Search

Gammarids are amphipods found worldwide distributed in fresh and marine waters. They play an important role in aquatic ecosystems and are well established sentinel species in ecotoxicology. In this study, we sequenced the transcriptomes of a male individual and a female individual for seven different taxonomic groups belonging to the two genera Gammarus and Echinogammarus: Gammarus fossarum A, G. fossarum B, G. fossarum C, Gammarus wautieri, Gammarus pulex, Echinogammarus berilloni, and Echinogammarus marinus. These taxa were chosen to explore the molecular diversity of transcribed genes of genotyped individuals from these groups. Transcriptomes were de novo assembled and annotated. High-quality assembly was confirmed by BUSCO comparison against the Arthropod dataset. The 14 RNA-Seq-derived protein sequence databases proposed here will be a significant resource for proteogenomics studies of these ecotoxicologically relevant non-model organisms. These transcriptomes represent reliable reference sequences for whole-transcriptome and proteome studies on other gammarids, for primer design to clone specific genes or monitor their specific expression, and for analyses of molecular differences between gammarid species., Measurement(s)transcription profiling assayTechnology Type(s)RNA sequencingFactor Type(s)sex • speciesSample Characteristic - OrganismGammarus • EchinogammarusSample Characteristic - Environmenthabitat Machine-accessible metadata file describing the reported data: 10.6084/m9.figshare.9777905

International audience; This study is part of a project aiming to improve our understanding of the toxicokinetics of perfluoro-alkyl substances (- PFOS, perfluorohexane sulfonate and perfluorononanoate - PFNA) in fish, using the rainbow trout (Onchorynchus mykiss) as model species. The project involves a combination of dietary exposure experiments and physiologically-based toxicokinetic (PBTK) modelling. We developed first a 10-compartment PBTK model to elucidate PFOS kinetics in adult rainbow trout. This PBTK model included various physiological characteristics: blood perfusion to each organ, plasmatic fraction, PFOS free fraction, and growth of individuals. The parameters were optimized using Bayesian inference. This development was based on an experimental dataset, where fish were first fed daily with spiked food, and allowed to depurate for an equivalent duration [1]. We evaluated the relative importance of different absorption as well as elimination routes, and the role of the entero-hepatic cycle. This model was then used to explore the effects of temperature on PFOS toxicokinetics. The modulation of physiological processes induced by temperature was introduced into the model structure thanks to Arrhenius law. These processes included food absorption, ventilation rate, growth, cardiac output, urinary, biliary and fecal clearances, and relative blood flow distribution to organs. We hypothesized that temperature could also affect PFOS tissue:plasma or tissue:blood partition coefficients. Simulations were performed for water temperatures of 7°C, 11 °C, 15 °C and 19°C and compared to experimental data [1, 2]. The simulation having the lowest Root Mean Squared Error was obtained when all physiological processes were modulated according to temperature. Overall, the model outcomes suggest that PFOS toxicokinetics was more affected by temperature induced changes in partition coefficients than in fish physiology. A sensitivity analysis is ongoing in order to confirm and refine this result. [1] Vidal A, Lafay F, Daniele G, Vulliet E, Rochard E, Garric J, Babut M. in press. Does water temperature influence the distribution and elimination of perfluorinated substances in rainbow trout (Oncorhynchus mykiss)? Environ Sci Pollut Res. [2] Goeritz I, Falk S, Stahl T, Schäfers C, Schlechtriem C. 2013. Biomagnification and tissue distribution of perfluoroalkyl substances (PFASs) in market-size rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem 32:2078-2088.

National audience; La réanalyse climatique FYRE (French Hydrometeorological REanalysis) Climate, une réanalyse ensembliste des précipitations et températures journalières à haute résolution couvrant la période 1871-2012, a permis d'améliorer la connaissance du climat passé en France (Devers et al., 2018). Cette réanalyse provient de l'assimilation « offline » d'observations historiques de précipitation et de température dans la reconstruction ensembliste SCOPE Climate (provenant d'une descente d'échelle statistique de la Twentieth Century Reanalysis, Caillouet et al., 2018). Le jeu de données FYRE Climate a été utilisé comme entrée d'une modélisation hydrologique sur 662 bassins versants faiblement anthropisés en France et couvrant une large partie du territoire. Le modèle pluie-débit conceptuel GR6J a été utilisé afin de produire une reconstruction ensembliste des débits en France depuis 1871. Lors de cette modélisation différentes sources d'incertitudes ont été prises en compte : (1) l'incertitude de la réanalyse climatique, représentée par les 25 membres de FYRE Climate, (2) l'erreur de mesure des débits, en perturbant les observations durant la calibration et (3) l'erreur du modèle hydrologique, en analysant les écarts entre débits simulés et observés. La reconstruction hydrologique ainsi produite est comparée à SCOPE Hydro (Caillouet et al., 2017) et Safran Hydro, deux jeux de données utilisant respectivement les forçages issus de SCOPE Climate et de la réanalyse déterministe Safran (sur la période 1958-2016, Vidal et al., 2010). La reconstruction hydrologique utilisant FYRE Climate et prenant en compte les trois sources d'incertitudes listées ci-dessus possède des corrélations journalières plus élevées ainsi que des erreurs inférieures à celles de SCOPE Hydro. La reconstruction des débits montre des performances similaires au jeu de données Safran Hydro, avec l'avantage de couvrir l'ensemble du XXe siècle et de fournir une incertitude associée. Cette étude montre la valeur ajoutée de la reconstruction hydrométéorologique en terme d'extension temporelle, de qualité et d'estimation de l'incertitude, et de caractérisation de la variabilité hydrologique passée en France.

International audience; Flash floods are amongst the most important natural hazards in Europe, particularly in the Mediterranean region creating economical damage and life risk. This study investigates the factors that control the hydrological response in a set of 13 catchments, partially nested and ranging from 0.17 to 42 km2, located in Ardèche, a Mediterranean region prone to flash floods. The period of interest is 2012-2015. The Claduègne catchment (42 km2), which contains the other 12 catchments, is part of the OHMCV observatory and the OZCAR research infrastructure. This catchment can be separated into two distinctively different parts with a basaltic plateau in the upstream part and sedimentary rock in the downstream part (each about 50%). The sub-catchments are distributed over these two parts and different land use forms (forest, cultivated soils and grasslands). For most of the sub-catchments, no discharge rating curve exists, so the focus of the analysis mainly lies on the time component of the hydrological response as well as on descriptors of the rainfall-runoff event and catchment properties. Several characteristic times were calculated for each catchment based on a set of 49 to 158 events depending on the catchment and various statistical methods were applied in order to estimate the importance of different factors on the hydrological response on two levels: (A) inter-comparison of the catchments with regards to their properties and averaged event characteristics. (B) comparison of the storm events for each catchment considered separately with regards to the rain characteristics, initial degree of wetness and seasonality. A simple estimate of lag time was shown to be able to characterize the dynamics of each catchment and provide an estimate of time of concentration, if calculated for a selection of impulsive events. A relationship between catchment size and parsimonious time step to describe rainfall forcing was also established. Besides the catchment's size, we found the geology to be an important factor influencing the catchments' response, with the catchments in the sedimentary part reacting faster. On the catchment-level, a strong seasonal variation of event characteristics was found. The most intense, shortest events take place in autumn, while the events in spring are particularly long and slow. Two sub-catchments have singular behaviors. One was shown to react much faster and one much slower than expected. This suggests additional in-situ investigations in order to understand the hydrological processes.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]ARCEAU [ADD1_IRSTEA]Hydrosystèmes et risques naturels; International audience; OZCAR, the French Critical Zone Research Infrastructure (RI) gathers around 20 observatories sampling various compartments of the Critical Zone, in France and in southern countries. Given their long history, most of them have developed their own data/metadata portals. There is however a need to gather all the data in a common Information System (Theia-OZCAR IS) to make the data visible, findable and easy to explore; allow their preservation and their citation; favor their reuse and sharing; be interoperable with European infrastructures by adopting international standards (controlled vocabularies and OGC compliant webservice). To build the portal, the project team visited the existing data centers and future users to understand current practices and collect future users' needs. This revealed the heterogeneous levels of data management maturity, variable names nomenclature and data exposition; and that finding data by variables was the point of interest of future users. The starting postulate when building the Theia-OZCAR IS is that databases stay next to data producers, which ensures the best data quality. A continuous information flow between observatories IS and the Theia-OZCAR IS will keep information up to date. Suppliers push data and metadata to the Theia-OZCAR IS, through a common data model, using an extraction script maintained locally and the data portal will provide an advanced search interface to explore the data as well as standardized services that can be harvested by other portals. The paper presents the approach and the first portal prototype, based on principles that can be reused by other disciplines/data portals.

SCOPE Climate (Spatially COherent Probabilistic Extended Climate dataset) is a 25-member ensemble of 142-year daily high-resolution reconstructions of precipitation, temperature, and Penman–Monteith reference evapotranspiration over France, from 1 January 1871 to 29 December 2012. SCOPE Climate provides an ensemble of 25 spatially coherent gridded multivariate time series. It is derived from the statistical downscaling of the Twentieth Century Reanalysis (20CR) by the SCOPE method, which is based on the analogue approach. SCOPE Climate performs well in comparison to both dependent and independent data for precipitation and temperature. The ensemble aspect corresponds to the uncertainty related to the SCOPE method. SCOPE Climate is the first century-long gridded high-resolution homogeneous dataset available over France and thus has paved the way for improving knowledge on specific past meteorological events or for improving knowledge on climate variability, since the end of the 19th century. This dataset has also been designed as a forcing dataset for long-term hydrological applications and studies of the hydrological consequences of climate variability over France. SCOPE Climate is freely available for any non-commercial use and can be downloaded as NetCDF files from https://doi.org/10.5281/zenodo.1299760 for precipitation, https://doi.org/10.5281/zenodo.1299712 for temperature, and https://doi.org/10.5281/zenodo.1251843 for reference evapotranspiration.

