New concepts for dynamic plant uptake models.

Models for the prediction of chemical uptake into plants are widely applied tools for human and wildlife exposure assessment, pesticide design and for environmental biotechnology such as phytoremediation. Steady-state considerations are often applied, because they are simple and have a small data need. However, often the emission pattern is non-steady. Examples are pesticide spraying, or the application of manure and sewage sludge on agricultural fields. In these scenarios, steady-state solutions are not valid, and dynamic simulation is required. We compared different approaches for dynamic modelling of plant uptake in order to identify relevant processes and timescales of processes in the soil-plant-air system. Based on the outcome, a new model concept for plant uptake models was developed, approximating logistic growth and coupling transpiration to growing plant mass. The underlying system of differential equations was solved analytically for the inhomogenous case, i.e. for constant input. By superposition of the results of n periods, changes in emission and input data between periods are considered. This combination allows to mimic most input functions that are relevant in practice. The model was set up, parameterized and tested for uptake into growing crops. The outcome was compared with a numerical solution, to verify the mathematical structure.

Comparison of prediction methods for the uptake of As, Cd and Pb in carrot and lettuce.

The New Model Framework (NMF) for uptake into crops is based on particle deposition and Transfer factors from soil to plant calculated from the BAse de donnees sur les teneurs en Elements Traces metalliques de Plantes Potageres (BAPPET) database. Besides NMF, approaches developed by the National Institute of Public Health and the Environment (RIVM), Hough, and the United States Environmental Protection Agency (US EPA), and the Contaminated Land Exposure Assessment (CLEA) approach were tested. Experimental data were assembled from the BAPPET database and Danish background data of As, Cd and Pb in soil, air and crops was collected. None of the models proved able to estimate the measured concentrations in plants from the BAPPET database with an absolute normalized error smaller than 70%. On average, the predictions had an error of 80-250%. However, when applying the models to the rural Danish background scenario, the NMF and other models predicted the concentrations in carrot and lettuce within the range of measured values. Regressions considering soil pH, organic matter and clay content were not superior to simple transfer factors. The transfer from air to plant is significant, at least under background conditions, and should be considered in the prediction methods.