Document Type : Research Paper
Infiltration and its modeling are important in design and implementation of irrigation projects, drainage, water-resources management, soil-conservation, hydrological studies, and artificial-recharge of groundwater and are influenced by different characteristics of soil and infiltrated water including the amount of salinity and sodium. The aim of this study was to evaluate the infiltration models of Kostiakov-Lewis, Kostiakov, Horton, SCS, and Philip using water salinity levels of 0.5, 3, 6, and 12 dS m-1 and SAR of 0.5, 10, 20, and 30. Infiltration was measured using double rings method with 5 replications on two calcareous soil-series of Daneshkadeh and Kooye-Asatid from Agricultural Research Station of Agricultural-College, Shiraz-University located in Bajgah region and the models were fitted to them. Calculating coefficient of determination (R2), efficiency percentage (EP) of model, and normalized root mean square error (NRMSE), the performance of each applied model was evaluated and ranking of models were performed. The model that ranked the highest was considered as the best and the most stable model. The Kostiakov-Lewis model with R2, NRMSE and EP of 0.997, 0.205 and 99.81 and SCS model with the aforementioned statistics of 0.992, 0.249and 98.76, respectively, were the most and the least suitable models in both of the studied soils and in almost all of the applied water quality treatments. Therefore, the Kostiakov-Lewis model can be used for modeling water infiltration into soil with an acceptable confidence in arid and semiarid regions, where waters with poor quality are often used.