Journal of Irrigation and Drainage Engineering

A method for assessing the impacts on streamflow resulting from withdrawing water from the stream channel or from shallow wells adjacent to the stream for irrigation is presented. Assessment is based on a detailed study at a base station on the stream. Monthly irrigation demands are estimated and added to measured streamflow to determine the natural streamflow. The natural streamflow is correlated with flow on a nearby watershed to develop an extensive series of monthly flows. The procedure produces probability distributions for average monthly flow for each month of the growing season, for different levels of irrigation usage. The flow distributions include the original variation and the additional variation produced by the irrigation withdrawal. The streamflow information developed can be combined with biological or water quality models to assess the impacts of reduced flow on instream biology, water quality, and treatment requirements of waste discharged to streams.

The flow features of a Venturi flume, constricted with sharp‐edged, thin‐plated elements are investigated. These elements, located at both side walls of a rectangular duct, as are usually found in large municipal sewers, are fixed at the bottom and at the ceiling by a simple device. Particular attention is paid to the head‐discharge equation for various constriction rates, the limit submergence, and the discussion of the internal flow mechanism. The theoretical approach accounts for the effects of fluid separation from the constriction elements and for the streamline curvature. Compared to the conventional discharge equation, the effects of the contraction geometry, and the relative energy head are thus included. The final result corresponds to an explicit equation for discharge once the contraction geometry is fixed and the upstream flow depth is recorded. Design curves allow a rapid and precise determination of discharge.

The primary source of error in properly calibrated, constructed, and installed flow measurement devices is due to reading error or uncertainty. Head reading uncertainty in small V‐notch flumes and submerged orifices is measured in the field as $\pm 3\mathrm{mm}$ with no consistent variation with reading. Elapsed time measurement uncertainty for volumetric measurements increases with the square root of the time. The sensitivity of flow measurement uncertainty to head or time reading uncertainty is proportional to the ratio of the device discharge equation exponent to the reading. Furrow flow measurement uncertainty varies with the device and flow rate, but generally exceeds $\pm 5\%$ and often exceeds $\pm 10\%\text{.}$ Maintaining uncertainty below $\pm 10\%$ requires flume measurements in the upper 50% of their range, orifice measurements with head readings greater than 13mm, and volumetric measurement elapsed times greater than 4 seconds.

Furrow infiltration and channel seepage are often measured with inflow‐outflow measurements. Inaccuracy in the flow measurement will cause a larger uncertainty in the calculated infiltration. The infiltration rate determination uncertainty increases rapidly as the percent of the inflow that is infiltrated decreases. The effect of measurement uncertainty on infiltration measurements can be estimated so that the confidence interval of a mean or the actual infiltration variability level can be determined.

A mathematical model for border irrigation is developed using the spatially lumped continuity equation and the Muskingum type storage relation. The storage parameter is determined from physical border characteristics. The model is calibrated and tested using twentyfour experimental data sets, and yields satisfactory results. The relative error in predicted advance is less than 16% and that in predicted recession is less than 7%. A short comparison with three other models shows that the Muskingum model is comparable in accuracy.

An important objective of any trickle system is a uniform distribution of water delivered through the emitters. Computation of flow distribution requires knowledge of the variables such as pressure, flow rate, length of lateral, characteristics of the orifices, and frictional loss in the system. Several studies have been reported that establish these relationships. In each study, the primary solution is based on a discharge that is uniform, although ramifications of the manufacturer's variability have been modeled based on the derived hydraulic profile. Herein, we present an alternative treatment that includes a spatially variable discharge function as part of the basic solution.

The trickle lateral analysis described in the companion paper is used to illustrate applications of a new analytical solution in system design. Examples are provided to point out the basic concepts and procedures based on this and related analytical solutions. Very simple and yet powerful design curves for different flow regimes are presented. Verification of the solution is also made by comparing the results with experimental measurements. The simplicity of the solution, coupled with the ease of including variable discharge, makes the analytical solution attractive for diversified uses in trickle system design.

The skew of a peak‐discharge frequency curve is an important determinant of the magnitude of events having small exceedance probabilities. Unfortunately, methods for estimating skews for ungaged locations provide highly inaccurate estimates. This appears to be the result of a lack of understanding of the primary causes of variation in skew. An analysis of existing hydrologic methods indicates that the skew assumed with these methods for ungaged locations is different than the mean skew for gaged sites in the same area. Furthermore, the design methods suggest that watershed storage is the primary factor affecting skew, with an increase in skew as the volume of watershed storage is decreased. Variables that show greater potential for providing accurate estimates of skew are discussed, and a relationship for estimating the effect of imperviousness on skew is provided.

