JWSS - Journal of Water and Soil Science

---

Increasing soil contamination by chemicals has become an issue of increasing environmental concern. Leaching of chemicals into and through the vadose zone creats serious problems due to the contamination of the soil matrix, soil solution and groundwater. Therefore, in order to study the effect of the preferential flow, macropores and organic matter on mobility and leaching of the metals such as cadmium lead, and zinc, an experiment was conducted as a factorial-split plot based on the completely randomized design with three replications. Three treatments of the undisturbed soil (U), the disturbed soil (D) and the disturbed soil containing 3 percent organic matter (O) were leached by the solutions with the concentration of 20 mg.L^-1 of Cd, Pb, and Zn for a month. Then the concentrations of Cd, Pb and Zn in the leachate were measured at different time intervals. The ANOVA results indicated that the metals had a significant difference in the leachate at 1% and the order of their mobility was: Zn>Pb>Cd. Also, there was a significant difference between different soil treatments at 1% and the concentration of the three metals in U and O treatments was more than their concentrations in D treatment. Furthermore, a significant difference between the time intervals of leaching (pore volumes) was observed at 1%. So that, Cd in leachate of U, O and D treatments indicated a significant difference after leaching for 3, 3 and 5 days, respectively (1%). But, Pb in the leachate of the three soil treatments after leaching for 11 days had a significant difference. Zn concentration only in O treatment had a clear trend at different time intervals of leaching and a significant difference was observed after leaching for 8 days.

---

Direct measurement of soil unsaturated hydraulic conductivity (K(h) or K(θ)) is difficult and time-consuming, and often in many applied models, predicting hydraulic conductivity is carried out according to measurements of soil retention curve and saturated hydraulic conductivity (Ks). However, using KS as a matching point in many procedures may result in over-estimation of unsaturated hydraulic conductivity in dry regions. Therefore, the unsaturated hydraulic conductivity at inflection point of retention curve (Ki) and Ks was used as a matching point to predict K(h). For measurement of K(h), 30 soil samples were collected based on variety of soil texture (8 texture classes from sandy to clay) and other chemical and physical properties. In addition to Ks, K(θ) values of undisturbed samples were measured using multi-step outflow method at matric suctions of 0.1, 0.2, 0.3, 0.5 0.7, 1 bar and inflection point of retention curve by using hanging water column and pressure plate. Then, the measured K(h), and water diffusivity (D(θ)) values were compared to the predicted values of van Genuchten and Brooks and Corey models (with Mualem and Burdine constraint). The results showed that for 80% of the samples, the van Genuchten–Mualem model with Ki was the best model for predicting K(h) (i.e. using Ki as a matching point in the van Genuchten–Mualem model resulted in best fitting to measured data). Also, in 6.7 % of samples (two sandy clay samples), Brooks and Corey-Mualem model with Ki and in 13.3 % soil samples (2 silty clay and 2 silty clay loam samples), van Genouchten–Mualem model had a best fitting to K(h) measured data. Furthermore, in 20 % samples (4 clay loam, and 2 silt loam textures), the accuracy and efficiency of van Genuchten–Mualem with Ki and van Genuchten–Mualem models in predicting K(h) were almost similar. According to t-Student test, the mean of RMSE and GSDER of van Genuchten–Mualem model with Ki was significantly less than van Genuchten–Mualem model at P < 0.01. In 90 percent of samples, van Genuchten-Mualem and Brooks and Corey-Burdine theory had the best fitting to the measured data of water diffusivity, but in some cases van Genuchten-Burdine model with Ki was the best model for predicting D(θ).

---

Direct measurement of soil hydraulic characteristics is costly and time-consuming. Also, the method is partly unreliable due to soil heterogeneity and laboratory errors. Instead, soil hydraulic characteristics can be predicted using readily available data such as soil texture and bulk density using pedotransfer functions (PTFs). Artificial neural networks (ANNs) and statistical regression are two methods which are used to develop PTFs. In this study, the multi-layer perceptron (MLP) neural network and backward and stepwise regression models were used to estimate saturated hydraulic conductivity using some soil characteristics including the percentage of particle size distribution, porosity, and bulk density. Data of 125 soil profiles were collected from the reports of basic soil science and land reclamation studies conducted by Khuzestan Water and Power Organization. The results showed that MLP neural network having Bayesian training algorithm with the greater coefficient of determination (R2=0.65) and the lower error (RMSE =0.04) had better performance than multiple linear regression model in predicting saturated hydraulic conductivity.

---

Soil tillage changes chemical and physical properties which can change the soil available water capacity. For understanding the effect of soil disturbance in cultivated soil on available water, parameter pedotransfer functions of these soils created and their results were compared with measured available water by moisture release curves. For this purpose 54 soil samples were taken from cultivated and non cultivated soils of Abyek-Qazvin area northwesthern Iran. Selected properties of these soils such as particle size distribution, bulk density, organic carbon, calcium carbonate, cation exchange capacity and pH were determined. Soil moisture curve of samples were obtained by pressure plates. Parameter pedotransfer functions were created by Mualem-van Genucten model for non cultivated soil and their available water were predicted. The predicted available water was compared with measured available water. The results illustrated that the predicted results had higher correlation coefficient for moisture of permanent wilting point compared to of field capacity moisture, As the correlation coefficient was inccreased from 0.67 to 0.83 and also the root mean square error (RMSE) reduced from 2.59 to 1.06. So the predicted available water was overestimated.