The QGIS plugin "Surcotes" permit to pass from a map of a well known flood to an extreme flood map. / Dans le cas où on dispose d'une carte d'une inondation de référence, mais pas du modèle de calcul, la méthode des surcotes permet d'estimer la zone inondée supplémentaire pour un débit supérieur au débit de référence. Le principe de calcul est l'application de la formule du régime uniforme dans quelques sections en travers de la vallée puis une interpolation en niveau d'eau afin d'établir la limite de la zone inondée par confrontation avec un modèle numérique de terrain. L'outil maintenant librement disponible se présente comme un plugin de Q-GIS, ce qui permet de visualiser les résultats et de choisir ainsi le meilleur paramétrage. La méthode repose sur la capacité de l'utilisateur à sélectionner tronçons et profils représentatifs et à caler leurs paramètres pour s'ajuster à la carte de référence. L'établissement d'une nouvelle carte d'inondation pour un autre débit s'effectue ensuite automatiquement avec un éventuel ajustement des paramètres aux caractéristiques de la vallée. Les limitations d'une telle méthode sont bien connues : nécessité d'une propagation de l'inondation selon un axe de vallée et absence d'aménagements ou de singularités topographiques qui écartent la ligne d'eau de celle du régime uniforme. La portée de ces limitations est ici réduite par l'utilisation d'une base constituée par une carte d'inondation de référence. Les erreurs sont donc limitées car, d'une part, le calcul ne porte que sur l'extension de la zone inondée et d'autre part, l'ajustement du modèle à la carte de référence oblige à choisir un jeu de paramètres qui puisse en partie compenser l'écart entre la vallée réelle et le canal idéal où le régime uniforme serait installé. En outre, l'utilisateur a le choix entre deux variantes de calcul, ce qui lui permet d'avoir une estimation de l'incertitude due à la méthode de calcul utilisée. Le plugin Surcotes est donc adapté à l'établissement de cartes des inondations extrêmes (dans le sens, plus rares que les crues référencées) que ce soit dans le cadre d'une étude préliminaire (diagnostic ou aménagement) ou d'une gestion de crise, cas où l'examen de multiples scénarios est privilégié par rapport à la précision sur l'extension de l'inondation d'un scénario.

National audience; Les contaminants organiques tels que les pesticides et les médicaments présents dans les cours d'eau peuvent se fixer sur les litières de feuilles, les sédiments et les matières en suspension. Ces substances potentiellement toxiques peuvent donc avoir une incidence sur le fonctionnement des écosystèmes lotiques. La détermination des niveaux de contaminations dans ces matrices solides environnementales est donc primordiale pour comprendre et appréhender les effets de ces contaminants dans les milieux aquatiques. Ces analyses de micropolluants à l'état de traces dans de telles matrices constituent un défi de taille, car elles requièrent des techniques analytiques efficaces pour atteindre une sensibilité et une spécificité appropriées. L'extraction et la purification avant analyse chromatographique sont des étapes cruciales qui nécessitent une optimisation afin de maximiser les rendements d'extraction et de minimiser les effets de matrices. En comparaison avec d'autres techniques classiques, la méthode QuEChERS (Quick, Easy, Cheap, Effective, Rugged and Safe), couramment utilisée pour la sécurité alimentaire depuis 2003, présente l'avantage d'être, simple à mettre en oeuvre, adaptable à tous types de matrices solides, plus respectueuse de l'environnement et permet d'obtenir des extraits relativement propres. L'objectif de cette étude est donc d'optimiser les étapes d'extraction et de purification par QuEChERS. La quantification de 15 pesticides agricoles dans les litières de feuilles et de 49 contaminants d'origine urbaine dans les sédiments est ensuite réalisée par chromatographie liquide ultra haute performance couplée à la spectrométrie de masse en tandem (UHPLC-MS-MS). Le premier groupe de 15 pesticides (2 herbicides et 2 métabolites, 6 fongicides et 5 insecticides) a été sélectionné en fonction de la fréquence de détection de ces composés dans les cours d'eau soumis aux contaminations agricoles diffuses. Le deuxième groupe de 49 substances est représentatif de pollutions urbaines (8 pesticides, 5 bétabloquants, 10 antibiotiques, 4 médicaments hypolipémiants, 2 bronchodilatateurs, 4 benzodiazépines, 2 antidépresseurs, 1 analgésique, 1 diurétique, 1 antidiabétique et 6 métabolites). Ces composés ont été choisis en fonction de leur présence dans les effluents de stations d'épuration. Tous ces composés présentent une large gamme de caractéristiques physico-chimique (-0.5 < log Kow < 5). Afin de sélectionner la méthode d'extraction la plus efficace pour chacun des deux groupes de micropolluants, nous avons tout d'abord comparé la méthode QuEChERS avec une méthode d'extraction liquide sous pression (PLE). Un plan d'expérience a été réalisé pour optimiser la masse et la nature des sels d'extraction et des sels de purification, les temps d'extraction et de centrifugation. Enfin, pour réduire les effets de matrice qui induisent une suppression de signal en UHPLC-MS-MS, des étapes supplémentaires de purification en "Pass through" sur Solid Phase Extraction (SPE) et de dilution des extraits ont été effectuées. Les extractions QuEChERS optimisées couplées à l'analyse par UHPLC-MS-MS permettent d'obtenir des rendements d'extraction compris entre 45 et 135 % en fonction des composés et les écart-types relatifs obtenus sont inférieurs à 20% (n=3). De faibles limites de quantification, allant de 2 ng/g sec à 200 ng/g sec selon le composé et la matrice ont été obtenues. Nous devons maintenant valider et déterminer les incertitudes de ces deux méthodes spécifiques et sensibles avant leur application pour des études de cas réelles.

/ Depuis la décennie 1970-1980, la Commission Internationale pour la Protection des Eaux du Léman (CIPEL) réalise périodiquement des campagnes d'analyse de micropolluants dans la chair de poissons, notamment des lotes (Lota lota). En 2018, 17 lotes et 28 gardons (Rutilus rutilus) ont ainsi été analysés, pour une large gamme de contaminants, notamment mercure (Hg), polychlorobiphényles (PCB), polybromodiphényl-éthers (PBDE), ainsi que 14 substances perfluoroalkylées et 12 retardateurs de flamme halogénés. Les mesures ont été interprétées en fonction des effets potentiels sur la faune piscivore, plutôt qu'en fonction de normes à visée sanitaire comme cela avait été le cas des campagnes précédentes. Le Léman apparait peu contaminé en ce qui concerne les PBDE, les chloroalcanes à chaîne courte, les PCB indicateurs (congénères 28, 52, 101, 138, 153 et 180), et la plupart des retardateurs de flamme chlorés et bromés alternatifs aux PBDE. En revanche, les concentrations moyennes dépassent les critères d'évaluation respectifs pour le mercure et les PCB de type dioxine (PCB-DL). Pour le PFOS, les concentrations moyennes sont inférieures au critère d'évaluation, mais quelques individus le dépassent. Un retardateur de flamme alternatif, le DBDPE, a également été détecté de manière systématique. Les concentrations de mercure et de PCB, qui ont fait l'objet d'un suivi de longue date notamment dans la chair de lotes, ont notablement baissé au fil des décennies, et semblent se stabiliser. Il est actuellement prématuré de statuer sur les tendances suivies par les PBDE et le PFOS dans le Léman. Des tendances comparables sont également observées dans d'autres lacs de la région comme le Lac Majeur, ou certains des Grands Lacs nord-américains, tandis que dans d'autres lacs ces dernières années ont vu les concentrations en mercure dans certaines espèces de poissons s'accroitre à nouveau. Le critère d'interprétation du mercure, qui est identique à la norme de qualité environnementale (NQE) retenue pour les eaux continentales dans l'Union Européenne, vise à protéger la faune piscivore des effets du méthyl-mercure [1]. Ce critère parait cependant problématique dans la mesure où il est systématiquement dépassé, alors même que les efforts visant à limiter les rejets de cet élément semblent avoir atteint leur limite, comme le suggère la tendance à long terme. Dès lors le niveau moyen actuel pourrait représenter une sorte de « bruit de fond anthropique » difficile à dépasser [2]. 1. Anonymous, Environmental Quality Standards (EQS) Substance Data Sheet - Priority Substance n°21 Mercury and its Compounds. 2015, European Commission: Brussels (B). p. 21. 2. Vignati, D.A.L., et al., Mercury environmental quality standard for biota in Europe: Opportunities and challenges. Integrated Environmental Assessment and Management, 2013. 9(1): p. 167-168.

1 .pdf copy (1 Pag.) of the abstract published by the Organization., The PIRAGUA project, funded by FEDER through the POCTEFA Programme of the EU, addresses the assessment of the hydrological cycle of the Pyrenees in the context of climate change. The goals of the project are to unify and homogenise the existing information, prospect future scenarios, develop indicators of change, and propose adaptation strategies with impact on the territory, with the ultimate goal of supporting investment aimed at adapting to climate change in relation to water resources.

/ 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.

/ Les résidus médicamenteux sont peu éliminés par les stations de traitement des eaux usées (STEU) et se retrouvent dans les écosystèmes aquatiques provoquant une contamination chronique persistante des eaux de surface et des sédiments. Parmi ces résidus, les antibiotiques peuvent également s'accumuler dans le biote et ont été détectés à tous les niveaux du réseau trophique (Aubertheau et al. 2017). Dans l'environnement, la présence d'antibiotiques est en particulier susceptible d'altérer les fonctions et la structure des communautés microbiennes et plus spécifiquement d'exercer une pression de sélection sur les bactéries, avec un risque avéré de développement d'antibiorésistances. Dans ce contexte, nous appliquons une approche multi-compartiments (eau de surface, périphyton, sédiment) et multi-sites incluant des écosystèmes lotique (rivière Arve) ou lacustre (lac Léman) afin de comparer la réponse des communautés microbiennes naturelles issues de sites présentant des expositions contrastées aux résidus de médicaments en provenance de STEU. Pour chaque site étudié, nous avons évalué le potentiel fonctionnel des communautés microbiennes du périphyton et des sédiments (photosynthèse, activités hétérotrophes extra-cellulaires) ainsi que leur tolérance à 2 antibiotiques (ciprofloxacine, sulfaméthazine) suivant une approche PICT et leur capacité à biodégrader les sulfonamides. Ces analyses biologiques seront complétées par l'étude du résistome et de la structure des communautés puis mises en relation avec les concentrations des résidus médicamenteux dans le périphyton, les sédiments et les eaux de surface. Les premiers résultats ont notamment mis en évidence dans le lac Léman une plus forte tolérance microbienne à la ciprofloxacine au site proche du rejet de STEU qu'au site plus protégé. Ces premiers résultats seront complétés et contribueront à une meilleure connaissance de la répartition des antibiotiques et des antibiorésistances dans les différents compartiments aquatiques ainsi qu'à une meilleure compréhension des liens entre structure du résistome et fonctions associées à ces résistances (tolérance ou biodégradation) chez les communautés microbiennes.

