Main authors: Susanne Klages, Nicolas Surdyk, Christophoros Christophoridis, Birgitte Hansen, Claudia Heidecke, Abel Henriot, Hyojin Kim, Sonja Schimmelpfennig
FAIRWAYiS Editor: Jane Brandt
Source document: »Klages, S. et al. 2018. Review report of Agri-Drinking Water quality Indicators and IT/sensor techniques, on farm level, study site and drinking water source. FAIRWAY Project Deliverable 3.1, 180 pp


Contents table
1. Definition based on literature review
2. What can we expect from monitoring?
3. Participative monitoring in FAIRWAY

1. Definition based on literature review

“Citizen science”

For many researchers, participative monitoring is a subset of citizen science that focuses on repeated data collection by non-scientists, often dedicated to identifying trends over time (Etrella and Gaventa, 1998). Although some researchers might disagree with that point of view, this categorization will be used (Eitzel et al., 2017, Lovett et al. 2007). Many definitions have been proposed for “citizen science”. One of the first attempt was in 1995 to describe expertise that exists among those who are traditionally seen as ‘lay people’ (Irwin, 1995). It is one of the broader applications of the term. The term citizen science was added recently to the Oxford English Dictionary in 2014 as:

“Scientific work undertaken by members of the general public, often in collaboration with or under the direction of professional scientists and scientific institutions.”

The main tenet of this concept is that members of the public are involved in science as researchers (Conrad, 2011). In order to extend this term to wider public than ‘citizen’ (which may imply a narrower definition, of a native or naturalised member of a state or nation), various terms were proposed: “community science” (Conrad, 2011), and Public Participation in Scientific Research” (PPSR) or “citizen and community science” (Eitzel et al., 2017).

In Europe, scientists as professionals is relatively new, emerging slowly throughout the 17th to 19th centuries, with specialization in science. Consequently, the distinction between “scientists” and “citizen scientists” is also relatively new. ‘Citizen scientist’ (meaning scientist independent of institutions) was used at least as early as 1912 (Eitzel et al., 2017). Before, the word “participative monitoring” was establish, the use of “unprofessional” collaborator was common. Many institutions asked vessel captains or crew medical officers to collect plants in newly discovered territory through the 16th and 17th centuries.

The participation of the public/citizens in science decreased during the 19th to 20th centuries with the increase of science specialisation. However, during the second half of the 20th century, thanks to technology development, citizens participations increased again. The fields in which citizen science is used are diverse: ecology, astronomy, medicine and much more. The point of a citizen science project is to facilitate scientist and citizen to collaborate towards a common goal. The collaboration that can occur through citizen science allows investigations at large scales and long time, that can lead to discovery scientists could not have achieved on their own.

Citizen scientists can help with opportunistic and observational studies that do not follow a strict design. These studies can be useful because of the large temporal or geographic scale of the data collection, the rarity of the phenomena observed (e.g., a rare species or infrequent weather event), or the timeliness of the observations (e.g., collecting information for crisis response, such as after earthquakes or oil spills), all of which make data collection difficult (McKinley et al., 2017).

Citizen scientist can participate in various ways in citizen science program by, 1) proposing programs, 2) analysing data and 3) collecting data. The last aspect (collecting data) is what refered to as “Participatory monitoring”.

Participative monitoring

Citizen observation of the environment and collection of data is a very ancient practice, being undertaken informallybefore the definitions of citizen science and citizen scientist were coined. But this practice, gradually fade until the 20th century, because the means of data collection was out of reach of many.

It was through association networks (e. g. Nature Conservatory, Earthwatch) that the practice of observing nature maintained its momentum in the second half of the 20th century. In the 1970s, some associations for the protection of birds used citizen science to developed programs of birds watching. Other citizen observation programs quickly follow in the course of the 1990s, involving mostly passionate, professional and amateur naturalists (Conrad and Hilchey, 2011). But, it was not until the 2000s with the advent of the internet, that the movement significantly accelerated and, above all, became open to a much wider audience.

Citizen observations of the environment today, covers a wide range of concerns. The quality of the environment, noise, air pollution, or even the quality of the water have become major concerns for citizens and new tools have faciliatitated their greater involved in address these issues. Technology developments have played an important role: the widespread use of smartphones and the ability to produce pollution sensors at a (relatively) low cost opened new horizons for citizen observation of the environment.

The notion of “participative monitoring” has several aspects in literature. As for "citizen science”, various definition are used such as “community based monitoring” (CBM). This expression includes a diversity of projects that involve citizen groups more or less organised in a collective observation and monitoring process of the environment (Conrad and Hilchey, 2011). More recently, participative monitoring programs have also included an increase of public education (Brossard et al., 2005) and/or an increase of the citizen’s involvement in the environmental decision linked to a specific project. Participative monitoring in its most inclusive form should include stakeholders in decision making but does not always do so even if volunteers tend to have the hope that their efforts will be used to assist in local decision making (Conrad and Daoust, 2008).

Participative projects have several characteristics, including the type of environment observed; the type of public mobilised and the program management. In the FAIRWAY project, we will focus on participative monitoring in a context of a scientific project, in the context of local/regional issue monitoring program.

Scientific project monitoring program: Scientific project monitoring programs have, as a principal characteristic, the generation of scientific knowledge. They either mobilise a well informed public, able to carry out a rigorous or data protoccols or or else citizens who have been trained to collect data according to the protocol.

These projects are usually driven by one or more research laboratories, often at the national level but also at local scale, and in partnership sometimes with associations or local organisations. Most of the large-scale ecosystem monitoring programs (e. g. bird monitoring programs) tend to be collaborative.

In the field of the environment, collaborative science programs are often developed around the theme of biodiversity. The Anglo-Saxon countries have generally been pioneers in this field. In Britain, for example, the Natural History Museum and the association Royal Society for the Protection of Birds have led for several decades programs of identification and monitoring of common birds by relying on a network made up of thousands of amateurs and professional naturalists (Bing et al., 2008).

The challenge is then to produce a suitable protocol that is 1) sufficiently scientifically rigorous, and 2) not too complex (and ideally enough fun) to enable broad public support.

These monitoring programs involve a central agency (most of the time a government agency or a governmentally funded agency) that requests information from volunteers. The purpose of monitoring by these volunteers is to provide early detection (by citizens) of issues of environmental concern, which can then be investigated/analysed by scientific experts (Whitelaw et al., 2003, Conrad and Daoust, 2008).

Although often successful in the short term, monitoring by volunteers are often funding dependent and cannot continue on their own without government or doner assistance. Also, these volunteers may not represent a very diverse stakeholder group as they have a vested interested in the issues being address in the monitoring program (i. e. only fishermen or only farmers; Conrad and Hilchey, 2011).

In most of the projects, the general principle is to rely on a network of volunteers (citizens) to follow a protocol of data collection with a scientific purpose. Most of the time, a website acts as an interface between scientists and the volunteers. The volunteers have to be able to recognise, count and locate individuals (animals or plants for example), with lots of information being downloaded from internet. Then they enter their comments on a website, the information is then sent directly to the researchers. The findings of the research are generally published on the same website, to inform citizen observers of the findings of the program (Figure 11.1).

D3.1 fig11.01
Figure 11.1

Local/regional activism monitoring programs: This type of programs often focus on local issues for which actions by governments should be initiated but has not occurred (Conrad and Daoust, 2008). It often focusses on specific issues and sometimes has no private sector or government support (Whitelaw et al., 2003). These participative monitoring projects are characterised by perspectives of joint public action. For these projects, the scientific goal is that data is generated that will inform local citizen to take action on a specific issue.

