Assessing the impacts of crop-rotation and tillage on crop yields and sediment yield using a modeling approach

This study was conducted in the Big Sunflower River Watershed (BSRW), north-west, Mississippi. The watershed has been identified as impaired waters under Section 303(d) of the Federal Clean Water Act due to high levels of sediment and total phosphorus. This excess is then transported to the Gulf of Mexico via the Yazoo River, further damaging the nation's water resources. The specific objectives of this study were to assess the impact of corn (Zea mays L), soybean (Glycine max (L) Merr., and rice (Oryza sativa, L) crop-rotations (corn after soybean, soybean after rice, continuous soybean) and tillage practices (conventional, conservation, no-till) on crop yields and sediment yield using the Soil and Water Assessment Tool (SWAT) model. The SWAT model was calibrated from January 2001 to December 2005 and validated from January 2006 to September 2010 for monthly stream flow with good to very good performance [coefficient of determination (R-2) values from 0.68 to 0.83 and Nash Sutcliffe Efficiency index (NSE) values from 0.51 to 0.63] using stream flow data from three spatially distributed USGS gage stations within the BSRW. The SWAT model was further calibrated for corn and soybean yields from research fields at Stoneville and validated using research fields at the Clarksdale experiment stations with fair to excellent statistics (R-2 values from 0.43 to 0.59 and NSE values from 0.34 to 0.96). The SWAT model simulation results suggested that corn yields were greater in the corn after soybean rotation under conventional tillage (mean = 9.88 Mg ha(-1)) than no-tillage (mean = 8.81 Mg ha(-1)) practices. However, tillage practices had no effects on soybean yield for the corn after soybean rotation. Soybean yields under conventional tillage practice indicated greater yields (mean =3.01 Mg ha(-1)) for the soybean after rice rotation than for soybean after corn. Continuous soybean under conventional tillage had the lowest simulated crop yield (mean = 2.07 Mg ha(-1)) and the greatest sediment yield (5.2 Mg ha(-1)) in this study. The cumulative (1981-2009) sediment yield at the end of the simulation period (2009) indicated a maximum difference of about 8 Mg ha(-1) between no-till and conventional tillage practices, with no-till contributing the lowest sediment yield. The cumulative difference of the sediment yield between no-till and conservation till was about 2 Mg ha(-1). The results of this study will help to better understand the impact of management practices on watershed crop management and water quality improvement within the BSRW. This information can be applied to other agricultural watersheds. (C) 2012 Elsevier B.V. All rights reserved.