Integrated agricultural systems modeling represents an effective research tool to meet the evolving challenges facing agricultural production and environmental quality. An integrated, process-based model was developed to simulate the impacts of the changing environment and different water and farming management practices on the hydrology, water quality, and crop growth and yield for artificially drained cropping systems. The new model; named DRAINMOD–DSSAT, was developed by integrating three different process based models: the hydrological model, DRAINMOD; the soil carbon and nitrogen (N) dynamics model; DRAINMOD-NII, and selected crop modules of the DSSAT–CSM model; CROPGRO and CERES-Maize. The integration of the three component models is implemented at the source code level and allows for daily interactions and feedback among simulated climatic conditions, soil water and nitrogen, and crop growth. DRAINMOD–DSSAT performance was evaluated using a 10-yr dataset collected from a corn–soybean production system on a subsurface drained field in Iowa, with corn receiving low, medium, and high N fertilization rates. The model was calibrated using the data collected from the high-N treatment, and validated for the other two treatments. Annual and monthly subsurface drainage outflows were predicted with modeling efficiencies (NSE) of 0.95 and 0.83, respectively. The NSE's for annual and monthly NO3–N mass losses were 0.87 and 0.70 for the high N-treatment, 0.93 and 0.86 for the medium N-treatment, 0.94 and 0.67 for the low N-treatment, respectively. Predicted and measured crop yields were accurately predicted with an absolute percent error less than 8% in 27 of the 30 simulated plot-years (3 plots×10yrs). Nitrogen removal in crop grain was reasonably predicted. This first model application suggested the potential capability of DRAINMOD–DSSAT of simulating the hydrology, water quality, and crop growth and yield for corn and soybean production on artificially drained fields in response to varying climatic conditions and nutrient management practices. Further research, using more intense field measurements, is needed to validate the model for its intended use.