The general principle of this type of participative monitoring is to mobilise a network of highly motivated volunteers, to collect data. They supply data to an organisation to develop an action program or intervention. As a result, the project manager is rarely a research laboratory, but more often an organisation (administration or association, for example) who seeks to deal with a problem in a given territory. Such programs can for example provide comments on the invasive species or allergens in order to deploy a program to fight against their dispersion or it can aim to develop an inventory in order to fight against a project planning (Bing et al., 2008).

The monitoring devices used in these projects are similar to those for scientific programs. On one side, the scientists (belonging to an administration or association for example) develop the data collection protocol. On the other side, volunteers collect data using the protocols and equipment supplied by the scientists. Volunteers provide their observations to scientists, directly, or through a website that allows the data entry. These data are then compiled and can lead to concrete action (plan of action, mobilisation, etc.; Figure 11.2).

D3.1 fig11.02
Figure 11.2

In some of this “bottom-up” monitoring programs, the local communities do not trust the data provided by the private companies and/or government organsiations and want to acquire data by themselves. In many instances,”official” monitoring programs also exist, which may have a high degree of technical credibility, yet generate little credit for the community (World Bank, 2008).

One reason for this is that most monitoring programs are top-down, with the public receiving information that has been collected, analysed, and reported by experts chosen by the project sponsor or company and presented in a way that the public may not understand. In many instances, the information may not even address the real concerns of the community; rather, it may be strictly oriented toward a organisation self interests in relation to compliance with regulations and legal commitments (also an important function of monitoring) (World Bank, 2008).

However, many failures of bottom-up community based monitoring groups are mentioned. These include lack of success due to little organisation credibility and capacity (Bradshaw, 2003). Others suggest that bottom-up participative monitoring programs tend to be unsuccessful on a more organisational level, perhaps due to poor goverance structures and no legislation or policy support (Conrad and Daoust, 2008).

Global concern monitoring program: The emergence of measurements carried out directly by citizens in open mode and using mobile technologies are gradually taking the shape of networks of “human sensors”. This category of participative monitoring includes devices geared towards the general public.

They fit into a perspective of democratisation of the tools for environmental monitoring and are present as complementary to official measures devices (alternative monitoring) to feed and eventually guide public policy. Moreover, they are mostly in the field of "low-dose", i. e. low impact/frequency pollution (pollution of air, water, sound), which they aim to measure and understand for a better control.

These devices provide the ability to have a global networks of sensors, their strength beingin the mass production of environmental data. Also, unlike the participative monitoring of "scientific program" or “local issue monitoring program”, this type of participative monitoring is little governed by a scientific protocol (which is based on the formulation and testing of hypotheses). They fit into a perspective of democratisation of the tools of environmental monitoring and present themselves as complementary to official measure devices (alternative monitoring) to feed and eventually guide public policy.

The results of measures are data visualisation tool and maps with with generaly little corrections or caculations between the measures and the maps/visualisations. The tools to visualise the measure repartition is generaly provide with the tools to collect the data (Figure 11.3).

D3.1 fig11.03
Figure 11.3

The complementarity of these devices with those of public administrations is based on two key principles:

  • the multiplication of the number of sensors/observers may support a measure based on a limited number of institutional sensors (e. g. sensors payed by government),
  • the production of measures of exposure for individuals rather than exposure based on a ‘representitive’ place: participants measure precisely what pollutants they are exposed to in their daily lives.

Issues for “human sensor” monitoring arise during data collection. These include data fragmentation, data inaccuracy, and lack of participant objectivity (Whitelaw et al., 2003). Studies are often lacking in experimental design and do not consider issues such as adequate sample size (through a priori power analysis, for example). This could generate mistrust (by the scientific or government community) in the credibility and capacity of “human sensor” monitoring data (Conrad, 2007).

2. What can we expect from monitoring?

Increase cost benefits of action

A key benefit of Participative Monitoring programs is a decrease in the cost of official (i. e. governmentally funded) monitoring programs (almost uniquely) in the field of research.

The general decrease in prices of connected tools and the increases of on-line tool are the main drivers behind the recent success of citizen science and participative monitoring programs. However, this is also the case for more ‘classic’ type monitoring programs. Two decades ago, groundwater level was computed manually every few days whereas now hourly data can be downloaded automatically on a central database. Innovative connected tools (a connected tool received instructions or parameters from the backend, and/or sends to this backend data points collected by its sensors) are not only present in the smartphone field but also in the field of conventional sensors (hourly measurement and hourly transmission are technically achievable and economicaly accessible).

Connected tools (automated probes) and participatives monitoring have different benefits. Automated approaches have the benefit of regular and frequent (e. g. daily or hourly) measurements whereas the participative monitoring has benefits that counterbalance the lower data frequency, such as engage volunteers and encourage them to become interested in local water resources. Anyhow, the costs are not necessarly lower (Little et al, 2016).

Participative monitoring seems to be a great solution in specific research programs where scientists do not want to (or cannot) spend time or money to realise a task. This kind of monitoring could be valuable in large countries or for monitoring in remote area. In Europe, several network (Lora, GPRS) can be used, so connected tools can be used instead of participative monitoring, and the issue is more a matter of costs than a matter of technical devices.

Modification of volunteers behaviour

Many monitoring programs rely on the idea that making people participate will make them more interested in the topic and more generally in science. However, if specific education and training actions are not planned in the project, the increase of interest is generally marginal as volunteers are usually already interested and motivated individualls people.

Increases of knowledge in a specific domain has been observed in some participative monitoring projects (bird watching) but in other programs the volunteers did not really understand what they measured and misunderstood the results (Land-Zandstra et al., 2016).

In fact, to improve the quality of their data (or increase the reception of their paper in a peer-review journal), the tasks asked to volunteers are generally very simple and do not improve their scientific knowledge. Only teaching and direct contact with scientists has been shown to increase participants general knowledge, and so monitoring activities alone cannot fulfill the hopes of increasing public knowledge. Where it is essential to increase environmental awareness, other more suitable and efficient methods. like environmental education, can be used to reach this goal (Garcia and Lescuyer, 2008).

Participatory methodologies in the agricultural sciences usually involve limited numbers of farmers, working in collaboration with researchers, and scaling is usually difficult when the aim is to reach a number of larger farmer groups (Beza et al, 2017). The selected farmers (the initial volunteer group) are already conscious of their dependence on their natural resources and are generally willing to change their practices (Dangles et al, 2010). In monitoring, there is no guarantee of environmental efficacy even if the monitoring phase is a success; research message is transfered to farmers who have the role of the adopters or rejectors of innovations developed by others (Probst et al., 2003).

Some of the stakeholders, as well as many farmers do not want to be volunteers because they see monitoring as a way to impose on them new ecological constraints.

Increase administrative action

Because of the common mistrust of public against private companies and government funded monitoring programs and vice versa, the administration mistrust of participative monitoring, monitoring not supervised by administration is rarely used outside of the research world.

Data quality is almost universally recognised as one of the problems that scientists working in case studies need to address (Riesh and Potter, 2014). For example, some studies have shown that monitoring based on visual counting approaches need to take into account specific risks of bias due to the technique itself, volunteers' competences (Crall et al, 2011), the sampling effort and volunteers' missing impartiality (Leopold et al, 2009).

Citizen science and participative monitoring has led to an important number of scientific publications but yet only few administrative decisions (Yank, 2005).

Examples of participative monitoring programs

The examples below are existing (or previously existing) programs that have reached a broad audience. The projects presented here are mostly from North America because a selection was made on the availability of accessible websites. The selection was also made to present an aspect as wide as possible of different projects on the water. Direct measurement programs for nitrates or pesticides are nevertheless rare.

