submerged single port outfalls. For a unidirectional multiport diffuser, the width of the effluent plume is observed to contract in shallow water. A model is proposed relating the dilution of a unidirectional diffuser to the dilution of a single jet using the contracted width. Measurements of the contracted width in shallow water are made to support the proposed model, which is verified by additional measurements of dilution. Three flow regimes, namely, deep, shallow, and vertically mixed, are identified based on the value of $D_0{F}_0/H$. Here, $D_0$ is the diameter of the jet, $F_0$ is the densimetric Froude number of the jet, and $H$ is the depth of receiving water. It is shown that shallow water depth affects the dilution in two ways—the dilution of individual jets is reduced, and the contraction in plume width leads to further reduction in dilution.

Radio frequency identification (RFID) tracer stones are an increasingly popular tool in sediment transport research. A key limitation of RFID tracers is the sensitivity of their detection zone shape to the orientation of the transponder. The variability in detection zone shapes reduces detection rates by obscuring tracer positions and creating complex interference effects when multiple tracer stones are within the antenna detection field. This paper presents two designs and prototypes of the “Wobblestone,” a synthetic tracer stone that ensures the transponder remains vertical regardless of the stone orientation. The designs feature a weighted inner ball containing the transponder that is free to rotate under gravity and correct to a vertical orientation. The prototypes were fabricated using a high-density polyurethane resin and corundum powder mixture. The prototypes achieved densities within the range of natural stone and confirmed the durability of the material. Performance tests verified the vertical orientation of the transponder regardless of the stone orientation and demonstrated an opportunity for the determination of tracer burial depth. The new tracers will reduce signal interference from closely spaced tracers by simplifying detection zone shapes and will allow for information on the vertical position of buried particles to be obtained.

A one-dimensional sediment transport model (SRH-1D) was updated to simulate a sediment sluicing and flushing plan for the Paonia Reservoir, located on Muddy Creek in western Colorado. The Paonia Reservoir has lost nearly 25% of its original capacity of 25.8 million cubic meters (20,950 acre-ft) to sediment deposition. In the fall of 2014, Paonia Dam became partially blocked with sediment and debris, and emergency action was necessary to maintain water deliveries. To analyze potential solutions to this problem, a proposed sediment sluicing and flushing plan was conceptualized to lower the reservoir pool in the early spring, and to sluice and flush sediment with high spring runoff flows through the outlet works before closing the gates to refill the pool for the irrigation season. SRH-1D was modified to include a user-defined set of reservoir operation rules, which was implemented to assess the effects of managing inflow and sediment deposition in the Paonia Reservoir. The numerical model was calibrated using the 2015 reservoir survey data and incoming and outgoing suspended sediment concentrations from April 2013 to June 2015. It was then used to predict the reservoir sedimentation with a one-season simulation. The numerical model shows that a spring sediment flush can be a useful method to pass sediment from a reservoir.

New laboratory data of local scour at submerged weirs are presented. The dimensionless scour depth downstream of a submerged weir in a uniform sediment bed is expressed as a function of dimensionless weir height, flow intensity, sediment coarseness, and weir slope. The existing and newly presented data are combined to evaluate the dependencies of downstream clear-water scour depth on each of the relevant dimensionless parameters. For live-bed scour, the limitations of the existing predictors are summarized.

Shallow-water models suffer from relatively high computational costs owing to the Courant–Friedrichs–Lewy (CFL) constraint and (in most cases) the adopted global minimum time step for variable updating. Here, an improved local time step (LTS) scheme is proposed by introducing slight but useful modifications to the Sanders LTS scheme for two-dimensional (2D) shallow-water modeling based on unstructured triangular meshes. The modifications include (1) avoiding an unrealistically large LTS in the case of a vanishing water depth for near-static cells and (2) avoiding spatial variation of the LTS level around a wet/dry front or dynamic/static front. A series of numerical tests show that the improved LTS gives stable and accurate solutions, along with significant reductions in the computational cost. Moreover, the computational overhead due to the aforementioned modifications is negligible. It is expected that the improved LTS scheme will find wide application in 2D shallow-water modeling.