Journal of Irrigation and Drainage Engineering

Weirs are devices installed in channels by means of which flow may be measured within a given degree of accuracy. Linear weir flow‐measuring devices introduce less relative error in discharge measurement. The design of a linear weir is such that the base approaches infinity; this makes its theoretical design physically unrealizable. Various attempts have been made to achieve a practical weir profile by cutting off the infinite bottom wings and providing base weirs of different shapes. Furthermore, all these designs use a constant discharge coefficient, whereas there is a considerable variation in the discharge coefficient with the head/weir‐height ratio. A linear weir profile that includes the variation of discharge coefficient and does not require base and datum corrections is presented herein. A design example has also been included.

Net reservoir evaporation is defined as the open water evaporation less the original evapotranspiration from the reservoir site. These rates are estimated on an average monthly basis by the method of Morton using sparse climatological data that include average annual precipitation, air temperature, dew‐point temperature, and sunshine ratio. Sunshine ratio data are substituted with the more generally available cloud cover data with little loss in accuracy. Generalized reservoir surface area‐volume relationships are derived by regression analysis separately for large and small reservoirs in Australia. They represent the surface area‐volume characteristics of a reservoir confined in a hypothetical valley whose shape is an averaging of the shapes of a large number of actual sites. The sensitivity of estimated storage sizes to errors in the evaporation rates and the reservoir surface area‐volume relationships is tested for five catchments located in the southeastern extremity of Australia. With regard to these parameters, estimated reservoir sizes are shown to be surprisingly robust.

Systems engineering research is conducted to analyze improvements in performance of a water‐storage and ‐distribution system in response to improved monitoring of irrigation demands. The Mahaweli system in Sri Lanka, a multireservoir system with both hydropower and irrigation objectives, has been selected as a case study. The operation of the Mahaweli system is simulated using a regression model obtained through implicit stochastic optimization. Spatially independent, cross‐correlated, and systematic errors in irrigation demands are considered. These three, types of errors have similar patterns in their effects on system operation. When the Mahaweli system is operated optimally, both energy shortages and irrigation shortages increased with increasing standard deviations of error. This indicates that improving the measurements of irrigation demand would be beneficial to both farmers and hydropower recipients.

International perspectives on irrigation organization provide new and useful ways of looking at old and increasingly difficult problems of drainage management in the United States. One particularly intractable issue at the state and federal levels is how to better manage irrigation‐induced water quality problems in California's San Joaquin Valley. Several recent and important studies recommend that irrigation and drainage responsibilities be consolidated at the district or regional level in order to more effectively address drainage management problems there. Another report, in contrast, suggests an autonomous, statewide drainage agency be established to handle these problems, leaving irrigation responsibilities to existing agencies. Analytic frameworks developed overseas identify a variety of organizational and institutional arrangements for better dealing with water quality problems, one of which is the creation of systemwide drainage authorities. A major implication of such analytic frarmeworks is that drainage and irrigation engineers must become much more publicly involved in designing improved organizational—not just technical—mechanisms for drainage treatment and disposal, both in the Valley and in California as a whole.

A major effort is under way to rehabilitate Egypt's vast irrigation system and to improve its management. To be effective, planning decisions must be in concert with irrigation needs at the farm level. Data on conventional farm irrigation practices provide an objective rationale for irrigation planning. Intensive field studies of irrigation practices on fields containing small‐level basins produce data descriptive of the magnitude and variability of application rate, application time (duration), applied depth, and application efficiency. Average application rates are between about 0.10 and $0.20(m^3/s)/\mathrm{ha}\text{.}$ Average application times are high, ranging between about 10 and 25 hr/ha. Applied depths average between 7 and 13 cm. In the alluvial clay soils, average application efficiencies are about 0.70–0.75. In sandy soils, the average measured application efficiency is about 0.45. The variability in the observed values of each of these irrigation parameters is high: Coefficients of variation typically range from about 0.50 to 1.00. The data reveal no clear dependence of application efficiency on application rate. Though the variance in the data is considerable, a power‐function dependence of application time on application rate is established. Thus, the delivery of larger flow rates to farm turnouts should produce savings in irrigation time and labor. However, larger flow rates cannot be expected to result in higher irrigation efficiencies unless farmers are assisted to understand better how to evaluate and manage irrigations under varied field conditions.

