Modeling complexity in simulating pesticide fate in a rice paddy

Abstract

Modeling approaches for pesticide regulation are required to provide generic and conservative evaluations on pesticide fate and exposure based on limited data. This study investigates the modeling approach for pesticide simulation in a rice paddy, by developing a component-based modeling system and characterizing the dependence of pesticide concentrations on individual fate processes. The developed system covers the modeling complexity from a “base model” which considers only the essential processes of water management, water–sediment exchange, and aquatic dissipation, to a “full model” for all commonly simulated processes. Model capability and performance were demonstrated by case studies with 5 pesticides in 13 rice fields of the California's Sacramento Valley. With registrant-submitted dissipation half-lives, the base model conservatively estimated dissolved pesticide concentrations within one order of magnitude of measured data. The full model simulations were calibrated to characterize the key model parameters and processes varying with chemical properties and field conditions. Metabolism in water was identified as an important process in predicting pesticide fate in all tested rice fields. Relative contributions of metabolism, hydrolysis, direct aquatic photolysis, and volatilization to the overall pesticide dissipation were significantly correlated to the model sensitivities to the corresponding physicochemical properties and half-lives. While modeling results were sensitive to metabolism half-lives in water for all fields, significances of metabolism in sediment and water–sediment exchange were only observed for pesticides with pre-flooding applications or with rapid dissipation in sediment. Results suggest that, in addition to the development of regional modeling scenarios for rice production, the registrant-submitted maximum values for the aquatic dissipation half-lives could be used for evaluating pesticide for regulatory purposes.