CATTFish The CATTFish, which stands for Conductivity and Temperature in your Toilet, uses an instrument that measures conductivity in water, allowing citizen in the US to monitor the quality of the water inside their home. It is designed with sensors that sit in a toilet tank. With a push of a button, it takes a measurement each time the tank refills after a flush. The main aim of this program is to measure a potential impact of hydraulic fracturing (

Rhode Island Water Quality Measure Program The heart of the program consists of weekly measurements of water quality taken by numerous trained volunteers. The program emphasises watershed scale monitoring, because the water quality of a given body of water is a reflection of the activities in the lands and waters that surround it and lie upstream. The program encourages citizen to understand the need to cooperatively manage and improve the water quality of water bodies within a watershed. In the program, the most common measured parameters are: water clarity, algal density, dissolved oxygen, water temperature, alkalinity and pH (

Streamselfie and Stream Tracker These two programs have the same objective: building of a map of a stream using photographs taken by volunteers. In Streamselfie, the aim of the map is to improved stream monitoring so community organisations involved in water monitoring are also part of the projet. The program aims at highlighting streams that are being monitored recently/at present and at developing a national inventory of streams that need to be monitored ( Stream Tracker aims at the improvement of mapping and monitoring smaller, intermittent streams through crowd sourced on-the-ground observations of streamflow presence and absence. Stream Tracker aims at filling in this information gap by combining a network of citizen scientists, sensors, and satellite imagery to track when and where streams flow (

NECi's Handheld Photometer The handheld photometers send nitrate and phosphate data to mobile phones when used with enzyme-based test kits. This tool can be used to obtain accurate nitrate or phosphate concentration in water, soil or plant tissue samples. The enzyme-based test kits provide reliable results while the photometer design ensures accurate translation of the quality data. The design of the photometer enables any citizen scientist to collect accurate nitrate data. The app software allows teams to effortlessly share results and coordinate projects with team members (

Roaring Fork Watershed Stream Temperature Monitoring This program is based on the citizen growing concern about stream temperatures increases (with flows expected to be lower than average for instance) and its potential impact on fishes and other aquatic species. Citizen scientists will volunteer to take water temperatures in the streams and rivers throughout the Roaring Fork Watershed (in US) so they can detect unusually high temperatures (

KSU "Citizen Science" The KSU "Citizen Science" program is designed to let community members do their own soil and water testing through the use of accurate test kits. Results can inform if a stream, lake, or another water supply meets water quality standards. They can also be used for the preliminary testing of drinking water, but a certified laboratory should perform follow-up testing if a problem is suspected. A helpful "how to" video for testing water in a local community or as part of a stream monitoring network is presented on the website (

Know Your Water: Sustainable Groundwater Research The primary objective of this project, run by a group of postgraduate students, is to model the distribution of modern groundwater across South Africa. In order to achieve this goal, an isotope tracer, tritium (the radioactive isotope of hydrogen), is measured in rainfall and groundwater samples. The sampling trip involves the collection of groundwater samples from pre-determined boreholes, where one can measure the depth to water as well as setting up rainfall collection stations. Citizen scientists who have received their sampling kits in the mail will sample rainfall and their groundwater from boreholes/springs and send them back to the project team for determination of the tritium activity.

Nitrate App A Nitrate App is tested currently in the Netherlands. A reference map is necessary. Surface water and groundwater are analysed with test strips (paper based sensors). Using the App, the result can be scanned and, if desired, shared. The App is in particular designed for people working professionally with water quality such as farmers, water authorities and water companies (

3. Participative monitoring in FAIRWAY

Evaluation of a device (tools)

Different tools (devices) that can be used for particpatory monitoring of pestcide and nitrates in groundwater and surface water, will be tested during the FAIRWAY project.

Currently in government monitoring programs, are spatially and temporal limited. Using volunteers in a participative monitoring action could increase the number of data available for a specific site (river or spring) or increase the number of points (stream, priezometer) followed.

Participative monitoring’s main interest is to gather a community around a a concept and/or an issue. In FAIRWAY, particpative monitoring will be address in two ways.

  1. Connected probes will be used to measure directly the concentration of nitrates in soils to help farmers to better understand fertilisation impact and then better calibrate fertilisation. The probes are maintained by the farmer and they will be provide with access to a website where there will be provided with support to help with the interpretation of the results collected
  2. Passive samplers will be evaluated for used in the measurement of pesticides (e.g. MCPA) concentrations instreams. If passive samples the evaluation demonstartes that passive samplers provide an accurate estaimate of the load of pesticides in the stream, then in the future thaey can be used to help farmers to better understand the impact of pesticides application of the drinking water quality in thier area. The management of the passive samplers during the FAIRWAY project will be done directly by farmers, as orginally hoped, but instead by other staksholders within the MAP of the specific case study catchments.
    »Use of passive samplers in drinking water catchments

These approaches toparticipatory monitoring are more valid for surface water than for groundwater (except if springs are present on the catchment).

Evaluation of a method

Education is propably the best way to change stakeholder behavior, with participative monitoring providing supplementary support for this. Significant improvements in stakeholder knowledge are not be expected as a result of this element of the FAIRWAY project, as the monitoring program will not be accompanied by and eductaional program. However, some training will be provided on site to upskill stakeholder on handling the device. Unfortunately, there will be less opportunity than expected to apply participative monitoring in case studies.


Note: For full references to papers quoted in this article see

» References


Main authors: Susanne Klages, Nicolas Surdyk, Christophoros Christophoridis, Birgitte Hansen, Claudia Heidecke, Abel Henriot, Hyojin Kim, Sonja Schimmelpfennig
FAIRWAYiS Editor: Jane Brandt
Source document: »Klages, S. et al. 2018. Review report of Agri-Drinking Water quality Indicators and IT/sensor techniques, on farm level, study site and drinking water source. FAIRWAY Project Deliverable 3.1, 180 pp


Contents table
1. Sensors for pesticide measurement in water
2. Sensors for nitrate measurement in water 
3. Automatic sampler techniques for pesticide and nitrate measurement in (soil-) water

1. Sensors for pesticide measurement in water

Optical Sensors

Optical sensors provides a facile, rapid and low-cost approach for sensitive detection of pesticide based on FL, UVevis, Raman, SPR or chemiluminescence signal variations. Generally, an optical sensor contains recognition unit that can interact specially with desired target pesticide and transducer component that is employed for signaling the binding event. Recognition elements including enzyme, antibody, molecularly-imprinted polymers, aptamer, and host-guest recognizer, draw increasing attention of scientific researcher to improve analytical performance of sensor. By combining the recognition units-assisted target response, the current well-established optical probes can be divided into four broad categories based on signal output formats:

  • fluorescence (FL),
  • colorimetric (CL),
  • surface-enhanced Raman scattering (SERS),
  • surface plasmon resonance (SPR),
  • chemiluminescence.

The optical sensors for pesticide detection based on various optical detection modes are fully described in a recent review (Yan et al., 2018) and are outlined below.

Fluorescence sensing strategy: With high sensitivity and simplification, fluorescence-based sensors as one of the most commonly used sensing candidate, have been widely applied in broad fields, including environmental monitoring ((Li et al., 2016, Guo et al., 2015, Wu et al., 2014), as the signal change can be collected vis spectrofluorophotometer and observed by naked eye on-site (Paterson and de la Rica, 2015, Wu et al., 2016, Zhang et al., 2011). As the development of advancing technologies, various kinds of materials have been widely employed for the fabrication of FL sensing platform, including fluorescent dyes (Strobl et al., 2017), semiconductors nanomaterials (Wu et al., 2013), metal nanomaterials (Chen et al., 2015, Wang et al., 2017), carbon materials (Yuan et al., 2016), and rare earth materials (Li et al., 2015). Meanwhile, it is very critical to choose and design a proper recognition unit that combined with FL probe for responding the fluorescent “turn off”, “turn on”, or “ratiometric” signal. Nsibande and Forbes reviewed the development of quantum dots-based FL probe for pesticide detection in terms of enzyme, molecularly-imprinted polymers (MIPs) and host-guest recognizer (Nsibande and Forbes, 2016). On the basis of the application of recognition elements, FL sensing strategies can be typically classified into several types: enzyme-mediated methods, antibody-assisted methods, MIPsbased methods, aptamer-based methods, host-guest complexes probe and other approach (see Yan et al., 2018 for details).

