Phorate is a broad-spectrum organophosphorus insecticide that rapidly transforms into a highly toxic metabolite upon soil application. These metabolites can be absorbed by crops leading to the production of noncompliant agricultural commodities. This study assessed the applicability of the diffusive gradient in thin film (DGT) technique, which uses a passive sampler that mimics diffusion-driven uptake, to predict the concentration of phorate available for lettuce uptake. Laboratory experiments were conducted to determine the diffusion coefficients and evaluate DGT accumulation under varying conditions including pesticide application rate, soil moisture, deployment depth, soil texture, and exposure duration. The results showed that metabolite formation enhanced diffusion, while soil concentration, moisture, and exposure time were key factors driving pesticide accumulation in the DGT. Thereafter, a pot experiment was performed with lettuce and DGT co-deployed to monitor the uptake at 7, 15, and 30 days. To examine the metabolite-enriched conditions, the soil was reused after 30 days with redeployed lettuce and DGT. The uptake patterns of lettuce and DGT were not strongly correlated; however, lettuce exhibited a markedly higher phorate uptake in metabolite-rich soils. Based on the pot data, a Random Forest model was developed to predict lettuce uptake. The model achieved strong performance, with R 2 values exceeding 0.97 and mean cross-validated R 2 values above 0.72. Although current datasets are insufficient to allow reliable predictions from DGT alone, the integration of DGT characteristics with expanded datasets could provide a robust framework for predicting crop uptake of soil pesticide residues. • Diffusive Gradients in Thin Films (DGT) assessed bioavailable phorate in soil. • Diffusion coefficients determined under varying soil and pesticide conditions. • Metabolite formation enhanced diffusion and influenced DGT accumulation. • Pot experiments revealed higher lettuce uptake in metabolite-rich soils. • Random Forest model predicted lettuce uptake with high accuracy (R 2 > 0.97).
Kim et al. (Thu,) studied this question.