Jan Bronders

Improving the management of nitrate pollution in water by the use of isotope monitoring: the δ15N, δ18O and δ11B triptych

In spite of increasing efforts to reduce nitrogen inputs into ground water from intensive agriculture, nitrate (NO3) remains one of the major pollutants of drinking-water resources worldwide, with NO3 levels approaching the defined limit of 50 mg l−1 in an increasing number of water bodies. Determining the source(s) of contamination in water is an important first step for improving its quality by emission control. The Life ISONITRATE project aimed at showing the benefit of a multi-isotope approach (δ15N and δ18O of NO3, and δ11B), in addition to conventional hydrogeological analysis, to track the origin of NO3 contamination in water. Based on land use and local knowledge, four distinct cases were studied: (1) natural soil NO3, (2) natural denitrification, (3) single source of NO3 pollution and (4) multiple sources of NO3 pollution. Our results show the added value of combining isotope information, compared to knowledge based on local authorities’ experience, land use and the ‘classical’ chemical approach, by efficiently identifying the number and type of NO3 source(s) for each watershed studied.

Quantification and characterization of glyphosate use and loss in a residential area.

Urban runoff can be a significant source of pesticides in urban streams. However, quantification of this source has been difficult because pesticide use by urban residents (e.g., on pavements or in gardens) is often unknown, particularly at the scale of a residential catchment. Proper quantification and characterization of pesticide loss via urban runoff require sound information on the use and occurrence of pesticides at hydrologically-relevant spatial scales, involving various hydrological conditions. We conducted a monitoring study in a residential area (9.5 ha, Flanders, Belgium) to investigate the use and loss of a widely-used herbicide (glyphosate) and its major degradation product (aminomethylphosphonic acid, AMPA). The study covered 13 rainfall events over 67 days. Overall, less than 0.5% of glyphosate applied was recovered from the storm drain outflow in the catchment. Maximum detected concentrations were 6.1 μg/L and 5.8 μg/L for glyphosate and AMPA, respectively, both of which are below the predicted no-effect concentration for surface water proposed by the Flemish environmental agency (10 μg/L), but are above the EU drinking water standard (0.1 μg/L). The measured concentrations and percentage loss rates can be attributed partially to the strong sorption capacity of glyphosate and low runoff potential in the study area. However, glyphosate loss varied considerably among rainfall events and event load of glyphosate mass was mainly controlled by rainfall amount, according to further statistical analyses. To obtain urban pesticide management insights, robust tools are required to investigate the loss and occurrence of pesticides influenced by various factors, particularly the hydrological and spatial factors.

Improving surface–subsurface water budgeting using high resolution satellite imagery applied on a brownfield

The estimation of surface-subsurface water interactions is complex and highly variable in space and time. It is even more complex when it has to be estimated in urban areas, because of the complex patterns of the land-cover in these areas. In this research a modeling approach with integrated remote sensing analysis has been developed for estimating water fluxes in urban environments. The methodology was developed with the aim to simulate fluxes of contaminants from polluted sites. Groundwater pollution in urban environments is linked to patterns of land use and hence it is essential to characterize the land cover in a detail. An object-oriented classification approach applied on high-resolution satellite data has been adopted. To assign the image objects to one of the land-cover classes a multiple layer perceptron approach was adopted (Kappa of 0.86). Groundwater recharge has been simulated using the spatially distributed WetSpass model and the subsurface water flow using MODFLOW in order to identify and budget water fluxes. The developed methodology is applied to a brownfield case site in Vilvoorde, Brussels (Belgium). The obtained land use map has a strong impact on the groundwater recharge, resulting in a high spatial variability. Simulated groundwater fluxes from brownfield to the receiving River Zenne were independently verified by measurements and simulation of groundwater-surface water interaction based on thermal gradients in the river bed. It is concluded that in order to better quantify total fluxes of contaminants from brownfields in the groundwater, remote sensing imagery can be operationally integrated in a modeling procedure.