Colorimetric sensing strategy: Owing to its convenience and simplicity, colorimetric (CL) sensing strategy has proven to be a powerful analytical approach for the analysis of variety of analyte, including ions (Wang et al., 2014), chemical warfare agents (Yue et al., 2016), small organic molecules (Liu et al., 2011) and biomarkers (Sun et al., 2014). A prominent merit of CL sensing is that their direct visualization output makes them promising candidates for point-of-care assays. Therefore, the key challenge for fabricating CL platform is transforming response behavior into visual color change. Reviewed the remarkable achievements of nanomaterials, AuNPs as fascinate signal transducer have been widely utilized to design CL sensors for pesticide detection. Xu et al. developed AuNPs-based probe for the directly monitoring of acetamiprid based on the strong affinity between cyano group and gold (Sun et al., 2011). The sensing mechanism was based on the state change of AuNPs from dispersion to aggregation. The concentration of acetamiprid can be qualitatively estimated from the color change (red to blue). The color change during nanoparticle aggregation is highly dependent on their distance and concentration. Chen et al. (2018) used citrate-stabilized AuNPs for the rapid detection of terbuthylazine and dimethoate by visualizing the color change. This AuNPs-based CL sensor showed high selectivity and good sensitivity for pesticide detection in real environment samples. Recently, a CL sensor array was constructed for identifying five OPs based on the dispersion-aggregation behavior of AuNPs by Fahimi-Kashani and Hormozi-Nezhad (Fahimi-Kashani and Hormozi-Nezhad, 2016). Apart from unmodified AuNPs, functionalised AuNPs have been utilized to improve selectivity for CL detection of pesticide as well. Sun et al. displayed p-amino benzenesulfonic acid functionalised AuNPs as signal reporter for detecting carbaryl (Sun et al., 2013). Based on the similar protocol, Kim et al. (2015) introduced imidazole into AuNPs-based probe to improve the sensitivity and shorten the detection time for quantitative analysis of diazinon. In addition, melamine (Liu et al., 2015), p-nitroaniline dithiocarbamate (Rohit et al., 2016) and guanidine acetic acid (Bhamore et al., 2016) were also served as ligand to decorate AuNPs for selective CL detection of pesticide. Despite many advantages of those aggregate sensors including easy-to-use and cost-effective, more endeavors are still needed to improve the sensitivity and selectivity. The combination of recognition elements is preferred as they address the above limitations. Thus, numerous efforts have been devoted to integrating the specific affinity of recognition units with the optical properties of metal nanoparticles for realizing pesticide analysis in a sensitive, selective and accurate manner. From perspective of recognition elements, CL sensing strategies can be typically summarised as four types: enzyme strategies, antibody assays, aptamer-based methods and other approaches (see Yan et al., 2018 for details).

Surface enhanced Raman scattering strategies: Raman spectroscopy can identify the chemical content of different molecular species via the collection of molecular vibrations, that is, Raman spectroscopy possess the capability of molecular “fingerprint” recognition for distinct molecule/analyte. Surface enhanced Raman scattering strategy (SERS) essentially integrated the molecular specificity of Raman spectroscopy with optical properties of plasmonic nanostructures (Gruenke et al., 2016). Owing to optical resonance properties of coinage-metal nanostructures, the local electromagnetic field can be significantly enhanced, accompanying the improvement of the SERS signal. Taking advantages of ultrafast analysis capabilities, label-free, high stability and nondestructive characterization, the application of SERS received numerous concern in the field from biomedical diagnosis to environmental monitoring (Cialla-May et al., 2017, Henry et al., 2016, Ali et al., 2016). By means of coinage-metal nanostructures, SERS can even achieve an ultra-sensitivity down to the single-molecule level, which offered new opportunities toward obtaining single molecule recognition (Ding et al., 2016, Zrimsek et al., 2017). Recently, the development of SERS technique for pesticide detection in the aspect of sensitivity, reproducibility, selectivity and portability was recently reviewed (Pang et al., 2016). The following are recent achievements in pesticide SERS strategy as a powerful analytical tool that have focused on the development of metal nanostructures-enhanced amplification. In this section, according to the coinage metal nanoparticles-based solid substrates, SERS nanoprobes are typically designed as gold substrate, silver substrate and Au@Ag bimetallic substrate (see Yan et al. 2018 for details).

Other detection strategies: Other detection techniques, such as surface plasmon resonance (SPR) strategy and chemiluminescence strategy, have also gained strong driving forces in the detection of pesticide due to their convenient manipulation and high efficiency. By taking advantage of the outstanding distinguish ability provided by recognition unit, SPR and chemiluminescence strategy possessed excellent sensitivity and selectivity for real-time monitoring (see Yan et al. 2018 for details on some of the attractive research on SPR and chemiluminescence strategy).

Outcome and perspectives: Continuous concerns over pesticide residues have provided a long-driven force to develop novel techniques. In the past decade years, thousands of research literatures have been published for the routine and convenient monitoring of pesticide to meet increasing market and social requirements. Yan et al. (2018) have recently reviewed various kinds of optical strategy that were ingeniously designed and successfully applied for the detection of pesticide, with a specific focus on the fluorescence, colorimetric and surface-enhanced Raman scattering sensing strategies. With the emergence of high affinity of recognition elements, as well as various novel signal transduction approaches, optical assay reveal good performance to quantify pesticide residues in complex environment and food matrices, especially in the simplification and visualization design, making them ideally suitable for on-site application.

On the basis of the discussed research, the stability, accuracy, sensitivity and selectivity of optical sensor can be improved as follows: (1) the development of recognition units with excellent distinguish capacity to offer selectivity and sensitivity toward targeted analytes. For example, bi-enzyme cascade catalytic format has the merit of multi-signal amplification, greatly improving the sensitivity. (2) the utilization of novel nanomaterials that employ as signal reporters, substrates and catalysts. Ratiometric probe with dual-emission can provide built-in correction to eliminate environmental effects, exhibiting advantage in terms of enhanced sensitivity and accuracy. Nanozymes possess lower cost, higher stability, and excellent recyclability in comparison with natural enzymes, which improved the stability of sensor. Furthermore, the integration of optical strategy into paper-based analytical devices can be constructed in simplicity and miniaturization, further promoting the commercialization of devices.

Even though optical sensor has a promising future in pesticide determination, there are sustainable challenges to be addressed in the field. Particularly, most optical sensors still retain at laboratory level of testing and verifying proof-of-concept, which have not been exploited in practical applications. In the aspect of recognition events, the stability of recognition units (such as enzymes, antibody and aptamer) can be easily influenced by environmental conditions, such as temperature and pH. Furthermore, the integration of recognition event into the analytical system is a vital step in the fabrication of a successful sensor. The conjugation between recognition elements and functionalised nanomaterials will inevitably increase the complexity, cost and time of optical sensor, especially suppress the distinguish ability of recognition elements. From the perspective of nanomaterials, nanomaterials-based analytical platforms are in the starting stage of development. The specificity and catalytic activity of current nanozymes are lower than that of natural enzymes, in turn impeding the use of nanozymes. The synthesis of functional materials/nanomaterials with relatively narrow size distribution will seriously influence the performance of sensors, because inhomogeneous distribution of nanoprobe can reduce analysis accuracy. Thus, future endeavors should directly focus on addressing above obstacles.

