Simulation of Nitrate and Ammonium Ions Leaching in a Sandy Loam Soil using Analytical and Numerical Models

Aquifers are vulnerable to pollution from industrial and agricultural activities, the share of the agricultural sector is higher than the rest. Nitrogen fertilizers have the highest consumption in agriculture. Nitrate ion with negative charge is not absorbed by soil particles; therefore, it is subjected to surface and ground water leaching which is more intensive in sandy loam. Analytical and numerical models are used to investigate nitrate transport between soil and groundwater and its effect on groundwater contamination. The application of these models depends on the determination of dispersion and retardation factors. Therefore, quantitative estimation of these factors is necessary to solve the problems related to solute and metal transport in the soil. The parameters were estimated by comparing laboratory and field data versus theoretical ones. The objective of this study is to determine dispersion and retardation factors of nitrate and ammonium ions with three different methods including breakthrough curve (BTC), least square and HYDRUS models in a saturated sandy loam soil. The study was conducted in the soil columns of 50 centimeter and 10.5 diameter with three replications. Before leaching, ammonium nitrate fertilizer was added to soil columns with concentration of 10 g per liter. Concentration of nitrate and ammonium in leached water at the end of soil column with time, commonly known as the breakthrough curve (BTC), is determined. BTC as the first method, resulting from a step input of solute is often of sigmoidal shape and the dispersion and retardation factors are determined with this curve. The second method was least square one. In this method an error function model that fits to a breakthrough curve is presented with two unknown parameters. The parameters can be estimated by using laboratory data and a least square method. The last method is HYDRUS model. In HYDRUS model, the convection-dispersion and mobile-immobile models through inverse modeling were used to estimate the parameters. Dispersion and retardation factors for nitrate ion were in the range of 0.09-0.15 and 0.44-0.48 and for ammonium ion were in the range of 0.042-0.053 and 0.24-0.37, respectively. For quick and accurate estimation of dispersion and retardation factors from a soil column data, three methods were used in this study. The models applied in this study have almost identical results in estimating dispersion and retardation factors. The nitrate ion has been absorbed into the soil particles more than ammonium ion and consequently has a lower dispersion and retardation factors and less leaching than ammonium ion.