International audience; The French part of the Meuse River catchment was identified as vulnerable to drought hazard by water managers, and this should be critically exacerbated in a changing climate where increasing drought magnitude is expected. Water managers need new statistical tools to better anticipate drought hazard, making best use of the available meteorological data and sometimes spatially scarce hydrological data. This issue is currently explored within the CHIMERE21 project supported by the Rhine-Meuse Water Agency, in which the impacts of climate change on water management are analysed. More specifically, simple and informative indicators are investigated to help decision in managing water scarcity on this international basin. In France, the meteorological services (Météo-France, MF) provide maps of diverse gridded hydro-meteorological indices at a 10-day or 1-month time-step directly to the Water Agencies, including maps of the Standardized Precipitation Index (SPI) and the Standardized Soil Wetness Index (SSWI). The SPI can be used to quantify meteorological drought while the SSWI can help characterising soil moisture drought. Hydrologists of DREAL Grand-Est (Direction Régionale de l'Environnement, de l'Aménagement et du Logement), the public body in charge of water management strategy in the Rhine-Meuse basins, are willing to use these maps for hydrological drought characterization in order to better understand drought hazard and adapt water management accordingly. To define drought hazard, DREAL Grand-Est generally uses a threshold approach, the threshold being calculated with a 3-day moving average minimum flow (VCN3) for different return periods (2, 5, 10, and 20 years). When river flow drops below one of these thresholds, a more or less severe drought state is declared and drought management measures are adopted consequently. The work presented here involves water managers and researchers to develop an innovative approach for drought characterization at the sub-catchment scale. It aims to use a logistic regression approach and build a relationship between the probability of reaching a given drought severity stage (VCN3 for different return periods) and observed hydro-meteorological indices (SPI and SSWI). The French Meuse River catchment is used as a case study to develop this approach. MF maps of SSWI and SPI are available on an 8-km grid. VCN3 indicators are spatially computed at the subcatchment scale based on observed discharge for 16 stations. Since the ultimate aim is to apply this method on ungauged catchments, regressions are built on a subset of these stations and assessed on the rest of them to estimate their transferability in space.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]BELCA [ADD1_IRSTEA]Systèmes aquatiques soumis à des pressions multiples [Departement_IRSTEA]Eaux [TR1_IRSTEA]BELCA [ADD1_IRSTEA]Systèmes aquatiques soumis à des pressions multiples; International audience; This data article associated with the manuscript "Comparative proteomics in the wild: accounting for intrapopulation variability improves describing proteome response in a Gammarus pulex field population exposed to cadmium" refers to the shotgun proteomics analysis performed on 40 Gammarus pulex animals sampled from the wild. Proteins were extracted, digested with trypsin, and the resulting peptides were identified by tandem mass spectrometry. Here, we present the list of proteins from males and the list of proteins from females that are differentially detected between the Brameloup and the Pollon populations. Data are available via ProteomeXchange with identifiers PXD013656 and PXD013712, respectively.

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.

/ L'adaptation in situ des communautés microbiennes aux contaminants est le résultat de différents processus favorisant la sélection des espèces tolérantes et la disparition des espèces sensibles. De nombreuses études ont démontré l'intérêt de caractériser cette capacité d'adaptation pour diagnostiquer un niveau d'exposition à un contaminant chimique suffisant pour induire une modification de la communauté microbienne étudiée[1], suivant le concept PICT (Pollution Induced Community Tolerance[2]). Récemment, la mesure de la tolérance au cuivre de communautés microbiennes benthiques issues de sédiments superficiels provenant de différentes rivières nous a permis de mettre en évidence une relation exponentielle entre la tolérance microbienne au cuivre et les niveaux de contamination in situ dans ce compartiment. Ces premiers résultats suggèrent l'existence d'une tolérance microbienne basale qui permettrait de dépasser l'approche relative (i.e. la comparaison de sites contaminés avec une référence) utilisée jusqu'à présent. Néanmoins, l'application de l'approche PICT au sédiment, peu pris en considération jusqu'alors, a mis en évidence une problématique propre à cette matrice. En effet, les caractéristiques de cette matrice (granulométrie, quantité et qualité de matières organiques...) peuvent affecter la bio-disponibilité des contaminants lors des tests de toxicité et influencer leurs résultats indépendamment de la tolérance microbienne. Afin de s'affranchir de cet « effet matrice », nous avons comparé plusieurs méthodes pour détacher les communautés microbiennes du sédiment en conservant au maximum leur viabilité et activité (prérequis pour pouvoir réaliser les tests de toxicité). L'homogénéisation mécanique dans un tampon pyrophosphate a donné les meilleurs résultats et son application à des communautés microbiennes exposées au laboratoire à différents niveaux de cuivre a permis de mettre en évidence une acquisition de tolérance en lien avec l'exposition.

Within continental waters, large lakes present a special interest for monitoring persistent contaminants, because of their physical characteristics, particularly a prolonged residence time. Moreover, professional fisheries take place in these large lakes, entailing human exposure to chemicals bioaccumulated. For these reasons it has long been acknowledged that monitoring persistent contaminants in fish is an essential component of environmental and health risk assessment in such large lakes. We report data on fish contamination by a range of perfluoroalkyl substances for seven European deep lakes located in populated areas, namely Lake Geneva, Lake Lugano and Lake Maggiore (all three being trans-national lakes), and Lake Iseo, Lake Como, Lake Garda and Lake Mergozzo in Italy. In addition, fish from two high altitude alpine lakes in Switzerland (>2000 m) were also sampled. These high altitude lakes were assumed to be references sites, as only atmospheric deposition should bring PFAS to them. All the other lakes are located in highly urbanized areas but without known industrial PFAS sources. A range of perfluorocarboxylic and perfluorosulfonic acids were analyzed in fillets from seven fish species occupying different habitats. Furthermore, in order to be able to assess wildlife as well as human exposure, different body fractions were analyzed separately in selected cases. Along with a summary of the results (in terms of concentration ranges and descriptive statistics), we will provide insights on the effect of seasons or type of PFAS source, and discuss about the relationships between concentrations in fillets and whole-fish.

/ La perturbation des communautés biologiques causée par la contamination chimique des écosystèmes aquatiques est encore mal comprise et l'émergence de nouveaux contaminants dans les cours d'eau pose la question de la vulnérabilité de ces écosystèmes face au développement des activités humaines. Aujourd'hui, la toxicité des substances et des milieux pour les populations aquatiques est évaluée en écotoxicologie par la mesure d'effets individuels (mortalité, baisse de fertilité...) ou subindividuels (biomarqueurs). Au sein du laboratoire d'écotoxicologie d'Irstea Lyon, j'ai contribué au développement d'une méthodologie de biosurveillance active qui procède par le suivi de telles réponses sur des individus encagés appartenant à l'espèce Gammarus fossarum (crustacé amphipode) choisie comme sentinelle des communautés d'invertébrés. L'opérationnalité de cette méthodologie de biotests in situ réside dans notre capacité à avoir pu intégrer la variabilité des conditions environnementales dans la lecture des réponses biologiques grâce à une approche de modélisation. Mon laboratoire dispose aujourd'hui d'indicateurs de contamination biodisponible et de toxicité pour de nombreuses rivières, à l'échelle des réseaux de surveillance nationaux. Ce type d'informations (rarement obtenues avec une telle dimension et à de telles échelles spatiales) renseigne sur la toxicité des milieux, mais certains verrous scientifiques (changement de niveau d'organisation biologique, échelles temporelles et spatiales, diversité et adaptation des populations) font que l'extrapolation de ces réponses pour prédire le risque pour les populations en place n'est pas triviale. Pour répondre à cette problématique, j'ai travaillé en premier lieu au développement de modèles de dynamique de populations qui tiennent compte de l'histoire de vie et de la phénologie de ces espèces d'invertébrés et permettent de traduire les réponses mesurées sur les traits de vie en termes d'impact démographique potentiel (indicateur populationnel). Par ailleurs, j'applique une approche de biologie comparative pour étudier la variabilité entre populations de gammaridés de la sensibilité à certains contaminants, leur capacité d'adaptation à la contamination chronique des milieux, ou encore l'éventuel signal phylogénétique qui pourrait sous-tendre ces hétérogénéités entre populations et espèces. Aujourd'hui, mon équipe poursuit ces questionnements en visant une compréhension moléculaire des vulnérabilités, des différences de sensibilité entre populations face au stress toxique. Pour avancer dans les capacités prédictives de ces approches d'écotoxicologie vis-à-vis des impacts à des niveaux d'organisation supérieurs, mon projet scientifique à court terme est de confronter les patrons de réponses observés aux échelles individuelles, leur extrapolation en termes d'effets populationnels potentiels, à l'observation de l'état des communautés écologiques en place. Pour cela, je peux notamment tirer bénéfice des bases de données d'indicateurs écotoxicologiques acquises à l'échelle nationale sur les réseaux de surveillance et aux données de bioindication disponibles sur ces mêmes réseaux qui peuvent permettre de dresser un état des communautés d'invertébrés en place. Cette démarche permettra de reposer les questions de différences de vulnérabilité au stress toxique entre espèces, leur capacité d'adaptation à la contamination, ceci à une large échelle spatiale et dans des contextes de multi-contamination représentatifs des cours d'eaux nationaux. Ce même type de raisonnement pourra également être posé à des échelles spatiales plus faibles, par exemple celle de bassins versants sur lesquels un lien entre pression chimique, toxicité et dégradation écologique pourra être établi au regard de sources de contamination identifiées. Mon projet veut ainsi contribuer plus largement à une meilleure prédiction de l'effet des perturbations anthropiques sur les écosystèmes aquatiques, et à l'évaluation du rôle des pressions chimiques toxiques dans la dégradation de la qualité biologique de ces milieux.

/ Les résidus médicamenteux sont peu éliminés par les stations de traitement des eaux usées (STEU) et se retrouvent dans les écosystèmes aquatiques provoquant une contamination chronique persistante des eaux de surface et des sédiments. Parmi ces résidus, les antibiotiques peuvent également s'accumuler dans le biote et ont été détectés à tous les niveaux du réseau trophique (Aubertheau et al. 2017). Dans l'environnement, la présence d'antibiotiques est en particulier susceptible d'altérer les fonctions et la structure des communautés microbiennes et plus spécifiquement d'exercer une pression de sélection sur les bactéries, avec un risque avéré de développement d'antibiorésistances. Dans ce contexte, nous appliquons une approche multi-compartiments (eau de surface, périphyton, sédiment) et multi-sites incluant des écosystèmes lotique (rivière Arve) ou lacustre (lac Léman) afin de comparer la réponse des communautés microbiennes naturelles issues de sites présentant des expositions contrastées aux résidus de médicaments en provenance de STEU. Pour chaque site étudié, nous avons évalué le potentiel fonctionnel des communautés microbiennes du périphyton et des sédiments (photosynthèse, activités hétérotrophes extra-cellulaires) ainsi que leur tolérance à 2 antibiotiques (ciprofloxacine, sulfaméthazine) suivant une approche PICT et leur capacité à biodégrader les sulfonamides. Ces analyses biologiques seront complétées par l'étude du résistome et de la structure des communautés puis mises en relation avec les concentrations des résidus médicamenteux dans le périphyton, les sédiments et les eaux de surface. Les premiers résultats ont notamment mis en évidence dans le lac Léman une plus forte tolérance microbienne à la ciprofloxacine au site proche du rejet de STEU qu'au site plus protégé. Ces premiers résultats seront complétés et contribueront à une meilleure connaissance de la répartition des antibiotiques et des antibiorésistances dans les différents compartiments aquatiques ainsi qu'à une meilleure compréhension des liens entre structure du résistome et fonctions associées à ces résistances (tolérance ou biodégradation) chez les communautés microbiennes.

