Similarities and differences between measured and predicted concentrations of pesticides in Dutch surface waters.

In order to have a thorough evaluation of the progress and effectiveness of Dutch crop protection policy, both model predictions and measured pesticide concentrations in surface waters are considered. To this purpose, monitoring data obtained by various water boards and other monitoring institutes were processed. Data were aggregated over a two year time period and over space (at 1x1 km-grid). A geographic view is given in the Dutch Pesticides Atlas (www.pesticidesatlas.nl). The model used for the predictions was the Dutch National Environmental Indicator NMI version 2 (www.nmi.alterra.nl) that has input data regarding spray drift data, crop interception, soil and climate and many more. Information on aggregation steps over time and space, grid sizes, information on crop areas was geared to one another for both instruments. Results on measured pesticide concentrations in surface waters and model predictions were compared to each other at the national scale. For this study, 10 different cases were selected covering a large range of pesticides' characteristics and pesticides' use. In 60% of the cases, the results were largely in agreement with each other when expressed as absolute numbers of measurements exceeding the environmental quality standard. This is very accurate and useful for policy purposes. Based on concentrations and on the order of magnitude, no significant agreement between measurements and model predictions was found. Differences were explained by various factors, and an overview of predominant systematic differences between the measurements and the model predictions was presented. Using both measurements and model predictions in supporting environmental policy evaluations is warranted, because of higher Weight-of-Evidence. Combining both can assist in optimizing the knowledge on pesticides behaviour, fate and ecological problems and therefore this is the preferred evaluation method.

Influence of pH-dependent sorption and transformation on simulated pesticide leaching.

The leaching of a substance is influenced by its physico-chemical characteristics as well as environmental conditions. In spatially distributed modelling the influence of soil properties on the half-life and the sorption constant of the substance might become important and can be taken into account. The GeoPEARL model includes options to account for sorption and transformation being dependent on soil characteristics. Using some of these options in calculations for a herbicide with both sorption and transformation dependent on the pH of the soil, the calculated leaching from an application in spring appeared to be higher than anticipated from calculations according to the so-called paired parameter approach, in which the leaching is assessed for pairs of sorption and transformation parameters at regular pH intervals. The reason for the higher leaching was that the most critical leaching conditions were not covered by the selected pH values. A 'paired approach' might however be useful as a first tier assessment of the leaching potential. The maximum leaching is expected with the highest DegT(50)/K(om) ratio, which might be obtained from plotting this ratio against the characteristic soil property. The leaching potential of the parent was more important for the leaching of the metabolite than the leaching potential of the metabolite itself. This should be accounted for in the evaluation procedure.

Pesticides in groundwater and drinking water wells: overview of the situation in the Netherlands.

In the Netherlands, many of the fresh groundwater resources are vulnerable to pollution. Owing to high population densities and intensive farming practices, pesticide residues are found in groundwater at many places. Hence a number of drinking water abstraction wells contain pesticides residues, causing considerable costs for purification. The Water Framework Directive (WFD) requires countries to assess the chemical status of groundwater bodies and set up monitoring plans for groundwater quality, including pesticides. 771 groundwater samples were taken from monitoring wells in 2006 and analysed for a broad list of pesticides in order to fulfil these requirements. Pesticide were detected in 27% of samples, while in 11% the WFD limit of 0.1 microg/l was exceeded. In this paper, these and earlier measurements are evaluated further, considering also measurements in drinking water wells, information about the origin of measured pesticides and calculated trends in use and emissions. The measurements in the monitoring wells showed that where pesticides are used, 15-55% (minimal and maximal estimation) of the wells in shallow groundwater (1 to 20 m below soil surface) contain pesticides residues at concentrations above 0.1 microg/l. When the metabolites BAM and AMPA are excluded (as not relevant in human toxicological terms), the estimation range is 7-37%. These patterns observed in shallow groundwater are reflected by the occurrence of pesticides in vulnerable abstraction wells that are used for the production of drinking water. The WFD requires the determination of both status and trends. The design of current monitoring network is evaluated from this perspective. Several recommendations are made for more adequate and efficient monitoring.

Mapping ground water vulnerability to pesticide leaching with a process-based metamodel of EuroPEARL.

To support EU policy, indicators of pesticide leaching at the European level are required. For this reason, a metamodel of the spatially distributed European pesticide leaching model EuroPEARL was developed. EuroPEARL considers transient flow and solute transport and assumes Freundlich adsorption, first-order degradation and passive plant uptake of pesticides. Physical parameters are depth dependent while (bio)-chemical parameters are depth, temperature, and moisture dependent. The metamodel is based on an analytical expression that describes the mass fraction of pesticide leached. The metamodel ignores vertical parameter variations and assumes steady flow. The calibration dataset was generated with EuroPEARL and consisted of approximately 60,000 simulations done for 56 pesticides with different half-lives and partitioning coefficients. The target variable was the 80th percentile of the annual average leaching concentration at 1-m depth from a time series of 20 yr. The metamodel explains over 90% of the variation of the original model with only four independent spatial attributes. These parameters are available in European soil and climate databases, so that the calibrated metamodel could be applied to generate maps of the predicted leaching concentration in the European Union. Maps generated with the metamodel showed a good similarity with the maps obtained with EuroPEARL, which was confirmed by means of quantitative performance indicators.