Journal of Irrigation and Drainage Engineering

Open‐channel transitions involving an expansion of width are a common feature of canals and flumes. Subcritical flow through an expansion transition can result in significant head loss due to separation of flow and subsequent eddy formation. In addition to more power generation in the case of power channels and increased service area for irrigation canals, the reduction of head loss is desirable from the viewpoint of increasing the life of the transition structure. This paper discusses a methodology based on optimal control theory for designing expansion transitions for gradually varied subcritical flow. Based on a large number of expansion transition designs, usable equations have been obtained for bed‐width, side‐slope, and bed profiles.

Two‐dimensional transient flow to ditches or drains in a heterogeneous, anisotropic porous medium is modeled for appropriate initial and boundary conditions using the finite differences method for the entire saturated‐unsaturated domain; numerically solving the governing ground water flow equation for the saturated zone below the water table; and using the Richards equation for the unsaturated zone. The water‐table rise computed by the composite flow model matches closely with observed water‐table rise in a tile‐drained experimental field. The composite flow model is superior to the saturated flow model (when based only on saturated flows), first, by accounting for the time of transfer of infiltration at the ground surface as recharge to the water table. Second, the general assumption of equal flux at the soil surface and at the water table is not required. The development of through flows has been studied in the presence of a clay layer in the unsaturated zone. Dependence of through flows on the thickness of clay layer and partial penetration has also been demonstrated. The model presented can also be used to calculate bank storage buildup during a flood and its subsequent release in a ditch or river having a ridge or a water divide parallel to the drain.

One of the key topics in irrigation engineering today is water management in irrigated command areas. This paper presents application of a comprehensive modeling approach based on the Système Hydrologique Européen (SHE), which allows an integrated modeling of all the relevant processes—overland flow, unsaturated zone flow, interception or evapotranspiration, ground‐water flow, and river‐channel flow to irrigation studies—at different scales ranging from a small plot to an entire command. Results of modeling studies the Barna command area in Madhya Pradesh, central India, are presented. The SHE simulated the advancing front of soil moisture as surface irrigation originated from the head end of a field, resulting in a nonuniform water distribution along the field. Furthermore, the effect of a shortage of water supply at the tail end of an irrigation canal was simulated. The study was carried out within the framework of an Indo‐European project cooperation resulting in the transfer of the SHE technology to the National Institute of Hydrology, Roorkee, India.

This paper examines the dynamic performance of the dual‐acting controlled‐leak (DACL) system for automating radial gate control. The dynamic performance of the DACL open‐channel level‐control system has been analyzed using classical linear control theory to facilitate the design of such systems to meet specific performance criteria. The mathematical analysis predicts that in normal closed‐loop operation, the system behaves like a second‐order control system. The response to disturbances is therefore a function of the system's natural frequency and damping ratio. Explicit expressions for the frequency and damping ratio were obtained in terms of system design parameters, and a design value of the damping ratio was recommended for optimal response in similar DACL systems. The analytical model was verified by tests conducted on a laboratory model arranged for downstream control. The results of these tests are presented in a companion paper.

This paper examines the dynamic performance of the dual‐acting controlled‐leak (DACL) system. The methodology and results of tests conducted on a DACL laboratory model configured for downstream control are reported. Water‐level responses in the controlled reach of the secondary channel to nominal step input changes in both the primary‐channel water level and set‐point level were measured. Laboratory test results showed that final values of the actual responses came to within about 1 mm of the predicted final values. This is in accord with control theory and previous findings that DACL systems can control water levels to within very narrow band widths. Use of the system response equations in the design of an irrigation turnout is demonstrated.

Infiltration was measured with a newly developed flow‐through infiltrometer at 20–27 sites in a 21 ha (52‐acre) cotton field in the northern region of the San Joaquin Valley of California. The results were comparable with infiltration measured with a neutron probe. The infiltrometer allowed water to flow through the infiltrometer when a rate test was not being conducted. Furthermore, the water level and quality inside the infiltrometer was the same as the water flowing past in the remainder of the furrow. Variability contributed by soil infiltration characteristics and intake opportunity time was quantified for preirrigation as well as for three postplant irrigations. The mean and variability of infiltration were greater for the preplant than the postplant irrigations, and soil variability contributed more to total infiltration variability than did the variability in intake opportunity time. Because the infiltration pattern was temporally correlated between postplant irrigations, representative sites rather than a complete field interrogation could be used to estimate mean and variability of infiltration.

