Evaluation of the DRAINMOD–N II model for predicting nitrogen losses in a loamy sand under cultivation in south-east Sweden

Abstract

The DRAINMOD–N II model (version 6.0) was evaluated for a cold region in south-east Sweden. The model was field-tested using four periods between 2002 and 2004 of climate, soil, hydrology and water quality data from three experimental plots, planted to a winter wheat–sugarbeet–barley–barley crop rotation and managed using conventional and controlled drainage. DRAINMOD–N II was calibrated using data from a conventional drainage plot, while data sets from two controlled drainage plots were used for model validation. The model was statistically evaluated by comparing simulated and measured drain flows and nitrate–nitrogen (NO 3–N) losses in subsurface drains. Soil mineral nitrogen (N) content was used to evaluate simulated N dynamics. Observed and predicted NO 3–N losses in subsurface drains were in satisfactory agreement. The mean absolute error (MAE) in predicting NO 3–N drainage losses was 0.16 kg N ha −1 for the calibration plot and 0.21 and 0.30 kg N ha −1 for the two validation plots. For the simulation period, the modelling efficiency ( E) was 0.89 for the calibration plot and 0.49 and 0.55 for the validation plots. The overall index of agreement ( d) was 0.98 for the calibration plot and 0.79 and 0.80 for the validation plots. These results show that DRAINMOD–N II is applicable for predicting NO 3–N losses from drained soil under cold conditions in south-east Sweden.