While remarkable progress has been made toward the design of optical sensor for pesticide detection, tremendous opportunities and new trends are emerging. Coupling newly developed recognition elements (nanobodies, peptide aptamers and so on) with functional materials/nanomaterials will afford exciting opportunities for the monitoring of pesticide, which can improve the performance of sensors. On the other hands, the integration of field-deployable devices with optical sensor perform promising on-site applications, with the aid of 3D printing technologies, improving the reproducibility and stability of sensors. By taking advantage of miniaturized device and wire-less networking, the recognition event of pesticide can be transformed into a measurable digital signal by hand-held devices, such as smartphone, then the detection results can deliver to the servers. Thus, the portable detecting platforms can be carried out outside of laboratory setting with minimal user involvement, paving the way for a new generation of analytical devices in real-time detection. Yan et al. (2018) envision that, therefore, optical sensors will assuredly act significant roles in future on-site monitoring of pesticide.

Electrochemical sensors

Electrochemical sensors based on carbon nanotubes: In Table 10.1, the most relevant works related to pesticide electrochemical monitoring using carbon nanotubes-based electrochemical sensors reported in recent years are summarised. From that, Wong et al. (2017) have made a general overview of the current scenario related to this research topic, and, as can be seen, a number of works have been reported using different electrode architectures for the detection of various target analytes. Electrochemical sensors designed with pristine carbon nanotubes or combinations of carbon nanotubes with other modifiers can be found amongst these. There are modified electrodes consisting of CNTs and ionic liquids (ILs), porphyrin, phthalocyanines, metallic nanoparticles, and others. Thus, Wong et al. (2017) discussed the technical issues and the main analytical features, as well as the future challenges of these reports in specific subsections, which were classified according to the type of electrode modifier.

The review after Wong et al. (2017) demonstrated that carbon nanotubes provided electrochemical sensors with relatively good analytical performance in pesticide determination. Pesticides from different classes were electrochemically quantified using carbon nanotubes-based electrochemical sensors. The main electrode modification strategies consisted of the incorporation of carbon nanotubes within the composition of carbon paste electrodes and the modification of the surface of glassy carbon electrodes using the classical dropping cast method. Carbon nanotubes were used alone or in combination with different types of modifiers, including conductive polymers, phthalocyanines, porphyrins, metallic nanoparticles, ionic liquids, and graphene, among others. In general, a typical result achieved from the modification of carbon paste or glassy carbon electrodes is the very high increment of the analytical signal and the displacement of the working potential closer to zero. Both of these effects are desired to ensure high sensitivity and good analytical selectivity. The revised works demonstrated the construction of analytical curves with good linear concentration ranges (typically two concentration decades or more) and low detection limits (at least at the micromolar level). Moreover, in most cases, a good stability of response, precision of measurement, and accuracy in the recovery of spiked environmental samples are proved. Therefore, the positive effects of the use of carbon nanotubes as electrode modifiers for the preparation of electrochemical sensors dedicated to pesticide monitoring is very well illustrated and demonstrated. From the well-established electrochemical sensing performance of carbon nanotubes-based sensors toward pesticides, a set of challenges should be investigated and overcome for the advance of this important research topic.

Table 10.1: Electrochemical sensors based on carbon nanotubes for the detection of pesticides (after Wong et al., 2017)

D3.1 tab10.01

An interesting approach for future investigations is the possibility of designing multiplexed arrays using microfluidic devices, with which different analytes could be simultaneously determined in different sensing points. This challenge is linked with a current and relevant trend in (electro)analytical chemistry, which is the miniaturization of the analytical devices, with minimization of the consumption of chemical reagents and waste generation, as well as the proposition of portable instrumentation for analysis in the field (outside of the lab doors). From an analytical point-of-view, the amperometric and voltammetric methods dedicated to the sensing of pesticides should to be subjected to more rigorous analytical tests in order to verify the selectivity and reproducibility (and improve them if necessary), long-term stability, and applicability in diversified matrice samples, once most of the electroanalytical methods are employed in an analysis of spiked water samples using bulk electrodes. The robustness of the electroanalytical methods must also be evaluated from the analysis of a great number of environmental samples. In terms of sensor architecture material, a current trend is the preparation of composites of carbon nanotubes with another allotropic carbon forms, such as carbon black, graphene, or diamond. These classes of carbon composite electrodes are very promissory for electroanalysis purposes, and future electrochemical investigations should be carried out on the sensing and biosensing of pesticides.


A biosensor is an analytical device, used for the detection of an analyte, combining a biological component (bioreceptor represented by biomolecules or synthetic molecules obtained using biological scaffolds) with a physico-chemical detector, as well as an associated electronic system, which amplifies, process and display the detected signal. A successful biosensor must use a highly specific biocatalyst, stable in various stirring, pH, and temperature conditions (most often enzymes) (Schöning and Poghossian, 2002), give a dose-dependent and, desirably, real-time response, be costeffective, portable, easy to use (Grieshaber et al., 2008).

Biosensors are used in a wide range of applications for the quick and easy detection of pesticides and water contaminants. Gheorghe et al. (2017) present an extensive review of biosensors including:

  • Electrochemical biosensing techniques used for pesticides detection
  • Optical and imaging biosensing methods
  • Immunosensors
  • Whole-cell based biosensors

Considering the electrochemical biosensing techniques, Ramnani et al. (2016) report that biosensors based on carbon nanostructures are suitable for the design of portable and point-of-use/field –deployable assay kits. Carbon allotropes such as graphene and carbon nanotubes, have indeed been incorporated in electrochemical biosensors for highly sensitive and selective detection of various analytes, due to their many advantages for such applications, like high carrier mobility, ambipolar electric field effect, high surface area, flexibility and compatibility with microfabrication techniques. A simple and sensitive electroanalytical method for cyclic voltammetry and differential pulse voltammetry determination using a magnetic nickel ferrite (NiFe2O4)/MWCNTs nanohybrid-modified GCE has been developed. The method was used to detect benomyl in real samples with satisfactory results (Wang et al., 2015).

Piezoelectric Biosensors, as immunosensors based on acoustic waves, are of emerging interest because of their good sensitivity, real-time monitoring capability, and experimental simplicity (Jia et al., 2012). Piezoelectric systems have emerged as ones of the most attractive biosensing assays for the biopesticides detection due to their simplicity, low instrumentation costs, possibility for real-time and label-free detection and generally high sensitivity. Piezoelectric crystals such as quartz vibrate with characteristic resonant frequency depending on their thickness and cut under the influence of an electric field. The resonant frequency will modify when different molecules adsorb or desorb from the surface of the crystal, and the induced changes are detected by an electronic circuit. Biosensors based on the quartz crystal microbalance have been reported in the literature for organophosphate and carbamate pesticide analysis (Marrazza, 2014).

As a part of developing new systems for continuously monitoring the presence of pesticides in groundwater, a microfluidic amperometric immunosensor was developed for detecting the herbicide residue 2,6-dichlorobenzamide (BAM) in water. A competitive immunosorbent assay served as the sensing mechanism and amperometry was applied for detection. Both the immunoreaction chip (IRC) and detection (D) unit are integrated on a modular microfluidic platform with in-built microflow-injection analysis (μFIA) function. The immunosorbent, immobilized in the channel of the IRC, was found to have high long-term stability and withstand many regeneration cycles, both of which are key requirements for systems utilized in continuous monitoring. Detection of BAM standard solutions was performed in the concentration range 0.0008-62.5 μg/L, which demonstrate the potential of the constructed μFIA immunosensor as an atline monitoring system for controlling the quality of groundwater supply (Uthuppu et al., 2015).

Paper-based sensors

Busa et al. (2016) present a review of paper-based analytical devices (μPADs) that incorporate different detection methods such as colorimetric, electrochemical, fluorescence, chemiluminescence, and electrochemiluminescence techniques for food and water analysis. In Table 10.2., different paper-based platforms are presented.

