Journal of Hydraulic Engineering

Computerized transient-flow models have been used with great success in the analysis of water-hammer events in topologically simple pipeline systems, and the performance of these models is well documented. This paper addresses the relatively unexplored area of transients in complex pipe networks. A new formulation permitting system demands to be represented as a distributed pipe flux is presented. This approach is compared with two conventional methods for modeling demands in pipe networks. The results of a field test conducted on August 29, 1990, by the City of Calgary Waterworks staff on one of the city's major transmission and distribution subsystems is presented. The results are compared with the behavior predicted by a network transient model. The computer model was generally in good agreement with the field test data, with all three demand models giving comparable results, particularly with respect to the initial downsurge and the first upsurge following the pump trip. However, the transient's long-term decay was poorly represented by all three demand models.

Residential water use is visualized as a customer-server interaction often encountered in queueing theory. Individual customers are assumed to arrive according to a nonhomogeneous Poisson process, then engage water servers for random lengths of time. Busy servers are assumed to draw water at steady but random rates from the distribution system. These conditions give rise to a time-dependent Markovian queueing system having servers that deliver random rectangular pulses of water. Expressions are derived for the mean, variance, and probability distribution of the flow rate and the corresponding pipe Reynolds number at points along a dead-end trunk line. Comparison against computer-simulated results shows that the queueing model provides an excellent description of the temporal and spatial variations of the flow regime through a dead-end trunk line supplying water to a block of heterogeneous homes. The behavior of the local flow field given by the queueing model can be coupled with water-quality models that require ultrafine temporal and spatial resolutions to predict the fate of contaminants moving through municipal distribution systems.

The capacity of a labyrinth spillway is a function of the total head, the effective crest length, and the crest coefficient. The crest coefficient depends on the total head, weir height, thickness, crest shape, apex configuration, and the angle of the side legs. Data and a procedure are presented for designing labyrinth weirs for angles between 6° and 35°, and for a range of heads. The design procedure allows the angle of the side legs and the number of cycles to be varied until the desired layout and capacity are achieved. The solution is presented in a spreadsheet format that automatically calculates the dimensions for the labyrinth. Even though the design procedure is quite accurate, it is recommended that the capacity and performance be verified with a model study. The model can evaluate factors not included in the design procedure, like aeration effects at low heads, unusual flow conditions in the approach channel, and flow conditions in the discharge channel.

There is a need for an accurate limit for the practical use of the kinematic wave form of the St. Venant equations for computing the time of concentration for sheet flow. Such a limit will help ensure accurate designs for small watersheds. The current practice of limiting the flow length in the kinematic-wave time-of-concentration equation is problematic. This paper presents and analyzes limiting criteria, including L , iL , and nL $\mathit{S}$, where L is the overland flow length, i is the rainfall intensity, and S is the slope of the surface. Data reported in the literature from 59 actual and experimental watersheds are empirically analyzed. The criterion nL /$\mathit{S}$ appears to be the best criterion, and it results in minimum standard errors of estimate when the upper limit of nL /$\mathit{S}$ is 100 (English units). Theoretical routing using a range of basin slopes, roughness coefficients, and flow lengths gives an optimum limit on nL $\mathit{S}$ of 80 (English system). Since the empirical analysis has a more direct relationship to the application of the kinematic wave model, the upper limit of 100 on the nL /$\mathit{S}$ criterion is recommended for design work. When the kinematic equation is employed to compute times of concentration for given values of n and S , the length of sheet flow should be chosen such that the nL $\mathit{S}$ criterion does not exceed the recommended limit of 100.

Experiments involving bed forms have usually been carried out in flumes, but in the present work a pressurized-conduit system was found to be appropriate for investigating sediment motion at high shear stress. The bed materials were three sands (1.1 mm, 0.6 mm, and 0.4 mm) and bakelite (1.0 mm and 0.7 mm). The grain-size Reynolds number was between 25.4 and 157.0. As the dimensionless shear stress (Shields parameter) was increased toward unity, the steepness of the bedforms diminished abruptly. This behavior indicates a sudden shift from the bedform regime to the upper-plane-bed (sheet flow) regime, rather than a gradual transition. The results for the upper-plane-bed regime show that this type of flow has a larger frictional resistance than that of a conventional rough boundary. The bedform results indicate that the frictional characteristics of the bedform regime cannot be expressed in terms of dimensionless shear stress alone. The observed behavior also suggests that, as a solid-transport mechanism, sand waves may be more efficient than upper-plane-bed flow.

In alluvial rivers, suspended matter is deposited in the pore space of the riverbed during low and medium discharges, and at high flows the armor layer breaks up and the riverbed is flushed. During this cycle, the hydraulic conductivity decreases with time and then suddenly increases, so one can speak of a characteristic graph of siltation. In this paper, the results of a 3-year investigation at the Laboratory of Hydraulics, Hydrology, and Glaciology of the Federal Institute of Technology in Zurich are summarized. A wealth of long-term tests allowed the derivation of several equations concerning the siltation of riverbeds, including the reduction of the hydraulic conductivity with time, the limit of siltation, the conditions for desiltation, and the consequent increase in hydraulic conductivity. Simultaneously, the influence of the significant variables is shown. This research makes it possible to quantify the siltation process as well as the effect of specific changes in the river regime and the catchment area.