Isotopologue ratios of N2O and N2 measurements underpin the importance of denitrification in differently N-loaded riparian alder forests.

Known as biogeochemical hotspots in landscapes, riparian buffer zones exhibit considerable potential concerning mitigation of groundwater contaminants such as nitrate, but may in return enhance the risk for indirect N2O emission. Here we aim to assess and to compare two riparian gray alder forests in terms of gaseous N2O and N2 fluxes and dissolved N2O, N2, and NO3(-) in the near-surface groundwater. We further determine for the first time isotopologue ratios of N2O dissolved in the riparian groundwater in order to support our assumption that it mainly originated from denitrification. The study sites, both situated in Estonia, northeastern Europe, receive contrasting N loads from adjacent uphill arable land. Whereas N2O emissions were rather small at both sites, average gaseous N2-to-N2O ratios inferred from closed-chamber measurements and He-O laboratory incubations were almost four times smaller for the heavily loaded site. In contrast, groundwater parameters were less variable among sites and between landscape positions. Campaign-based average (15)N site preferences of N2O (SP) in riparian groundwater ranged between 11 and 44 ‰. Besides the strong prevalence of N2 emission over N2O fluxes and the correlation pattern between isotopologue and water quality data, this comparatively large range highlights the importance of denitrification and N2O reduction in both riparian gray alder stands.

The role of ponds in pesticide dissipation at the catchment scale: The case of the Save agricultural catchment (Southwestern France).

Pesticides are a major source of pollution for ecosystems. In agricultural catchments, ponds serve as buffer areas for pesticide transfers and biogeochemical hotspots for pesticide dissipation. Some studies have highlighted the specific impact of ponds on the dynamics of pesticides, but knowledge of their cumulative effect at the watershed scale is scarce. Hence, using a modelling approach, we assessed the cumulative role of ponds in pesticide transfer in an agricultural basin (Southwest of France, 1110 km2). The Soil and Water Assessment Tool (SWAT) model was used to model the Save basin, including 197 ponds selected with a Multi-Criteria Decision Aiding Model based on their pesticide interception capacities. The daily discharge, the suspended sediment loads and two herbicide loads (i.e. S-metolachlor and aclonifen) in dissolved and particulate phases were accurately simulated from January 2002 to July 2014 at a daily time step. The presence of ponds resulted in a yearly mean reduction at the watershed outlet of respectively 61 % and 42 % of aclonifen and S-metolachlor fluxes compared to the simulations in the absence of ponds. Sediment-related processes were the most efficient for pesticide dissipation, leading to a mean dissipation efficiency by ponds of 51.0 % for aclonifen and 34.4 % for S-metolachlor. This study provides a first quantification of the cumulative role of ponds in pesticide transfer at the catchment scale in an intensive agricultural catchment.

Screening potential toxicity of currently used herbicides in the freshwater amphipod Gammarus fossarum based on multi-level biomarker responses to field-realistic exposures.

Herbicides are widely used to control weeds and maximize crop growth. Because of agricultural runoff, these chemicals are potentially hazardous to aquatic wildlife. However, their ecotoxicity and resulting disturbance in individual performance remain scarcely documented in freshwater crustaceans. This study aimed to screen the potential toxicity of currently used herbicides in the ecosystem engineer Gammarus fossarum using multi-level biomarkers. In microcosms, gammarids were exposed for 72 h to 12 herbicides individually (quinmerac, mesotrione, bentazone, isoproturon, chlortoluron, metazachlor, chloridazone, diflufenican, flufenacet, aclonifen, prosulfocarb and metolachlor) at a field-realistic concentration (i.e. 10 µg/L). The sublethal effects were assessed by monitoring several biochemical, physiological and behavioural traits. In exposed gammarids, alterations in behavioural activities were observed, i.e. increased locomotion and respiration as a general trend. Moreover, biochemical biomarkers suggested herbicide-dependent disruptions in moulting, antioxidant responses and cell integrity. Integrating multi-metric variations through statistical analyses allowed us to identify herbicide clusters likely to trigger common sets of biological responses. Depressed antioxidant defence at the cell level and impaired respiration at the individual level were the predominant toxic effects of herbicides, related to their hydrophobic feature. Furthermore, establishing relationships between sublethal alterations in gammarids and acute lethality or chronic toxicity values defined for regulatory purposes supports the relevance of these alterations as early warnings of toxicity. Our findings demonstrate that currently used herbicides have unexpected toxicological effects in a non-target wild animal, with possible long-term alterations in population dynamics and associated ecological functions, which constitute promising diagnostic tools for risk assessment in agricultural areas.

