Journal of Hydraulic Engineering

Prediction of snowmelt runoff is one of the most important problems in the determination of optimum use of water resources. In this study, the combined snowmelt and rainfall runoff during the snowmelt period is predicted using a combination of the tanks and autoregressive models, with parameters identified by the Kalman filtering. The proposed method can accurately predict the snowmelt runoff. The input data used are rainfall, snowfall, runoff and air temperature.

Water resource engineers often have to relate qualitative dependent variables to one or more independent variables, which may or may not be quantitative. In such circumstances, the use of conventional regression analysis would encounter a number of difficulties. This paper introduces a statistical method called logistic regression which is specially developed for such conditions. The method is applied to a hydraulic problem of relating scouring potential in a channel to depth and velocity of flow. Whether or not the methodology could become a useful addition in water resources engineering analyses further investigations and applications are necessary.

The “Solver” algorithm, developed as an outgrowth of work on cost allocation in water distribution systems, is a simple technique that solves a number of interesting problems in water distribution system analysis. Problems related to mixing water from different sources within the distribution network, travel time from any source to any node of the network, and development of the cost of service to any node in the network can be solved assuming steady‐state conditions, given a prior solution of the hydraulics (flow in each link) of the network. All these problems are formulated as the solution of simultaneous linear equations. The Solver algorithm has been coded in Fortran and incorporated within the structure of a multi‐purpose system of computer programs that allows for the storage, display and manipulation of data associated with node‐link networks describing water distribution systems. These programs are identified by the name Water Supply Simulation Model. The formulation of the Solver algorithm is described in detail, with examples of its use.

Formulas proposed to date for calculating ultimate scour depth under jets, such as issue from free dam overfalls and flip buckets, are examined. The accuracies of the formulas are evaluated by using each to process sets of scour data from prototypes and models of such prototypes. It is established that scour depth is as adequately calculable using only unit flow q and head drop H as using more complex considerations, but that those formulas most applicable for model purposes are not those best for prototypes. It is demonstrated that where bed particle size d is also considered, the use of the mean particle size $d_m$ is more appropriate than the $d_{90}$ size. Chute friction loss allowances and jet impact angle are also examined in terms of improved accuracy; the relevance of Froude law scaling for this type of scour is verified. Lastly, a new formula for predicting ultimate scour depth, which includes an allowance for tailwater depth h is presented and shown to give improved accuracy for both models and prototypes.

A persistent difficulty in modeling continental shelf and estuarine circulations is that associated with the correct specification of conditions on the open boundaries. This paper deals with the application of a new form of radiation condition on the open boundaries, which permits the mean subtidal and tidal forcing to be prescribed and, yet, allow transients generated inside the region to be transmitted outwards. A numerical model of shelf circulation is formulated, using vertically integrated equations, which accounts for tidal and transient wind forcing, as well as the effect of density gradients and the offshore large‐scale oceanic circulation. The shelf model is dynamically linked to a storm model. The shelf circulation model is applied to the Middle Atlantic Bight in a study of its response to a migratory storm superimposed on the subtidal and tidal circulation. The major estuaries of the region—Chesapeake Bay, Delaware Bay and Long Island Sound—are included in the domain, albeit at a resolution coarser than desirable. The model results are consistent with those of earlier observational and modeling studies.

A computer model is described to predict unsteady and nonuniform flows in alluvial streams. The model uses a generalized expression for evaluating the energy slope in mobile boundary flows. Therefore, it is possible to adopt different friction factor relationships into the model without affecting its structure. It is also possible to upgrade the model easily as new expressions for alluvial bed roughness become available. The model has been tested for laboratory and field conditions and shows good agreement with measured data. It is capable of predicting long‐term changes in riverbed slope and thus is wellsuited for estimating change in river regime caused by certain modifications to river flow, such as river diversions, meander cutoffs, river training works, dams and hydroelectric power plants.

TWO models were tested in order to determine the adequacy of a channelization scheme for Salt River near Sky Harbor International Airport, Phoenix, Arizona. One fixed‐bed model was used to study velocity patterns, flow stages, bank freeboards and bank protection requirements. One distorted movable‐bed model was utilized to study erosion and deposition patterns and bank stability in the proposed scheme. Incipient motion of bed particles and tractive force theory for stability of banks were utilized for modeling the various complicated parts of the movable‐bed model. Results were compared with mathematical model results, and good agreement was observed. Recommendations for improving the channelization scheme were given.

With the proliferation of low‐cost graphics systems and small computers, high‐quality computer graphic representation of the modeling of geophysical flow is feasible. In this work, storm surge activity on Lake Erie is simulated by a Leendertse‐type model and the output from this model (water surface elevations) are pictured as three‐dimensional surfaces in an animated sequence. Four methods for constructing such images are surveyed: A wire‐frame plot on a line‐drawing device; a representation as a set of flat panels on a polygondrawing raster display; a smooth shaded rendering on a point‐addressable monochrome display; and a full‐color rendering with “highlights” on a larger color palette display. Each of these methods are assessed in terms of speed, economy, reliability, the ability to render all relevant scales of activity, and the ability to be comprehended readily and unambiguously.

TWO procedures were developed to simulate salinity intrusion in the Mississippi River. The first one is based on the arrested salinity wedge theory and is applicable to quasi‐steady‐state flow conditions. It simulated very well the experimental results of Keulegan. It also simulated the positions of salt wedge tips and their interfaces in the Mississippi River with proper adjustment of the interfacial friction factor. The second one is a steady‐dynamic routing model, which routes salt water in a cascade of reservoirs in the river bottom. Densimetric critical flow is assumed over a reservoir ridge. Error in storage definition is included in a calibration coefficient for the saltwater flow computations over the reservoir ridges. The model satisfactorily simulated durations of salt wedge presence along the lower Mississippi for the existing conditions and after deepening the channel to 55 ft.

A forecasting technique for predicting river flow resulting from combined snowmelt and rainfall is presented The technique incorporates Kalman filtering techniques to reflect Uncertainty in the measured data as well as errors in the system model. A primary emphasis is given to how the forecasting algorithm is applied to a case study area, thus demonstrating the utility of the technique when applied to real‐world data. Methodologies are presented which can be used to calculate the covariance matrices associated with the Kalman filter algorithm utilized by the forecasting procedure.

Experiments were conducted to study the circulation patterns and mixing processes in a stratified, lid‐driven cavity flow. The ratio of the cavity depth to width used was 1:1 and that of span to width was 3:1. The flow is different from those reported in the literature in up to four ways, viz., recirculation is a dominant feature, temperature control is used to obtain the stratification, the upper and lower boundaries are held at different constant temperatures, and the flows considered are, at most, only partially turbulent. For the geometric configuration used, the following may be concluded about the flow. The flow is strongly three‐dimensional. The number of circulation cells expected in this stratified flow can be estimated from the value of the bulk Richardson number, ${\mathrm{Ri}}_b\text{.}$ Given a local Richardson number, Ri, the entrainment rate, E, is found to be proportional to ${\mathrm{Ri}}^{-1.4}\text{,}$ indicating a shear‐induced mixing process related to the recirculation. Taylor‐Görtler‐like vortices above the thermocline are significant contributors to the mixing process.

The performances of four of the most commonly used methods for discharge calculation in compound channels of varying roughness, were assessed by comparison with laboratory data. The four methods are based on splitting the section into main channel and floodplain subdivisions using either horizontal or vertical interfaces. The accuracy of the four methods in calculating overall section discharge generally increases at greater floodplain depth and with smoother floodplains. However, the distribution of flow within the section is generally poorly modelled, which leads to very erroneous values for momentum and kinetic energy flux.