Agricultural nitrogen budget for a long-term row crop production system in the Midwest USA

In the Midwestern United States, subsurface drainage (commonly known as tile drains) systems have been extensively used for sustaining agricultural production. However, the tile drains have raised concerns of facilitating the transport of agricultural chemicals from the fields to receiving waters. Data from a long-term field experiment in the Little Vermilion River (LVR) watershed of east-central Illinois, USA, shows that the tile drain systems have contributed to increased nitrate N (NO₃-N) to the receiving water body, Georgetown Lake Reservoir, over time. We conducted more than 10 years of research on fate and transport of NO₃-N in tile drain water, surface runoff and soil N. Corn (Zea mays L.) and soybean (Glycine max L.) were planted in rotation for this watershed. We evaluated N balance (inputs and outputs) and transfer (runoff and leaching) components from three sites with both surface and subsurface flow stations within this watershed, and N budgets for individual sites were developed. Nitrogen fertilizer application (average 192 kg ha⁻¹ y⁻¹) and soil N mineralization (average 88 kg ha⁻¹ y⁻¹) were the major N inputs for corn and soybean, respectively in this watershed. Plant N uptake was the major N output for both crops during this entire study period. Annual N uptake for the LVR watershed ranged from +39 to +148 (average +93) kg ha⁻¹ and −63 to +5 (average −32) kg ha⁻¹, respectively, for corn and soybeans. This data indicates that most of the soil mineralized N was used during soybean production years, while corn production years added extra N in the soil. Surface runoff from the watershed was negligible, however, subsurface leaching through tile drains removed about 18% of the total rainfall. Average NO₃-N concentrations of leaching water at sites A (15 mg L⁻¹) and B (16.5 mg L⁻¹) exceeded maximum contaminant level (MCL; 10 mg L⁻¹) throughout the experiment. However, NO₃-N concentrations from site E (6.9 mg L⁻¹) never exceeded MCL possibly because 15–22% lower N was received at this site. We estimated that the average corn grain yield would need to be 28% higher to remove the additional N from this watershed. Our study suggests that N application schemes of the LVR watershed need to be reevaluated for better N management, optimum crop production, and overall environmental sustainability.