Impacts of neonicotinoids on biodiversity: a critical review.

Neonicotinoids are the most widely used class of insecticides in the world, but they have raised numerous concerns regarding their effects on biodiversity. Thus, the objective of this work was to do a critical review of the contamination of the environment (soil, water, air, biota) by neonicotinoids (acetamiprid, clothianidin, imidacloprid, thiacloprid, thiamethoxam) and of their impacts on terrestrial and aquatic biodiversity. Neonicotinoids are very frequently detected in soils and in freshwater, and they are also found in the air. They have only been recently monitored in coastal and marine environments, but some studies already reported the presence of imidacloprid and thiamethoxam in transitional or semi-enclosed ecosystems (lagoons, bays, and estuaries). The contamination of the environment leads to the exposure and to the contamination of non-target organisms and to negative effects on biodiversity. Direct impacts of neonicotinoids are mainly reported on terrestrial invertebrates (e.g., pollinators, natural enemies, earthworms) and vertebrates (e.g., birds) and on aquatic invertebrates (e.g., arthropods). Impacts on aquatic vertebrate populations and communities, as well as on microorganisms, are less documented. In addition to their toxicity to directly exposed organisms, neonicotinoid induce indirect effects via trophic cascades as demonstrated in several species (terrestrial and aquatic invertebrates). However, more data are needed to reach firmer conclusions and to get a clearer picture of such indirect effects. Finally, we identified specific knowledge gaps that need to be filled to better understand the effects of neonicotinoids on terrestrial, freshwater, and marine organisms, as well as on ecosystem services associated with these biotas.

Main conclusions and perspectives from the collective scientific assessment of the effects of plant protection products on biodiversity and ecosystem services along the land-sea continuum in France and French overseas territories.

Preservation of biodiversity and ecosystem services is critical for sustainable development and human well-being. However, an unprecedented erosion of biodiversity is observed and the use of plant protection products (PPP) has been identified as one of its main causes. In this context, at the request of the French Ministries responsible for the Environment, for Agriculture and for Research, a panel of 46 scientific experts ran a nearly 2-year-long (2020-2022) collective scientific assessment (CSA) of international scientific knowledge relating to the impacts of PPP on biodiversity and ecosystem services. The scope of this CSA covered the terrestrial, atmospheric, freshwater, and marine environments (with the exception of groundwater) in their continuity from the site of PPP application to the ocean, in France and French overseas territories, based on international knowledge produced on or transposable to this type of context (climate, PPP used, biodiversity present, etc.). Here, we provide a brief summary of the CSA's main conclusions, which were drawn from about 4500 international publications. Our analysis finds that PPP contaminate all environmental matrices, including biota, and cause direct and indirect ecotoxicological effects that unequivocally contribute to the decline of certain biological groups and alter certain ecosystem functions and services. Levers for action to limit PPP-driven pollution and effects on environmental compartments include local measures from plot to landscape scales and regulatory improvements. However, there are still significant gaps in knowledge regarding environmental contamination by PPPs and its effect on biodiversity and ecosystem functions and services. Perspectives and research needs are proposed to address these gaps.

Do pesticides degrade in surface water receiving runoff from agricultural catchments? Combining passive samplers (POCIS) and compound-specific isotope analysis.

Pesticides lead to surface water pollution and ecotoxicological effects on aquatic biota. Novel strategies are required to evaluate the contribution of degradation to the overall pesticide dissipation in surface waters. Here, we combined polar organic chemical integrative samplers (POCIS) with compound-specific isotope analysis (CSIA) to trace in situ pesticide degradation in artificial ponds and agricultural streams. The application of pesticide CSIA to surface waters is currently restricted due to environmental concentrations in the low µg.L-1 range, requiring processing of large water volumes. A series of laboratory experiments showed that POCIS enables preconcentration and accurate recording of the carbon isotope signatures (δ13C) of common pesticides under simulated surface water conditions and for various scenarios. Commercial and in-house POCIS did not significantly (Δδ13C < 1 %) change the δ13C of pesticides during uptake, extraction, and δ13C measurements of pesticides, independently of the pesticide concentrations (1-10 µg.L-1) or the flow speeds (6 or 14 cm.s-1). However, simulated rainfall events of pesticide runoff affected the δ13C of pesticides in POCIS. In-house POCIS coupled with CSIA of pesticides were also tested under different field conditions, including three flow-through and off-stream ponds and one stream receiving pesticides from agricultural catchments. The POCIS-CSIA method enabled to determine whether degradation of S-metolachlor and dimethomorph mainly occurred in agricultural soil or surface waters. Comparison of δ13C of S-metolachlor in POCIS deployed in a stream with δ13C of S-metolachlor in commercial formulations suggested runoff of fresh S-metolachlor in the midstream sampling site, which was not recorded in grab samples. Altogether, our study highlights that the POCIS-CSIA approach represents a unique opportunity to evaluate the contribution of degradation to the overall dissipation of pesticides in surface waters.