Hydraulic friction loss in pipelines directly affects pipe and pump sizing, as well as the hydraulic balance of networks. All friction head‐loss equations have an uncertainty in the estimation of pipe interior surface roughness. When a single resistance coefficient based only on the pipe material is used, different equations may give significantly different estimates of frictional head losses depending on pipe size and water flow rate. In complex pipe network systems, minor differences in equations or coefficients may accumulate and result in vastly different predicted head losses or flow rates in individual pipes. This study presents mathematical relationships and nomographs to relate the friction factors of three widely used pipe friction equations and to determine the magnitude of the differences in calculated head losses. Friction factors are developed to be used with the Hazen‐Williams and the Manning equations for PVC and cast‐iron pipes that give similar calculated head losses to those of the Darcy‐Weisbach equation. The variation of the friction factors with diameter and flow conditions is presented, and computer implementation of the variable friction factors is discussed.

Adequate and reliable records of rainfall depth‐duration‐frequency amounts are often not available. Use of estimates of probabilities from limited data can reduce errors due to either underdesign or overdesign of structures and can reduce damage to irrigation facilities and hydraulic structures from floods and excessive runoff. Estimates of extreme rainfall events may also make possible an increased utilization of rain at the locations where it falls. Methods are presented for estimating rainfall intensities with a minimum of data. Equations are given for use with 20‐yr return period monthly rainfall amounts and with daily rainfall extremes. Depth‐duration and depth‐frequency ratios of extreme rainfall amounts have been found to be remarkably uniform within large and diverse geographical areas. This appears to be true for all of Africa. Large expenditures are being made to improve the collection of hydrologic and meteorologic data in the Sahel of Africa. Known ratios and relationships presented herein can improve the use of the existing long records and of data currently being collected. Greater emphasis on training in the use of the climatic data base is now available, and the need to be improved is recommended.

The Salt River drainage basin is an agricultural watershed of $2\text{,}150{\mathrm{km}}^2$ in western Wyoming. Starting in 1971, several irrigation projects were completed that converted surface irrigation systems to sprinkler irrigation systems on approximately one‐half of the irrigated land. This conversion resulted in less total water being diverted from streams for the sprinkler system than the previous surface system. Salt River stream flows were analyzed to determine if there were differences between the sprinkler systems prior to and after conversion. Mean monthly flows in May and June increased significantly (58.7%) following the conversion to sprinklers. The Salt River flows were also compared with flows of the Greys River, a nonagricultural watershed, using the double‐mass analysis. The test again showed higher May and June flows. Lower fall flows were evidently a consequence of irrigation practices rather than climatological factors. Annual flood peaks increased significantly following the conversion to sprinklers. The mean annual flood peak increased by 47.0%.

Field conditions were reproduced in the laboratory in order to investigate redevelopment techniques. Segmental perspex models were recharged with water of controlled quality and temperature containing different types of suspended solids of known size. Both graveland nongravel‐packed recharge wells were tested. It was found that the most efficient method of redeveloping a clogged recharge well was by multiple reversals of flow direction (i.e., pumping and recharging the aquifer). A purely intermittent pumping action following the first reversal of flow had practically no increased beneficial effect on the recovery recharge rates. It was also concluded that neither increasing the driving head under which the reversal is carried out nor increasing the duration of the reversal of flow has a marked beneficial effect. Nongravel‐packed wells were easier to redevelop than gravel‐packed ones and the use of backwashing tubes in the gravel pack did not prove highly successful in restoring the well efficiency.

A terminology is presented with which all variety of farm irrigation water demands may be identified and described. Only the relationship between the water demands and the conveyance system supplying the water is considered. An examination of previous classification efforts is described. From the results of these studies a comprehensive terminology is developed and demonstrated.

Research to determine crop water-production functions normally involves random block designs or a line sprinkler source. Both of these setups have disadvantages that can largely be overcome by using modified center pivots. The ôtraveling tricklerö center-pivot modification uses a series of small adjustable emitters (tricklers) to apply water. This system can be used for both row and grain crops. These emitters can be adjusted to apply different amounts of water on different bands of the center pivot giving a balanced design that offers more control than traditional experimental designs.

The Su-Brooks retention function was linearized by transforming the functon into logarithmic form. The resulting equation had a form that allows determination of the parameters using linear multiple regression techniques. Although the linearized form did not fit the experimental data as well as when the parameters were determined from a nonlinear regression method, the differences were minor. The greater robustness of a linear regression procedure compared to a nonlinear procedure and that simple hand-held calculators can be employed to determine the parameters makes the linearization procedure an attractive alternative to obtaining Su-Brooks retention parameters compared to nonlinear regression techniques. In addition, the described method makes it considerably easier to obtain parameters of the Su-Brooks function compared to the method originally proposed by Su and Brooks and takes all of the experimental data into consideration.