Search

/ Les bandes tampons végétalisées constituent une solution intéressante pour limiter les transferts de pesticides par ruissellement et réduire la pollution diffuse de l'eau. Pour être pleinement efficaces, elles doivent être adaptées au contexte dans lequel elles sont implantées. BUVARD Online est un outil disponible gratuitement en ligne (https://Buvard.irstea.fr), qui permet de dimensionner des bandes enherbées en tenant compte des conditions locales sur la France métropolitaine.

International audience; In France and more generally over Europe, significant amounts of pollutants are measured in surface water, partly due to the use of pesticides by agriculture. In the European Water Framework Directive, Europe is advocating the development of best management practices to reduce pesticide transfers to the river network once they are applied in the watershed. This includes implementing vegetative filter strips (VFS), that ensure the interception and the mitigation of contaminant transfers arising from fields. VFS are now mandatory along rivers in many countries, due to their recognized effectiveness to limit pesticide and sediment transfer by surface runoff (Asmussen et al., 1977; Dosskey, 2001). However, the general effectiveness of these buffers to reduce runoff transport of pesticides highly depends on pedologic characteristics, climatic conditions, and cultural practices. It is thus necessary to use mechanistic models that represent processes occurring on a vegetative filter strip, such as VFSMOD (Vegetative Filer Strip Modeling, Muñoz-Carpena et al., 1999). These models are relevant tools to design properly the buffers accounting for local conditions, although they are rarely used in France, since they are considered too complex for operational use (Carluer et al., 2017). In France, in order to help decision-makers use physically-based modelling, Irstea developed the modeling toolkit BUVARD (BUffer strip for runoff Attenuation and pesticides Retention Design tool). It consists of several steps including analyzing the watershed and its characteristics (soil, climate, cultural practices), and running dynamical models, in particular the mechanistic model VFSMOD adapted to French conditions (Muñoz-Carpena et al., 2018, Lauvernet and Muñoz-Carpena, 2018). At the end this toolkit delivers the optimal VFS width considering the needed filter efficiency (for example, 70% of runoff reduction). However, this very complete method assumes that the user provides detailed field knowledge and data (type of soil of the contributive area and of the VFS, rainfall rate, water table depth, slope, etc.), which are not easily available in many practical applications. Moreover, the variety of tools, which rely on several interfaces or several programming languages, makes it relatively difficult to take over the design procedure. We get to situations where the tool is used, without any uncertainty quantification nor sensitivity analysis, although they should be performed together with the tool's simulations (Saltelli et al., 2008). The next step in seeking to increase the operational scope of the modeling toolkit was to use metamodeling techniques. By reducing the computational cost of the modeling toolkit, the metamodel of BUVARD will make it possible to apply the tool on new watersheds with far fewer input parameters (6 against 70), and to determine the output uncertainty and sensitivity to input parameters in other climatic and agronomic conditions at low cost. Metamodeling is still rarely used in the water quality domain, since processes related to pesticide transfer are highly nonlinear. The VFS sizing tool BUVARD and the physical processes it represents (water and pesticide transfer at surface/subsurface) includes high non-linearities, due to the dependence on qualitative inputs (or categorical variables). Indeed, two major inputs, the typeof soil of the VFS and the type of rainfall event, have been defined in BUVARD for operational purposes, as substitute to functional inputs (rainfall hyetograph) and to correlated inputs that are the hydrodynamics properties of the soil (saturated hydraulic conductivity, porosity, and van Genuchten parameters). Qualitative inputs generate discontinuities in the model's response that many methods are unable to deal with, removing the smoothness of the model's output that is generally a necessary condition to build a metamodel (Zang and Notz, 2015). In this study, we adapted kriging to mixed variables (qualitative and quantitative), by testing several covariance kernels for a mixture of qualitative and quantitative inputs. Their performances are compared to a linear model and to a generalized additive model (GAM) that have been often used in water quality metamodeling. The methods are validated with the physically-based simulations conducted on a full factorial test design. It will be shown that the adapted kriging is very efficient and weakly dependent on the sampling size of the experimental design. These metamodels will then be used to perform uncertainty quantification and global sensitivity analysis of the VFS efficiency in a French watershed in the Beaujolais vineyard region. Polynomial Chaos Expansion do not allow to account for qualitative variables to our knowledge, but they have the advantage of being very efficient on complex models, and to compute Sobol indices directly from polynomial chaos expansions (Le Gratiet et al., 2017). We will compare the analysis from the PCE on each group of modality to the one performed with the kriging with an adapted covariance kernel. The final aim of this study is to give the users a complete tool accounting for uncertainty and sensitivity of the model outputs to design their vegetative filter strips. Asmussen, L.E., White, A.W., Hauser, E.W., Sheridan, J.M., 1977. Reduction of 2,4-D Load in Surface Runoff Down a Grassed Waterway1. Journal of Environment Quality 6, 159. Carluer, N.; Lauvernet, C.; Noll, D., Muñoz-Carpena, R. Defining context-specific scenarios to design vegetated buffer zones that limit pesticide transfer via surface runoff Science of The Total Environment , 2017, 575, 701 - 712 Dosskey, M.G., Helmers, M.J., Eisenhauer, D.E., 2011. A design aid for sizing filter strips using buffer area ratio. Journal of Soil and Water Conservation 66, 29-39. Hastie, Tibshirani and Friedman, 2009. The Elements of Statistical Learning (2nd ed.). Springer-Verlag. Le Gratiet, L.; Marelli, S. & Sudret, B. Metamodel-Based Sensitivity Analysis: Polynomial Chaos Expansions and Gaussian Processes Handbook of Uncertainty Quantification, Springer International Publishing, 2017, 1289-1325 Lauvernet, C., Muñoz-Carpena, R. Shallow water table effects on water, sediment, and pesticide transport in vegetative filter strips -- Part 2: model coupling, application, factor importance, and uncertainty. Hydrology and Earth System Sciences, 2018, 22, 71-87 Muñoz-Carpena, R., J.E. Parsons, et J.W. Gilliam, 1999. Modeling hydrology and sediment transport in vegetative filter strips. Journal of Hydrology 214:111. Muñoz-Carpena, R.; Lauvernet, C., Carluer, N. Shallow water table effects on water, sediment, and pesticide transport in vegetative filter strips -- Part 1: nonuniform infiltration and soil water redistribution. Hydrology and Earth System Sciences, 2018, 22, 53-70 Rasmussen, C.E..and Williams, C.K.I. Gaussian Processes for Machine Learning. The MIT Press, 2006. ISBN 0-262-18253-X. Saltelli, A.; Ratto, M.; Andres, T.; Campolongo, F.; Cariboni, J.; Gatelli, D.; Saisana, M., Tarantola, S. Global Sensitivity Analysis: The Primer John Wiley & Sons, 2008 Zhang, Y., Notz, W. I., 2015. Computer experiments with qualitative and quantitative variables: A review and reexamination. Quality Engineering 27 (1), 2-13.

