Performance and Sensitivity Analysis of Infiltration Equations under Different Initial and Boundary Conditions in Furrow Irrigation

In management of surface irrigation, infiltration function is important. Quantitative understanding of this phenomenon is essential in preventing soil erosion and water losses. The main objective of this study was to estimate the coefficients of infiltration equations (Kostiakov-Lewis, Philip, and Horton) and evaluate the sensitivity and performance of these equations under various initial (initial soil moisture) and boundary conditions (water head on soil surface). Therefore, two-dimensional infiltration for furrow irrigation were simulated by changing the initial soil moisture (before irrigation) and water head on soil surface (during irrigation) using the solution of the Richards’ equation (HYDRUS model). To determine the coefficients of infiltration equations, outputs of the HYDRUS model (cumulative infiltration over time) were fitted. Evaluating the performance of infiltration equations via statistical evaluation indicators showed that infiltration empirical equations (Kostiakov-Lewis and Horton) had better performance than the physical equation (Philip) to estimate the cumulative infiltration under various initial and boundary conditions. Kostiakov-Lewis equation ranked the first in all cases of estimating infiltration by simulations, followed by Horton and Philip equations. Sensitivity assessment of the coefficients of the equations showed the power coefficient (b) of Kostiakov-Lewis equation as the most sensitive coefficient, while the final infiltration rate coefficient ( 15ff"> ) in Horton equation was the least sensitive. The results also showed that Horton equation was the most sensitive equation in evaluation of infiltration equations sensitivity.