/ Le code ARFAN (« Areal RainFall ANalyses ») permet l'analyse multidurées de chroniques de pluie à pas de temps fixe. Il peut traiter les chroniques ponctuelles, ou des « pluies de bassin » qui en sont déduites (méthode des polygones de Thiessen 'dynamique' recalculant les coefficients de pondération en fonction des lacunes, calcul géostatistique par simulations conditionnelles, permettant d'estimer l'incertitude [code externe SAMPO]). L'échantillonnage est réalisé en maxima annuels et par dépassement de seuils (avec contrôle du choix du seuil et des critères d'indépendance, et validation de l'échantillon obtenu). L'ajustement des lois de probabilité reconnues donne les quantiles. Répéter sur la chronique moyennée sur plusieurs durées produit les courbes Intensité-durée-Fréquence décrivant le régime des pluies fortes (de bassin le cas échéant). Ces IdF sont classiquement utilisées pour construire des « pluies de projet probabilistes », moyennant le choix a priori d'une forme, symétrique ou non (triangle, double triangle, hyétogramme « moyen »). Cependant, cet objet est subjectif et très réducteur. La pluie « de bassin » suppose une hypothèse d'homogénéité spatiale des pluies ; or la variabilité temporelle et spatiale des phénomènes est considérable y compris au sein du bassin, et gouverne l'effet concret de tous dispositifs. Une stratégie de gestion des crues, basée ou non sur des ouvrages à dimensionner, doit considérer cette variabilité des phénomènes. Un progrès possible est de travailler par simulation continue : générer des chroniques de débit à partir de longues chroniques de champs de pluie stochastiques. La méthode est plus exigeante à mettre en oeuvre, mais restitue une variabilité riche et réaliste, nécessaire pour évaluer véritablement l'aléa d'inondation et l'effet d'ouvrages sur cet aléa. ARFAN est un outil de recherche ouvert et évolutif, destiné à tester des méthodes. Une version stable sera aussi disponible pour des utilisations pédagogiques et à l'attention des services de l'État et des bureaux d'étude. Ces travaux bénéficient du soutien du Ministère français chargé de l'Environnement (convention Multirisques) et de partenariats (PHC franco-algérien Tassili ; université autonome de Querétaro, Mexique). SAMPO a été soutenu par divers acteurs (ANR Floodscale, CCR, EdF, Sintef).

International audience; Streamflow estimation in real time is required for data assimilation by flood forecasting systems. Streamflow time series are established using 'rating curves' that approximate the stage-discharge relationship. A major problem is that the river bed at many hydrometric stations can evolve due to morphogenic floods, leading to rating changes, or 'shifts'. The methods proposed in the literature for updating the rating curve are based on a retrospective statistical analysis of calibration data (or gaugings). The aim of the proposed method is the automatic detection of shifts and the estimation of their magnitude in real time, using both observational data and hydraulic knowledge of the site. It is built on an existing Bayesian framework for estimating probabilistic stage-discharge rating curves with quantitative uncertainties. Before starting a real-time application, a retrospective analysis is performed to detect past changes typically based on available gaugings. The retrospective results are then used to calibrate the methods for detecting and estimating shifts and their magnitudes in real-time. To this aim, the method uses different sources of information available in real time, e.g. stage records, estimates of the sediment transport that affects the bed evolution, changes in the shape of recession curves, changes in correlations with neighbouring hydrometric stations, etc. The proposed method was applied to the Ardèche River at Meyras, France, a river with a gravel-bed degrading during each flood. Both retrospective analysis and real-time estimation of the stage-discharge rating curves yielded encouraging results, with consistent detection of shifts and estimation of their magnitudes. Further research includes testing the method at more challenging sites and the use of other useful information available in real-time. The final goal is to provide real-time streamflow estimates with uncertainties that can be assimilated in models and decision making.

[Departement_IRSTEA]Eaux [ADD1_IRSTEA]Hydrosystèmes et risques naturels; International audience; Several research groups have been developing algorithms to calculate river discharges using satellite measurements only. However, the use of ground measurements is useful to greatly reduce the uncertainties. We have developed and tested an original Bayesian method to estimate streamflow at hydrometric stations using satellite measurements of water surface elevation (stage), width and slope as input data. The tests were conducted using an existing set of synthetic data defined in the context of the future SWOT satellite mission for a range of rivers. Typical sets of ground discharge measurements (1-24 gaugings) were created by sampling these synthetic data and adding measurement errors. We found that the reach-averaged cross-sectional geometry of each site can usually be approximated by a single rectangular, trapezoidal, parabolic or triangular shape, or a combination of two shapes for low and high flows. Based on such cross-sectional shapes, simple stage-width-slope-discharge models can be derived based on the Manning equation for fairly steady and uniform flows. A few gaugings in each flow segment of the model are necessary to estimate the parameters and produce discharge records with acceptable uncertainty. Our results show that it is possible to apply methods used for ground hydrometric stations to virtual stations created by satellite remote sensing throughout large river networks. Fewer ground discharge measurements are required because the satellite provides the water surface width and slope in addition to the stage. The proposed Bayesian approach bridges the gap between traditional ground-based hydrometry and river hydraulics remote sensing, and observational strategy can be optimized based on uncertainty quantification. Figure: Estimates of a parameter of the model: the mean elevation h0 of the Sacramento River bed using 2-24 gaugings. Boxplots show 95% probability intervals and the medians. Prior distribution is shown in white (0 gauging).

International audience; The estimation of sand fluxes in large rivers is an important issue because of its significant role on the river morphodynamic. However, these fluxes are very difficult to measure and compute because of their large spatial and temporal variability. Equations can be used to compute the sediment transport capacity but the actual sediment transport may be lower than this capacity when sand supply is limited, especially for gravel bed rivers like the Rhône River, France. This study presents the results of a comparison between recent flux measurement surveys on the Rhône River in Lyon and a simple physically based model for computing the sand transport capacity. Flux measurements were achieved combining ADCP measurements and direct sampling using a Niskin sampler. Seven field surveys were carried out, most of them during a 10-year flood in 2018. The range of investigated discharges varies from 900 m3/s to 2700 m3/s during different phases of the flood. Each experiment is made of a set of 12 samples distributed on a cross section, velocities and bathymetry being acquired with an ADCP. The analyses were carried out by sieving and filtering. On the other hand, sediment transport is estimated using a semi-empirical sediment transport formula for suspended load. The model is based on stage-discharge relationships to estimate the main hydraulic parameters, such as the bed shear stress. This model was implemented on two cross-sections with relatively different hydrodynamic conditions, and results were compared with measurements. It is interesting to observe that the transport capacity is reached for the less dynamic cross-section, which appears to control sand transport over the whole reach.We established a suspended sand rating curve for this cross section and experimental results were correctly predicted by the model. However, uncertainties need to be evaluated before establishing reliable sand fluxes time series and budgets.

International audience; The development of populated areas near mining activities, an important source of metal contamination, poses a challenge for urban sustainability and public health, which must be addressed in a multidisciplinary way. Street dusts are a relevant indicator of environmental pollution, as they are an important exposure pathway and record recent metal accumulation. Risk assessment is an effective tool to evaluate and manage these risks. The majority of these analyses, however, do not evaluate the particle size used to represent environmental exposure, despite its importance, historically using a broad range.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]ARCEAU [ADD1_IRSTEA]Hydrosystèmes et risques naturels; International audience; The Rhône river basin is a highly populated area in southeastern France, with a high concentration of alpine glaciers which have been rapidly retreating in the last decades as the climate warms. In order to anticipate potential issues and conflicts concerning water resources in the high alpine valleys, there is a strong need for local-to-regional projections of glacier and snow hydrological contributions. In this study, we present ALPGM (ALpine Parameterized Glacier Model), a fully parameterized glacier model, which makes use of the most recent available data and reanalysis to simulate the evolution of all the ~660 glaciers in the French Alps for the 1984-2100 period. It computes glacier-wide yearly surface mass balances (SMB) using machine learning, and it updates the glacier geometry by redistributing the glacier-wide SMB values along the glacier elevations according to a glacier-specific parameterized function.

International audience; ARFAN, pour Areal RainFall ANalyses, est un prototype d'outil d'analyse de chroniques de pluie au pas de temps fixe, développé en Python3 pour automatiser les traitements habituels et proposer des traitements spécifiques, avec des sorties graphiques adaptées. Il permet en particulier d'estimer des pluies "de bassin", pour l'instant avec la méthode sommaire mais répandue des polygones de Thiessen, réajustée en fonction des lacunes. La chronique de bassin peut ensuite être échantillonnée (max annuels, sup-seuil) pour en tirer des IdF de bassin ainsi qu'une forme "moyenne" d'événement. Ceci permet de construire des "pluies de projet", qui, comme les "crues de projet" sont bien pratiques en ingénierie, mais en fait très critiquables ; les approches par "simulation continue" sont préférables car elles respectent la variabilité spatiale et temporelle des pluies.

