Goodness of Fit for Eight Water Infiltration Models

Abstract

Eight infiltration models were investigated for their capability to describe water infiltration into soils. The models were; Philip's, Modified Philip's, Kostiakov's, Talsma and Parlange's, Horton's, and Lewis and Milne's. Field measurements of water infiltration were made using single ring infiltrometers on field plots established on a Norwood soil (Typic Udifluvents) with four long‐term (27 yr) winter cover crop treatments followed by cotton (Gossypium hirsutum L.) each year. The four treatments were; (i) common vetch (Vicia sativa), (ii) wheat (Triticum aestivum L.) + 60 kg/ha N; (iii) 60 kg/ha N, and (iv) check (no fertilizer or winter cover). Measured infiltration rates for the winter cover treatments were significantly higher than for the check plots. With the exception of Lewis and Milne's, all models showed good agreement with measured infiltration. Model parameters obtained by best fit of measured data, were significantly different among the winter cover treatments. Based on R2 values, Horton, Kostiakov and Philip equations provided best predictions over all other models. Soil bulk density and saturated hydraulic conductivity for two soil depths were not affected by the winter cover treatments.

Prediction of Wetting Front Movement During One‐Dimensional Infiltration into Soils

“…The pressure and conductivity were determined as functions of the water content from transient flow data. Measured wetting front depths and the depths predicted from the modified Green-Ampt and Philip models closely agree, as shown inFigure 4 Davidoff and Selim [1986]. performed field infiltration experiments on Norwood fine to silty sandy loam with an initial saturation of about 0.37.…”

“…The need for the complete K(h) curve in most solution schemes has limited the ability to validate the models. Attempts at validation typically fit proposed models to cumulative infiltration or wetting front depth measurements by adjusting the parameters rather than using independently measured parameter values [Davidoff and Selim, 1986;Hogarth et al, 1989;Whisler and Bouwer, 1970]. Of the models described the Green-Ampt formulation requires the least parameter characterization, which explains in large part its success: the reduction in measurement…”

Evaluation of Infiltration Models in Soils of Two Land Use Systems in Calabar, Nigeria

“…Infiltration characteristics can mostly be evaluated under either ponded or rainfall conditions or predicted using infiltration models (Haverkamp et al, 1988;Majaliwa and Tenywa, 1998). Several studies have been conducted to quantify the infiltration process (Green and Ampt, 1911;Kostiakov, 1932;Horton, 1940;Philip, 1957a;Talsma and Parlange, 1972;Rao et al, 2006), as well as evaluate infiltration models either for the purpose of validation to establish the model parameters or comparison of model efficiencies and applicability for different soil conditions (Ahmed, 1982;Bach et al, 1986;Davidoff and Salim, 1986;Obiechefu, 1991;Topaloglu, 1999;Wudduvira et al, 2001;Igbadun and Idris, 2007;Egbai et al, 2011) because the models are strongly subjected to soil spatial variability (Dashtaki et al, 2009). Al-Azawi (1985), reported that a lack of knowledge of the model parameters for different soils and locations makes the use of these models difficult.…”

Site-Dependence Performance of Infiltration Models

“…The value of a, c and m cab be obtained from curve-fittings (Davidoff and Selim 1986). While the above discussed models have been well addressed in the literature (e.g.…”