Dynamic Multicrop Model to Characterize Impacts of Pesticides in Food

A new dynamic plant uptake model is presented to characterize health impacts of pesticides applied to food crops, based on a flexible set of interconnected compartments. We assess six crops covering a large fraction of the worldwide consumption. Model estimates correspond well with observed pesticide residues for 12 substance-crop combinations, showing residual errors between a factor 1.5 and 19. Human intake fractions, effect and characterization factors are provided for use in life cycle impact assessment for 726 substance-crop combinations and different application times. Intake fractions typically range from 10(-2) to 10(-8) kg(intake) kg(applied)(-1) Human health impacts vary up to 9 orders of magnitude between crops and 10 orders of magnitude between pesticides, stressing the importance of considering interactions between specific crop-environments and pesticides. Time between application and harvest, degradation half-life in plants and residence time in soil are driving the evolution of pesticide masses. We demonstrate that toxicity potentials can be reduced up to 99% by defining adequate pesticide substitutions. Overall, leafy vegetables only contribute to 2% of the vegetal consumption, but due to later application times and higher intake fractions may nevertheless lead to impacts comparable or even higher than via the larger amount of ingested cereals.