No abstract

Since a few years, land use management aims to reduce and control water erosion processes in watersheds but there is a lack of quantitative information on the contribution of the sources of transported sediment. This is most important in agricultural areas where soils are sensitive to erosion. The geology of these areas is often characterized by large expanses of relatively homogeneous quaternary silts. The possibility of distinguishing the sources of erosion according to their geological substratum is thus very delicate. This information is important because its lack can lead to the mis-implementation of erosion control measures. To address this request, a confluencebased sediment fingerprinting approach was developed on the Canche river watershed (1274 km²; northern France), located in the European loess belt, an area that is affected by diffuse and concentrate erosion processes. Suspended particulate matter was collected during five seasonal sampling campaigns using sediment traps at the outlet of each tributary and confluence with the main stream of the Canche river. The final composite fingerprint was defined using physico-chemical and statistical analyses. The best tracer parameters for each tributary were selected using stepwise discriminant function analyses. These parameters were introduced into a mass balance mixing model incorporating Monte-Carlo simulations to represent the uncertainty. Estimates of the overall mean contributions from each tributary were quantified at different temporal scales. The annual sediment flux tributaries contributions range from 3 to 22% at the outlet of the Canche river, and annual sediment flux range from 0.87 to 40.7 kt yr -1 . The Planquette and the Créquoise tributaries appear to be those producing the largest sediment flux. In contrast, tributaries with the highest number of erosion control on their area exhibit the lowest values of sediment flux. Our results indicate a positive impact of recent land management policies in the Canche river watershed.

Three configurations of single-phase polymer passive samplers made of polyoxymethylene (POM), silicone rubber, and polyethylene (PE) were simultaneously calibrated in laboratory experiments by determining their partitioning coefficients and the POM diffusion coefficients and by validating a kinetic accumulation model. In addition, the performance of each device was evaluated under field conditions. With the support of the developed model, the device properties are discussed with regard to material selection and polymer thickness. The results show that a sampler's properties, such as its concentration-averaging period and ability to sample a large amount of polycyclic aromatic hydrocarbons, are widely affected by material selection. Sampler thickness also allows modulation of the properties of the device but with a much lower magnitude. Selection of the appropriate polymer and/or thickness allows samplers to be adapted either for quick equilibration or for the kinetic accumulation regime and promotes either membrane or water boundary layer control of the kinetic accumulation. In addition, membrane-controlled or equilibrated compounds are quantified with greater accuracy because they are not corrected by the performance reference compounds approach. However, the averaged concentrations cannot be assessed when compounds reach equilibrium in the sampler, whereas membrane-controlled devices remaining in the kinetic accumulation regime provide averaged concentrations without requiring performance reference compound correction; detection limits are then increased because of the higher mass transfer resistance of the membrane. Environ Toxicol Chem 2016;35:1708-1717. © 2015 SETAC.

A passive sampler inspired from previous devices was developed for the integrative sampling of a broad range of contaminants in the water column. Our primary objective was to improve the performance of the device to provide accurate and averaged pollutant water concentrations. For this purpose, an agarose diffusive gel was used as the boundary layer that drives the analyte uptake rate. Contrary to conventional passive samplers, the developed device does not require the sampling rates to be corrected for exposure conditions (e.g. hydrodynamic flow) because the diffusive gel boundary layer selected was sufficiently large to control the pollutant diffusion rate from the aqueous phase. The compounds diffusion coefficients in agarose gel and the gel thickness are the only required data to accurately calculate the time weighted averaged water concentration of pollutants. The performance of the developed sampler was evaluated in the laboratory under two contamination scenarios and in the field in 8 contrasting exposure sites for a selection of 16 emerging pollutants and pesticides. The results show that detection limits of this method are environmentally relevant and allow the determination of the averaged pollutant concentrations. Additionally, the ability of the device to sense very short contamination pulses (5-320min) was evaluated through a theoretical approach and laboratory tests. Results show that the device is suitable for sampling contamination pulses as short as 5min without deviation from the actual average concentrations of pollutants.

The diffusivity of 145 compounds in polydimethylsiloxane (PDMS) material was determined in the laboratory using a film stacking technique. The results were pooled with available literature data, providing a final data set of 198 compounds with diffusivity (D ) spanning over approximately 5 log units. The principal variables controlling the diffusivity of penetrants were investigated by comparing D within and between different homologous series. The dipole moment, molecular size, and flexibility of penetrants appear to be the most prevalent factors controlling a compound's diffusivity. A nonlinear quantitative structure-property relationship is proposed using as predicting variables the molecular volume, the number of rotatable bonds, the topological polar surface area, and the number of O and N atoms. The final relationship has a correlation coefficient of R = 0.81 and a mean absolute error of 0.26 m s (log unit), approaching the average error for the experimentally determined values (0.12 m s ). The model, based on a heuristic approach, is ready for use by analytical chemists with no specific background in theoretical chemistry (notably for passive sampler development). Environ Toxicol Chem 2018;37:1291-1300. © 2018 SETAC.