Evaluation and comparison of infiltration models for estimating infiltration capacity of different textures of irrigated soils: Case of Northern Ethiopia.

Abstract

Determination of infiltration capacity is a very important parameter during the design and evaluation of irrigation systems. Accurate estimation of infiltration rates helps in optimizing irrigation scheduling, preventing soil erosion, and improving water use efficiency. This study was conducted to evaluate and compare selected infiltration models for estimating water infiltration rates of five soil textures in the Shillanat-iv- irrigation scheme in northern Ethiopia. Soil samples were taken from selected sites in the irrigation scheme for determining soil textural classes using the hydrometer texture laboratory analysis and the USDA textural triangle. Soil textural map of the irrigation area was prepared using inverse distance weight interpolation technique in ArcGIS version 10.4. The double ring infiltrometer was used to measure the infiltration rates of different soil textures in the irrigation scheme. Six selected infiltration models namely Kostiakov, Modified Kostiakov, Revised Modified Kostiakov, Philip, Horton, and Novel models were used to estimate infiltration rates for five soil textural classes namely, clay loam, loam, sandy clay loam, clay and sandy loam soils. To evaluate the performance of the models, infiltration rate was measured in 38 points of the study area, out of which 70% of the data was calibrating model parameters and 30% of the data was used for model validation. Parameters values of the infiltration models were optimized using the least-squares errors in SPSS statistical software. Five statistical parameters including the Coefficient of determination (R2), Maximum absolute error (MAE), Bias, Root mean square error (RMSE) and Percentage average error (PAE) were used to evaluate the performance of the infiltration models. Results indicated that the Revised Modified Kostiakov’s, Modified Kostiakov’s, and Novel’s infiltration models had better capability in estimating infiltration rates for clay loam, loam and sandy loam soil textures respectively. Similarly, the Hortons’s model had better performances in estimating infiltration rates of both sandy clay loam and clay soil textures compared to other models. In the design of the irrigation system, the best fitted infiltration models can be used for estimating the infiltration rate of soil textures in the selected irrigation scheme and other sites with similar soil textures.