AN ARTIFICIAL NEURAL-NETWORK MODEL FOR SIMULATING PESTICIDE CONCENTRATIONS IN SOIL

The simulation of pesticide concentrations in soil requires knowledge of complex physico-chemical processes that pesticides undergo, in both unsaturated and saturated zones. Generally, conventional models are u sed for this purpose. This article reports on the use of artificial ne ural networks (ANNs) to simulate pesticide concentrations in agricultu ral soils. The main advantages of ANN modeling are significantly fewer input parameters and a very short execution time. An ANN model can be executed in real-time, while the sprayer is working in the field, in order to adjust application rates to the real extent of the problem. I n this study, an ANN model was built and trained with inputs of: accum ulated daily rainfall, soil temperature, potential evapotranspiration, as well as tillage practices and the number of days elapsed after pes ticide application. The outputs of the ANN model were the daily accumu lated amounts of pesticide levels in the soil. The results were compar ed with the data collected in 1992 and 1993 from an agricultural field in Ottawa, Canada. The results show the benefits of ANNs in predictin g pesticide concentrations in agricultural soils. In this study, only six input parameters are required with fast execution. The ANN-based m odel can be very helpful in making quick and appropriate decisions dur ing real-time application of pesticides. In this study, the performanc e of ANNs was investigated when the amount of available training data was limited. The results indicated that the performance of ANNs was go od, in spite of limited data, with the values of root-mean-square erro r and standard deviation being generally lower than 0.2 mu g/g. Howeve r the performance of ANNs could be improved with more training data ob tained from field experiments.