Predictability of Philip and Kostiakov Infiltration Models under Inceptisols in the Humid Forest Zone, Nigeria

Abstract

Knowledge of the infiltration of water into a soil is very important for efficient soil and water management and conservation, especially when the water supply is through rainfall. For efficient irrigation water management, once field infiltration values are constant and the curve established for a particular soil, it is possible to determine during irrigation how long it will take to infiltrate a certain amount of water. Ring infiltrometer measurements were carried out at 10 m intervals down a 5% Inceptisol toposequence of 100 m length in a humid forest in southern Nigeria. The aim was to determine the infiltration capacity of the soil with slope positions and to fit the infiltration data into the Philip and Kostiakov infiltration models in order to quantify the hydrological behavior of the soil and the ability of these models to predict infiltration into the Inceptisols of a humid forest. The nonlinear least square procedure was employed to determine the parameters of the models—namely, the transmissivity and
sorptivity of Philip’s model and the initial infiltration and the index of soil sorptivity reflecting the rate of declining infiltration capacity of Kostiakov’s model. The results indicated that the initial infiltrationand cumulative infiltration ranged from 2.0 to 7.50 cm min^-1 and from 84.90 to 555.20 cm after 6 h of elapsed time, respectively. The coefficient of determination was near unity indicating the models were a good fit for data and could be used to predict infiltration for the studied soil. Transmissivity values ranged from 0.14 to 1.29, putting the soil conductivity class between ‘very slow’ and ‘slow’. Sorptivity was very high with values in the range 1.20 to 8.08. The index of sorptivity of soil ranged from 0.65 to 0.88 indicating fairly high soil sorptivity and the initial infiltration of Kostiakov’s model ranged from 0.03 to 1.09. Goodness of fit was used to compare the measured and predicted infiltration values and showed that 10% and 50% disparities existed for initial infiltration and 30% and 70% disparities for cumulative infiltration under Philip’s and Kostiakov’s models, respectively. It can be deduced that Philip’ s model was more suitable than Kostiakov’s model for predicting water infiltration in Inceptisols of the humid forest zone of Nigeria.