Vegetation buffers like vegetative filter strips (VFSs) are often used to protect water bodies from surface runoff pollution from disturbed areas. Their typical placement in floodplains often results in the presence of a seasonal shallow water table (WT) that can decrease soil infiltration and increase surface pollutant transport during a rainfall-runoff event. Simple and robust components of hydrological models are needed to analyze the impacts of WT in the landscape. To simulate VFS infiltration under realistic rainfall conditions with WT, we propose a generic infiltration solution (Shallow Water table INfiltration algorithm: SWINGO) based on a combination of approaches by Salvucci and Entekhabi (1995) and Chu (1997) with new integral formulae to calculate singular times (time of ponding, shift time, and time to soil profile saturation). The algorithm was tested successfully on five distinct soils, both against Richards's numerical solution and experimental data in terms of infiltration and soil moisture redistribution predictions, and applied to study the combined effects of varying WT depth, soil type, and rainfall intensity and duration. The results show the robustness of the algorithm and its ability to handle various soil hydraulic functions and initial nonponding conditions under unsteady rainfall. The effect of a WT on infiltration under ponded conditions was found to be effectively decoupled from surface infiltration and excess runoff processes for depths larger than 1.2 to 2 m, being shallower for fine soils and shorter events. For nonponded initial conditions, the influence of WT depth also varies with rainfall intensity. Also, we observed that soils with a marked air entry (bubbling pressure) exhibit a distinct behavior with WT near the surface. The good performance, robustness, and flexibility of SWINGO supports its broader use to study WT effects on surface runoff, infiltration, flooding, transport, ecological, and land use processes. SWINGO is coupled with an existing VFS model in the companion paper (Lauvernet and Muñoz-Carpena, 2018), where the potential effects of seasonal or permanent WTs on VFS sediment and pesticide trapping are studied.

