Estimating 42 pesticide sampling rates by POCIS and POCIS-MIP samplers for groundwater monitoring: a pilot-scale calibration.

Pesticides occur in groundwater as a result of agricultural activity. Their monitoring under the Water Framework Directive is based on only a few spot-sampling measurements per year despite their temporal variability. Passive sampling, which was successfully tested in surface water to provide a more representative assessment of contamination, could be applied to groundwater for a better definition of its contamination. However, few reliable calibration data under low water flow are available. The objective of our study thus consisted in determining sampling rates by two types of passive samplers, a POCIS (polar organic chemical integrative sampler) for polar pesticides, and a POCIS-MIP sampler based on a receiving phase of molecular imprinted polymers, specific for AMPA and glyphosate under low flow conditions as exist in groundwater. To our knowledge, this is the first time that sampling rates (sampling rate represents the volume of water from which the analyte is quantitatively extracted by the sampler per unit time) are estimated for groundwater applications. Our calibrations took place in an experimental pilot filled with groundwater and with low water flow (a few metres per day). Pesticide uptake in POCIS showed good linearity, with up to 28 days before reaching equilibrium. Two types of accumulation in POCIS were noted (a linear pattern up to 28 days, and after a time lag of 7 to 14 days). Sampling rates for 38 compounds were calculated and compared with those available in the literature or obtained previously under laboratory conditions. The values obtained were lower by a factor 1 to 14 than those estimated under stirring conditions in the literature, whereas water flow velocity (m s