Herbicide Accumulation and Evolution in Reservoir Sediments
Total Page：16
File Type：pdf, Size：1020Kb
Open Archive Toulouse Archive Ouverte (OATAO) OATAO is an open access repository that collects the work of Toulouse researchers and makes it freely available over the web where possible. This is an author-deposited version published in: http://oatao.univ-toulouse.fr/ Eprints ID: 8317 To link to this article: DOI:10.1016/j.scitotenv.2008.12.064 http://dx.doi.org/10.1016/j.scitotenv.2008.12.064 To cite this version : Devault, Damien A. and Gérino, Magali and Laplanche, Christophe and Julien, Frédéric and Winterton, Peter and Merlina, Georges and Delmas, François and Lim, Puy and Miguel Sánchez-Pérez, José and Pinelli, Eric Herbicide accumulation and evolution in reservoir sediments. (2009) Science of the Total Environment, vol. 407 (n° 8). pp. 2659-2665. ISSN 0048-9697 Any correspondence concerning this service should be sent to the repository administrator: [email protected] Herbicide accumulation and evolution in reservoir sediments Damien A. Devault a,⁎, Magali Gérino a, Christophe Laplanche a, Frédéric Julien a, Peter Winterton c, Georges Merlina a, François Delmas b, Puy Lim a, José Miguel Sánchez-Pérez a, Eric Pinelli a a Laboratoire d'Ecologie fonctionnelle ECOLAB UMR 5245 CNRS-UPS-INPT, Ecole Nationale Supérieure Agronomique de Toulouse, Institut National Polytechnique, Avenue de l'Agrobiopole, BP 32607 Auzeville Tolosane, 31326 Castanet-Tolosan Cedex, France b Unité de Recherche REQUE, Groupement Cemagref de Bordeaux, 50 Avenue de Verdun, Gazinet, 33 612 Cestas Cedex, France c Université Paul Sabatier, 118, Route de Narbonne, 31062 Toulouse Cedex 04, France abstract The aim of the present study was to understand the effect of reservoir configurations on sediment pesticide fate. Two dams were selected on the River Garonne, in southwest France: Carbonne and Golfech, both with reservoirs subject to accumulation of herbicide-contaminated sediment. They are situated upstream and downstream respectively of an agricultural and urban area: the Mid-Garonne. The results presented include pesticide concentrations and C/N ratios in the smaller sediment particles (b2 mm) and values of oxygenation Keywords: and herbicide concentrations in the water. S-triazines The dynamic behaviour of sediment in the reservoirs is discussed. The present study shows that the Substituted ureas theoretical lifespan (weak remanence in vitro) and the results actually observed in the sediment are Anilides conflicting. Pesticide contamination in Carbonne indicates conservation, even accumulation, of herbicide POPs molecules while in Golfech transformation processes clearly dominate. The hydromorphological position of Water Golfech reservoir, i.e. located at the junction of two rivers with contrasting hydrological regimes and very Sediment different oxygenation conditions, leads to accelerated pesticide desorption or degradation. Unfortunately, Dams this configuration is rare. 1. Introduction s-triazines. So, they are considered to be short-lived and milder even though very few conservation data exist. The 40,000 dams in the world present various proven disadvan- The aim of the present study was (i) to gain a fuller understanding tages (Giles, 2006). Sediment, especially coming from upstream, can of the phenomena affecting herbicide remanence in sediment, (ii) to be the vector of pollutants, including herbicides (Devault et al., 2007), investigate the effect on pesticide fate of two hydromorphologically affecting the food networks (Frery et al., 2001). The pollutants and physico–chemically contrasting large reservoirs on the same accumulate in the dam essentially after having arrived from upstream river continuum and (iii) to determine which dam layout will via the river (Doggett and Rhodes,1991) and pesticide accumulation in minimize pesticide impact on the environment. sediment causes strong contamination especially of the benthos (Armstrong et al., 1980). Hormonal and behavioural disruption due to 2. Study area pesticides can prove (Scott and Sloman, 2004) to be lethal at the top of the food net (Heger et al., 1995). Although herbicides such as s- France is the second agricultural producer state in the world and triazines, substituted ureas and anilides account for 47% of the world the first in Europe (UIPP, 2008). France is the fourth pesticide pesticide trade, they have been the object of a few environmental consumer state in the world and the first in Europe (UIPP, 2008). studies (Gao et al., 1997, 1998) due, in particular, to methodological The Mid-Garonne has been largely described elsewhere (Devault et al., problems especially for substituted ureas and anilides. Those two 2007) for human settlement, pedo-morphologic, hydrologic and use families of pollutants are more labile in vitro than the abundantly- characteristics. The Garonne River is the third largest river in France documented persistent organic pollutants (POPs) or, to a lesser extent, (Fig. 1) and structures the Mid-Garonne, an area included between two dams on the Garonne River: Carbonne and