Thresholds of tile runoff generation and nitrogen transport in intensively managed landscapes

Abstract:

Nonlinear changes in rainfall-runoff generation commonly represent the activation of runoff mechanisms or hydrologic pathways. In watersheds with limited human influence, threshold runoff responses have been observed relative to a combination of antecedent soil moisture and total event precipitation. The ability to link subsurface runoff response patterns to runoff generation mechanisms would similarly improve our understanding of water and nutrient transport in human-modified agricultural catchments. In the Midwestern U.S., the installation of subsurface drainage (“tiles”) to maintain optimal growing conditions for crops has altered hydrologic connectivity between landscapes, groundwater, and streams, with consequences for pollutant transport and transformation. In this study, we identify patterns of rainfall-runoff tile response for an artificially drained agricultural field in Illinois and evaluate how antecedent conditions, including soil moisture, groundwater level, and the presence or absence of crops, control non-linearities. We then link runoff thresholds to observed patterns of nitrate export. Data show that tile runoff displays a threshold relationship with antecedent soil moisture and total event precipitation but with more spread compared to forested hillslope catchments. Analyses of both field data and model simulations using Dhara, a coupled surface-subsurface flow and soil-vegetation interaction model, indicate that antecedent groundwater level is an important control on tile runoff activation and explains the spread. Finally, our results show that event nitrate export can be predicted using runoff thresholds because relatively chemostatic responses lead to a linear dependence of load on total event discharge. However, event-scale nitrate concentration dynamics depends upon additional antecedent conditions imposed by management (i.e., fertilization timing and crop type).