International audience; Vegetation buffers like vegetative filter strips (VFS) are often used to protect water bodies from surface runoff pollution from disturbed areas. Their typical placement in bottomland often results in the presence of a seasonal shallow water table (WT) that can decrease soil infiltration and increase surface pollutant transport during a rainfall/runoff event. Simple and robust components of hydrological models are needed to analyse the impacts of WT in the landscape. To simulate VFS infiltration under realistic rainfall conditions with WT, we propose a generic infiltration solution (Shallow Water table INfiltration algorithm: SWINGO) based on a combination of approaches by Salvucci and Entekhabi (1995) and Chu (1997) with new integral formulae to calculate singular times (time of ponding, shift time, and time to soil profile saturation). The algorithm was tested successfully on 5 distinct soils both against Richards’s numerical solution and experimental data in terms of infiltration and soil moisture redistribution predictions, and applied to study the combined effects of varying WT depth, soil type, and rainfall intensity and duration. The results show the robustness of the algorithm and its ability to handle various soil hydraulic functions, and initial non-ponding conditions under unsteady rainfall. The effect of a WT on infiltration under ponded conditions was found effectively decoupled from surface infiltration/excess runoff processes for depths larger than 1.2 to 2 m, shallower for fine soils and shorter events. For non-ponded initial conditions, the influence of WT depth also varies with rainfall intensity. Also, we observed that soils with a marked air entry (bubbling pressure) exhibit a distinct behaviour with WT near the surface. The features and good performance of SWINGO support its coupling with an existing VFS model in the companion paper, where the potential effects of seasonal or permanent WTs on VFS pollutant transport and control are studied.

International audience; In the context of the European Water Framework Directive (WFD, 2000/60/EC), which aims to achieve a good ecological and chemical status for all natural aquatic environments, it is necessary to better quantifying pesticide transfers in agricultural watersheds. Modeling is a relevant tool to study the processes occuring in the fate of pesticides, and identifying sound managing solutions to water contamination. Physically-based and spatially distributed models are particularly useful to represent in detail processes and interactions between the soil surface and subsurface. In particular, they can help to assess the role of landscape elements such as vegetative buffer strips, ditches, ... . The present study aims to test a recently added reactive transport to the coupled surface water/groundwater 3D model CATHY, in order to represent pesticide transfers at the watershed scale. Contaminant reactions implemented are linear adsorption and degradation (first order kinetics). The model has been successfully validated on laboratory data and submitted to a sensitivity analysis as a further step of validation. In this study, we tested the model on field conditions, simulating a 150 m x 40 m hillslope in the Morcille catchment (Beaujolais, France). The hillslope is predominantly covered with a chemically weeded vineyard crossed by 4 very shallow ditches gathering the runoff towards a concrete channel. The flow coming from the vineyard plot in then conducted on a 25 m long vegetative buffer strip. The site is instrumented with piezometers, lysimeters, flow and solute concentration measurement devices. Real field conditions makes the model's application more difficult because of the complexity in representing the interactive processes in a large domain, combined to uncertainty on input parameters that is important in field experiments. However, such challenging modeling allows a more comprehensive understanding of solute transfers and gives some keys to evaluate the efficiency of mitigation elements of the landscape.