Limitations in the use of compound-specific stable isotope analysis to understand the behaviour of a complex BTEX groundwater contamination near Brussels (Belgium)

The application of compound-specific stable isotope analysis (CSIA) was evaluated to characterise a complex groundwater contamination. For this purpose, δ13C and δ2H analysis of benzenes and alkylated derivatives were used to interpret both the impact of different sources on a contaminant plume and the presence of degradation processes. The different contaminant sources could be distinguished based on their combined δ13C–δ2H signature of the benzene, toluene, ethylbenzene and xylenes (BTEX) dissolved in the groundwater. Despite this source differentiation, plume characterisation was not possible due to the complex mixing of the respective contaminant plumes. Furthermore, the original isotope signatures of the sources were not preserved across these plumes. To estimate the level of in situ biodegradation independently from concentration data, the Rayleigh equation was used. Although current literature identifies the application of CSIA as very promising in the frame of characterising organic groundwater pollution, this study has indicated that this approach can be limited with respect to successfully distinguish the different plumes and their relation to the known source zones.

Use of Compound-Specific Nitrogen (d 15 N), Oxygen (d 18 O), and Bulk Boron (d 11 B) Isotope Ratios to Identify Sources of Nitrate-Contaminated Waters: A Guideline to Identify Polluters

The use of various isotopes (d15N, d18O & d11B) to identify the sources of nitrate (NO3 −) present in natural waters is described. Then a new guideline of how to apply the multi-isotope approach is presented. This guideline is written for policy makers and scientists who are involved in the different steps and processes related to nitrate contaminated waters including monitoring and data interpretation. NO3 − is a common pollutant in water (both surface and groundwater). In several water bodies over Europe, point measurements identify that the level of this pollutant is higher than the reference value of 50 mgL−1, defined by the European Union (EU) Water Framework Directive 2000/60/EC (European Parliament, 2000). This directive also states that all waters have to reach a “good status” (i.e., good quality) by 2015. This statement implies that EU member states have to take actions to achieve this goal. One of the major obstacles with NO3 − contamination in water is the identification of the corresponding source(s) of pollution, a prerequisite for properly designing appropriate actions and remediation. Recent studies have proven the added value of analyzing compound specific isotopic signature (CSIA) of nitrate (both nitrogen (d15N), oxygen (d18O) and bulk boron (d11B) isotopic composition) to define the origin/source of NO3 − in waters. This definition is possible because different sources of nitrate have distinct isotopic signatures. The recent EU-LIFE ISONITRATE project demonstrated the benefit of the multi-isotope approach, while the presented guideline to implement this method is one of the outcomes of this project. More details on the scientific results of ISONITRATE are available at http://isonitrate.brgm.fr/.

Decision support for water quality management of contaminants of emerging concern

Water authorities and drinking water companies are challenged with the question if, where and how to abate contaminants of emerging concern in the urban water cycle. The most effective strategy under given conditions is often unclear to these stakeholders as it requires insight into several aspects of the contaminants such as sources, properties, and mitigation options. Furthermore the various parties in the urban water cycle are not always aware of each others requirements and priorities. Processes to set priorities and come to agreements are lacking, hampering the articulation and implementation of possible solutions. To support decision makers with this task, a decision support system was developed to serve as a point of departure for getting the relevant stakeholders together and finding common ground. The decision support system was iteratively developed in stages. Stakeholders were interviewed and a decision support system prototype developed. Subsequently, this prototype was evaluated by the stakeholders and adjusted accordingly. The iterative process lead to a final system focused on the management of contaminants of emerging concern within the urban water cycle, from wastewater, surface water and groundwater to drinking water, that suggests mitigation methods beyond technical solutions. Possible wastewater and drinking water treatment techniques in combination with decentralised and non-technical methods were taken into account in an integrated way. The system contains background information on contaminants of emerging concern such as physical/chemical characteristics, toxicity and legislative frameworks, water cycle entrance pathways and a database with associated possible mitigation methods. Monitoring data can be uploaded to assess environmental and human health risks in a specific water system. The developed system was received with great interest by potential users, and implemented in an international water cycle network.

