New approaches to uncertainty analysis for use in aggregate and cumulative risk assessment of pesticides.

Risk assessments for human exposures to plant protection products (PPPs) have traditionally focussed on single routes of exposure and single compounds. Extensions to estimate aggregate (multi-source) and cumulative (multi-compound) exposure from PPPs present many new challenges and additional uncertainties that should be addressed as part of risk analysis and decision-making. A general approach is outlined for identifying and classifying the relevant uncertainties and variabilities. The implementation of uncertainty analysis within the MCRA software, developed as part of the EU-funded ACROPOLIS project to address some of these uncertainties, is demonstrated. An example is presented for dietary and non-dietary exposures to the triazole class of compounds. This demonstrates the chaining of models, linking variability and uncertainty generated from an external model for bystander exposure with variability and uncertainty in MCRA dietary exposure assessments. A new method is also presented for combining pesticide usage survey information with limited residue monitoring data, to address non-detect uncertainty. The results show that incorporating usage information reduces uncertainty in parameters of the residue distribution but that in this case quantifying uncertainty is not a priority, at least for UK grown crops. A general discussion of alternative approaches to treat uncertainty, either quantitatively or qualitatively, is included.

Variation in the level of protection afforded to birds and crustaceans exposed to different pesticides under standard risk assessment procedures.

First-tier risk assessment for pesticides is often based on the quotient of the toxicity divided by the predicted environmental concentration or dose. This ratio is compared to a fixed assessment factor (AF) to decide whether the pesticide is to be allowed on the market or whether further research is needed. Often, a high value (e.g., the 90th percentile) is assumed for the predicted environmental concentration, and the lowest available value is chosen to represent toxicity; yet, the real level of protection is not known. Therefore, it is also not known whether the first tier is conservative enough or too conservative. By using 2 large toxicity databases and assuming a log-logistic species sensitivity distribution for each pesticide, the percent of species not covered by the AF is estimated in the scenario, where exposure is at the maximum level allowable in the first tier. In the case of crustaceans, the median estimate of the fraction of species not covered by the AF of 100 in the first-tier scenario is 3.4%, on average, for 72 pesticides. In other words, on average, 3.4% of the crustacean species will be exposed above their median lethal concentration (LC50) and median lethal dose (LD50) value in 10% of receiving surface waters that receive the maximum allowable exposure to an individual pesticide. The estimated level of protection varies widely between pesticides. For 10% of the pesticides, the estimated fraction of species not covered is ≥10% (maximum=41.4%). For 28% of the pesticides, 99.9% of the species will have the assumed level of protection. For birds, the median estimate of the fraction of species exposed above their median lethal dose for the first-tier scenario (AF=10) is 3.0% on average, when the AF is applied to the lower of the toxicity values for the 2 standard test species. For 11% of the pesticides, the median estimate is ≥10% (maximum=15.7%). When the AF is applied instead to the geometric mean of the toxicity values for the 2 standard species, the median estimate of the fraction of species not covered by the AF is increased to 7.4% on average; for 31% of the pesticides, this fraction is ≥10% (maximum=33.4%). This variation in the level of protection should be considered when defining the assumptions, assessment factors, and decision criteria in regulatory risk assessment.

Can transgenic maize affect soil microbial communities?

The aim of the experiment was to determine if temporal variations of belowground activity reflect the influence of the Cry1Ab protein from transgenic maize on soil bacteria and, hence, on a regulatory change of the microbial community (ability to metabolize sources belonging to different chemical guilds) and/or a change in numerical abundance of their cells. Litter placement is known for its strong influence on the soil decomposer communities. The effects of the addition of crop residues on respiration and catabolic activities of the bacterial community were examined in microcosm experiments. Four cultivars of Zea mays L. of two different isolines (each one including the conventional crop and its Bacillus thuringiensis cultivar) and one control of bulk soil were included in the experimental design. The growth models suggest a dichotomy between soils amended with either conventional or transgenic maize residues. The Cry1Ab protein appeared to influence the composition of the microbial community. The highly enhanced soil respiration observed during the first 72 h after the addition of Bt-maize residues can be interpreted as being related to the presence of the transgenic crop residues. This result was confirmed by agar plate counting, as the averages of the colony-forming units of soils in conventional treatments were about one-third of those treated with transgenic straw. Furthermore, the addition of Bt-maize appeared to induce increased microbial consumption of carbohydrates in BIOLOG EcoPlates. Three weeks after the addition of maize residues to the soils, no differences between the consumption rate of specific chemical guilds by bacteria in soils amended with transgenic maize and bacteria in soils amended with conventional maize were detectable. Reaped crop residues, comparable to post-harvest maize straw (a common practice in current agriculture), rapidly influence the soil bacterial cells at a functional level. Overall, these data support the existence of short Bt-induced ecological shifts in the microbial communities of croplands' soils.

Prediction of ecological no-effect concentrations for initial risk assessment: combining substance-specific data and database information.

A new method is proposed to derive predicted no-effect concentrations (PNECs) for initial risk assessments for aquatic ecosystems from a limited set of single-species toxicity data. The method includes three steps. First, acute toxicity data are divided by an acute-to-chronic assessment factor to obtain chronic toxicity data. Subsequently, chronic toxicity data are averaged to obtain an average hazardous concentration (HC50). Finally, the HC50 is divided by an interspecies assessment factor to obtain a PNEC. Both assessment factors are derived as probability distributions from an extensive ecotoxicological database. The interspecies assessment factor is combined with substance-specific toxicity data using Bayesian statistics. The proposed method optimizes the use of the available ecotoxicological information and it produces an uncertainty estimate of the PNEC. Sample calculations indicate that the proposed method may provide a good alternative for currently applied methods for initial risk assessment, particularly if few toxicity data are available.