Analysis of best management practices implementation on water quality using the Soil and Water Assessment Tool (SWAT)

Nitrates and phosphorus are two important nutrients required for plant growth. Unfortunately, high discharge levels of these pollutants have led to major problems, affecting both health and the ecological balance. Discharged nutrients entering the Gulf of Mexico from the Mississippi River accelerate the growth of algae. When the algae die, the decomposing microorganisms consume all the dissolved oxygen in the water which thus creates a dead-zone. A vast majority of these nutrients can be traced to agricultural watersheds in the Midwest that are artificially drained in order to make the land suitable for agriculture. The Little Vermillion River (LVR) watershed in east-central Illinois (predominately Vermillion County) is a mainly agricultural tile-drained watershed, where the dominant soil series are Drummer silty clay loam and Flanagan silt loam. Using previously collected data from the watershed outlet (1998-2000) and land management scheduling, a Soil and Water Assessment Tool (SWAT) model was calibrated and validated on a daily basis. Afterward the model was calibrated and validated using the observed data, three common Best Management Practices (BMPs) were evaluated. These BMPs include residue reduced tillage operations, vegetative filter strips, and wetlands. These BMPs were tested on a yearly (1980-2009) basis in order to determine their statistical significance and the associated pollutant reduction potential. SWAT is able to calibrate most storm events relatively well, but large storm events are vastly underestimated on a daily basis. Base flow also tends to be underestimated. These flow problems are correlated with underestimation of daily nitrate discharge for both peak events and base flow. Since SWAT assumes that mineral phosphorus discharge occurs from watershed in surface flow only and not tile flow, mineral phosphorus could not be calibrated. For BMPs, no tillage operation performed the best in reducing nitrates discharge from the watershed. This is thought to be caused by improved soil structure that results from not disturbing the soil. Although filter strips have been proven effective in reducing pollutants discharged from other types of watersheds, they are simulated to not be effective in LVR because most nitrates are discharged via tile drainage. Similarly, wetlands also show little effectiveness because of SWAT modeling tile flow to bypass the wetlands into the reach. Combining filter strips and wetlands with no tillage management operations results in little improvement over only no tillage management