Modelling non-systemic pesticide residues in fruits with initial deposit variability and weather effects.

A flexible and generic model was developed to predict the decline of residues of a non-systemic pesticide for both single and multi-spray situations as well as for different tree canopy zones. The model predicts not only the average residue levels, but also the confidence interval of the residues through either a deterministic or a stochastic approach. This generic model includes several key aspects of residue fates in the environment: initial deposit, physical loss and growth dilution. The model considers a tree canopy in three distinct zones for which initial deposition of pesticides may differ. In addition to predicting the average residue within each zone, it also estimates the 95 and 99% confidence intervals of residues on individual fruit within each zone. For the purpose of evaluation, this model was parameterized specifically for captan, one of the most important non-systemic fungicides used to control disease in horticultural crops. The observed average initial deposit for each zone was used in the evaluation. The overall correlation between predicted average residues and those observed on apple fruit in two applications was 0.93. Confidence intervals were also predicted accurately.