U.S. Department of Agriculture: Agricultural Research Service, Lincoln, Nebraska


Date of this Version



Agricultural Water Management 132 (2014) 10– 22


This article is a U.S. government work, and is not subject to copyright in the United States.


Well tested agricultural system models can improve our understanding of the water quality effects of management practices under different conditions. The Root Zone Water Quality Model (RZWQM) has been tested under a variety of conditions. However, the current model’s ability to simulate pesticide trans-port to subsurface drain flow over a long term period under different tillage systems and application rates is not clear. Therefore, we calibrated and tested RZWQM using six years of data from Nashua, Iowa. In this experiment, atrazine was spring applied at 2.8 (1990–1992) and 0.6 kg/ha/yr (1993–1995) to two 0.4 haplots with different tillage (till and no-till). The observed and simulated average annual flow weighted atrazine concentrations (FWAC) in subsurface drain flow from the no-till plot were 3.7 and 3.2 µg/L, respectively for the period with high atrazine application rates, and 0.8 and 0.9 µg/L, respectively for the period with low application rates. The 1990–1992 observed average annual FWAC difference between the no-till and tilled plot was 2.4 µg/L while the simulated difference was 2.1 µg/L. These observed and simulated differences for 1993–1995 were 0.1 and 0.1 µg/L, respectively. The Nash–Sutcliffe model performance statistic (EF) for cumulative atrazine flux to subsurface drain flow was 0.93 for the no-till plotted sting years (1993–1995), which is comparable to other recent model tests. The value of EF is 1.0 when simulated data perfectly match observed data. The order of selected parameter sensitivity for RZWQM simulated FWAC was atrazine partition coefficient > number of macro-pores > atrazine half life in soil > soil hydraulic conductivity. Simulations from 1990 to 1995 with four different atrazine application rates applied at a constant rate throughout the simulation period showed concentrations in drain flow for the no-till plot to be twice those of the tilled plot. The differences were more pronounced in the early simulation period (1990–1992), partly because of the characteristics of macro-pore flow during large storms. The results suggest that RZWQM is a promising tool to study pesticide transport to subsurface drain flow under different tillage systems and application rates over several years, the concentrations of atrazine in drain flow can be higher with no-till than tilled soil over a range of atrazine application rates, and atrazine concentrations in drain flow are sensitive to the macro-pore flow characteristics under different tillage systems and rainfall timing and intensity.