Plant uptake of organic pollutants from soil: bioconcentration estimates based on models and experiments.

The role of terrestrial vegetation in transferring chemicals from soil and air into specific plant tissues (e.g., stems, leaves, and roots) is still not well characterized. We provide here a critical review of plant-to-soil bioconcentration ratio (BCR) estimates based on models and experimental data. This review includes the conceptual and theoretical formulations of the BCR, constructing and calibrating empirical and mathematical algorithms to describe this ratio and the experimental data used to quantify BCRs and calibrate the model performance. We first evaluate the theoretical basis for the BCR concept and BCR models and consider how lack of knowledge and data limit reliability and consistency of BCR estimates. We next consider alternate modeling strategies for BCR. A key focus of this evaluation is the relative contributions to overall uncertainty from model uncertainty versus variability in the experimental data used to develop and test the models. As a case study, we consider a single chemical, hexahydro-1,3,5-trinitro-1,3,5-triazine, and focus on variability of bioconcentration measurements obtained from 81 experiments with different plant species, different plant tissues, different experimental conditions, and different methods for reporting concentrations in the soil and plant tissues. We use these observations to evaluate both the magnitude of experimental variability in plant bioconcentration and compare this to model uncertainty. Among these 81 measurements, the variation of the plant-to-soil BCR has a geometric standard deviation (GSD) of 3.5 and a coefficient of variation (CV; i.e., ratio of the arithmetic standard deviation to the mean) of 1.7. These variations are significant but low relative to model uncertainties, which have an estimated GSD of 10, with a corresponding CV of 14.