Evaluation of predicted long-term water quality trends to changes in n fertilizer management practices for a cold climate

Objectives of this study were to calibrate and validate a water quality model for monthly flow and NO₃-N losses, and evaluate a set of alternative nutrient management practices to reduce NO₃-N losses in an agricultural watershed. A dynamic watershed scale spatial modeling approach that uses ADAPT, a field scale model and GIS was calibrated to predict monthly flow and NO₃-N losses from a sub-watershed of Seven Mile Creek in south-central Minnesota. It is a 4029-ha watershed with over 85% of the total area under agriculture. Calibration and validation of the model were done using monitoring data from 2000-2002 and 2003-2004, respectively. For the calibration period, the model predicted mean monthly flow and NO₃-N losses of 0.38 m³/s and 4.04 kg/ha, respectively, against measured flow and NO₃-N losses of 0.48 m³/s and 3.77 kg/ha, respectively. For the validation period, the predicted mean monthly flow and NO₃-N losses were 0.29 m³/s and 2.93 kg/ha, respectively, against measured flow and NO₃-N losses of 0.18 m³/s and 1.37 kg/ha, respectively. Long-term simulations were made for a wide range of climatic conditions between 1955 and 2004 to evaluate the effects of fertilizer management practices on the NO₃-N losses. A 35% reduction in NO ₃-N losses was observed when application rate and timing were changed from a fall application of 179.3 kg/ha to a spring application of 112 kg/ha.