Three models were evaluated for their accuracy in simulating pesticide runoff at the edge of agricultural fields: Pesticide Root Zone Model (PRZM), Root Zone Water Quality Model (RZWQM), and OpusCZ. Modeling results on runoff volume, sediment erosion, and pesticide loss were compared with measurements taken from field studies. Models were also compared on their theoretical foundations and ease of use. For runoff events generated by sprinkler irrigation and rainfall, all models performed equally well with small errors in simulating water, sediment, and pesticide runoff. The mean absolute percentage errors (MAPEs) were between 3 and 161%. For flood irrigation, OpusCZ simulated runoff and pesticide mass with the highest accuracy, followed by RZWQM and PRZM, likely owning to its unique hydrological algorithm for runoff simulations during flood irrigation. Simulation results from cold model runs by OpusCZ and RZWQM using measured values for model inputs matched closely to the observed values. The MAPE ranged from 28 to 384 and 42 to 168% for OpusCZ and RZWQM, respectively. These satisfactory model outputs showed the models' abilities in mimicking reality. Theoretical evaluations indicated that OpusCZ and RZWQM use mechanistic approaches for hydrology simulation, output data on a subdaily time-step, and were able to simulate management practices and subsurface flow via tile drainage. In contrast, PRZM operates at daily time-step and simulates surface runoff using the USDA Soil Conservation Service's curve number method. Among the three models, OpusCZ and RZWQM were suitable for simulating pesticide runoff in semiarid areas where agriculture is heavily dependent on irrigation.
Evaluation of Three Models for Simulating Pesticide Runoff from Irrigated Agricultural FieldsXuyang Zhang* and Kean S. Goh O ff-site movement of pesticides from applied agricultural areas has been recognized as one of the major contributors to the contamination of surface waters worldwide (Schulz, 2004;Gangbazo, 1999;Humenik et al., 1987;Line et al., 1997;Loague, 1998). Pesticides move into surface water via drift, surface runoff, or subsurface flow. Among these routes, surface runoff generated by rainfall events has attracted the most attention (Schulz, 2004). In semiarid regions, such as California, pesticide runoff occurs not only during the rainy season but also during the dry growing season between March and October when the crops are irrigated and pesticides are applied. Surface runoff generated by irrigation events has been identified as a major cause for the detection of pesticides in agricultural areas of California during the dry season (Starner et al. 2005;Starner, 2009;Foe, 1995).To assess the ecological risks of pesticides in surface water, mathematical models that predict exposure to pesticides have been increasingly used in addition to water quality monitoring. Prediction of pesticide loss at the edge of a field is fundamental to exposure assessment both at local and watershed scales. In agricultural lands, field applicatio...