International audience; In the context of the European Water Framework Directive (WFD, 2000/60/EC), which aims to achieve a good ecological and chemical status for all natural aquatic environments, tools to help understand and quantify pesticide transfers in agricultural watersheds are necessary. Models which are physically based and spatially distributed can be particularly useful for representing in detail processes and interactions between the soil surface and subsurface and thus to evaluate the management of landscape elements remediation. The present study aims to test and validate a recently added reactive transport to the coupled surface water/groundwater model CATHY, in order to represent pesticide transfers. Contaminant reactions implemented in CATHY for this study are linear adsorption and degradation (first order kinetics). The advection part of the model is solved according to the finite volume method and reactions are computed on volumes using a sequential non-iterative approach. The CATHY model has been tested on laboratory data and with a Morris sensitivity analysis and is applied now in real field conditions, on a vegetative buffer strip monitored by Irstea in a vineyard catchment (Beaujolais, France). The site is instrumented with lysimeters, flow and solute concentration measurement devices. This test represents a complex step into the model validation, initial and boundary conditions are not fully controlled, and field parameters measurement are not fully known.

International conference on sensitivity analysis of model output, La Réunion, FRA, 30-/11/2016 - 03/12/2016; International audience; Irstea developed a methodology which allows designing site-specific VFS by simulating their efficiency to limit transfers among a hillslope. The modeling toolkit BUVARD (BUffer strip for runoff Attenuation and pesticides Retention Design tool) consists in several steps like analyzing the watershed and its characteristics (soil, climate, cultural practices), and running dynamical models, in particular the mechanistic model VFSMOD (Munoz-Carpena et al., 1999). At the end this toolkit delivers the optimal VFS width considering the needed filter efficiency (for example, 70% of runoff reduction).The present study aims at making possible to simplify this very complete method, to determine its sensitivity to input parameters in new climatic and agronomic conditions by reducing the computational cost of the modeling toolkit. A metamodel (or surrogate model) developed on local conditions would allow to perform GSA with low cost yet ensuring it is based on physics. It would help users understanding the most important processes in the VFS they want to design and would increase the operational scope of the modeling toolkit. We performed a much smaller sampling of input parameters using the Latin Hypercube Sampling method optimized with a maximin criteria. The metamodel is then be based on a Kriging approach with a very good quality of prediction. It is a promising tool to perform uncertainty analysis and sensitivity analysis at alow cost.

International audience; The modeling toolkit BUVARD (BUffer strip for runoff Attenuation and pesticides Retention Design tool) consists in several steps like analyzing the watershed and its characteristics (soil, climate, cultural practices), and running dynamical models, in particular the mechanistic model VFSMOD. At the end this toolkit delivers the optimal VFS width considering the needed filter efficiency (for example, 70% of runoff reduction).The present study aims at making possible to simplify this very complete method, to determine its sensitivity to input parameters in new climatic and agronomic conditions by reducing the computational cost of the modeling toolkit. A metamodel (or surrogate model) developed on local conditions would allow to perform GSA with low cost yet ensuring it is based on physics. It would help users understanding the most important processes in the VFS they want to design and would increase the operational scope of the modeling toolkit. We performed a much smaller sampling of input parameters using the Latin Hypercube Sampling method optimized with a maximin criteria. The metamodel is then be based on a Kriging approach with a very good quality of prediction. It is a promising tool to perform uncertainty analysis and sensitivity analysis at alow cost.

