Journal of Hydraulic Engineering

Discharges into ambient water bodies by means of a submerged single‐port jet flow can exhibit a great diversity of flow patterns, depending on the geometric and dynamic characteristics of the environment and the discharge flow. A rigorous classification scheme has been developed—based on a length‐scale representation of all dynamic features—that classifies the possible flow configurations into 35 generic flow classes. The classification is applicable to an ambient cross‐flow with different types of density stratification, and to discharge conditions with arbitrary jet orientation angles and positive or negative buoyancy. The scheme has been validated with a wide variety of available laboratory and field data. The classification procedure has been implemented in the expert system CORMIX for analysis of the pollutant mixing zone, but also can be used for the general discharge design and selection of appropriate predictive models.

Discharges into water bodies from submerged multiport diffusers are characterized by a variety of possible flow configurations, ranging from internally trapped layers in deep stratified environments to vertically fully mixed plumes under shallow conditions. The length‐scale classification for single‐port discharges in the companion paper is extended to submerged multiport configurations in arbitrary ambient conditions. The classification procedure defines 32 generic flow classes in three major categories: internally trapped flow in linear ambient stratification, buoyant flows in uniform ambient layers, and negatively buoyant flows in uniform ambient layers.

Depositional deltas form in the headwaters of most reservoirs. The deltas are characterized by a developing foreset slope that eventually approaches the submerged angle of repose. Deltas extend both upstream and downstream; downstream growth seriously depletes reservoir storage, while upstream evolution raises local ground‐water levels and increases flood frequency. Flow separates as it passes downstream over the delta lip into the deeper part of the reservoir. Because of this flow separation and attendant recirculation, traditional finite difference modeling approaches are invalid near the steep foreset slope and cannot model delta growth accurately. A method of numerically fitting a vertical shock face to the evolving delta is developed and illustrated. Conditions upstream of the shock are described with the traditional St. Venant equations; downstream conditions are constant. A one‐dimensional mobile‐bed computer model is developed and compared to a simulated reservoir in a laboratory flume. The simulated delta closely matches the growth and propagation rate of the observed delta.

The Saint‐Venant equations describing unsteady flow in open channels and the continuity equation for the conservation of sediment mass are numerically solved to determine the aggradation and degradation of channel bottom due to an imbalance between water flow and sediment discharge. For this purpose, the MacCormack explicit finite‐difference scheme is introduced. This scheme is second‐order accurate, handles shocks and discontinuities in the solution without any special treatment, and allows simultaneous solution of the water and sediment equations, thereby obviating the need for iterations. The sediment transport relationship in any form may be included in the computations. Computational procedures are outlined for incorporating the typical boundaries for hydraulic engineering applications. The mathematical model presented here is applied to predict (1) Bed‐level changes due to sediment overloading; (2) development of longitudinal profile due to base‐level lowering; (3) and bed‐level changes associated with the migration of knickpoints. The computed results are compared with the available experimental data obtained on laboratory flumes. The agreement between the computed and experimental results is satisfactory.

Longitudinal dispersion of substances transported in rivers is caused by variations of the advective velocity over the river cross section and by transverse mixing. This two‐dimensional nature of the process is not accounted for by the usually applied one‐dimensional advection‐diffusion equation, which therefore is not valid in the so‐called initial period immediately downstream of a substance input site. An enhanced one‐dimensional model is presented based on a very much simplified transverse velocity profile, with a flow velocity of zero near the riverbanks and a constant value in the center of the river. Dispersion is modeled by the exchange of mass between the two zones, in which no transverse spatial gradients are considered. Dispersion can be estimated in this model by two dimensionless parameters that vary only moderately from one river to another. The behavior of the model in the asymptotic range and in the initial period, the limitations of the model, and its advantages compared to the one‐ and two‐dimensional models are discussed. The main advantage of the model is that it is not more complex than the one‐dimensional model but still yields a good description of substance transport for about 80% of the initial period.

SO far, various calculation models for the vertical distribution of suspended sediment concentration have been produced by several investigators from different theories. The limitations of all these models imply that it is possible to find a more reasonable model, for which each previous model can be included as special case. The formulation of a reasonable general model is the purpose of this paper.

Boussinesq equations describing one‐dimensional unsteady, rapidly varied flows are integrated numerically to simulate both the sub‐ and supercritical flows and the formation of a hydraulic jump in a rectangular channel having a small bottom slope. For this purpose the MacCormack (second‐order accurate in space and time) and two‐four (second‐order accurate in time and fourth‐order in space) explicit finite‐difference schemes are used to solve the governing equations subject to specified end conditions until a steady state is reached. The inclusion of initial and boundary conditions is discussed, and the importance of the Boussinesq terms is investigated. Complete test results for a range of Froude numbers are presented that may be used by other researchers for the verification of mathematicalmodels. A comparison of the computed and measured results shows that the agreement between them is satisfactory for the fourth‐order finite‐difference scheme although the second‐order scheme does not accurately predict the location of the jump. These simulations show that the Boussinesq terms have little effect in determining the location of the hydraulic jump.