A hybrid monitoring and modelling approach to assess the contribution of sources of glyphosate and AMPA in large river catchments

Large river catchments with mixed land use capture pesticides from many sources, and degradable pesticides are converted during downstream transport. Unravelling the contribution of pesticide source and the effect of degradation processes is a challenge in such areas. However, insight and understanding of the sources is important for targeted management, especially when water is abstracted from the river for drinking water production. The river Meuse is such a case. A long-term monitoring data set was applied in a modelling approach for assessing the contribution of waste water treatment plants (WWTPs) and tributaries (sub-basins) to surface water contamination, and to evaluate the effect of decay on the downstream concentrations of glyphosate and AMPA at the point of drinking water abstraction. The results show that WWTPs are important contributors for glyphosate and AMPA in large river catchments with mixed land uses. In the studied area, the river Meuse in the Netherlands, the relative contribution of WWTP effluents is above 29% for glyphosate and around 12% for AMPA. Local industries are found to be potentially big contributors of AMPA. Glyphosate entering the river system is gradually converted to AMPA and other degradation productions, which results in downstream loads that are considerably lower than the sum of all influxes. In summer when the travel time is longer due to lower discharge, the first order decay of glyphosate in the river Meuse is estimated to result in about 50% reduction of the downstream glyphosate concentrations over a river stretch of 250km. The contribution of glyphosate decay to the observed AMPA concentrations ranges between 2% and 10%. Contributions are sensitive to seasonal variations in discharge that influence the concentrations through dilution and degradation.

Model-based Scenario Analysis of the Impact of Remediation Measures on Metal Leaching from Soils Contaminated by Historic Smelter Emissions

A spatially distributed model for leaching of Cd from the unsaturated zone was developed for the Belgian-Dutch transnational Kempen region. The model uses as input land-use maps, atmospheric deposition data, and soil data and is part of a larger regional model that simulates transport of Cd in soil, groundwater, and surface water. A new method for deriving deposition from multiple sites was validated using soil data in different wind directions. Leaching was calculated for the period 1890 to 2010 using a reconstruction of metal loads in the region. The model was able to reproduce spatial patterns of concentrations in soil and groundwater and predicted the concentration in shallow groundwater adequately well for the purpose of evaluating management options. For 42% of the data points, measurements and calculations were within the same concentration class. The model was used for forecasting under a reference scenario, an autonomous development scenario including climate change, and a scenario with implementation of remediation measures. The impact of autonomous development (under the most extreme scenario of climatic change) amounted to an increase of 10% in cumulative Cd flux after 100 yr as compared with the reference scenario. The impact of remediation measures was mainly local and is less pronounced (i.e., only 3% change in cumulative flux at the regional scale). The integrated model served as a tool to assist in developing management strategies and prioritization of remediation of the wide-spread heavy metal contamination in the region.

Flux-based risk management strategy of groundwater pollutions: the CMF approach

A site- and receptor-specific risk management strategy for groundwater pollution based on the measurement of contaminant mass flux is proposed. The approach is useful and compatible with the demands formulated in the European Water Framework Directive, its Groundwater Daughter Directive and the regulations applicable in the EU member states. The proposed CMF method focuses on the following: (1) capture zones, (2) the location of control planes, (3) the definition of the maximum allowed contaminant mass discharge and (4) contaminant mass flux measurements. For every control plane, such a maximum allowed contaminant mass discharge is derived and is crucial for the receptor risk management strategy. The method is demonstrated for a large area of groundwater pollution present in the industrial area of Vilvoorde–Machelen located in Flanders, Belgium.

Examples of Risk Management in Flanders for Large Scale Groundwater Contamination

With respect to soil and groundwater remediation in Flanders, regulators have recently adopted a new, risk-based, policy. This policy is illustrated by two examples of risk management applied for large scale ground-water contamination. The first case involves a benzene and MTBE groundwater contamination threatening a shallow drinking water production site. Source zone treatment combined with pump & treat plume interception may be the only risk-based remediation strategy applicable in such cases. The second example is a risk management plan designed for the redevelopment of a brownfield located near Brussels, Belgium. In this area, an extensive ground-water contamination of monoaromatic and chlorinated aliphatic hydrocarbons is present in the Quaternary aquifer drained by the river Zenne. The risk management plan involves the combination of intensive treatment of source zones and plume treatment.