عنوان مقاله [English]
Infiltration is one the most important components of hydrologic cycle for utilization and managements of water resources. This phenomenon is affected by several factors and its measurement is difficult. Therefore, several models have been introduced for the simulation of infiltration. Previous studies in evaluation of infiltration models show that each model can be preferred over others in a specific condition. Considering this fact, six infiltration models including Philip, Horton, Green-Ampt, SCS, Kostiakov and Luise-Kostiakov and thier parameters were evaluated. Gonbad catchment in Hamedan province was selected and it was divided into four homogenous hydrologic units. Then infiltration was measured during the dry season by double ring. The results showed that Philip model was the most accurate for estimating of the infiltration and Kostiakov model is also the second one. Regression coefficients of Philip model were betwean 0.975 to one, mean error -0.017 to +0.017 and the maximum root mean square error was 0.22. Regression coefficients of Kostiakov model were between 0.956 to 0.998, mean error -9.3 to +0.003 and the maximum root mean square error was 14.25. In addition, the correlation relations between parameters of these models and two soil texture indices were carried out. The findings show that except for SCS model, the other parameters correlate with these two indices in a 5% level and at least one parameter of each model has high correlation with them. Regression coefficient for A in Philip model, B in Green-Ampt, k in Horton, b in Kostiakov and b in Luise-Kostiakov were 0.99, 0.95, 0.99, 0.999 and 0.96 respectively. Results show a lower correlation between the other parameters and the percentage of clay and sand. Based on these results, it is suggested that the regression relations of models’ parameters for different conditions of soil texture and antecedent moisture in a variety of basins should be evaluated and determined.