A differential approach is applied to multiple outlet pipes with constant and continuously variable outflow. This approach simulates pressure distributions along uniform sprinkle and trickle irrigation laterals and manifolds, as well as gated pipes. The effect of ground slope and velocity head on the pipeline hydraulics were considered. An alternate simplified procedure that neglects velocity head effects and assumes that the lateral is laid on level ground is also included. Several mathematical relations are derived for computing maximum, minimum, and average pressure heads and their locations of occurrence along a lateral. The hydraulics of the common types of manually and mechanically moving sprinkler irrigation systems such as wheel line, linear move, and center‐pivot systems are considered in the present analysis. When compared to pressures and locations of occurrence computed using a discretized, stepwise computation procedure, values computed using the developed equations are generally within 3% for maximum, minimum, and average pressures, and are generally within 5% for the locations of occurrence. The proposed equations are useful when applied for design and evaluation purposes.

Areal, vertical, and temporal differences in salinity characteristics of shallow, unconfined ground water were related to irrigation, cropping practices, and surficial geology. Nine wells were monitored monthly for a period of up to 20 years for electrical conductivity (EC), $\mathrm{Ca}+\mathrm{Mg}\mathrm{and}\mathrm{Na}$ content, and the sodium adsorption ratio (SAR) was calculated from the ionic concentrations. In the last year of the study, the samples were occasionally analyzed also for K, Mg, and major anion contents. The ionic composition of the ground water was dominated by $\mathrm{Ca}+\mathrm{Mg}\mathrm{and}{\mathrm{SO}}_4$. Spatial, vertical, and seasonal variabilities were affected by cropping practice and depth to glacial till. Ground‐water salinity (EC) near the water table was lower under annual than perennial cropping and had a higher mean EC and higher mean seasonal EC in soils having till at less than 1.5 m from the surface than in those with till at depths greater than 1.5 m. Irrigation applications and precipitation events caused dilution of the poor‐quality ground water resulting in vertical water‐quality gradients in the top part of the aquifer. The vertical extent of the gradient depended mainly on the nature and thickness of the vadose zone. Temporal fluctuations were greater during the irrigation season than the off‐season. Ground‐water salinity near the water table since initiation of the study has trended toward higher EC values, probably due to improvements in irrigation efficiency.

Water‐table depth, piezometric head, and precipitation were measured at nine sites in southern Alberta over a period of up to 28 years to determine the effect of irrigation on long‐term groundwater levels. Irrigation and precipitation, mainly rainfall, directly recharged the unconfined aquifer. Temporary mounding of water occurred immediately following irrigation applications, causing short‐term lateral groundwater movement. Downward hydraulic heads occurred during the spring and summer seasons and upward ones in the fall‐winter period. High water tables were most frequent during May–July, when irrigation water was applied and precipitation was high. Highest levels generally occurred in soils with glacial till at depths less than 1.50 m from the surface. Water tables receded during the fall‐winter period, most frequently to depths greater than 2.00 m, due chiefly to closure of irrigation canals. Above‐average precipitation was generally reflected in that year's average annual water‐table depth, but there was a one‐year lag before drought effects were evident. Irrigation method and management also affected average water‐table depth. Depth to water table did not change appreciably over the duration of the study, increasing slightly but significantly at four sites and decreasing similarly at four other sites. As increasing water‐table levels predominated at sites with till within 1.50 m of the surface, special management schemes need to be developed for these soils should salinization become a problem.

The water infiltration rate during sprinkle irrigation is known to decrease due to the impact effects of water droplets on soil‐surface structure. In this paper, a numerical simulation model that employs full three‐dimensional Navier‐Stokes equations and the finite difference procedure is utilized for studying pressure and shear distribution on a soil surface following sprinkler droplet impact. Variables include impact velocity (800 cm/s and 580 cm/s), angle of impact (90°, 60°, and 45°), and surface condition (with and without a water layer). The results indicate that droplets impacting at oblique angles rather than vertically have less impact pressure and less impact force, but a higher shear velocity. For example, the shear velocity, which is a major contributor to soil surface crusting, is approximately 50% higher for a 60° impact than for a vertical impact. The effects of a water layer are to lessen the peak‐impact pressure and the shear velocity, but to increase the impact force.