International audience; Identifying organic contaminants present in aquatic environments is of major concern for water quality monitoring. New analytical and data treatment tools have been developed to meet this challenge. The sample preparation before analyses is the first crucial step to be optimized before using suspect and non-target-screening approaches for qualitative evaluations of water pollution by contaminants from various chemical and usage families. As part of the action of Aquaref, which is the French reference laboratory for environmental monitoring in the Water Framework Directive, different solid phase extraction conditions (SPE, n=3 per condition) were tested towards UHPLC-HRMS (separation by UHPLC carried out with a C18) analysis. The SPE protocols varied according to the extraction phase (Oasis HLB or multilayer cartridge), the elution solvent (methanol or methanol and dichloromethane) and sample pH (pH 3 or 7). To ensure data quality, blank solvents, extracted blanks and quality control (QC) were also injected. The QC sample corresponds to a pool of 5 µL of each sample extract and is injected repeatedly throughout the injection sequence to monitor the analytical repeatability. By suspect screening strategy, most of the investigated compounds were found in all SPE extracts. The results revealed the presence of contaminants such as pesticides, pharmaceuticals, sweeteners, or anti-corrosive agents. For pesticides, approved (e.g. diuron) and not approved (e.g. terbutryn) compounds were both detected. Concerning the pharmaceuticals, drugs acting on the nervous system were most common (e.g. oxazepam), as well as betablockers (e.g. metoprolol). The without any a priori non-target approach highlighted that the SPE conditions affect the quality and the interpretation of the data. The number of signals (features) detected by each method varies, as does the number of signals associated with background noise, for each sample preparation method tested. These results support the suspect screening results. Data treatment revealed that the two SPE with Oasis HLB extraction phase at pH 7 (elution with methanol or methanol and dichloromethane) lead to detection of more signals with m/z above 600 Da. Based on the retention time distribution of signals detected for each SPE methods, the multilayered cartridge allowed the extraction of more polar compounds than the other SPE methods. The comparison of all the signals detected with each SPE condition revealed that 31% to 57% of them are specific to only one SPE method. The evaluation of sample preparation methods prior to suspect and non-target screening is required before data treatment to assess the presence of contaminants in water.

International audience; Transfer and fate of pesticides are highly influenced by the presence of discontinuities, such as grass strips, ditches, roads, etc. that can accelerate or slow down and dissipate water and contaminant flows. Those landscape features must thus be integrated into watershed management plans. Taking these elements into account in modelling tools is a useful way to evaluate their efficiency on pesticide transfers at the watershed level. However, if processes in landscape features have already been widely modelled locally, models generally don't reach the catchment scale. The PESHMELBA model aims at developping a spatially-explicit modeling tool of water and contaminants circulation and fate at the scale of small catchments in order to optimise the landscape organisation. The model takes into account the spatial organisation of the landscape by representing explicitly the existing features, their location and shape. The final aim of this modelling tool is to compare different scenarios in order to assess the influence of agricultural/landscape management practices on water quality. As a preprocessing step, a meshing representative of the landscape configuration is designed. Instead of being regular as usually done, each mesh unit coincides with one homogeneous landscape element. The final meshing is composed of surface elements (plots, plots with grass strips, grassland, etc.) and linear elements (landscape discontinuities such as ditches, banks or hedgerows, etc.). GeoMELBA, a dedicated tool for landscape discretization was developped to deduce the meshing from different GIS layers provided by the user. Soil types and land uses are overlayed and intersected to deduce surface homogeneous units and linear elements are resliced to fit surface elements boundaries. In addition to shaping the area of application of each modelling unit, this method also aims at defining connectivity between landscape features accordingly to specific flow rules. Dominant processes ruling water and contaminants circulation on each landscape mesh are partly represented by existing and validated models. For example, the Richards equation is solved with the fast solution from \cite{Ross2003} to represent water infiltration, and VFSMOD \cite{Munoz-Carpena1993} is used for grass strips modelling. New modules have also been designed to handle specific elements such as ditches. Each model stands for one or several processes governing water and pesticide transfers and fate in an element. Those modelling units present different levels of complexity in term of representation, from conceptual to physically-based approaches. In order to implement the spatial and temporal couplings, the different modules are connected in OpenPALM, an open-source coupler \cite{Fouilloux1999}. Although this coupler is not hydrology-oriented and doesn't have existing modules for hydrological processes as some existing modelling frameworks, its flexibility and large range of possibilities for coupling models allow to obtain a spatialised model of the whole catchment. In this current phase of construction of the PESHMELBA model, each landscape feature and its dominant processes are integrated one by one and the physical consistency of each module is evaluated. In this study, we will show an application of the PESHMELBA model based on la Fontaine du Theil catchment (France) soil profiles and climatic events. A simplified landscape composed of ditches and plots (fig. \ref{fig:ex}, left) is set up to carry out preliminary tests that aim at assessing the correct functionning of the unitary models and interaction between them. The preprocessing step with GeoMELBA consists in defining the surface and linear elements and assessing the connectivities. Processes that are represented in this example are vertical infiltration, lateral saturated transfers, root extraction, and transformation of pesticides. The first test application simulates water and solute transfers in linears (ditches) and plots on a small virtual catchment with an homogeneous silty-loam soil, after a 8h rainfall event. They highlight two different response speeds for pesticide transfers associated to their pathes: solutes are quickly transferred to outlet by the ditch network whereas transfers are slower along plots. These results show physical consistency and a very good mass balance for water and solutes. As a conclusion, processes still have to be implemented and their evaluation have to be performed, but this method is promising to represent water and pesticides fate in different complex landscapes.

International audience; Identifying organic contaminants present in aquatic environments is of major concern for water quality monitoring. During the last 15 years, application of suspect and non-target analyses to study the presence of emerging contaminants and their transformation products in aquatic environment has increased to meet this challenge. The sample preparation is the first crucial step to be optimized before using these approaches for qualitative evaluations of water contamination. The main objectives of our study - based on a metabolomics workflow strategy - were to evaluate the quality of the data obtained with the different sample preparation tested (repeatability, number of relevant signals, polarity range of the analytes detected, ...). A large surface water sample (15 L) was collected from an urbanized river. Different solid phase extraction conditions (SPE, n=3 per condition) were tested towards UHPLC-HRMS (Ultra High Performance Liquid Chromatography - C18 column - hyphenated to High Resolution Mass Spectrometry) analysis. The SPE protocols varied according to the extraction phase (Oasis HLB® or multilayer cartridge), the elution solvent (methanol (MeOH) or MeOH and dichloromethane (DCM)) and sample pH (pH 3 or 7). To ensure data quality, analytical blanks, SPE blanks and quality control (QC) were also injected. The data were processed based on a metabolomic workflow using XCMS before comparison of features between sample preparation methods. Data were sorted out depending on (i) signal-to-noise ratio (S/N) compared to analytical blank, (ii) S/N compared to SPE blanks, (iii) analytical repeatability (obtained by coefficient of variation calculation for QC features), and (iv) extraction repeatability. Only signals with a fold change higher than 50 were kept and further processed to structural elucidation. Data treatment revealed the Oasis HLB® extraction at pH 7 with MeOH elution is the least repeatable one, unlike the multilayer extraction at pH 7 with MeOH elution which is the most repeatable. General comparison of the data revealed that the number of features detected for each method varies, with a higher number of features for the two SPE with Oasis HLB® extraction phase at pH 7 (elution with MeOH or MeOH and DCM). The extraction with Oasis HLB® - especially at pH 7 - revealed a higher detection of features with a m/z above 600 Da compared to the multilayered cartridge which seems better to recover compounds with a m/z between 50 and 200 Da. Based on the features retention time distribution comparison, the multilayer cartridge allowed the extraction of more polar compounds than the other SPE. The structural elucidation of features with a fold change higher than 50 highlighted presence of pharmaceuticals in the water sample (e.g. celiprolol - a betablocker, lamotrigine - an antiepileptic). To our knowledge, this methodology is poorly applied to evaluate analytical repeatability and has never been carried out to assess sample preparation methods for water suspect and non-target screening studies. Such an approach allows to better assess the quality of the data, especially for structural elucidation.

This paper describes an integrated, high-resolution dataset of hydro-meteorological variables (rainfall and discharge) concerning a number of high-intensity flash floods that occurred in Europe and in the Mediterranean region from 1991 to 2015. This type of dataset is rare in the scientific literature because flash floods are typically poorly observed hydrological extremes. Valuable features of the dataset (hereinafter referred to as EuroMedeFF database) include i) its coverage of varied hydro-climatic regions, ranging from Continental Europe through the Mediterranean to Arid climates, ii) the high space-time resolution radar-rainfall estimates, and iii) the dense spatial sampling of the flood response, by observed hydrographs and/or flood peak estimates from post-flood surveys. Flash floods included in the database are selected based on the limited upstream catchment areas (up to 3000 km2), the limited storm durations (up to 2 days), and the unit peak flood magnitude. The EuroMedeFF database comprises 49 events that occurred in France, Israel, Italy, Romania, Germany, and Slovenia, and constitutes a sample of rainfall and flood discharge extremes in different climates. The dataset may be of help to hydrologists as well as other scientific communities because it offers benchmark data for the identification and analysis of the hydro-meteorological causative processes, evaluation of flash flood hydrological models and for hydro-meteorological forecast systems. The dataset also provides a template for the analysis of the space-time variability of flash flood-triggered rainfall fields and of the effects of their estimation on the flood response modelling. The dataset is made available to the public as a "public dataset" with the following DOI: (https://doi.org/10.6096/mistrals-hymex.1493).

[Departement_IRSTEA]Eaux [ADD1_IRSTEA]Systèmes aquatiques soumis à des pressions multiples; International audience; Genotoxic evaluation has been developing for a couple decade among ecotoxicological assessment approaches in the aquatic field. It offers some prospect for understanding delayed effects on the offspring and the population dynamics (provided genetic mutations affect gametic genome). However, the modification of the DNA sequence by genetic mutation (as a result of primary DNA damage) is not the only impact of toxic substances on the genome. For example, epigenetic effects, defined as hereditary effects on the DNA function, may add up to mediated effects by genetics way. Among these marks, DNA methylation is very studied by scientists. As such, it is important to examine epigenetic changes in the context of population level perspective. This provides a complementary approach to effects on the primary structure of the genome for taking into account the time interval between the exposure to environmental contaminants and their effects. This issue is a real challenge in ecotoxicology. Moreover, the epigenetic allows improving the understanding of changes of life history traits (reproduction, growth and development). These parameters are suitable for assessing the toxicity and are known to be regulated by epigenetic. Finally, epigenic could also help better understand the variability of effects related to experimental conditions between the studies. Therefore, epigenetic marks have an innovative nature for the evaluation of environmental risks. In this regard, we have investigated the measurement of genomic DNA methylation level as a possible stress biomarker in the ecologically relevant species Gammarus fossarum. First, the basal level was explored by studying the difference of overall DNA methylation between male, female and juvenile. Then, we evaluated the effects of natural factors on genomic DNA methylation level as starvation and exposed to different temperatures (8, 12 and 18°C) for 7 days and 1 month. To know the variability of DNA methylation in populations of Gammarus fossarum, three French populations from the same genetic strain, were studied. Lastly, the epigenetic biomarker response to a chemical stress was evaluated in a field experience. We encaged gammarids stemming from a reference unpolluted station in sited impacted by various human activities.

