A pH‐Based Pedotransfer Function for Scaling Saturated Hydraulic Conductivity Reduction: Improved Estimation of Hydraulic Dynamics in HYDRUS. Issue 1 (1st November 2019)

• Record Type: Journal Article
• Authors: Ali, Aram; Biggs, Andrew J.W.; Šimůnek, Jirka; Bennett, John McL.
• Is Part Of: Vadose zone journal. Volume 18:Issue 1(2019)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-11-01
• Subjects: Soil science -- Periodicals; Zone of aeration -- Periodicals; Groundwater flow -- Periodicals; Groundwater flow; Zone of aeration; Periodicals; Electronic journals; 631.4
• DOI: 10.2136/vzj2019.07.0072
• Languages: English
• ISSNs: 1539-1663

Abstract

Abstract : Core Ideas: Soil‐specific pedotransfer functions can be incorporated into the HYDRUS model. Electrolyte concentration reduces the adverse effect of pH and Na on soil structural stability. Clay content is important in governing soil hydraulic reduction dynamics due to pH. Hydraulic conductivity is a key soil property governing agricultural production and is thus an important parameter in hydrologic modeling. The pH scaling factor for saturated hydraulic conductivity ( K s ) reduction in the HYDRUS model was reviewed and evaluated for its ability to simulate K s reduction. A limitation of the model is the generalization of K s reduction at various levels of electrolyte concentration for different soil types, i.e., it is not soil specific. In this study, a new generalized linear regression model was developed to estimate K s reduction for a larger set of Australian soils compared with three American soils. A nonlinear pedotransfer function was also produced, using the Levenberg–Marquardt optimization algorithm, by considering the pH and electrolyte concentration of the applied solution as well as the soil clay content. This approach improved the estimation of the pH scaling factor relating to K s reduction for individual soils. The functions were based on K s reduction in nine contrasting Australian soils using two sets of treatment solutions with Na adsorption ratios of 20 and 40; total electrolyte concentrations of 8, 15, 25, 50, 100, 250, and 500 mmolc L −1 ; and pH values of 6, 7, 8, and 9. A comparison of the experimental data and model outputs indicates that the models performed objectively well and successfully described the K s reduction due to the pH. Further, a nonlinear function provided greater accuracy than the generalized function for the individual soils of Australia and California. This indicates that the nonlinear model provides an improved estimation of the pH scaling factor for K s reduction in specific soils in the HYDRUS model and should therefore be considered in future HYDRUS developments and applications.