To design the irrigation systems, especially, the surface irrigation system, it’s necessary to know about soil infiltration characteristics. Infiltration depends on soil properties, and irrigation methods. One of the presented equations for soil water infiltration is the Kostiakovʹs equation, which is used by researchers around the world due to its simplicity and obtaining acceptable results. For this purpose, Kostiakovʹs equation of infiltration was used in this study. In this research, an efficient method was proposed for estimating the infiltration coefficients and Manning roughness coefficient based on comparison between the real advance curve and itʹs calculated using the numerical solution of the Saint-Venant equations (full hydrodynamic model). Subsequently, the result of the proposed method was compared with Elliot and Walkerʹs two-point method and IPARM computer model. In this proposed method, the numerical solution was calculated by coding and optimizing in MATLAB software. In optimization process, the objective function, was the difference between calculated and observed advance curves and decision variables were the infiltration coefficients of the Kostiakovʹs equation and the Manning roughness coefficient. To eva‎luate the accuracy of the proposed method, two series of furrow irrigation field data were used. The computed infiltration volume using all these above-mentioned methods and the actual infiltration volume were compared. The actual infiltration volume was obtained from the difference between input and tail water volumes and the computed infiltration volume was determined by replacing the intake opportunity at any point of furrow in the infiltration equation. This proposed model with a value of percentage error 8.16 had the least error in estimating the total volume of infiltrated water in six irrigation events. However, The IPARM computed model and the Elliott and Walker two-point method indicated results of 39 and 49 percent, respectively. Additionally, correlation coefficient between of the computational infiltration volume using this proposed method and the actual volume, had an average value of 0.71, which is the most correlation compared to the other two methods. Based on t-test, the difference between the three proposed methods, the IPARM computer model, and Elliott and Walkerʹs two-point method is significant in estimating the infiltration coefficients at the level of five percent. Finally, the sensitive analyzing the simulated advanced phase, related to the infiltration coefficients of the Kostiakov equation and Manning roughness coefficient, indicated that the simulated advanced phase is more sensitive to the infiltration coefficients and less sensitive to Manning roughness.

محمد شايان نژاد

منوچهر حيدر پور، محمد علي حاج عباسي