This study reports on the occurrence and behaviour of six pesticides and one metabolite in a small stream draining a vineyard catchment. Base flow and flood events were monitored in order to assess the variability of pesticide concentrations according to the season and to evaluate the role of sampling frequency on the evaluation of fluxes estimates. Results showed that dissolved pesticide concentrations displayed a strong temporal and spatial variability. A large mobilisation of pesticides was observed during floods, with total dissolved pesticide fluxes per event ranging from 5.7x10(-3) g/Ha to 0.34 g/Ha. These results highlight the major role of floods in the transport of pesticides in this small stream which contributed to more than 89% of the total load of diuron during August 2007. The evaluation of pesticide loads using different sampling strategies and method calculation, showed that grab sampling largely underestimated pesticide concentrations and fluxes transiting through the stream.

No abstract

No abstract

Grassed buffer strips are one way to reduce pesticide transfer by surface runoff from farmed fields to streams. Numerous experimental studies have demonstrated that this type of development is very effective in various conditions. The results are nevertheless very variable. This variability is partially explained by the multiplicity of processes and contributive factors and by the fact that the latter are dynamic in nature and site-specific. Several results also show that the intercepted products are not systematically immobilized within the strip but can move in the subsurface, thus affecting the overall effectiveness of the system. As a consequence of this complexity, the present guidelines for the sizing and siting of grassed strips are still only qualitative or are the result of empirical approaches. The present review analyses the available results and defines needs for further research. This concerns several basic processes determining the pesticide interception by grassed strips and numerical models necessary to integrate the complexity of interacting processes and formulate reliable managing guidelines.

No abstract