Golfech. The fifth largest city in France, Toulouse with a population of 1 million is located in the Mid-Garonne. ⁎ Corresponding author. Tel.: +33 5 62 19 39 01. Carbonne dam (Fig. 2) is the last of the hydroelectric installations of E-mail address: [email protected] (D.A. Devault). the Pyrenean Piedmont. It forms the upstream limit of the Mid-Garonne, doi:10.1016/j.scitotenv.2008.12.064 Fig. 1. Location of the sampling sites and aerial (Géoportail, 2007) views of Golfech reservoir (a) and Carbonne reservoir (b). a simple dam across the river with no other particular features: the vast pesticide to be spread is herbicide and its uses are urban as well as majority of dams in France have the same configuration. Upstream of agricultural. Unfortunately, no data of local pesticide consumption or Carbonne, the Pyrenean Piedmont is primarily used for livestock while use is available to estimate the quantities entering the river. However, field crops start to become preponderant near the reservoir even if such data were available, interpretation would be subject to (8,000,000 m3), located in the transition zone. The second dam in this caution because of the slow release of compounds into surface water study, built for the production of hydro-electricity in 1968, is at Golfech from contaminated groundwater and soils. (Fig. 3). It is located at the junction of the Garonne with its largest affluent: the Tarn River and forms the Golfech reservoir 3. Materials and methods (25,000,000 m3), which is the downstream limit of Mid-Garonne. Locating a dam on a confluence is very unusual, occurring only once in Water was taken at the surface and kept at 4 °C in darkness until the whole of the 296 dams in the basin. The configuration of the dams is analysis. Raw water was analysed for both reservoirs. Filtered water their only structural difference: Golfech and Carbonne are both was analysed only for Golfech reservoir. constant-head dams, accommodate fishing and sailing, have the same The sediment was taken with an Eckmann bucket at the sites length (3.5 km) and water residence time. Their reservoirs are both indicated in Figs. 2 and 3. At Carbonne reservoir, six samples were protected as zones of high ecological richness and for this reason, they taken while at Golfech, to respect the ratio of reservoir areas, eleven are never drained. The present study seeks to understand the processes were studied. Sampling sites were chosen in Carbonne and especially which condition the fate of pesticides in the sediments of the reservoirs, Golfech to follow the main Garonne riverbed. Sampling and analysis each with its own configuration. were performed using a method validated in Devault et al., 2007 which Large quantities of pesticides are applied on the slopes in the Mid- reports the justification for the choice of herbicide families studied, the Garonne, subjected to significant levels of erosion. The main type of high significance of large samples (400 cm2, 1 kg of homogenised Fig. 2. Geographic distribution of herbicide contamination in Carbonne reservoir. sediment samples) and the high performance of the analytical method 3.1. Sampling period (high accuracy with 1 ng/g-detection limit and high reproducibility). Triplicates were performed but, as for Devault et al. (2007), repeat- The sediment samples were taken simultaneously on March 16, ability was indiscernible to the instrumental background noise. 2005, at the lowest winter water level. The monthly spreading Fig. 3. Geographic distribution of herbicide contamination in Golfech reservoir. Fig. 4. Water oxygen (mg/L) in Golfech reservoir. X, Y and Z are the spatial coordinates of the reservoir (Z is the depth). intensity of pesticide use in March corresponds to the annual average obtained on the Garonne branch of the reservoir, the Tarn branch intensity of pesticides spreading. The temperature remained at typical and at a point between the confluence and the dam (Fig. 4). winter values (around 5 °C). 3.2.2. Sediment 3.2. Sampling treatment Herbicide content was analysed as explained elsewhere (Devault et al., 2007). Briefly, fine (lower than 63 µm-mesh) and coarse (between 63 µm 3.2.1. Water and 2 mm-mesh) fractions of dry sample were extracted by ASE 200 then Extraction was performed on samples where the pH had been purified on Florisil cartridges and quantified with an HP5290 GC-MS. adjusted to 7, filtered through GFF membrane. Solid-Phase Extraction Whole sediment, i.e. all particles smaller than 2 mm, was reconstituted by (Supelclean vacuum device) of 200 ml samples was performed with calculation. Herbicides from the standard Mix44 provided by Cluzeau Info silica cartridges grafted with Supelco C18. Labo (Sainte-Foy-la-Grande, France) and produced by Dr Ehrenstorfer Analysis of raw water is detailed in Devault et al., 2007. The three- Gmbh were also quantified. dimensional profile of the water oxygen content was measured on site Carbon was measured with a CS-500 Determinator (Eltra GmbH, with an oxygen electrode. The Golfech profile includes results Neuss, Germany) and nitrogen by the Kjeldahl method.