Accuracy of the Hydrus Model to Predict Nitrate Leaching as a Result of Applying Treated Sewage Sludge to Soils of Different Textures

ABSTRACT

Due to its organic matter and nutrient content, sewage sludge presents a good potential for use in reforestation and reclamation of degraded lands. However, its application poses water contamination risks, due to the high mobility of compounds such as nitrate (carcinogenic). On the other hand, mathematical modeling is becoming an important tool for the evaluation of risks of water contamination by nutrients and pesticides. The aim of the present study was to evaluate the global accuracy of the Hydrus 1-D model (version 3.0) to predict nitrate leaching within the soil profile. For this purpose, experimental data from a laboratory leaching study, considering three different soils and submitted to four doses of sewage sludge, were fed into the Hydrus model, after calibrating the parameters of the Van Genuchten-Mualen equation, using the percolated volumes from the experiment as the objective function. Mean nitrate concentrations in the leachate, collected at the bottom of the experimental columns, were then compared with nitrate concentrations predicted by the Hydrus model. The mean relative error between the observed (experiment) and calculated (model) values was -1.97%, and the R² was 0.96. The Nash-Sutcliffe coefficient was E=0.95, indicating a very high model accuracy in the prediction of nitrate leaching, under the studied conditions

Keywords:
Mathematical modeling; Hydrus 1-D; Leaching; Nitrate