Table 10.2: Summary of pesticides and insecticides for food and water analyses on paper-based platforms (after Busa et al., 2016)

D3.1 tab10.02

[1] Wang, S.; Ge, L.; Li, L.; Yan, M.; Ge, S.; Yu, J. Molecularly imprinted polymer grafted paper-based multi-disk micro-disk plate for chemiluminescence detection of pesticide. Biosens. Bioelectron. 2013, 50, 262–268.
[2] Sicard, C.; Glen, C.; Aubie, B.; Wallace, D.; Jahanshahi-Anbuhi, S.; Pennings, K.; Daigger, G.T.; Pelton, R.; Brennan, J.D.; Filipe, C.D.M. Tools for water quality monitoring and mapping using paper-based sensors and cell phones. Water Res. 2015, 70, 360–369.
[3] Nouanthavong, S.; Nacapricha, D.; Henry, C.; Sameenoi, Y. Pesticide analysis using nanoceria-coated paper-based devices as a detection platform. Analyst 2016, 141, 1837–1846.
[4] Liu, W.; Kou, J.; Xing, H.; Li, B. Paper-based chromatographic chemiluminescence chip for the detection of dichlorvos in vegetables. Biosens. Bioelectron. 2014, 52, 76–81.
[5] Liu, W.; Guo, Y.; Luo, J.; Kou, J.; Zheng, H.; Li, B.; Zhang, Z. A molecularly imprinted polymer based a lab-on-paper chemiluminescence device for the detection of dichlorvos. Spectrochim. Acta. A Mol. Biomol. Spectrosc. 2015, 141, 51–57.
[6] Badawy, M.E.I.; El-Aswad, A.F. Bioactive paper sensor based on the acetylcholinesterase for the rapid detection of organophosphate and carbamate pesticides. Int. J. Anal. Chem. 2014, 2014, 536823.
[7] Sun, G.; Wang, P.; Ge, S.; Ge, L.; Yu, J.; Yan, M. Photoelectrochemical sensor for pentachlorophenol on microfluidic paper-based analytical device based on the molecular imprinting technique. Biosens. Bioelectron. 2014, 56, 97–103
[8] Su, Y.; Ma, S.; Jiang, K.; Han, X. CdTe-paper-based Visual Sensor for Detecting Methyl Viologen. Chin. J. Chem. 2015, 33, 446–450.

With the goal to devise portable and easy measuring techniques and considering the increasing use of smartphones, the number of μPAD strategies that incorporate mobile or smartphones for target measurements is increasing. For instance, Sincard et al. (2015) describe a combination of paper-based sensors as an ultra-low cost approach for large-scale monitoring of water quality. The paper-based analytical device (mPAD) produces a colorimetric signal that is dependent on the concentration of a specific target, including organophosphate pesticides in water. A mobile phone equipped with a camera for capturing images of two mPADs e one tested with a water sample and the other tested with clean water that is used as a control, and an on-site image processing app that uses a novel algorithm for quantifying color intensity and relating this to contaminant concentration (Figure 10.1). The mobile phone app utilizes a pixel counting algorithm that performs with less bias and user subjectivity than the typically used lab-based software, ImageJ. The use of a test and control strip reduces bias from variations in ambient lighting, making it possible to acquire and process images on-site. The cell phone is also able to GPS tag the location of the test, and transmit results to a newly developed website,™, that displays the quantitative results from the water samples on a map. We demonstrate our approach using a previously developed mPAD that detects the presence of organophosphate pesticides based on the inhibition of immobilized acetylcholinesterase by these contaminants. The objective of this paper is to highlight the importance and potential of developing and integrated monitoring system consisting of mPADs, cell-phones and a centralised web portal for low-cost monitoring environmental contaminants at a large-scale.

D3.1 fig10.01
Figure 10.1

2. Sensors for nitrate measurement in water

In their upcoming review on spectroscopic methods for determination of nitrite and nitrate in environmental samples, Singh et al. (2019) extensively described the different laboratory methods referring 229 publications on the topic.

According to Azmi et al. (2017), many researchers in the field of potentiometry, electrochemical, and biosensors have focused on miniaturising their detection systems to enhance the capability of nitrate in-situ measurement. The performance of miniaturised sensor systems is comparable to that of conventional systems.

Potentiometry sensors

Basically, the conventional architecture of the system consists of two electrodes known as the working electrode and the reference electrode; a salt bridge, and a voltmeter. Figure 10.2(a) illustrates the architecture of the conventional potentiometry system. Meanwhile Figure 10.2(b) illustrates that of the miniaturised potentiometry system.

The advantages of this technique are its low cost (Hassan et al., 2007, Zhang et al., 2015, Mendoza et al., 2014, Paczosa-bator et al., 2014), non-destruction of sample, portable device (Zhang et al., 2015, Mendoza et al., 2014, Chang et al., 2013, Hassan et al., 2007, Santos et al., 2004) with fast response/feedback (Zhang et al., 2015, Li and Li, 2010, Paczosa-bator et al., 2014, Chang et al., 2013, Santos et al., 2004) and the requirement of minimum sample pretreatment.

Research of the potentiometry system has followed several avenues. Early work by Hassan (1976) was concerned with organic nitrate ions and nitramine determination based on the reaction with mercury sulphuric acid mixture. Mendoza et. al. (2014) characterised a nanobiocomposite as Ion Selective Electrodes (ISE) for nitrate ion determination in water. Mahajan et. al. (2007) developed a polymeric membrane by means of Zn (||) complex-based electrodes that work as anion carriers for nitrate anion determination in water. Li and Li (2010) and Nuñez et al. (2013) predict the nitrate contamination level in water based on an artificial neural network (ANN) algorithm.

D3.1 fig10.02
Figure 10.2
D31 tab1003
Table 10.3

Azmi et al. (2017) remind that the use of a membrane helps the potentiometry system to be selective to nitrate ions and is one of the factors that affects the system’s limit of detection (LOD). Bendikov and Harmon (2005) mentioned that doped polypyrrole (PPy(NO3-)) is a highly selective membrane in an ISE system for nitrate determination in water. They revealed that conductive polymer polypyrrole is widely used due to its high conductivity ability and it being relatively stable. As a result, Zhang et al. (2015) took the initiative to apply doped polypyrrole as a sensitive membrane material for the potentiometry system. The polypyrrole could improve selectivity, simplify the recipe procedure, and reduce toxicity compared to the conventional non-porous polyvincyl chloride (PVC) ISE (Zhang et al., 2015). Moreover, this study successfully demonstrated that the use of carbon nanostructure materials between the membrane and the substrate layer in the electrode structure of potentiometric system could prevent the water formation that led to instability. Meanwhile, Mahajan et. al. (2007) developed a polymeric membrane that was made of zinc (||) complex for selective nitrate determination in water. The finding demonstrates that the output of a potentiometry system using zinc (||) complex membrane exhibits better selectivity for nitrate ions than for other inorganic anions. They highlighted the advantages of zinc (||) complexes, such as stable detection reproducibility and being highly sensitive to nitrate. Wardak (1976) developed an active membrane component using trihexyltetradecylphosphonium chloride (THTDPCl) for polymeric membrane. THTDPCl could enhance the PVC membrane sensitivity by reducing electrical resistance.

The majority of potentiometric nitrate sensors that integrated either true-liquid or liquid polymeric membranes are bulky due to the tubular design with internal reference electrode and internal reference electrolyte solutions. Thus, a micro-fabricated planar potentiometric sensor was introduced (Hassan et al., 2007, Calvo-lópez et al., 2013). The micro-scale sensor could provide several advantages such as small size, simple design, low cost and mass production. Various materials are introduced to produce a micro-scale potentiometric sensor chip. Such materials are screen-printed thick film, silicon transducer chip, silicon nitride base chip and metal printed flexible polyimide film. Current miniaturised micro scale sensors for nitrate detection demonstrate a good response towards nitrate ions.

