Journal of Irrigation and Drainage Engineering

Soil hydraulic properties are usually expressed as functional relationships between the soil hydraulic conductivity and water content or matric potential, and between the matric potential and water content. Both functions are highly nonlinear, and direct in-situ measurements are time consuming and involve considerable uncertainty. Therefore, methods based on soil hydraulic models, i.e., functions chosen a priori and containing parameters estimated from measurements, are of interest. The most widely used models are those of Gardner in 1958, Russo in 1988, Brooks and Corey in 1964, and van Genuchten in 1980. For such models, an equation is derived that relates the developed expression for the first moment, M, of the diffusivity function with a combination of measured parameters: saturated hydraulic conductivity, capillary length, saturation and residual water content, and soil sorptivity. The M expressions are derived for the three models in the form of explicit functions of their parameters. Three overlapping M ranges are determined where the selection of a respective model is preferable.

Traditional finite-difference and finite-element solutions to Richards' equation can exhibit stability problems and mass balance errors when sharp saturation fronts migrate across material interfaces. Several semianalytical solutions to Richards' equation have been developed for layered media, but none allow for arbitrary constitutive relationships between capillary pressure, water saturation, and relative water permeability. This paper develops a more general semianalytical solution to Richards' equation that can simulate unsaturated flow in layered media and utilize arbitrary constitutive relationships. Based on a combination of the Runge-Kutta and shooting methods, initial-boundary-value problems are solved for layered systems without experiencing the types of stability problems commonly associated with numerical models. The proposed numerical scheme is computationally efficient and numerically stable, and it compares favorably with other analytical solutions. Results of two example simulations demonstrate the versatility of the method for solving both horizontal and vertical unsaturated flow problems. The primary disadvantages of the numerical approach are that it is limited to one-dimensional flow and that it requires the development of specific iteration schemes for different categories of boundary-value problems.

In application of previous analytical approaches for drip lateral hydraulic computation, it is generally assumed that the emitter outflow along the lateral is constant. This basic assumption may cause significant errors in hydraulic computation. In this present paper an accurate analytical approach is developed. In the suggested analysis the emitter outflow is considered to be variable along the lateral. The proposed method is demonstrated over a practical range of variation of drip irrigation parameters. A comparison test indicated that the suggested analytical approach produces results that follows closely the accurate numerical solution.

The Soil Conservation Service (SCS) method is widely used to estimate runoff from small- to medium-sized watersheds. The most critical assumption of the SCS method is that the ratio of the actual retention to the potential retention is the same as the ratio of actual runoff to potential runoff, but this assumption has not been theoretically nor empirically justified. This paper shows that the exact relationship between rainfall and runoff in the SCS method can be derived theoretically if two simple but reasonable assumptions are made: (1) The spatial variation of infiltration capacity has an exponential distribution; and (2) the temporal variation of rainfall rate also follows an exponential distribution. A theoretical basis for the SCS method allows an independent validation of the method by testing how rainfall intensity and infiltration capacity actually vary in time and space, respectively.

A study was performed to examine the effects of flume settlement on the accuracy of discharge measurements, using the cutthroat flume in a free-flowing condition. A series of 111 tests was performed on 10.2 cm × 0.91 m, 30.5 cm × 1.37 m, and 61 cm × 1.37 m flumes. Each cutthroat flume was systematically adjusted to simulate settlement in the lateral, longitudinal, and combined lateral-longitudinal directions. Discharge inaccuracies were recorded for each settlement condition. Flume discharge measurement errors ranged from ±3% to ±23% for lateral, longitudinal, and combined lateral-longitudinal slopes of up to 4%. The greatest errors occurred when the lateral slope and/or the longitudinal slope reached or exceeded 3%. An algorithm was developed to adjust the flow measurements for settlement. A procedure was devised for determining the true flow measurement to within ±3% for free-flowing conditions.

Physical solutions for the flow through circular channels due to increase in bed elevations are obtained, satisfying the condition of choke-free flow. In this analysis, three different cases of channel transitions, commonly found in practice, are considered. They are (1) when the channel bed is raised with the formation of a flat base; (2) when the channel bed is raised due to change in position of the centerline; and (3) when the channel bed is raised due to decrease in channel diameter. Considering a smooth and gradual transition zone, the equations of energy and continuity are solved for the subcritical approaching flow regime to determine the maximum limiting rise in bed elevation. Using the continuity equation of the upstream flow, the ranges of the upstream flow depths are also obtained for the choke-free condition.