Pesticide risk mitigation by vegetated treatment systems: a meta-analysis.

Pesticides entering agricultural surface waters threaten water quality and aquatic communities. Recently, vegetated treatment systems (VTSs) (e.g., constructed wetlands and vegetated ditches) have been proposed as pesticide risk mitigation measures. However, little is known about the effectiveness of VTSs in controlling nonpoint source pesticide pollution and factors relevant for pesticide retention within these systems. Here, we conducted a meta-analysis on pesticide mitigation by VTSs using data from the scientific literature and the European LIFE ArtWET project. Overall, VTSs effectively reduced pesticide exposure levels (i.e., the majority of pesticide retention performances was >70%). A multiple linear regression analysis of 188 retention performance cases identified the two pesticide properties, organic carbon sorption coefficient value and water-phase 50% dissipation time, as well as the VTS characteristics overall plant coverage and hydraulic retention time for targeting high efficacy of pesticide retention. The application of a Tier I risk assessment (EU Uniform Principle) revealed a higher toxicity reduction for hydrophobic and nonpersistent insecticides compared with less sorptive and not readily degradable herbicides and fungicides. Overall, nearly half (48.5%) of all pesticide field concentrations ( = 130) failed Tier I standard risk assessment at the inlet of VTSs, and 29.2% of all outlet concentrations exceeded conservative acute threshold levels. We conclude that VTSs are a suitable and effective risk mitigation strategy for agricultural nonpoint source pesticide pollution of surface waters. Further research is needed to improve their overall efficacy in retaining pesticides.

Individual performances and biochemical pathways as altered by field-realistic exposures of current-use fungicides and their mixtures in a non-target species, Gammarus fossarum.

Persistent fungicides, which are widely applied to agricultural soils to protect crops, are frequently detected in freshwaters because of hydraulic transfer, possibly resulting in unintentional adverse effects on wildlife. However, the ecotoxicity of fungicides in aquatic species remains scarcely assessed at environmentally relevant concentrations, and there is scant information available on their combined effects. This study aims at investigating multi-level changes elicited by two currently used fungicides, boscalid and tebuconazole, in the amphipod Gammarus fossarum. In microcosms, gammarids were exposed for 72 h to fungicides tested individually or in binary mixture at 0.01, 0.1 and 1.0 µg/L to monitor individual performances (locomotion, respiration and amplexus formation) and biochemical parameters (involved in energy metabolism, growth, moulting and cell stress). This range of exposure concentrations was field-realistic and largely lower than local environmental quality standards for the protection of aquatic wildlife. Overall, results showed that single and combined exposures altered the mobility and respiratory activity of individuals. At the cellular level, boscalid inhibited energy-based biomarkers whereas tebuconazole led to cytotoxicity associated with reduced antioxidant defences. In binary mixtures, the biochemical responses were mainly driven by the presence of boscalid. Multi-variable analyses, integrating individual alterations and cellular impairments alike, confirmed the relevance of the multi-level approach in forecasting the toxicity of fungicides and their mixtures towards other aquatic species. This study demonstrates dose-related sublethal effects of fungicides on multiple functional traits in an ecosystem engineer under realistic exposure scenarios, highlighting the harmful signs of these toxicants. Such sublethal alterations could thus constitute reliable tools for the early diagnosis of the organisms' health and the ecological status of agriculturally impacted hydrosystems.

Denitrifying bio-cathodes developed from constructed wetland sediments exhibit electroactive nitrate reducing biofilms dominated by the genera Azoarcus and Pontibacter.