A companion paper by Chang and Hills, 1992, describes the stress distribution in a soil surface during water droplet impact according to a numerical simulation model. This project was undertaken to relate field data to the simulated stress values. In this paper, sprinkler droplet impact angle and surface water layer depth were studied in relation to a soil's infiltration rate. Droplet impact angles were monitored at different distances from the head of a conventional medium‐size sprinkler. The test variables included nozzles with diameters of $3.2\mathrm{mm}\left(\frac{1}{8}\mathrm{in}\text{.}\right)\text{,}4.0\mathrm{mm}\left(\frac{5}{32}\mathrm{in}\text{.}\right)\text{,}\mathrm{and}4.8\mathrm{mm}\left(\frac{3}{16}\mathrm{in}\text{.}\right)$, and operation pressures of 200 kPa, 300 kPa, and 400 kPa. The impact angles ranged from approximately 90° near the nozzle to approximately 40° near the wetted perimeter. Impact angles were not significantly affected by nozzle size; however, increase in pressure caused the angles near the wetted perimeter to increase. Infiltration tests were then conducted on three soil types with droplet impact angles of 90°, 60°, and 45°. The basic infiltration rate for all soil types decreased in the following order of impact angle: no impact, 90°, 45°, and 60°. These results correlate with the shear stresses predicted for those conditions by the simulation model.

On the west side of California's San Joaquin Valley, researchers are experimenting with agroforestry systems to help solve the agricultural drainage problem. Eucalyptus tree plantations are used to consume the shallow ground water and saline drainage waters from croplands. The long‐term efficacy of this practice is being evaluated. This paper presents a growing body of data on soil water and salt fluxes in a tile‐drained 9.43 ha eucalyptus plantation. The irrigation water used in this study has an electrical conductivity of 10 dS/m, 12 ppm boron, 400 ppb selenium, and a sodium adsorption ratio of 11. The tree plantation was successful in lowering the water table from 0.6 m to 2.3 m and consumptively using the saline irrigation water. But after several years of drain‐water reuse, a substantial buildup of salinity and boron has occurred throughout the soil profile to the extent that the trees are unable to fully extract the available soil water. The estimated leaching fraction of 16% in the 1990 irrigation season will need to be increased so that the trees can survive and continue to consume saline drainage water from croplands.

Laboratory experiments were conducted to evaluate the hydraulic characteristics of a simple portable flume. The flume consists of a pipe installed axially inside a trapezoidal channel with side slopes of 1/1. The ratio of the diameter of the pipe to channel bottom width was approximately 90% or more. The presence of the column reduces the cross section of the flow, creating a critical‐flow condition. A gage is installed at the upstream side of the column to measure the depth of the water upstream of the critical‐flow section. The reading on the gage is directly related to the flow rate. Based on the parameters measured in the laboratory, a computer model is developed for calibration of this type of flume. The computer model is tested using data collected from flumes installed in the field. The computer model predicted the measured flow rates with a maximum error of 5.1%. The maximum submergence limit was measured at 81%.

A numerical procedure is developed for quickly and accurately finding the location and flow depth of a flow transition (critical control section) in open channels with discharge gradually varying along the channel length. General equations are presented for trapezoidal, triangular, and rectangular cross sections. A numerical analysis of the procedure shows that the geometry of the channel cross section is an important factor determining the existence of a unique solution to the flow‐transition problem in open channel flow. A procedure is described to decide if solutions exist and, if they do, to compute them and select an acceptable solution as the starting point for water‐surface profile calculations. Test runs on a computer are performed to verify the speed and accuracy of the proposed procedure. Examples involving the finding of location of flow transitions in open channels are used to illustrate the usefulness of the procedure. The findings from this study are beneficial to hydraulic engineers solving problems related to the design of lateral spillway channels and gutters for conveying stormwater runoff; to hydrologists using spatially varied flow equations to describe flow processes in natural channels; and to the enhancement of algorithms for spatially varied flow computations in hydrologic simulation models such as the CREAMS field‐scale model and the WEPP watershed model.

The electrical conductivity (salinity) and boron concentrations of subsurface agricultural drainage effluent are analyzed. The effluent is from drainage systems ranging from 0 to 40 years of age, located on irrigated saline soils in the San Joaquin Valley, California. Data collected from 1962 to 1990 verifies that the equation developed between 1959 and 1962 is valid for use in predicting the quality of effluent that will be discharged from artificial subsurface drainage systems installed in irrigated agricultural soils of the San Joaquin Valley, California. The work is useful in that it demonstrates the predictability of the quality of subsurface drainage effluent from drains installed in irrigated land. Such relationships are important in arid areas of the world where reclamation of irrigated land is required, particularly when there is growing concern about the sustainability of irrigated agriculture on saline soils and the environmental impact of saline drainage effluent.