Furrow irrigation with physically based spatially varying infiltration

The variability in infiltration is well documented, but it is not considered by most conventional simulation models of surface irrigation because of the difficulty of characterizing the variability. The focus of this research was to study the effects of considering the spatial variability of infiltration on furrow irrigation performance. To facilitate this study, a kinematic wave furrow irrigation model with a physically based spatially varying or constant infiltration sub-model was developed. The space intervals in the finite difference approximation of the continuity equation were predetermined to coincide with the spatial variability of infiltration.;A simple furrow infiltration sub-model based on the Green and Ampt equation and the one-dimensional lateral movement of the wetting front in a direction perpendicular to the vertical, was used to predict the amount of water infiltrating into a fine textured soil. The semi-variogram of the infiltration function was determined and kriging used to calculate different values of infiltration within strips along a furrow or blocks on the field. An approach to determine the sampling intensity required to characterize the infiltration properties of a furrow irrigated field with a specified level of precision is proposed.;The simulation model was used to study the effects of varying levels of precision of the infiltration function on the predicted performance parameters of irrigations. In addition, comparisons were made between results predicted by models with constant infiltration and spatially varying infiltration with actual results. The results indicate that simulation models of surface flow irrigation that incorporate spatial variability of infiltration predict irrigation performances closer to the actual performance than those that assume spatially constant infiltration. The performance parameters predicted when infiltration was assumed to be spatially constant were worse than the actual parameters. Most of the performance parameters were found to be insensitive to the level of precision of kriged estimates of the infiltration. Geostatistics was shown to be a satisfactory tool for describing spatial variability in any direction of the field. This approach leads to a reduction in data requirement and means that spatial variability considerations will become a practical proposition in irrigation design and evaluation.