International audience; Owing to their ecological importance, freshwaters provide important services which leads to a strong societal demand concerning the preservation of their quality. They are the receptors of many contaminants emitted by human activities and more specifically by wastewater treatment plants effluents. Water resources of the Meuse watershed are used on both sides of the French-Belgian border which involves a coherent and joint management. In that framework, the DIADeM project suggests developing and spreading out a cross-border multidisciplinary approach to improve the diagnosis and the chemical and biological (biomarkers) monitoring of freshwaters using the Meuse river as a case study. In the past, results from chemical surface water monitoring of the Meuse has revealed the occurrence of numerous substances and more particularly pharmaceuticals. Overall, the project suggests coupling chemical and biomarkers analysis on caged organisms (a crustacean, a mollusk, a moss and a fish species) with predictive mathematical population level models. In order to calibrate and validate these models, a lotic mesocosm experiment was set up. Five substances were chosen: diclofenac, carbamazepine, naproxen, paracetamol and irbesartan. An environmental realistic mixture M of the five substances was tested along with MX10 and MX100. The study was carried out in twelve 20 m long lotic channels. The mesocosms were set up with artificial sediments, macrophytes, periphyton, benthic and pelagic invertebrates, decomposers and one fish species (Gasterosteus aculeatus), zebra mussels, Fontinalis antipyretica and Gammarus fossarum were also caged in the mesocosms. After 3 months of stabilization, treatment lasted 5 months. Periphyton biomass, macrophyte biovolume, zebra mussel biomarkers and growth, G. fossarum survival, reproduction and growth, F. antipyretica biomarkers and growth, zooplankton and invertebrate abundance and diversity, and fish individual physiological responses along with population dynamics were the main measured biological endpoints. The concentrations of each substance in water was monitored monthly along with some physico-chemical parameters. The overall experimental design will be presented along with the results related to the monitoring of substance concentrations in water, physico-chemical parameters, macrophyte biovolume, invertebrate community response, fish larvae densities. A brief discussion of the direct and/or indirect effects will then be performed.

This study considers an "offline" data assimilation method using the Ensemble Kalman Filter to build a precipitation reanalysis over France. The method is here applied for reconstructing the 2009-2012 period, using past observation density of 1871, 1900 and 1950. The methodology allows taking two main features of precipitation into account: (1) an anisotropic localization matrix based on the climatological background information, (2) a Gaussian transformation applied to daily precipitation. Results show a reduced error and a reduced uncertainty compared to background reconstructions, even with few observations, thus demonstrating the added value of data assimilation.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]ARCEAU [ADD1_IRSTEA]Hydrosystèmes et risques naturels [ADD2_IRSTEA]Systèmes aquatiques soumis à des pressions multiples; International audience; Due to a reduce cost, availability of many favorablefavourable locations, easy access due to proximity, the number of small reservoirs has increased. The cumulative impact of reservoirs in a catchment is considered as the modifications induced by the reservoir network taken as a whole. The impact may exert on the flow regimes and sediment, nutrient and contaminant transfer, and thereby modify the ecological behaviour of the aquatic environment, the continuity of rivers and the habitats of organisms living in them. The cumulative impact is not necessarily the sum of individual and local modifications, because reservoirs may be inter-dependent. This is the case for instance in cascading reservoirs along a stream course. The cumulative impact is not straightforward to estimate, even solely considering hydrological impact, in part due to the difficulty to collect data on the functioning of those reservoirs. However, there are evidences that the cumulative impacts are not negligible. This work is dedicated to a review of the studies dealing with the cumulative impact of small reservoirs on hydrology, focusing on the methodology as well as on the way the impacts are reported. It is shown that similar densities of small reservoirs can lead to different impacts on the quantitative water resource in different regions. This is probably due to the hydro-climatic conditions, and makes it difficult to define simple indicators to provide a first guess of the cumulative impact. The impacts vary also on time, with a more intense reduction of the river discharge during the dry years than during the wet years. This is certainly an important point to take into account in a context of climate change.

International audience; Low-flow statistics support the definition of legal thresholds used for estimating environmental flows, and for designing water structures or maximum abstraction levels. In France, the most used low-flow statistics is the annual monthly minimum flow with a 5-year return period (called QMNA5). Such low-flow statistics are moreover usually estimated from observations over the last few decades. Streamflow observations are indeed barely available before the end of the 1960s in France. This works aims at demonstrating that such low-flow estimates are highly dependent on the period chosen for calculation, as a consequence of the large multidecadal climate - and thus hydrological - variability. The basis for this analysis is the SCOPE Hydro dataset, a 25-member ensemble daily streamflow reconstruction for more than 600 near-natural catchments in France covering the period 1871-2012 (Caillouet et al., 2017). The SCOPE Hydro dataset had been derived through a hydrometeorological reconstruction approach that combines a statistical downscaling of the Twentieth Century Reanalysis (Compo et al., 2011) and catchment-scale hydrological modelling. A complementary dataset is the Safran Hydro dataset, derived from the same hydrological models but using the Safran meteorological reanalysis from 1958 onwards (Vidal et al., 2010). The QMNA5 low-flow statistics are here computed for all catchments over 7 consecutive 20-year periods, from (1) observations, (2) the Safran Hydro dataset, starting with 1971-1990, and (3) the SCOPE Hydro dataset, starting with 1871-1890. QMNA5 values are estimated by fitting a lognormal distribution, and in a probabilistic way to take account of sampling uncertainty. The performance of SCOPE Hydro is first assessed by its ability to reproduce low-flows statistics as derived from both observations and Safran hydro over the recent overlapping periods. Results of the long-term analysis show a large multidecadal variability of QMNA5 estimates over the 7 periods covering the end of the 19th century and the whole 20th century. They notably show significantly higher values between the most recent period (1991-2010) and some earlier periods, notably 1871-1990 and 1911-1930. These results are consistent with the identification of extreme spatio-temporal low-flow events at the scale of France previously derived from the SCOPE Hydro dataset (Caillouet et al., 2017). They critically question current practices for estimating legal streamflow thresholds, which generally use a short and recent period for determining low-flow statistics and associated uncertainty.

Most permafrost is located in the Arctic, where frozen organic carbon makes it an important component of the global climate system. Despite the fact that the Arctic climate changes more rapidly than the rest of the globe, observational data density in the region is low. Permafrost thaw and carbon release to the atmosphere are a positive feedback mechanism that can exacerbate global warming. This positive feedback functions via changing land–atmosphere energy and mass exchanges. There is thus a great need to understand links between the energy balance, which can vary rapidly over hourly to annual timescales, and permafrost, which changes slowly over long time periods. This understanding thus mandates long-term observational data sets. Such a data set is available from the Bayelva site at Ny-Ålesund, Svalbard, where meteorology, energy balance components and subsurface observations have been made for the last 20 years. Additional data include a high-resolution digital elevation model (DEM) that can be used together with the snow physical information for snowpack modeling and a panchromatic image. This paper presents the data set produced so far, explains instrumentation, calibration, processing and data quality control, as well as the sources for various resulting data sets. The resulting data set is unique in the Arctic and serves as a baseline for future studies. The mean permafrost temperature is −2.8 °C, with a zero-amplitude depth at 5.5 m (2009–2017). Since the data provide observations of temporally variable parameters that mitigate energy fluxes between permafrost and atmosphere, such as snow depth and soil moisture content, they are suitable for use in integrating, calibrating and testing permafrost as a component in earth system models.The presented data are available in the Supplement for this paper (time series) and through the PANGAEA and Zenodo data portals: time series (https://doi.org/10.1594/PANGAEA.880120, https://zenodo.org/record/1139714) and HRSC-AX data products (https://doi.org/10.1594/PANGAEA.884730, https://zenodo.org/record/1145373).

National audience; La MO des milieux aquatiques soumis à des pressions urbaines importantes possède des propriétés physico-chimiques différentes de la MO des eaux de surface pas ou peu impactées par les rejets anthropiques : plus faible proportion d'acides fulviques et humiques, molécules de tailles plus petites, plus faible aromaticité, enrichissement en azote (Imai et al., 2002 ; Pernet-Coudrier et al., 2011). De même, les caractéristiques de la MO d'origine anthropique peuvent varier en fonction de l'origine des eaux (e.g. domestique, industrielle, pluviale, ruissellement) et du type de traitement appliqué ou non (Michael-Kortadou et al., 2015). Les différences de nature entre MO naturelle et anthropique sont aussi observées au travers de leurs interactions avec certains métaux qui sont plus complexés par la MO d'origine anthropique (e.g. Louis et al., 2014 ; Matar et al., 2015). Cependant, les connaissances sur la nature et les propriétés physico-chimiques de la MO d'origine anthropique, ainsi que sa réactivité et son devenir sont limitées. En effet : - de nombreuses études sur la MO d'origine anthropique traitent des rendements épuratoires à partir d'analyses globales (type DCO ou DBO) qui sont limitées pour décrire la nature de la MO ; - la MO est souvent représentée par les substances humiques qui représentent en moyenne 40 à 60% de la MO naturelle mais qui sont peu représentatives de la MO anthropique (Pernet-Coudrier et al., 2008) ; - la MO est souvent analysée après des étapes d'extraction et de purification qui permettent de séparer la MO selon certaines propriétés mais qui sont longues à mettre en oeuvre et ne permettent pas l'analyse d'échantillons en grande quantité. Nous proposons d'initier une étude sur l'établissement de relations entre les différentes sources de MO anthropique et sa nature ou ses propriétés physico-chimiques. Le potentiel de la combinaison de différentes techniques analytiques complémentaires est étudié au travers du développement d'une stratégie de caractérisation intégrée de la MO basée sur la combinaison de descripteurs pertinents. Pour cela, des eaux usées brutes et traitées (n=26) ont été collectées en amont et aval de différentes stations de traitement d'eaux usées et de traitements tertiaires (traitements par boues activées, filtres plantés de roseaux, biofiltres, zones de rejets végétalisées...). La MO dissoute de ces échantillons a été caractérisée par des techniques permettant des analyses sans prétraitements de l'échantillon : spectrophotométrie UV-visible, fluorescence 3D, chromatographie liquide couplée à la spectrométrie de masse haute résolution, chromatographie d'exclusion stérique couplée à détection UV. La variabilité de la composition/nature de la MO anthropique et de ses propriétés physico-chimiques est mise en évidence. Ce travail préliminaire permettra de commencer la constitution d'une base de données inédite à partir d'un jeu d'échantillons couvrant une très large gamme de types de rejets urbains.