To limit the nitrate contamination of ground and surface water, stimulation of denitrification by electrochemical approach is an innovative way to be explored. Two nitrate reducing bio-cathodes were developed under constant polarization (-0.5 V vs SCE) using sediments and water from a constructed wetland (Rampillon, Seine-et-Marne, France). The bio-cathodes responded to nitrate addition on chronoamperometry through an increase of the reductive current. The denitrification efficiency of the pilots increased by 47% compared to the negative controls without electrodes after polarization. 16S rRNA gene sequencing of the biofilms and sediments evidenced the significant and discriminating presence of the Azoarcus and Pontibacter genera in the biofilms from biocathodes active for nitrate reduction. Our study shows the possibility to promote the development of efficient Azoarcus-dominated biocathodes from freshwater sediment to enhance nitrate removal from surface waters.

Electrochemical analysis of a microbial electrochemical snorkel in laboratory and constructed wetlands.

Microbial electrochemical snorkel (MES) is a short-circuited microbial fuel cell applicable to water treatment that does not produce energy but requires lower cost for its implementation. Few reports have already described its water treatment capabilities but no deeper electrochemical analysis were yet performed. We tested various materials (iron, stainless steel and porous graphite) and configurations of snorkel in order to better understand the rules that will control in a wetland the mixed potential of this self-powered system. We designed a model snorkel that was studied in laboratory and on the field. We confirmed the development of MES by identifying anodic and cathodic parts, by measuring the current between them and by analyzing microbial ecology in laboratory and field experiments. An important application is denitrification of surface water. Here we discuss the influence of nitrate on its electrochemical response and denitrification performances. Introducing nitrate caused the increase of the mixed potential of MES and of current at a potential value relatively more positive than for nitrate-reducing biocathodes described in the literature. The major criteria for promoting application of MES in artificial wetland dedicated to mitigation of non-point source nitrate pollution from agricultural water are considered.

High denitrification potential but low nitrous oxide emission in a constructed wetland treating nitrate-polluted agricultural run-off.

Constructed wetlands (CW) can efficiently remove nitrogen from polluted agricultural run-off, however, a potential caveat is nitrous oxide (N2O), a harmful greenhouse gas and stratospheric ozone depleter. During five sampling campaigns, we measured N2O fluxes from a 0.53 ha off-stream CW treating nitrate-rich water from the intensively fertilized watershed in Rampillon, France, using automated chambers with a quantum cascade laser system, and manual chambers. Sediment samples were analysed for potential N2 flux using the HeO2 incubation method. Both inlet nitrate (NO3-) concentrations and N2O emission varied significantly between the seasons. In the Autumn and Winter inlet concentrations were about 11 mg NO3--N L-1, and < 6.5 mg NO3--N L-1 in the Spring and Summer. N2O emission was highest in the Autumn (mean ± standard error: 9.7 ± 0.2 µg N m-2 h-1) and lowest in the Summer (wet period: 0.2 ± 0.3 µg N m-2 h-1). The CW was a very weak source of N2O emitting 0.32 kg N2O-N ha-1 yr-1 and removing around 938 kg NO3--N ha-1 yr-1, the ratio of N2O-N emitted to NO3--N removed was 0.033%. The automated and manual chambers gave similar results. From the potential N2O formation in the sediment, only 9% was emitted to the atmosphere, the average N2 N 2O ratio was high: 89:1 for N2-Npotential: N2O-Npotential and 1353:1 for N2-Npotential: N2O-Nemitted. These results indicate complete denitrification. The focused principal component analysis showed strong positive correlation between the gaseous N2O fluxes and the following environmental factors: NO3--N concentrations in inlet water, streamflow, and nitrate reduction rate. Water temperature, TOC and DOC in the water and hydraulic residence time showed negative correlations with N2O emissions. Shallow off-stream CWs such as Rampillon may have good nitrate removal capacity with low N2O emissions.

Single and combined effects of insecticides on multi-level biomarkers in the non-target amphipod Gammarus fossarum exposed to environmentally realistic levels.

