Nutrient-rich, highly acidic leachate draining from a coastalphosphogypsum storage site located at a phosphate fertilizer manufacturing company posed potentially harmful environmental effects to the coastal ecosystem. This study evaluated the chemical characteristics of the phosphogypsum and the hydrogeological characteristics of the surroundings. Hydraulic and chemical properties of the leachate draining from the site were also evaluated. The leachate is nutrient-rich and very acidic and discharges into the sea. The leachate and seawater are hydraulically connected through highly-permeable riprap placed at the toe of the stack. The chemical quality of the leachate and its drainage rate showed large variations depending on the location of the well, indicating heterogeneous hydraulic conditions. A vertical barrier system consisting of horizontal drainage wells and vertical leachate-collecting wells is suggested, in order to contain and collect the leachate. A back-up system of soil-bentonite cut-off walls is suggested to control leachate discharge to the sea.

A statewide total gaseous mercury (TGM) monitoring campaign was conducted from January 1997 to December 1999 in the State of Connecticut, U.S.A. Eight monitoring sites with different characteristics of geographical location (coastal vs. interior) and land use (rural vs. urban) were included in the monitoring program. Statistical procedures were utilized to evaluate the temporal trend and spatial distribution of the TGM concentration in the State, and the influence of long-range transport from non-local sources. The statewide mean TGM concentration was 2.08 ng m⁻³. The annual mean concentration had no significant differences among the three years of measurements for all the sites. Weak seasonal variations were detected in the State with higher ambient TGM concentration found in warmer seasons. Urban areas in general had higher TGM concentrations than rural areas. The effect of site location of the monitoring sites on TGM concentration was interacted with land use characteristics. Waterbury site with extremely high concentration measurements was the major cause for this interaction. The long-range transport of TGM from remote sources showed an important influence on local ambient concentrations, by explaining over 45% of the total variance of the ambient TGM concentration for most sites. Local sources were responsible for the extremely high TGM concentration in the Waterbury site. The TGM concentrations at Voluntown, Hammonasset and Avery Point in southeast Connecticut were likely to be affected by some local sources.

Bulk rainfalls were sampled during ten months in the Eastern Channel (Northern France). Chloride and sodium are the heavily loaded major ions in wet deposition, indicating a high influence of seasalt. However, the presence of heavy metals is independent of seasalt emissions and various anthropogenic sources should have an impact on their atmospheric concentrations.The comparison between heavy metals concentrations, always ranked as [Zn] ≫ [Pb] > [Cu] ≫ [Cd], with other values obtained during the same period on European coastal or on urban sites, confirms that ourwestern European station is in a semi-rural area where no local source strongly influences the abundance of atmospheric heavy metals. This semi-rural site could be disturbed by medium and/or long range transportepisodes.The observation of literature data confirms a decrease of Cd, Cu and Pb concentrations in rain waters from the beginning of the nineties. On the other hand, no sharp evolution has been observed for zinc concentrations.The structuring of data by clustering has permitted to extract five clusters, including four different types of origins. Two marine clusters, with relatively low levels of heavy metals, are opposite to three other clusters with higher heavy metal levels and a more important impact of continental polluted sources. In 47% of cases, the heavy metals collected at Cap Gris-Nez could originate from British Islands or north western Europe.Heavy metals fluxes show that this rural coastal European site has a level of contamination higher than those measured in remote area. The wet depositions of heavy metals to the English channel and the southern bight of the north sea are not negligible in the evaluation of global fluxes for this area, representing from 20 to 70% of the total input (riverine, direct and atmospheric inputs).

To date there have only been a few studies that measured mercury emissions from background substrate worldwide, and only a small amount of mercury flux data, from background substrate, exists for the Western United States. Because of this, the database of mercury emissions from background units < 0.1 mg kg⁻¹ mercury) is incomplete. This study focused on the collection of in-situ mercury flux data from representative lithologic units in Nevada. Measured mercury fluxes from substrate with background mercury concentration throughout Nevada were low (mean 2.0 ± 4.1 ng m⁻² hr⁻¹), and ranged from –3.7 to 9.3 ng m⁻² hr⁻¹. The mean measured mercury flux is slightly higher than those measured from background substrate from various locations throughout the world. The mean mercury flux from in-situ mercury measurements from substrate located near altered geologic units across Nevada was 15.5 ± 24.2 ng m⁻² hr⁻¹. These mercury fluxes are higher than the values applied in published global models for naturally enriched geologic units.