45e congrès du Groupe Français des Pesticides, Versailles, FRA, 27-/05/2015 - 29/05/2015; National audience; De nombreuses études sur les zones tampons végétalisées ont permis de mettre en évidence leur capacité à atténuer les transferts de pesticides vers les eaux de surface. En particulier, les bandes enherbées ont montré leur efficacité à réduire le ruissellement en favorisant son infiltration dans le sol de la zone tampon. En revanche, on trouve très peu de références concernant le devenir des pesticides une fois infiltrés dans les bandes enherbées, et notamment leur éventuel transfert vers une nappe superficielle sous-jacente. Il est primordial d'évaluer cette situation dans le cas d'une position topologique de la bande enherbée en bas de pente, du fait de sa proximité avec le cours d'eau. Or, c'est un cas très fréquemment rencontré dans les préconisations liées aux bonnes pratiques environnementales.Dans le contexte du Beaujolais de coteaux, cette situation est exacerbée par la présence de réseaux de rigoles artificielles qui collectent et concentrent le ruissellement au sein même des parcelles de vignes, entrainant des transferts de surface rapides et conséquents. Nous nous interrogeons donc sur le risque de contamination d'une nappe superficielle sous-jacente à une bande enherbée de bas de versant, du fait de l'infiltration importante des écoulements de surface qu'elle engendre en son sein. Pour cela, nous avons effectué un traçage hydro-chimique par simulation de ruissellement afin de travailler dans des conditions semi-contrôlées. Le site expérimental se situe sur le bassin versant de la Morcille, dans le Beaujolais (69). Il est composé d'une placette enherbée isolée hydrauliquement (en surface) de 6 m de longueur sur 4 m de largeur, instrumentée par Irstea depuis 2003.Le protocole a consisté en l'apport en amont de la placette et sur toute sa largeur, d'un volume d'eau de concentration connue en ions (Bromures de Potassium) et en pesticides (Tébuconazole, Diméthomorphe, Isoproturon), afin de simuler un ruissellement provenant de la parcelle en amont équivalent à un épisode pluvieux de période de retour 2 ans. Nous avons suivi la répartition du ruissellement en surface de la placette, la part infiltrée jusqu'à 50 cm de profondeur (au moyen de plaques lysimétriques) et les variations de la nappe sous-jacente située à 2 m de profondeur (au moyen d'un transect de piézomètres). Des échantillons ont été prélevés dans chacun de ces compartiments tout au long de l'expérimentation et des échantillonneurs passifs ont permis un suivi intégratif des concentrations dans la nappe.Lors de ce traçage, le ruissellement s'est intégralement infiltré dans les 3 premiers mètres de longueur de placette. L'abattement de concentrations en traceurs (ions bromures) et en pesticides dans le ruissellement infiltré à 50 cm de profondeur est assez faible en amont de placette mais il est plus important à 4 m de l'injection (58% d'abattement en concentration pour les ions bromures et entre 75 et 95 % d'abattement pour les pesticides apportés). L'écart d'abattement entre bromures et pesticides pourrait être attribué aux phénomènes de rétention auxquels les pesticides sont soumis, notamment dans la zone racinaire de la bande enherbée. On observe que l'eau d'infiltration rejoint rapidement la nappe (en moins d'une heure) où elle engendre un pulse de contamination dans sa zone supérieure (20 premiers centimètres sous le toit de la nappe), alors que la progression des solutés en profondeur de la nappe semble bien moindre. Par ailleurs, cette forte réactivité de surface s'accompagne cependant d'un abattement important des concentrations moyennes dans la nappe sur les douze premières heures suivant le début de la simulation, (abattements de 86% pour les bromures et entre 91 et 98% pour les pesticides comparativement aux concentrations initiales dans le ruissellement). Par ailleurs, on observe une hiérarchisation nette des taux d'abattement des pesticides, en lien avec leurs propriétés d'adsorption. Les capteurs passifs utilisés, bien que ne fournissant à ce stade de leur développement que des informations semi-quantitatives, ont cependant permis d'évaluer l'évolution spatiale et temporelle de la contamination de l'amont vers l'aval et de la surface à plusieurs mètres de profondeur de la nappe.Dans ce contexte spécifique sur sols sableux du Nord Beaujolais, on peut conclure qu'une bande enherbée interceptant du ruissellement concentré permet de réduire significativement le risque de transfert par ruissellement jusqu'au cours d'eau et permet une infiltration rapide dans le sol. Malgré une dilution et rétention forte dans le sol, la contamination parvient quand même à la nappe qui peut ensuite transporter les pesticides jusqu'au cours d'eau et entrainer une contamination chronique du cours d'eau. Ces observations mettent donc en avant l'intérêt mais aussi les limites de tels dispositifs rivulaires, qui, dans les situations à risque (nappe proche de la surface, sol très perméable) pourront nécessiter de recourir à des dispositifs de limitation des transferts plus en amont.

