ARS | Publication request: APEX Model Assessment of Variable Landscapes on Runoff and Dissolved Herbicides

Submitted to: ASABE Annual International Meeting
Publication Type: Proceedings/Symposium
Publication Acceptance Date: June 29, 2008
Publication Date: June 29, 2008
Citation: Mudgal, A., Baffaut, C., Anderson, S., Sadler, E.J., Thompson, A.L. 2008. Apex model assessment of variable landscapes on runoff and dissolved herbicides. In: Proceedings of the American Society of Agricultural and Biological Engineers Annual International Meeting, June 29 - July 02, Providence, Rhode Island. Paper No. 084498.

Interpretive Summary: Slope profiles in fields can have significant effects on runoff, soil erosion and deposition as well as transport of pesticides and nutrients. Simulation models are one way in which these effects can be assessed. This study evaluated the effects of slope profile on runoff and dissolved atrazine (an herbicide used on corn and sorghum) using the farm- and field-scale Agricultural Policy/Environmental eXtender (APEX) model. Fourteen agricultural plots (18 x 189 m2 [60 x 620 ft2]) in the Goodwater Creek watershed, a 7250 ha (16,000 acres) agricultural area in north-central Missouri, were simulated with APEX. Plots were under three different tillage and herbicide management systems for three grain crop production systems. Each plot¿s profile was characterized by three slope elements: summit (usually at the top of a slope), backslope (usually in the main part or side of a slope), and footslope (usually at the bottom of a slope). Runoff was measured and samples were collected at the bottom of these plots from 1997 to 2002 when corn was growing. Runoff samples were analyzed for dissolved atrazine. Data from 1997 to 1999 were used to adjust the parameters of the APEX model and data from 2000 to 2002 were used to verify that the model correctly simulated runoff and atrazine concentrations without further parameter adjustment. The model was then used to simulate different orders of the slope elements as well as different lengths of each element, keeping the steepness of each element constant. Simulated results indicate that as the length of the backslope increased, so did the volume of runoff discharged and the atrazine concentrations at the plot outlet. In addition, the highest level of simulated runoff occurred when the backslope position was located just before the outlet. Results from this study will be helpful to managers in placement of conservation practices on sensitive landscapes for improvement in water quality.

Technical Abstract: Variability in soil landscapes and their associated properties can have significant effects on erosion and deposition processes that affect runoff and transport of pesticides and nutrients. Simulation models are one way in which the effects of landscapes on these processes can be assessed. This study evaluated the effects of variations in landscape position on runoff and dissolved atrazine using a calibrated APEX model. Fourteen agricultural plots (18 x 189 m2) in the Goodwater Creek watershed, a 7250 ha agricultural area in north-central Missouri, were simulated with the farm- and field-scale Agricultural Policy/Environmental eXtender (APEX) model. Plots were under three different tillage and herbicide management systems for three grain crop production systems. Each plot contained three landscape positions: summit, backslope, and footslope along with two transition zones. Runoff was measured and samples were collected from 1997 to 2002 during the corn year of the crop rotations. Runoff samples were analyzed for dissolved atrazine. The APEX model was calibrated and validated with event data from each plot during the corn growing years from 1997 to 1999 and 2000 to 2002, respectively. APEX reasonably simulated runoff and dissolved atrazine concentrations with annual coefficients of determination (r2) values ranging from 0.60 to 0.98 and 0.52 to 0.97, and Nash-Sutcliffe efficiency (NSE) values ranging from 0.46 to 0.94 and 0.45 to 0.86 for calibration and validation, respectively. The calibrated model was then used to simulate variable sequencing of landscape positions and associated soil properties as well as variable lengths of landscape positions. Simulated results indicate that as the length of the backslope increased, so did the volume of runoff discharged and the atrazine concentrations at the plot outlet. In addition, the highest level of simulated runoff occurred when the backslope position was located adjacent to the outlet. Results from this study will be helpful to managers in placement of conservation practices on sensitive landscapes for improvement in water quality.