Aquatic media are ultimate recipients of various contaminants including pesticides pervasively applied in agrosystems. Characterizing the ecotoxicity of pesticides and their mixtures to aquatic wildlife at field-realistic levels is thus crucial for environmental risk assessment. This study aims at assessing the effects of two current-use insecticides, imidacloprid and chlorpyrifos, on Gammarus fossarum using multi-level biomarkers. In microcosms, gammarids were exposed for 72 h to insecticides tested individually or in mixture at 0.01, 0.1 and 1 µg/L of each chemical. Multi-metric responses were assessed at the individual level (behavioural traits: locomotion, respiration and amplexus formation) and the cellular level (enzymes involved in growth, moulting, digestion and cell stress). The results showed insecticide-elicited behavioural and biochemical responses from the lowest concentration of 0.01 µg/L. Overall, single exposures stimulated behavioural traits and inhibited enzymatic activities, highlighting subtle impacts at different organizational levels but these were not dose related. For binary mixtures, antagonistic effects (i.e. less-than-additive) on biomarkers were mainly observed when compared with single exposures. Multi-variable analyses indicated the complementarity of behavioural and biochemical biomarkers in identifying sublethal biological alterations and dose-dependent multiple action sites of insecticides. Besides, the mortality observed only for the mixture at 1 µg/L demonstrated a high lethal potential of insecticides in a simple binary combination. To conclude, this study demonstrates disturbances in individual performances and cellular impairments occurring at environmentally realistic exposure levels in a non-target wild species. Since the sublethal effects, such as those identified with this multi-biomarker approach, could lead to long-term alterations in population dynamics of agricultural areas, they constitute promising early endpoints for risk assessment of insecticides.

Nitrogen-rich organic soils under warm well-drained conditions are global nitrous oxide emission hotspots.

Nitrous oxide (N2O) is a powerful greenhouse gas and the main driver of stratospheric ozone depletion. Since soils are the largest source of N2O, predicting soil response to changes in climate or land use is central to understanding and managing N2O. Here we find that N2O flux can be predicted by models incorporating soil nitrate concentration (NO3-), water content and temperature using a global field survey of N2O emissions and potential driving factors across a wide range of organic soils. N2O emissions increase with NO3- and follow a bell-shaped distribution with water content. Combining the two functions explains 72% of N2O emission from all organic soils. Above 5 mg NO3--N kg-1, either draining wet soils or irrigating well-drained soils increases N2O emission by orders of magnitude. As soil temperature together with NO3- explains 69% of N2O emission, tropical wetlands should be a priority for N2O management.

Author Correction: Nitrogen-rich organic soils under warm well-drained conditions are global nitrous oxide emission hotspots.

The original version of this Article contained an error in the first sentence of the Acknowledgements section, which incorrectly referred to the Estonian Research Council grant identifier as "PUTJD618". The correct version replaces the grant identifier with "PUTJD619". This has been corrected in both the PDF and HTML versions of the Article.

Dynamics and mitigation of six pesticides in a "Wet" forest buffer zone.

Pesticide pollution is one of the main current threats on water quality. This paper presents the potential and functioning principles of a "Wet" forest buffer zone for reducing concentrations and loads of glyphosate, isoproturon, metazachlor, azoxystrobin, epoxiconazole, and cyproconazole. A tracer injection experiment was conducted in the field in a forest buffer zone at Bray (France). A fine time-scale sampling enabled to illustrate that interactions between pesticides and forest buffer substrates (soil and organic-rich litter layer), had a retarding effect on molecule transfer. Low concentrations were observed for all pesticides at the forest buffer outlet thus demonstrating the efficiency of "Wet" forest buffer zone for pesticide dissipation. Pesticide masses injected in the forest buffer inlet directly determined concentration peaks observed at the outlet. Rapid and partially reversible adsorption was likely the major process affecting pesticide transfer for short retention times (a few hours to a few days). Remobilization of metazachlor, isoproturon, desmethylisoproturon, and AMPA was observed when non-contaminated water flows passed through the forest buffer. Our data suggest that pesticide sorption properties alone could not explain the complex reaction mechanisms that affected pesticide transfer in the forest buffer. Nevertheless, the thick layer of organic matter litter on the top of the forest soil was a key parameter, which enhanced partially reversible sorption of pesticide, thus retarded their transfer, decreased concentration peaks, and likely increased degradation of the pesticides. Consequently, to limit pesticide pollution transported by surface water, the use of already existing forest areas as buffer zones should be equally considered as the most commonly implemented grass buffer strips.

Pesticide contamination interception strategy and removal efficiency in forest buffer and artificial wetland in a tile-drained agricultural watershed.