/ Les composés perfluorés (PFAS) possèdent des propriétés physico-chimiques uniques, telles qu'une excellente stabilité thermique et chimique et une bonne résistance à la dégradation biologique, expliquant leur ubiquité dans l'environnement. Les milieux aquatiques sont particulièrement exposés et de nombreuses études ont démontré la bioaccumulation de certains de ces composés par les vertébrés aquatiques. Pourtant, les données liées aux mécanismes d'absorption, de distribution, de métabolisation et d'élimination des PFAS demeurent limitées. Il est donc primordial de mieux comprendre ces processus ainsi que leurs déterminants, d'autant plus que la production et l'utilisation industrielles des PFAS évoluent rapidement. Dans ce contexte, une approche par la modélisation parait pertinente puisqu'elle permet d'intégrer l'ensemble des processus dynamiques influençant la toxicocinétique des composés d'intérêt. Les modèles toxicocinétiques à base physiologique (PBTK) se révèlent être des outils puissants pour comprendre et expliquer le comportement d'une substance et prédire leur devenir au sein d'un organisme. Cependant, les applications des modèles PBTK aux PFAS sont encore inexistantes pour les vertébrés aquatiques. Nous avons donc développé un premier modèle PBTK, basé sur la physiologie de la truite arc-en-ciel (Oncorhynchus mykiss), décrivant le comportement du perfluorooctane sulfonate (PFOS), composé de référence, majoritaire dans les matrices environnementales malgré l'arrêt de sa production depuis quelques années. Celui-ci, inspiré principalement de deux modèles mécanistiques (Ng et Hungerbühler 2013 ; Nichols et al., 2004), a été calibré à partir d'un jeu de données obtenu grâce à une exposition par voie alimentaire chez la truite arc-en-ciel adulte. Des simulations de la cinétique du PFOS ont été réalisées pour les phases d'absorption et de dépuration, tout en prenant en compte la croissance des organismes. Ainsi, nous avons pu étudier la contribution de chaque organe concernant la distribution du PFOS et celle des différentes voies d'élimination (urine, bile, fèces, branchies), encore controversées dans la littérature. Par ailleurs, plusieurs perspectives d'extension du modèle sont envisagées : effets de la température de l'eau sur les processus d'absorption et d'élimination et application à d'autres PFAS, tels que le perfluorohexane sulfonate, alternative industrielle au PFOS.

Passive sampling has been developed to obtain, at lower cost, more representative evaluations of the average concentrations of contaminants in aquatic environments. An innovative silicone-based passive sampler (named PACSiR, for Polar Apolar Composite Silicone Rubber) was designed to address sampling, detection and quantification of a wide range of organic contaminants (native or metabolites) present at ultratrace levels in waters. The PACSiR passive samplers, of small size (2 cm length), were applied in different aquatic media (surface or underground waters), typically for periods of 1 week, to evaluate the contamination by a wide range of pesticides and metabolites. They were also tested for their capacity to monitor several hormones and pharmaceuticals. The nature and mass of contaminants sorbed on the passive samplers exposed in situ were used as qualitative or comparative measures, to highlight the presence of compounds or to easily assess trends and gradients of contamination in aquatic environments. For quantitative purposes, the estimation of the average concentration of contaminants in water required the prior determination of kinetic parameters (sampling rate, equilibration time) through laboratory experiments. We demonstrated that this reactive, single-use and low cost passive sampler can be used for multiple research or operational purposes, thanks to the different levels of data acquisition, from qualitative to quantitative approach.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]ARCEAU [ADD1_IRSTEA]Hydrosystèmes et risques naturels; International audience; OZCAR (Critical Zone Observatories: Research and Applications) is a national distributed infrastructure associatingmost of the French observation sites dedicated to the observation and monitoring of the Critical zone, the thinouter veneer of Earth's continents extending from the top of the vegetation canopy down to groundwater. OZCAR includes more than 60 highly instrumented sites/observatories for long-term measurements of biological,chemical and physical parameters of groundwater, river water, glaciers, soils, and wetlands in France and overseas.Each observatory focuses on one or more components of the CZ but importantly OZCAR is covering most of thelateral and vertical compartments of the CZ from mountains to costal areas.These observatories have been set up formerly to answer a local/regional scientific issue (ie. flood event prediction,acid rain resilience. . . ) but they all share the overarching question issue of predicting the response of the CZ toperturbation ranging from the second to the millions of year.They all aim at seeking and understanding the dynamic architecture of the CZ, identifying the elementaryprocesses in order to determine mass and energy budgets, to be implemented in numerical predictive models.OZCAR's crosscutting scientific approach to sites and disciplines using the breadth and richness of its observatoriesfrom mountains to costal areas will allow:- Concepts and instruments to be exchanged- To test hypotheses- To construct realistic predictive models of Earth's surface evolution in response to global change at the localscale.Through environmental data portal and modeling platforms, OZCAR is not only a research infrastructure opento the scientific community, it is also aiming at advising policy makers and stakeholders on the water, soil, and biodiversity resource and the landscape scale.

Passive sampling has been developed as an alternative to grab or average automated sampling to obtain, at lower cost, more realistic estimates of the average concentrations of contaminants in aquatic environments. This technique allows the in situ accumulation of chemicals from large amounts of water, resulting in ultratrace level detection and smoothed integrative sampling over exposition periods ranging from days to months. In small agricultural watersheds, pesticide concentrations quickly fluctuate due to episodic rainfall events and resulting transfers from fields to rivers via various pathways (runoff, infiltration, lateral flows...). In this context, an innovative silicone-based passive sampler (named PACSiR, for Polar Apolar Composite Silicone Rubber) was designed to address sampling, detection and quantification of a wide range of organic contaminants (native or metabolites) in waters. The PACSiR passive samplers, of small size (2 cm length), were applied in different aquatic media (surface or underground waters) to estimate the levels of contamination by pesticides of different families. The nature and mass of contaminants sorbed on the passive samplers exposed in the field were used as qualitative or comparative measures, to highlight the presence of compounds or to easily assess trends and gradients of contamination in aquatic environments. For quantitative purposes, the estimation of the average concentration of contaminants in water required the prior determination of kinetic parameters (sampling rate, equilibration time) through laboratory experiments. In this study, PACSiR passive samplers were deployed in situ, typically for periods of 1 week, to evaluate the contamination of agricultural watercourses by a wide range of pesticides and metabolites under different agro-pedo-climatic conditions. These small tools were also successfully used to assess the variation of pesticide concentrations in time and space in subsurface waters thanks to the use of piezometers. In typical small agricultural watersheds, comparative studies between upstream and downstream sites allowed to highlight clear gradients of contamination in rivers and to put forward the role of wetland buffer zones. Finally, we could determine average concentrations with their associated uncertainties for about 30 pesticides and metabolites that were previously calibrated in laboratory. The values achieved with this new passive sampler were in accordance with average concentrations obtained with active grab or automated sampling techniques. We demonstrated that this reactive, single-use and low cost passive sampler can be used for multiple research or operational purposes, thanks to the different levels of data acquisition, from qualitative to quantitative approach.

[Departement_IRSTEA]Eaux [ADD1_IRSTEA]Systèmes aquatiques soumis à des pressions multiples; International audience; The first step to evaluate the effects of contamination on organisms is to study toxicokinetics. The bioaccumulation of contaminants by aquatic species is a variable phenomenon, since it depends on the characteristics of the environment, the properties of the contaminants and the species. Different toxicokinetic models have been developed to describe the accumulation of contaminants in aquatic food webs. In these models, the organism is often considered as a single compartment: the bioaccumulation is then described as the balance between uptake and elimination processes. The absorption process can involve both dissolved or trophic route. The diet of aquatic organisms is known to be an important route of bioaccumulation of contaminants. The elimination process includes excretion, metabolism and dilution by growth. To date, there are few models focusing on persistent organic contaminants. Furthermore, estimating models' parameters is generally done through a frequentist approach in two steps: first by estimating parameter(s) related to depuration and then estimating parameter(s) related to accucumulation. The problem by doing this is that depuration during the accumulation phase is neglected, while this process occurs in the two phases. The aim of our study is to propose a Bayesian framework to estimate the parameters of a biodynamic model together by considering simultaneously accumulation and depuration data. The posterior distribution obtained for all parameter will enable a more accurate assessment of bioaccumulation uncertainty. We illustrate our approach with the freshwater benthic invertebrate Gammarus fossarum exposed for 7 days to a sediment spiked with PCB153 and transferred to a clean media for 7 more days. The PCB153 concentrations in Gammarus fossarum increased from an initial concentration of 0.32 to 12.36 ng.g-1 ww (wet weight) at the end of accumulation step. When gammarids were transferred into a clean media, the PCB153 concentration in organisms decreased to 6.41 ng.g-1 ww at day 14. The bioaccumulation model assuming first-order kinetics was fitted to the data using Bayesian inference. The inference process quickly converged and thin posterior distributions were obtained for each parameter, meaning that data brough enough information to estimate preciselly each parameter. The median model predictions and their 95% credibility intervals showed a good fit of the model to the data.

In this paper, we present and analyze a novel global database of soil infiltration measurements, the Soil Water Infiltration Global (SWIG) database. In total, 5023 infiltration curves were collected across all continents in the SWIG database. These data were either provided and quality checked by the scientists who performed the experiments or they were digitized from published articles. Data from 54 different countries were included in the database with major contributions from Iran, China, and the USA. In addition to its extensive geographical coverage, the collected infiltration curves cover research from 1976 to late 2017. Basic information on measurement location and method, soil properties, and land use was gathered along with the infiltration data, making the database valuable for the development of pedotransfer functions (PTFs) for estimating soil hydraulic properties, for the evaluation of infiltration measurement methods, and for developing and validating infiltration models. Soil textural information (clay, silt, and sand content) is available for 3842 out of 5023 infiltration measurements ( ∼ 76%) covering nearly all soil USDA textural classes except for the sandy clay and silt classes. Information on land use is available for 76% of the experimental sites with agricultural land use as the dominant type ( ∼ 40%). We are convinced that the SWIG database will allow for a better parameterization of the infiltration process in land surface models and for testing infiltration models. All collected data and related soil characteristics are provided online in *.xlsx and *.csv formats for reference, and we add a disclaimer that the database is for public domain use only and can be copied freely by referencing it. Supplementary data are available at https://doi.org/10.1594/PANGAEA.885492 (Rahmati et al., 2018). Data quality assessment is strongly advised prior to any use of this database. Finally, we would like to encourage scientists to extend and update the SWIG database by uploading new data to it.