Chlorothalonil, a halogenated benzonitrile compound, is one of the most widely used fungicides in the world. Anaerobic microcosm assays were established to evaluate the combined effect of the initial content of carbon (6.3, 9.45 and 12.6 mg g⁻¹), nitrogen (0.6, 1.8 and 3 mg g⁻¹)and chlorothalonil (432, 865 and 1298 ηg g⁻¹) on the biodegradation of this fungicide by microbiota from an agricultural tropical soil. A Box-Behnken experimental design was used and chlorothalonil depletion was followed by HPLC with UV detection. The initial carbon content and fungicide dose were found to have a significant effect on removal efficiency. After 25 days of incubation, a high chlorothalonil depletion was observed in all biologically active microcosms (56–95%) although abiotic loss in a sterile blank was also notable (37%). The results suggest a high potential for chlorothalonil biodegradation under anaerobic conditions by indigenous microbial communities from soil that has been continuously exposed to high doses of the fungicide.

L`intensification de l`agriculture en Bretagne ces 20 dernières années a notamment généré une augmentation des apports azotés au milieu naturel, et a ainsi provoqué une augmentation des concentrations en nitrate dans les eaux de surface. Bien souvent au-delà les normes sanitaires, cette augmentation entraîne des surcoûts (assainissement, respect des directives européennes, coût environnemental). Les recherches menées au sein de l`équipe TERREAU (Cemagref Rennes), notamment dans le cadre de 2 thèses, ont pour but de mieux comprendre les mécanismes de transformation de l`azote lors de son transfert de la parcelle agricole à l`exutoire du bassin versant. Ces travaux sont nécessaires à la mise en oeuvre et l`amélioration des modèles agro-hydrologiques spatialisés à l`échelle du bassin versant, qui sont les outils les plus élaborés d`aide à l`aménagement et à la reconquête de la qualité de l`eau.Un des axes de recherche est la mesure spatialisée, dans un bassin versant agricole, de la dénitrification hétérotrophe. En effet, en Bretagne, on constate une déconnexion entre les pratiques agricoles et les flux d`azote. Quelle que soit la méthode de calcul, il y a en général une différence entre le solde du bilan de l `azote et le flux annuel mesuré dans l `eau à l`exutoire (abattement). Une partie de cet abattement est attribuable à la dénitrification hétérotrophe dans les zones humides, mais de fortes incertitudes existent encore sur sa quantification à l`échelle du paysage. Grâce à un suivi exhaustif des pratiques agricoles et des flux d`azote du bassin versant de la Fontaine du Theil, complété par des mesures in-situ régulières et réparties dans tout le bassin de la dénitrification, un modèle agro-hydrologique spatialisé permettra une estimation mieux contrainte et plus fiable de la dénitrification et des autres termes du bilan azoté. Cette étude est réalisée en lien avec de nombreux programmes de recherche portés par le CAREN auquel l`unité GERE est associée.L`autre axe de recherche se focalise sur les transferts et transformations de l`azote au sein du système zone riparienne / cours d`eau / sédiments. En effet, il existe des relations importantes mais non quantifiées des évolutions. L`un des objectifs de cette seconde thèse est la quantification fine des interactions entre les petits cours d`eau et leurs sédiment à travers des bilans de matière (flux d`azote et carbone) entrée-sortie sur des tronçons de réseau hydrographique. Ces études des flux portent sur différents types de substrats sédimentaires et différentes concentrations en nitrates dans la colonne d`eau. Une relation simple est recherchée entre la rétention d`azote, la concentration en nitrate et la typologie des sédiments, afin de permettre une intégration dans les modèles. Ces deux axes de recherche, de part leur complémentarité spatiale, permettront à terme de modéliser plus finement les processus d` auto-épuration et de rétention de l`azote à l`échelle du bassin-versant.Nous présenterons les méthodes employées pour l`évaluation des flux, c`est à dire les techniques de traçage et les techniques d`évaluation des émissions gazeuses utilisées (traçage au brome et blocage à l`acétylène), et les premiers résultats obtenus.

