Calibration of DRAINMOD in South Dakota for Houdek Soil Series (State Soil of South Dakota)

Abstract. Subsurface drainage has been increasing in South Dakota over the last several years. One of the challenges in drainage water management is the accurate estimation of subsurface drainage flow. The objective of this study was to calibrate DRAINMOD for a South Dakota State soil series to perform long term simulations for developing drainage design criteria, and to compare them to existing design criteria developed for the eastern U.S. DRAINMOD requires the following inputs: weather data, soil properties, site characteristics, drainage system parameters, and input related to crops. A site location near Flandreau, SD in Moody County with a Houdek soil was used for calibrating DRAINMOD. The model was calibrated using Soil Survey Geographic Dataset (SSURGO) soil input with Rosetta pedotransfer functions (PTF) to estimate the required soil input parameters, measured weather data, and observed water table depth and tile flow. The calibrated model showed satisfactory performance in simulating daily water table fluctuation throughout the growing season (April through October 2005) with average absolute error (AAE) of 11.2 cm and Nash-Sutcliffe efficiency (NEF) of 0.68. The results will be used to perform long-term simulations to determine the optimum drainage design intensity to maximize crop yield for a common soil in South Dakota.