International audience; In Europe, a significant presence of contaminants is found in surface water, partly due to pesticide applications. Vegetative filter strips or buffer zones (VFS), often located along rivers, are a common best management practice (BMP) to reduce non point source pollution of water by reducing surface runoff. However, they need to be adapted to the agro ecological and climatic conditions, both in terms of position and size, in order to be efficient. The TOPPS-PROWADIS project involves European experts and stakeholders to develop and recommend BMPs to reduce pesticide transfer by drift or runoff in several European countries. In this context, Irstea developed a guide accompanying the use of different tools, which allows designing site-specific VFS by simulating their efficiency to limit transfers using the mechanistic model VFSMOD. This method which is very complete assumes that the user provides detailed field knowledge and data, which are not always easily available. The aim of this study is to assist the buffer sizing by using a unique tool with a reduced set of parameters, adapted to the available information from the end-users. In order to fill in the lack of real data in many practical applications, a set of virtual scenarios was selected to encompass a large range of agro-pedo-climatic conditions in Europe, considering both the upslope agricultural field and the VFS characteristics. As a first step first, in this work we present scenarios based on North of France climate consisting of different rainfall intensities and durations, hillslope lengths and slopes, humidity conditions, a large set of field rainfall/runoff characteristics for the contributing area, and several shallow water table depths and soil types for the VFS. The sizing method based on the mechanistic model VFSMOD was applied for all these scenarios, and a global sensitivity analysis (GSA) of the VFS optimal length was performed for all the input parameters in order to understand their influence and interactions, and set priorities for data collecting and management. Based on GSA results, we compared several mathematical methods to compute the metamodel, and then validated it on an agricultural watershed with real data in the North-West of France . The analysis procedure allows for a robust and validated metamodel, before extending it on other climatic conditions and typical european watersheds . The tool will allow comparison of field scenarios, and to validate/improve actual existing placements and VFS sizing

The objective of this study is to evaluate the influence of soil hydrodynamic characteristics variability on surface and subsurface flows at a vegetative buffer strip scale, using mechanist modeling. Cathy (CATchment HYdrology, Camporese et al. 2010) is a research partial-differential-equation-based model, solving Richards equation in 3 dimensions for water fluxes in the soil, and a simplified scheme of Navier-Stokes equation for surface runoff. Its particularity is to handle interactions between surface and subsurface, which is a key point concerning water but also solute transport in vegetative filter strips. Balance between runoff and infiltration, flow pathways, water content, are very sensitive to hydrodynamic characteristics, especially saturated hydraulic conductivity (Ksat). This soil property is very difficult to measure and to describe at a fine scale, since it is highly variable spatially in the 3 dimensions. Models described by PDE such as Richards equation need a value of Ksat at each soil layer and each node, though simpler conceptual modeling run with average values of larger cells or storages, using some 'representative Ksat' at a larger scale. This kind of models however can simulate with high quality the processes despite the simplifications they make on parametrization. Using a mechanist and physically-based modeling, we evaluate the influence of Ksat high spatial variability on fluxes, by comparison with observations from an experimental vegetative filter strip. It should allow to understand until which degree of simplification one can describe hydrodynamic characteristics in modeling for more conceptual models.