Pesticide pollution is a major threat to aquatic ecosystems that can be mitigated through complementary actions including buffer zones (BZs). This paper discusses the results of 3 yr of field-scale monitoring of the concentration and load transfer of 16 pesticides out of a tile-drained catchment (Bray, France) and their reduction through two BZ: an artificial wetland (AW) and a forest buffer (FB). Typically, the highest concentrations were measured in the first flows following pesticide applications or resuming after periods of low or no flow. An open/close water management strategy was implemented to operate the parallel BZ based on pesticide applications by the farmer. The strategy was efficient in intercepting molecules whose highest concentrations occurred during the first flows following application. Inlet vs. outlet pesticide load reductions ranged from 45% to 96% (AW) and from -32% to 100% (FB) depending on the pesticide molecule and the hydrological year. Partly reversible adsorption was a dominant process explaining pesticide removal; whereas, degradation occurred for sufficiently long water retention time. Apart from the least sorbing molecules (e.g., isoproturon), BZ can partially remove pesticide pollution.

Restoration of ponds in rural landscapes: modelling the effect on nitrate contamination of surface water (the Seine River Basin, France).

Ponds were ubiquitous features of the traditional rural waterscape in the Seine watershed, as shown by the 18th century Cassini map. Using the result of a water quality survey at the entrance and the outlet of a small pond receiving agricultural drainage water, the Seneque/Riverstrahler biogeochemical model was shown to accurately simulate the observed 30% reduction in nitrogen fluxes crossing this pond. The model was then used to simulate the effect of various scenarios of pond restoration (inspired by their 18th century geographical distribution as revealed by the Cassini map) on surface water nitrate contamination at different spatial scales. In regions with an impermeable lithological substrate, the restoration of ponds at a density of 5% of the agricultural area would reduce the riverine nitrogen export by up to 25% on an annual basis. It is suggested that such waterscape management, used in conjunction with more preventive measures, can be a useful means to reduce nitrate contamination of water resources.

Selected pesticides adsorption and desorption in substrates from artificial wetland and forest buffer.

Buffer zones such as artificial wetlands and forest buffers may help decrease non-point-source pesticide pollution from agricultural catchments. The present study focuses on understanding the role of the substrates mainly found in such buffer zones for pesticide adsorption and desorption. Radiolabeled [(14)C]isoproturon, [(14)C]metazachlor, and [(14)C]epoxiconazole were used to measure adsorption and desorption isotherms on wetland sediments and plants and forest soil and litter from two sites in France. Wetland sediments and forest soil exhibited the most important potential for pesticide adsorption. Wetland plants and forest litter also showed high adsorption coefficients and were associated with highly hysteretic desorption, particularly for the moderately mobile isoproturon and metazachlor. Adsorption of the highly hydrophobic epoxiconazole was strong and associated with weak desorption from all substrates. Calculated sorption coefficients were larger than those classically measured on soils. Isoproturon, metazachlor, and epoxiconazole K(OC) sorption coefficients ranged from 84 to 372, 131 to 255, and 1,356 to 3,939 L/kg, respectively. Therefore, specifically collecting buffer zone substrate sorption data is needed for modeling purposes. Results showed that forests and wetlands present potential for pesticide retention. This may be enhanced by planting vegetation and leaving dead vegetal material in buffer zone design.

Design of experiments and detailed uncertainty analysis to develop and validate a solid-phase microextraction/gas chromatography-mass spectrometry method for the simultaneous analysis of 16 pesticides in water.

A solid-phase microextraction (SPME)/gas chromatography (GC)-mass spectrometry (MS) multiresidue analytical method was developed for 16 pesticides presenting different physicochemical properties including diphenyl ether, triazine, ureas, acetamides, benzofuran, thiocarbamate, pyridine carboxamides, chloronitrile, piperedine, and azoles. Optimization was achieved by means of the design of experiments methodology. Extraction temperature, extraction time, desorption temperature, and NaCl addition were the factors exhibiting the most significant effects on pesticide extraction. Validation was carried out through model adequacy and specificity tests, limits of quantification and detection determination, and full uncertainty assessment on the whole analytical method. Good first- and second-order model adequacy was found for pesticide calibration. LOQs were in the 0.05-0.5 microg L(-1) range and specificity recoveries varied from 75% to 140%. These results were considered acceptable for our research purposes on highly concentrated agricultural flows. Uncertainty calculations accounted for several steps: standard preparation, calibration model selection, and use. On average, real sample concentration uncertainties were lower than 10%, indicating that the analytical method performed very well. Its application to 61 real water samples confirmed the presence of some pesticide concentrations in relation to farmer use, whereas other molecules were usually either not detected or not quantified.