behrouz mehdinejadiani

Soil water characteristic curve (SWCC) represents the quantative relationship between soil water content and matric suction. Direct measurement of this curve is laborious, time consuming and expensive. Hence, several experimental, mathematical and analytical models have been presented to describe the SWCC. In this research, 11 SWCC models were calibrated using volumetric water content and matric suction data of 27 soil samples of Qorveh-Dehgolan plain agricultural lands. Solver Toolbox in Excel used to calibrate the models and six other soil samples data used for result validation. The R2, RMSE, NRMSE and MBE statistics were used to assess the prediction accuracy of the models. According to the obtained results, for clay and loam textures, Libardi et al and Simmons et al models, for silty clay loam texture, Brooks and Corey model, for clay loam texture, Brooks and Corey and Campbell models, for silty clay texture, Van Genuchten m=1-2/n and Brooks and Corey models and for sandy loam texture, power model were proposed to predict soil water characteristic curve in the soils of the studied lands.

Hydrodynamic dispersion rate of solutes in soil is considered as the major parameter for pollution and solutes transport in soil, which is related to pollutant transport distance. As fractal geometry theory and geostatistical theory are capable of explaining and predicting the distance-related phenomena, this research used fractal geometry and geostatistics method for determining dispersivity.
Solutes transport experiment was carried out at 16 points of soil vertical column with a diameter of 10 centimeters and a length of 1 meter and BTCs were extracted at the depth of 6, 12, 18, 24, 30, 36, 42, 54, 48, 60, 66, 72, 78, 90, 84, 96 centimeters from the model bottom. CDE equation was then fitted with the BTCs with respect to the fractal assumptions on dispersivity coefficients.
Findings: With respect to phosphorus absorption experiments in soil, phosphorus adsorption isotherm had the best fitting at 4, 12, 25, 50, 70 mg/l of phosphorus concentrations. The results showed that both methods are capable of predicting changes and increase of dispersivity coefficient in soil column after performing a mean-comparison test. However, fractal geometry method estimated values at a higher accuracy.
Discussion and Result showed that, dispersivity along the sample followed the exponential relation. The regression coefficients of the fractal and geostatistical models in predicting dispersivity values were 0.97 and 0.84, respectively.

In this research, the transport process of a non-reactive contaminant through layered sandy soils and dispersion coefficient behavior in different configurations of these soils were investigated. Tracer tests included pollution and leaching experiments conducted in a sandbox with dimensions of 130 cm×10 cm×60 cm. Three types of homogeneous sandy soils (coarse, medium and fine) as a control and six types of layered sandy soils were studied. The results showed that in the homogeneous sandy soils, the dispersion coefficient increased with increasing particle size of sand. For example, the dispersion coefficients of the coarse sandy soil in the pollution and leaching experiments were 8.46 and 7.91 orders of magnitude larger than the dispersion coefficients of the fine sandy soil, respectively. In the layered sandy soils, the type of configuration and the order of layers at each configuration influenced dispersion coefficient. In the pollution and leaching experiments, the least values of dispersion coefficients were equal to 12.50 cm2/min and 17.16 cm2/min, respectively, which were observed in the perpendicular configuration with stratification of coarse-medium-fine. However, the most values of dispersion coefficients in the pollution and leaching experiments were equal to 32.56 cm2/min and 37.31 cm2/min, respectively, which were obtained in the parallel configuration with stratification of coarse-fine- medium (from bottom to top). The dispersion coefficients of the layered sandy soils differed significantly from those of the homogeneous sandy soils. The transport process in the layered sandy soils, compared to the homogeneous sandy soils, had a more deviation from Fickian transport.

An Accurate estimation of contaminant transport parameters is one of the most important steps for simulation of contaminant transport in porous media. In this work, the mean pore water velocity and the dispersion coefficient in three kinds of saturated porous media were estimated and their dispersivity variations with scale were investigated. The values of the mean pore water velocity and the dispersion coefficient were obtained by fitting the analytical solution of advection-dispersion equation to experimental breakthrough curves by using software CXTFIT2.1. The required data were collected by conducting tracer tests on a sand tank of internal dimensions 250 ×10 ×60 cm3. The tests were conducted in three kinds of porous media (homogeneous sandy soil, randomly heterogeneous sandy soil and natural heterogeneous sandy soil) under three hydraulic gradients 0.017, 0.025 and 0.034. The results showed that the dispersivities of each three porous media at each three hydraulic gradients increased with scale. Also, it was observed that with increasing the heterogeneity degree, the dispersivity variations with scale increased. Under hydraulic gradient 0.017, the dispersivity of the homogenous porous medium increased as a nonlinear function and in two other hydraulic gradients, its variations followed the linear functions. In the studied heterogeneous porous media, under hydraulic gradient 0.017, the dispersivity variations were erratic. However, in two other hydraulic gradients, its variations were as a power function or an exponential function. Therefore, in general, in a homogeneous aquifer and under high hydraulic gradients and in a heterogeneous aquifer, one can approximate the variations of the dispersivity using a linear function and a power function or an exponential function, respectively.