Field evaluation of infiltration models in lawn soils

Abstract

Water infiltration into soil is a critical process in designing irrigation systems, especially those used for the application of wastewater to land. Although some research was conducted to evaluate the performance of proposed infiltration models, few efforts were given to the study of infiltration models specifically in lawn soils. This paper was to compare five existing classic infiltration models with field data collected from lawn soils, to estimate and assess those models’ parameters, and to evaluate their prediction ability for lawn irrigation systems. A field study was conducted in three lawn soils using a double-ring infiltrometer. The performances of infiltration models proposed by Philip, Kostiakov, Mezencev, the United States Department of Agriculture Natural Resources Conservation Service (originally the United States Soil Conservation Service), and Horton were investigated with consideration of sum of squared error, root mean square error, coefficient of determination, adjusted coefficient of determination, mean error, absolute value of mean error, model efficiency, and Willmott’s index of agreement. Mezencev model and Horton model performed better than the other three infiltration models. The research indicated that three-parameter models had the best description of the relationship between cumulative infiltration and time in the researched lawn soils.