International audience; In Europe, environmental agencies do the finding a significant presence of contaminants in surface water, which is partly due to pesticide applications. Vegetative filter strips (VFS), often located along rivers, are a common tool among other buffer zones to reduce non point source pollution of water by reducing surface runoff. However, they need to be adapted to the agro-pedo-climatic conditions, both in terms of position and size, in order to be efficient. This is one of the roles of TOPPS-PROWADIS project which involves European experts and stakeholders to develop and recommend Best Management Practices (BMPs) to reduce pesticide transfer by drift or runoff in several European countries. In this context, Irstea developed a guide accompanying the use of different tools, which allows designing VFS by simulating their efficiency to limit transfers. It needs the user to define both a scenario of incoming surface runoff and the buffer zone characteristics. First, the contributive zone (surface, length, slope) is derived from the topography by a GIS tool, HydroDem. ; 2nd, the runoff hydrograph coming in the buffer zone is generated from a rainfall hyetogram typical of the area, using Curve Number theory, taking into account soil characteristics. The VFS’s optimal width is then deduced for a given desired efficiency (for example 70% of runoff reduction), by using VFSMOD model, which simulates water, suspended matters (and pesticides) transfer inside a vegetative filter strip. Results also indicate if this kind of buffer zone is relevant in that situation (if too high, another type of buffer zone may be more relevant, for example constructed wetland). This method assumes that the user supplies quite a lot of field knowledge and data, which are not always easily available. In order to fill in the lack of real data, a set of virtual scenarios was tested, which is supposed to cover a large range of agro-pedo-climatic conditions in Europe, considering both the upslope agricultural field and the VFS characteristics. These scenarios are based on: 2 types of climates (North and South-west of France), different rainfall intensities and durations, different lengths and slopes of hillslope, different humidity conditions, 4 soil types (silt loam, sandy loam, clay loam, sandy clay loam), 2 crops (wheat and corn) for the contributive area, 2 water table depths (1m and 2.5m) and 4 soil types for the VFS. The sizing method was applied for all these scenarios, and a sensitivity analysis of the VFS optimal length was performed for all the input parameters in order to understand their influence, and to identify for which a special care has to be given. Based on that sensitivity analysis, a metamodel has been developed. The idea is to simplify the whole toolchain and to make it possible to perform the buffer sizing by using a unique tool and a smaller set of parameters, given the available information from the end users. We first compared several mathematical methods to compute the metamodel, and then validated them on an agricultural watershed with real data in the North-West of France.

International audience; In France, agriculture uses large quantities of fertilizer and pesticides. Water contamination by pesticides is highlighted by monitoring networks, at local and national levels. Control and reduction of contamination are major issues, for the protection of drinking water resources and aquatic ecosystems. Thus, understanding and quantifying the mechanisms involved in mobilization, transfer and dissipation of these substances can help to perform risk of water contamination diagnosis, and to estimate the effectiveness of corrective solutions.

/ L'implantation de bandes enherbées constitue une des méthodes correctives possibles pour réduire les pollutions diffuses des eaux de surface par les produits phytosanitaires. Le dispositif expérimental présenté ici a pour objectif de compléter la compréhension des processus déterminant le potentiel épuratoire d'un sol enherbé vis à vis d'un ruissellement contaminé, notamment en subsurface. Les questions spécifiques au transfert subsurfacique portent sur la rapidité des écoulements verticaux, sur la quantité de produits transportée au delà de la zone racinaire, sur l'hétérogénéité amont-aval des flux au sein du système enherbé, et sur l'existence d'éventuels flux latéraux. Le dispositif a été mis en place sur une placette de 6m*4m dans le Beaujolais viticole, et permet de suivre les flux entrant et sortant de la bande enherbée en surface, et de mesure les flux percolés à 50 cm de profondeur à 4 emplacements disposés le long d'un transect longitudinal.