[Departement_IRSTEA]Eaux; International audience; Suspended particulate matter (SPM) flows in rivers are mainly due to soil erosion and anthropogenic activities. They contribute to the transport of a large amount of contaminants and can induce impacts on water quality and river ecosystem. To better manage these inputs in river systems, it is essential to identify the origin of sediments. In that way, SPM fluxes monitoring or fingerprinting approaches in rivers are increasingly addressed. In the frame of the Rhône sediment observatory (OSR) program, the Rhône River, which is the main sediment input to the Mediterranean Sea, is studied through several stations of measurement. Since 7 years, SPM were collected on the Rhône River and its tributaries for contrasting hydrological conditions allowing developing fingerprinting approaches. The aims of this study were to use major and trace element concentrations in the conservative fraction of SPM from the Rhône and its tributaries in order (i) to determine the actual relative contribution of SPM fluxes, with uncertainty, from tributaries to the Rhône River and (ii) to determine the historical SPM inputs of the tributaries by applying this approach on a sediment core. To determine the origin of sediment at given time and space, the use of conservative parameters is essential to avoid any tracer's transformations during particles transport. To assess element concentrations in the conservative fraction of SPM, samples were extracted by a total mineralization (HNO3, HCl, HF) and a soft extraction (HCl 1M). The discrimination of SPM and sediment sources were realized by using a Kruskal-Wallis test and a Discriminatory Factory Analysis to select element concentrations that could discriminate the main tributaries of the Rhône (Ain, Arve, Bourbre, Fier, Guiers, Saône, Isère, Durance). 17 discriminant elements were then integrated into a mixing model with uncertainty analysis using the Monte Carlo method. Our method allowed to correctly determining SPM origin in the Upper Rhône and downstream the confluence of the Rhône and Saône rivers, for contrasting hydrological conditions. These results were successfully confronted to SPM dynamics estimated by a hydro-sedimentary 1-D model or by using SPM and discharges data obtained via the OSR stations network. Determination of the SPM origin at the outlet of the Rhône is in progress in order to, first, integrate tributaries signatures from the downstream part of the Rhône and, second, to apply this method on a sediment core sampled at the outlet of the Rhône River. The results of the core analyses will allow defining a historical profile of sediment inputs at the outlet of the Rhône catchment.

International audience; Accurate assessment of low streamflows is important to the management of water supply, ecological conditions in streams, and protection of water quality. This study compares observed trends in low streamflow across the conterminous U.S. at near-natural (Hydroclimatic Data Network-2009) stream gages to trends at stream gages which have experienced substantial basin alteration (Agricultural, Regulated, and Urban), while accounting for the natural persistence present in low-flow time series. This work provides insight into regional climate-related trends in low streamflows and changes driven, in part, by human activities. Trends were tested for low 1-, 3-, and 7-day streamflows, number of zero-flow days, and annual deficit metrics at 2,482 U.S. Geological Survey stream gages across the conterminous U.S. For all time periods tested (50, 75, and 100 years through 2015), higher low streamflows over time were consistently observed throughout much of the northeastern quadrant of the conterminous U.S., as were lower flows in the southeastern and northwestern U.S. Reference gages with minimal human impacts generally indicated lower low-flow conditions over time for much of the U.S. In agricultural basins, trends predominantly indicated higher low-flow conditions over time which may be associated with increased precipitation and activities related to annual row crop cultivation. Trends in regulated basins were generally toward higher low flows, lower annual deficits, and fewer days of zero flow, though results were less regionally coherent. Geographic distributions of trends at urban gages were very similar to those for the reference gages but generally had a greater percentage of significant upward trends in low streamflows, fewer days of zero flow, and decreases in annual deficits which may be attributed to urban landscape features and patterns of water use and waste-water discharge that can increase base flow.

International audience; The COST-Action SMIRES brings together scientists from various disciplines to foster a common understanding of hydrology, hydrochemistry and ecology of intermittent rivers and ephemeral streams. Such rivers are characterized by episodes of stagnant waters or drying-off, occurring either annually during the drought season or exceptionally during extreme drought years. Since intermittency can be seen as a special case of low-flow regimes that reach zero-flow, relevant low flow characteristics exist but need to be adopted to characterise ecologically relevant features of the particular flow regime. In this paper we present a set of metrics for characterizing flow intermittency in an ecologically relevant way, stemming from interdisciplinary discussions among SMIRES experts of various disciplines. The indices are compatible with standard low flow indices of the WMO manual on low flow estimation and prediction (Gustard and Demuth, 2008) and the handbook Hydrological Drought (Tallaksen and Van Lanen, 2004), extending these indices to noflow conditions. The metrics characterise the statistical distribution including mean and variance of the proportion of no-flow years, the annual number of no-flow-days, the duration of no-flow spells and the timing of seasonality of onset and termination of no-flow spells. In addition, the dynamics of drying-off and rewetting are characterized. An R software package (Gauster and Laaha, 2017) has been compiled that enables uniform calculation of indices at the annual, seasonal and monthly scale, suitable for a broad range of hydrological regimes. The paper concludes with illustrating examples of how the metrics can be applied in eco-hydrological studies of rivers that cease to flow.

/ La caractérisation de la toxicité des différents effluents d'une chaine de traitements complémentaires de station d'épuration, et l'impact du rejet final sur le milieu récepteur ont été évalués à partir de deux méthodologies complémentaires. La caractérisation de la toxicité des effluents a été réalisée à partir de bioessais sur des organismes exposés en continu dans un laboratoire ex-situ dans lequel les eaux usées ont été amenés. L'impact du rejet a été mesuré à partir d'une exposition dans la rivière d'organismes encagés, et de la réalisation de ces mêmes bioessais.

International audience; To prevent flood risks due to bed elevation, dam flushing operations are regularly conducted on the upstream part of the Rhône River to flush away sediments accumulated over time upstream the dams. The extreme increase of suspended particulate matter concentrations (up to 15 g/L) suggests a potential release of contaminants. In the aim of evaluating spatial and temporal variability of dissolved metals concentrations during this event, a sampling strategy combining grab and passive sampling by DGT was set on 3 different sites. During the event, the temporal monitoring of dissolved metals showed that Cd, Cu, Fe, Hg, Pb, and Zn concentrations were not altered, whereas those of As, Co, Mn, Mo, Ni, Sb and U were multiplied by 2 to 22. These dissolved contaminants originated in part from remobilized particles, as the increases could not be explained by the sole contribution of interstitial water from sediment. Moreover, DGT measurements confirmed the low and stable concentrations for Cu, Fe, Pb and Zn and demonstrated a mean concentration for As, Co, Mn and Ni which reflects the concentration increase during the event. Finally, speciation analyses showed that before the event, As was only on the oxidized form (AsV), whereas the As release during the operations was mainly in reduced form (AsIII), which is more toxic for aquatic ecosystems.; Pour réduire le risque inondations causé par l'élévation du lit, des opérations de gestion sédimentaire sont menées régulièrement sur le Rhône pour évacuer les sédiments accumulés au cours du temps en amont des barrages. La hausse drastique des teneurs en matières en suspension (jusqu'à 15 g/L) suggère un possible relargage de contaminants. Afin d'évaluer la variabilité spatiale et temporelle des concentrations en métaux dissous lors de cet évènement, une stratégie d'échantillonnage combinant prélèvements ponctuels et déploiement d'échantillonneurs passifs de type DGT a été mise en place sur 3 sites. Le suivi temporel des métaux dissous a montré que les concentrations en Cd, Cu, Fe, Hg, Pb, et Zn n'étaient pas modifiées durant l'évènement, alors que celles en As, Co, Mn, Mo, Ni, Sb et U ont été multipliées de 2 à 22 fois. Ces contaminants dissous proviennent en partie des particules remobilisées, puisque ces hausses ne peuvent pas être expliquées par la seule contribution de l'eau interstitielle du sédiment. De plus, les mesures DGT ont confirmé les concentrations faibles et stables de Cu, Fe, Pb et Zn et montré pour As, Co, Mn et Ni un niveau moyenné reflétant l'augmentation des concentrations durant l'évènement. Enfin, les analyses de spéciation ont montré qu'avant l'évènement, l'As était présent uniquement sous forme oxydée (AsV), alors que l'apport d'As durant l'évènement était majoritairement sous forme réduite (AsIII), plus toxique pour l'écosystème aquatique.

/ Située sur l'Aire d'Alimentation de Captage (AAC) de Villevieux (eau potable de la ville de Lons-le-Saunier, Jura), la Madeleine est une rivière fortement exposée aux pesticides agricoles provenant de cultures diverses (céréales, maïs, colza, vigne). Le service des eaux, gestionnaire de l'AAC, a mis en place entre 2011 et 2016 un suivi de la qualité chimique des eaux des rivières du captage par prélèvements ponctuels, qui a notamment soulevé des interrogations sur la contamination des eaux dès les sections amont de ces rivières, et sur la contribution des différents affluents. Néanmoins, un diagnostic précis de la contamination en pesticides est difficile à réaliser à cause du manque de représentativité temporelle des prélèvements ponctuels réalisés, et devant l'absence de prise en compte de la toxicité potentielle associée. C'est ainsi qu'au printemps et à l'automne 2017, périodes propices aux transferts de pesticides vers les cours d'eau, des outils de diagnostic de la qualité chimique et biologique ont été déployés pendant un mois sur 3 stations de la Madeleine de façon à intégrer les apports à la contamination de deux principaux affluents. En plus de la variabilité saisonnière, une variabilité temporelle plus fine a été évaluée par le déploiement de certains outils sous forme de séquences de plusieurs tests (séquences d'un mois constituées de 4 ou 2 tests de 7 ou 15 jours respectivement). Le diagnostic a été réalisé à l'aide : i) d'échantillonneurs passifs (Tige Silicone Polaire et POCIS permettant l'analyse de 70 substances actives ou métabolites de pesticides) : placés dans le cours d'eau, ils permettent un échantillonnage en continu, et améliorent ainsi la représentativité de la mesure de la qualité du milieu par l'obtention d'une concentration intégrée sur le temps d'exposition de l'outil (7 ou 14 jours). ii) de la bioviosurveillance active qui permet de mesurer des effets toxiques aux échelles moléculaires et individuelles chez des organismes Gammarus fossarum transplantés (encagements). Les marqueurs choisis ont été la survie, l'inhibition de l'activité acétylcholinestérase, le taux d'alimentation et les marqueurs de reproduction. Nous présenterons dans ce poster les premiers résultats de ces expérimentations complémentaires qui mettent en évidence des variabilités spatiales et temporelles (hebdomadaires et saisonnières) de la contamination en pesticides et de l'impact toxique, et permettent ainsi de compléter le diagnostic de la qualité de l'eau de la Madeleine.