Simulating nitrogen dynamics in agricultural soils fertilized with pig slurry and urea

Within the framework of an interregional project in the Emilia Romagna region of northern Italy, the coupled MACRO¡SOILN model was chosen to estimate soil protective capacity against pollutants. The aim of our study was to evaluate the model to better identify key parameters and processes that influence N losses in agricultural soils. Nitrate N content was monitored in soil under corn (Zea mays L.) fertilized with urea and/or pig slurry, in two field experiments performed on four different soils: a Fienili clay, a Barco-like silt, a Sant'Omobono silt loam, and a La Boaria silty clay soil. Measurements were compared with model predictions. For all soils, nitrate content was underestimated on average by 24 to 88% at lower N rates; it was overestimated by 1 to 104% at higher N rates. The root mean square error (RMSE) was equal to 81.1%. Simulation of crop N uptake and soil water flow, estimation of the ammonia losses at pig slurry spreading, and N transformation parameter setting were considered as possible error sources. The calibration of crop N uptake gave rise to good model efficiency index values. The RMSE for the simulation of soil water content varied between 9.8 and 20.2%. A more accurate setting of the ammonia losses and of the feces transformation parameter values could allow the RMSE for the simulation of soil nitrate content to be reduced by no more than 10 to 15%. It is possible for the model not to include the simulation of processes that could have relevant effects on the soil N dynamics.