The miniaturisation of ISEs, while maintaining their selectivity and sensitivity, is a crucial step in the next phase of ISE evolution. Traditionally, in so-called coated-wire ISEs, the ion-selective membrane is placed directly on a solid electronically conductive support, thereby removing the need for an inner solution. However, in these devices, it was observed that the long-term potential stability was quite limited, and they were useful only in specific applications such as capillary electrophoresis or in flow-injection analysis. An important breakthrough in ISE design was achieved by the application of conducting polymers (CPs) as a solid contact layer, i.e. a mediating layer between the electronically conducting substrate and ionically conducting ISE membrane, which was possible due to the mixed conductivity of CPs. Various conductive polymers have been examined as possible internal contact materials that could simultaneously stabilise the overall electrode potential and remove the need for an inner filling solution.

Basically, the conventional architecture of the system consists of two electrodes known as the working electrode and the reference electrode, a salt bridge, and a voltmeter.

Electrochemical sensors

The electrochemical detection of nitrate and nitrite can be divided into a number of categories. Fortunately, these can be broadly grouped within the distinctions of voltammetric and potentiometric systems. Electrochemical systems have the ability to convert the measurement of nitrite ions into the current signal, potential difference and impedance, respectively. In electrochemical systems, various types of electrode were introduced for nitrate detection. Table 10.4 summarises the system performance based on different types of material.

D3.1 tab10.04
Table 10.4
D3.1 fig10.03
Figure 10.3

The electrochemical method is widely used due to its high sensitivity to nitrate, simple operation, easy to miniaturise and low power-consumption. However, the conventional electrochemical cell is too massive to be a portable and durable device. Research into electrochemical systems for nitrate detection has followed several avenues (Andreoli et al., 2011, Bhansali and Bhansali, 2013, Can et al., 2013). This is due to the demand for portable devices for continuous monitoring of nitrate concentration in aqueous solutions.

Several researchers have developed a microfludic base associated with electrochemical sensors for miniaturisation and portable purposes (Li et al., 2011, Li et al., 2012, Li et al., 2013). This combination has promoted many advantages such as the small configuration of electrodes that can be integrated within a microfluidic platform, requiring a minimum instrumentation, small volume of sample, fast response time, and low cost. Moreover miniaturised electrochemical detection is reliable, selective, and highly sensitive to the measured sample. The current architecture of miniaturised electrochemical sensor is designed based on the planar form or flatten of structure. According to Azmi et al. (2017), the performance of miniaturised electrochemical sensor demonstrated good LOD that is comparable to the conventional size of electrochemical system. Figure 10.3(a) illustrates the conventional electrochemical system architecture. Meanwhile, Figure 10.3(b) illustrates the miniaturised electrochemical system architecture.


A biosensor is one of the direct methods used for nitrate detection in water. In a biosensor system, the concentration of targeted ion in an analyte solution can be determined by employing the biological material, detection system and signal conditioning circuit. The analyte solution is directly exposed to a biological material. The biological material interacts with the targeted ion in the analyte solution. Information on the interaction process is then translated into an electrical signal such as voltage or current by a detection system. The signal is harvested by the signal conditioning circuit in the biosensor system. The signal conditioning circuit such as a digital data acquisition system will recondition the acquired data before being analysed. The concentration of nitrate ion is estimated based on the output signal of the proposed detection system.

Over the last decade, the miniaturisation of biosensor system has been carried out to characterize and quantify the bio molecules. The reduction size of the sensor system can promote lower material cost, lower the power consumption and the system weight. In most biosensors and also chemical and gas sensors, the trace of detection reversible redox species should be implemented by using very small amounts of samples, to descend upon the nanolitre or picolitre range.

Nitrate biosensors have been developed over the last two decades considering the advantage of enzymes that are strongly substrate-selective. Nitrate reductase (NR) is used in the fabrication of nitrate biosensors. However, its multiredox centre responsible for the biological conversion of nitrate to nitrite is generally not very active, and is deeply embedded in the protein structure, thus preventing the direct electron transfer with the electrode.

Carbon nanotubes (CNTs) have emerged as a new class of nanomaterials that are receiving considerable interest owing to their ability to promote electron transfer reactions with enzymes showing low electroactivity. The high conductivity of this carbon material has led to improving electrochemical signal transduction, while its nano architecture imposes an electron contact between the redox centres. CNTs can donate and accept electrons in a wide range of potentials and could therefore be used as mediators in biosensor systems. As a result, Can et al. (2012) investigated the performance of carbon nanotube/polypyrrole/nitrate reductase biofilm electrodes for nitrate detection.

Table 10.5 summarises the different types of biological material, detection systems, LOD and applications of biosensor systems for nitrate ion detection.

Table 10.5: Types of biological materials, detection systems, LOD (a review by Azmi et al., 2017)

D3.1 tab10.05

Paper-based sensors

Paper-based sensors, so-called paper-based analytical devices (PADs), are a new alternative technology for fabricating simple, low-cost, portable and disposable analytical devices for many application areas environmental monitoring. The unique properties of paper which allow passive liquid transport and compatibility with chemicals/biochemicals are the main advantages of using paper as a sensing platform.

Current paper-based sensors are focused on microfluidic delivery of solution to the detection site whereas more advanced designs involve complex 3-D geometries based on the same microfluidic principles (Figure 10.4). Although paper-based sensors are very promising, they still suffer from certain limitations such as accuracy and sensitivity (Liana et al., 2012). However, it is anticipated that in the future, with advances in fabrication and analytical techniques, that there will be more new and innovative developments in paper-based sensors. In the Netherlands, a monitoring tool based on this technology is tested (Nitrate-app), the measurement is paper based, A phone application scans and analyzes nitrate strips on the paper.

D3.1 fig10.04
Figure 10.4

Traditional electrochemical sensors often suffer from the effects of fouling due to the adsorption of oxidation products on the electrode surface. That is why paper-based, inexpensive, disposable electrochemical sensors have been developed for nitrite analysis. For example, Wang et al. (2017) present a system based on a simple and efficient vacuum filtration system. Taking advantage of the physicochemical properties of graphene nanosheets and gold nanoparticles, the mass transport regime of nitrite at the paper-based electrode was thin layer diffusion rather than planar diffusion. In comparison with the electrochemical responses of commercial gold electrodes and glassy carbon electrodes (GCE), a considerably larger current signal is seen at the paper-based sensing interface, which significantly improved its sensitivity for nitrite detection. In particular, the paper-based electrode was a disposable sensing device, so that it effectively avoided the fouling effect arising from the adsorption of oxidation products. According to Wang et al. (2017), the paper-based sensing platform made it possible to determine nitrite in environmental and food samples in an accurate, convenient, inexpensive, and reproducible way, indicating that the proposed system is promising for practical applications in environmental monitoring and public health.

3. Automatic sampler techniques for pesticide and nitrate measurement in (soil-) water

Automatic water sampling systems exist for pesticides and nitrates measured in water samples from ground- or surface waters or extracted from soils.. It is essentially a pump controlled by a clock or other automatic trigger, so that water samples can be pumped from a water source into a bottle at some pre-determined time or event and later collected for analysis (Figure 10.5). Such devices can be settled to collect water in the satured zone (piezometer), in streams, rivers and lakes. They can be portable or require an indoor environment. Experimental systems have also been designed to sample percolating water through the vadose/unsaturated zone. These in-house systems mainly consist of sucion cups connected to a controlled pump (Hamon et al., 2006; Farsad et al., 2012).

