Skip to main content
ARS Home » Research » Publications at this Location » Publication #280566

Title: Evaluation of revised subsurface tile drainage algorithms in SWAT for a cold climate

Author
item Moriasi, Daniel
item Gowda, Prasanna
item Arnold, Jeffrey
item MULLA, DAVID - University Of Minnesota
item ALE, SRINIVASULU - Texas A&M University

Submitted to: Annual International SWAT Conference
Publication Type: Proceedings
Publication Acceptance Date: 4/9/2012
Publication Date: 7/20/2012
Citation: Moriasi, D.N., Gowda, P., Arnold, J.G., Mulla, D.J., Ale, S. 2012. Evaluation of revised subsurface tile drainage algorithms in SWAT for a cold climate. Annual International SWAT Conference, July 16-20, 2012, New Delhi, India. Avaliable: http://swat.tamu.edu/media/69009/swat-proceedings-2012-india.pdf.

Interpretive Summary: Subsurface tile drains in agricultural systems of the Mid-west U.S. are a major contributor of nitrate loadings to the Mississippi River Basin and contribute to hypoxic conditions in the northern Gulf of Mexico. Development of strategies to reduce nitrate loadings from these agricultural systems require better understanding of the role of subsurface tile drains on flow. In this study, tile drainage algorithms in the Soil and Water Assessment Tool (SWAT) model were revised and long term (1983-1996) monitoring data on subsurface tile drain flow was used to evaluate the revised model. Measured tile drain flow was obtained from three continuous corn plots located in the University of Minnesota’s Agricultural Experiment Station near Waseca, Minnesota. Water, crop, and nutrient management practices on these plots were typical of the Upper Mid-western U.S., where tile drains are essential for agricultural production by draining water from shallow water tables to allow timely tillage and planting operations. Tile flow computations in SWAT are heavily driven by water table depth, which is a function of soil water movement. However, the traditional method, which computes the retention parameter in SWAT as a function of soil profile water content, generally over predicts runoff in poorly drained soils such as those in the Mid-west U.S. The retention parameter is used to compute daily curve number (CN) for estimating surface runoff. This paper presents 1) modifications made to potential maximum soil moisture retention parameter algorithms in SWAT to account for the effects of tile drainage on the computation of surface runoff using the CN method in poorly drained agricultural watersheds and 2) calibration and validation of the modified SWAT model for subsurface tile drain flow for the Upper Mid-west using long term monitoring data. The retention parameter was increased to account for the effect of tile drainage, which is not accounted for in the CN tables. Predicted annual water budgets, including surface runoff, were similar to those reported for the same study area with the Agricultural Drainage and Pesticide Transport (ADAPT) Model. Comparison of monthly tile drain flows predicted by SWAT to measured data indicated excellent agreement. Monthly calibration and validation Nash-Sutcliffe efficiency (NSE) values of 0.77 and 0.78, respectively, the percent bias (PBIAS) values of -1% and 5%, respectively, and root mean square error (RMSE) values of 2.9 mm and 2.0 mm, respectively, were obtained with the modified SWAT. The validated revised tile flow algorithms in SWAT will be useful for modeling the impacts of tile drain spacing and depth on nitrate losses.

Technical Abstract: Subsurface tile drains in agricultural systems of the Mid-west U.S. are a major contributor of nitrate loadings to the Mississippi River Basin and contribute to hypoxic conditions in the northern Gulf of Mexico. Development of strategies to reduce nitrate loadings from these agricultural systems require better understanding of the role of subsurface tile drains on flow. In this study, tile drainage algorithms in the Soil and Water Assessment Tool (SWAT) model were revised and long term (1983-1996) monitoring data on subsurface tile drain flow was used to evaluate the revised model. Measured tile drain flow was obtained from three continuous corn plots located in the University of Minnesota’s Agricultural Experiment Station near Waseca, Minnesota. Water, crop, and nutrient management practices on these plots were typical of the Upper Mid-western U.S., where tile drains are essential for agricultural production by draining water from shallow water tables to allow timely tillage and planting operations. Tile flow computations in SWAT are heavily driven by water table depth, which is a function of soil water movement. However, the traditional method, which computes the retention parameter in SWAT as a function of soil profile water content, generally over predicts runoff in poorly drained soils such as those in the Mid-west U.S. The retention parameter is used to compute daily curve number (CN) for estimating surface runoff. This paper presents 1) modifications made to potential maximum soil moisture retention parameter algorithms in SWAT to account for the effects of tile drainage on the computation of surface runoff using the CN method in poorly drained agricultural watersheds and 2) calibration and validation of the modified SWAT model for subsurface tile drain flow for the Upper Mid-west using long term monitoring data. The retention parameter was increased to account for the effect of tile drainage, which is not accounted for in the CN tables. Predicted annual water budgets, including surface runoff, were similar to those reported for the same study area with the Agricultural Drainage and Pesticide Transport (ADAPT) Model. Comparison of monthly tile drain flows predicted by SWAT to measured data indicated excellent agreement. Monthly calibration and validation Nash-Sutcliffe efficiency (NSE) values of 0.77 and 0.78, respectively, the percent bias (PBIAS) values of -1% and 5%, respectively, and root mean square error (RMSE) values of 2.9 mm and 2.0 mm, respectively, were obtained with the modified SWAT. The validated revised tile flow algorithms in SWAT will be useful for modeling the impacts of tile drain spacing and depth on nitrate losses.