Prediction of water quality effect on saturated hydraulic conductivity of soil by artificial neural networks

Abstract

This study was conducted to investigate the impact of water salinity (ECw) and sodicity (SARw) on saturated (Ks) and relative (Kr) hydraulic conductivities in two clay (C) and sandy clay loam (SCL) soils. The results showed that the Ks decreased with increasing SARw, and in all of water quality treatments, the Ks of SCL soil was higher than that of the C soil. Sodicity effect (even at high SARw) on the Kr of clay soil was minimized by high salinity. Although Kr of both soils similarly responded to ECw and SARw, microstructure of clay soil was more sensitive to water quality. Effect of ECw on soil structure was greater than that of SARw. In order to assess the applicability of artificial neural networks (ANNs) in estimating Ks and Kr, two types of FFBP and CFBP ANNs and two training algorithms, namely Levenberg–Marquardt (LM) and Bayesian regulation, were employed with two strategies of uniform threshold and different threshold functions. Multiple linear regressions were also used for Ks and Kr prediction. Based on the ANN results of second strategy, best topology (4–5–4–1) was belonged to CFBP network with LM algorithm, LOGSIG–LOGSIG–TANSIG threshold functions, and values of MAE and R2 are equal to 0.1761 and 0.9945, respectively. Overall, the efficacy of ANNs is much greater than regression method for Ks prediction.