Dans le cadre de l`opération Ferti-Mieux « le Don, l`eau, la vie », les équipes de la Chambre d`agriculture de Loire Atlantique et du Cemagref se sont associées depuis 1997 pour quantifier le risque de pollution des eaux superficielles par l`azote et les produits phytosanitaires issus de l`activité agricole, et amorcer la proposition de solutions correctives sur ce bassin versant. Pour cela, nous travaillons sur trois bassins emboîtés, le Don (650 km2), le Cétrais (35 km2), et le micro-bassin du Coisbrac (0,5 km2). Pour appuyer les propositions de solutions correctives, nous avons choisi de montrer leurs effets sur les sols et l`eau, en nous appuyant sur le micro bassin de démonstration du Coisbrac. Sur ce micro-bassin, les agriculteurs ont accepté de modifier leurs pratiques. Nous constatons, dès la première année, que la mise en place d`outils d`aide à la fertilisation raisonnée se traduit par une diminution nette des apports d`engrais azotés aux cultures sur le micro-bassin du Coisbrac. Cette année a de plus vu des rendements exceptionnels sur céréales. Ces deux facteurs se sont conjugués pour diminuer les quantités d`azote potentiellement transférables vers les eaux. Cette diminution est cependant tempérée par la minéralisation de l`azote organique du sol et les arrière-effets des pratiques des années antérieures, qui sont dorénavant prépondérants dans les quantités d`azote minéral du sol.

Dans le cadre de l`opération Ferti-Mieux « le Don, l`eau, la vie », les équipes de la Chambre d`agriculture de Loire Atlantique et du Cemagref se sont associées depuis 1997 pour quantifier le risque de pollution des eaux superficielles par l`azote et les produits phytosanitaires issus de l`activité agricole, et amorcer la proposition de solutions correctives sur ce bassin versant.Pour appuyer les propositions de solutions correctives, nous avons choisi de montrer leurs effets sur les sols et l`eau, en nous appuyant sur un micro bassin de démonstration. Sur ce micro-bassin (bassin du Coisbrac), les agriculteurs ont accepté de modifier leurs pratiques, selon les recommandations formalisées lors des premières années d`étude.

Le projet européen AgriBMPWater a pour objectif de réaliser une comparaison de différentes modifications de pratiques agricoles élaborées pour améliorer la qualité de l`eau. Cette comparaison est réalisée sur le coût des modifications, leur efficacité attendue, et leur acceptabilité potentielle. Un des résultats novateurs de ce projet est que différencier les politiques de restauration de la qualité de l`eau selon les types d`agriculteurs permet d`accroître l`efficacité de ces politiques tout en diminuant leur coût et en augmentant leur acceptabilité. La traduction sur le terrain de ce résultat de recherche pose cependant de nombreuses questions. C`est cette traduction, sur le bassin versant du Don, qui est exposée ici.

[Departement_IRSTEA]RE [TR1_IRSTEA]RIE / PHYLEAU | Passive capillary samplers, which sample water from the vadose zone via a hanging water column in a fiberglass wick, have shown potential to provide better estimates of actual soil percolation fluxes than alternative field methods. Unsaturated and saturated flows (water and solutes) are extracted continuously and without external vacuum generator from a nondisturbed soil volume, through a significant area (typically 600cm2). In order to achieve a minimal disturbance of the native flow regime, the wick type (hydraulic conductivity and section), length and number and the contact area have to be dimensioned to match as close as possible the expected soil pressure/flow conditions. First, the Hydrus 2D code (Simunek et al., 1999) solving the Richards equation for simulating two dimensional unsaturated flow was used to evaluate the uncertainties in flux estimation by such passive capillary samplers. Two sources of uncertainties were examined. Those associated with the theoretical assumptions of the dimensioning of the wick and those associated to experimental uncertainties. A numerical experimentation was conducted on two reference soils submitted to a 11 days actual hyetograph. Results showed that an analytical dimensioning method proposed in the literature is relevant. But significant errors on the observed fluxes occur when the soil and wick properties do not exactly match, which is the common case since the range of available characteristics of fiberglass wicks is limited. Also, uncertainties in the hydraulic conductivity properties of the soil, in wick length appear to have an important influence on the representativity of the wick fluxes against the actual soil drainage fluxes. Elsewhere, eight such samplers have been installed for an in situ long term experimentation ,initiated this year in the Ardière Watershed (France). The objective is to measure water and pesticides percolation fluxes, at a 50cm depth under a grassed strip receiving contaminated runoff. Flow rate and pesticide concentration in surface runoff water are also measured so as to determine the soil boundary condition above each sampler. Soil water content and soil matric potential are measured at three different depth very close to each instrumented profile in order to explain the volumes extracted by each sampler. We present here the experimental methodology and first results obtained with simulated runoff events.