Many schemes and devices have been developed and tested during the past three decades for automatic control of canal networks. Most of these have used local flow criteria to determine gate settings. Computer simulation of transient canal hydraulics has also been used to help test some of the ideas for canal automation, and in some cases the simulations have been applied directly to canal operation. In this paper, a centralized gate‐scheduling logic for a transient hydraulic model is presented, and a sample comparison between the centralized logic, and one form of a local‐gate‐scheduling scheme, is given. The centralized approach to canal operation is shown to be significantly more stable than the local approach, with minimal fluctuations in target flow levels.

A computerized water ordering system is developed to quickly determine the distribution of the water demands for an irrigation water delivery system. The inputs required for the program are water‐order requests by system users at specific locations with specified starting and ending times. The water‐ordering status for fields can be quickly changed and then summarized and printed as part of the daily water orders. Sets of utilities to create, edit, and maintain the files required for the water‐ordering program are also described. Additional program features include daily demand summaries at pump stations and control structures, and year‐to‐date and seasonal summaries of the water ordered by each field in the project. The programs have been used to make water orders for over 24,300 ha (60,000 ac) or irrigated land on the Navajo Indian Irrigation Project since 1983. Features and use of the programs are described.

Several researchers have shown that on‐farm runoff collection and supplemental irrigation systems can increase and stabilize crop production in many regions of the world. The objective of this study is to develop simulation capabilities to determine optimum pond size and irrigation strategies for supplemental irrigation systems. A comprehensive basin‐scale soil and water‐resource model is modified to simulate crop yield, supplemental irrigation, furrow diking, and economics. The simulation model is linked with a golden‐section search optimization scheme to optimize farm pond storage with respect to net profits from crop production under supplemental irrigation. The golden section is a direct search method and is applicable to optimization problems where an analytic expression for the objective function is unavailable. The model also produces monetary‐return frequency distributions for various pond sizes and is capable of determining the effects of various management strategies on crop production and water and sediment yields. The frequency distributions give decision makers a risk assessment tool for designing the pond. Simulated hydrologic components and crop yields are compared with measured values, showing the model performs well under various climate, soil, crop and management conditions. Finally, the model is applied to two basins in Texas, and results are analyzed.

This laboratory study investigates the local scouring phenomenon in sand that occurs downstream of box culverts. Although culvert hydraulics can include special cases, such as culverts on hydraulically steep slopes for which special energy dissipators at an outlet must be provided to handle supercritical flows, this study was restricted to culverts on horizontal or mild slopes that operate freely and do not require special energy dissipators at their outlets. The principal factors governing this form of local scouring were found to be the discharge rate, the culvert width, the tailwater depth, the downstream channel width, and the bed‐material properties. The experimental program investigated the effects of these principal variables on the local scour hole characteristics. The investigation also reviewed some well‐known empirical formulas for the prediction of maximum scour depth, under a variety of hydraulic conditions. One such relationship for maximum scour depth in a stone bed downstream of a culvert outlet, was modified in this study to make it applicable to scour in sand beds.

Critical depth can be defined as (1) The depth at which fluid velocity equals the speed of a gravity wave; (2) the depth at which the Froude number equals one; (3) the depth at which specific force corrected for velocity profile is minimized; or (4) the depth at which specific energy corrected for velocity profile is minimized. When the velocity profile is not uniform (when a boundary layer exists), the definitions yield alternative results. The greatest critical depth is that minimizing specific energy corrected for velocity profile. A general solution for alternative definitions of critical depth is developed for several channel cross sections having an exponential velocity profile. A relative critical-depth equation is derived relating the profile-corrected solution to one without correction. For many engineered and natural channels, the difference is only a few percent, but profile-corrected critical depths can be significantly higher than uncorrected solutions in the case of channels with large velocity gradients. Design implications are discussed.