Simulating nitrogen leaching and turnover in a subsurface-drained grassland receiving animal manure in Northern Germany using DRAINMOD-N II

The primary objective of this study was to evaluate the nitrogen fate and transport simulation model DRAINMOD-N II for drained permanent grassland soils. Since the plant component of DRAINMOD-N II was developed for annual row crop systems, it was modified using an empirical approach to consider perennial grasses. The model was field-tested using a 12-year (1989-2000) data set of field hydrology, non-reactive tracer transport, and carbon (C) and nitrogen (N) dynamics from a tile drained grassland research site (Infeld, North-West Germany). Model calibration was performed using the first half of the data set, followed by validation using the second half. Model simulations were visually and statistically compared to field measurements. Modified DRAINMOD-N II successfully predicted drain flow during the validation period. The model could also simulate the observed dynamics of weakly preferential tracer transport by using a high longitudinal dispersivity. Furthermore, the model well described organic carbon (OC) dynamics during calibration. Since there were no OC measurements during the second half of the study, OC model predictions were not validated. Simulated and measured N losses via drain flow were in agreement during model calibration. The model, however, substantially under-predicted N loads during the validation period. Two reasons were hypothesized for the relatively poor performance of DRAINMOD-N II during the validation period. First, the simplistic approach used in the modified DRAINMOD-N II to quantify grass biomass yield did not take into account the effect of soil water dry or wet stresses on plant growth and biomass yield. This hypothesis could not be tested since there were no measurements of grass biomass yield. Secondly, the effect of the calibration errors in N process rates increased with the difference in precipitation patterns between the calibration and validation periods. This is despite the fact that calibrated transformation rates were within published ranges. While inconclusive, these results indicates that a more robust approach for quantifying grass biomass and N uptake may be needed for the current version of DRAINMOD-N II in order to successfully simulate C and N dynamics in drained grassland. (c) 2007 Elsevier B.V. All rights reserved.