Classically, water samples should be stored in solvent-washed or brand-new (amber) glass bottles verified as uncontaminated, sealed with aluminium foil or Teflon, fitted with new plastic screw-caps and chilled immediately to less than 4°C in a refrigerator (Kennedy et al., 1998). Organic solvent (e. g. dichloromethane) can be added immediately where convenient to limit volatilisation or hydrolysis, although care to prevent leakage is essential. Extraction of water samples with organic solvent should be made within 48 hours and immediately on receipt. Even so, it can be anticipated that samples containing endosulfan isomers will loose chemicals by volatilisation if jars are not properly sealed, ideally with Teflon. A loss of chemicals can also occur by hydrolysis if the pH of the water is above 8.

Due to its instability in over time, automatic water sampling devices require regular human intervention, e. g. to collect water samples (every 24 hours), to fit in new sampling bottles/lysimeters, to change device batteries and for other maintenance work.

D31 fig1005
Figure 10.5
D31 fig1006
Figure 10.6

Another way consist to directly sample the analytes of interest (pesticides and nitrates) rather than water. This is the aim of passive sampling technologies, which have been developed to monitor pollutants in the aquatic environment. The advantage of passive samplers is, that they sample in situ without disrupting the environment. They can be used in ground- and surface water. Thanks to their phase or selective membrane, these devices allow to integrate sampling over the time and as a result to concentrate molecules.(Figure 10.6:). Their capacity of accumulation allows to improve the sensibility of the analytical process and so to detect concentrations of micropollutants in concentrations measured in µg/L or even ng/L. Sampling proceeds without the need for any energy sources other than this chemical potential difference. Several types of devices are used depending on targeted compounds. Pesticides can for example be sampled by SPME, SLM, sorbant devices, SPMD, PDBS, POCIS, TRIMPS, dialyse membranes, Chemcatcher, TLC and PISCES devices. These tools require to remain submerged and do not respond well to dry episodes. They are considered from now on as complementary tools with the discrete water sampling techniques.


Note: For full references to papers quoted in this article see

» References


Main authors: Susanne Klages, Nicolas Surdyk, Christophoros Christophoridis, Birgitte Hansen, Claudia Heidecke, Abel Henriot, Hyojin Kim, Sonja Schimmelpfennig
FAIRWAYiS Editor: Jane Brandt
Source document: »Klages, S. et al. 2018. Review report of Agri-Drinking Water quality Indicators and IT/sensor techniques, on farm level, study site and drinking water source. FAIRWAY Project Deliverable 3.1, 180 pp


Contents table
1. The process of prioritisation of indicators
2. Survey of ADWIs already used in the FAIRWAY case studies 
3. First step in prioritistion of indicators in FAIRWAY

In this article, an overview on principles and aims of a priorisation process is given , followed by a summary on the outcome of a survey among FAIRWAY case studies on indicators used and an explanation of the stepwise priorisation process chosen in FAIRWAY.

1. The process of prioritisation of indicators

The absence of a properly documented indicator selection process is not a minor issue: Niemeijer and de Groot (2008) explain, that the choice of indicators highly influences conclusions as to whether environmental problems are serious or not, whether conditions are improving or degrading, and in which direction causes and solutions need to be sought. The authors propose to use the enhanced DPSIR-framework to frame the indicator-selection: causal chains are linked to form a causel network, similar to a flowchart. These are according to Niemeijer and de Groot (2008) the steps to steps to build a casual network:

  1. Broadly define the domain of interest.
  2. Determine boundary conditions that can help determine which aspects to cover and which to omit.
  3. Determine the boundaries of the system.
  4. Identify (abstract) indicators covering the factors and processes involved.
  5. Iteratively map the involved indicators in a directional graph.

Figure 8.1. shows the ideal process for indicator selection.

D3.1 fig08.1
Figure 8.1

The following elements and criteria are of relevance for the process:


Contextualisation describes the preliminary choices and assumptions (Bockstaller et al., 2008) and includes a definition of the purpose of the analysis, the desired level of operation (farm, region, member state…), the temporal analysis scales and also the involvement of stakeholders (Lebacq et al., 2013).

Agricultural relevance

According to CORPEN (2006), indicators for nitrate pollution that describe or estimate the condition of a plot (e. g. soil cover, nitrogen budgets and model-derived indicators) are more relevant than the indicators that only describe fertilisation practices (e. g. phased fertilisation). Indicators of high relevance should be preferred. Indicators of low relevance should be used only as part of a set. These sets are, however, difficult to interpret: the larger the number of indicators, the more likely they give divergent assessments. To avoid this, single indicators can be combined within a chart or an index. Annex 6 in »FAIRWAY Project Deliverable 3.1 shows estimates on plot level of the relevance of a range of indicators evaluating the potential of nitrates pollution of ground- and surface waters.

Data availability within case studies/official statistics

Often, limitation of data availability compelled data driven approaches to focus on agricultural practices and hence on means-based indicators. Model-based and effect-based data indicators require context-specific data, e. g. climate and soil characteristics or specific on-site measurements, that are for some reason not available. A solution may be the use average data as default values, for a region or a sector (Lebacq et al., 2013).


With increasing scale, direct methods are getting too expensive and are replaced by indirect methods. Annex 7 in »FAIRWAY Project Deliverable 3.1 shows for plot, farm and regional level how the feasibility of indicators for the potential of nitrate pollution of ground- and surface waters was evaluated for France by CORPEN (2006).

According to Lebacq et al. (2013), criteria for a prioritisation of indicators can be summarised under the main categories

  • relevance,
  • practicability and
  • end user value.

In data-driven approaches, most means-based indicators and some intermediate indicators, such as nutrient surplus, can be used to assess environmental themes, because they are based on farmers’ practices. As means-based indicators often posess a low quality of prediction of envirnmental impacts, in order to increase accuracy, Bockstaller et al. (2008) propose to use a combination of indicators for the same theme. This procedure may be complicated in practice and requires an aggregation process but allows to focus on significant variables to develop simplified indicators.

In order to be able to compare indicators, functional units are applied (Thomassen et al. 2008):

  • the expression of impacts per amount of product (i.e. liter of milk, kilogram of meat) is related to the function of market goods production,
  • the expression per hectare of agricultural land refers to the function of non-market goods production, such as environmental services (Basset-Mens and van der Werf 2005).

Indicators concerning global impacts, e. g., greenhouse gas emissions, should be expressed per unit of product, while indicators related to local impacts, e. g., eutrophication potential, should be expressed per hectare (Halberg et al. 2005a).

Indicators differ with respect to the compartments considered (i.e. soil, surface water, groundwater and air) and effects taken into account. Therefore, the results obtained can strongly depend on these factors (Oliver et al., 2016). The evaluation of indicators should consider multiple applications and wide range of applicability. They should also take into account the synergistic effects of applying different pollutants (e. g. pesticides) and they should consider the application method and the level of application (regional, field scale).

2. Survey of ADWIs already used in the FAIRWAY case studies

The aim of this section of FAIRWAY is to to prioritise and evaluate data-driven indicators for the monitoring of the impact of agriculture activities on nitrates and pesticides in drinking water.

A questionnaire on ADWIs already in use was compiled and sent to all FAIRWAY case studies. Case study leaders were asked to choose out of a set list those indicators for drinking water pollution by nitrates and pesticides used in their case study. They were also asked to indicate the level (plot, farm, regional or higher) on which data for the calculation of these indicators are available. Case study leaders were also asked for further suggestions on ADWIs. The results of this survey are enclosed in this report as Annex 2 in »FAIRWAY Project Deliverable 3.1 and also available as »Milestone 3.1 from the FAIRWAY website.

Main results of this survey are as follows:

  • The aim, size and structure of the different case studies are different, and so are the ADWIs in use.
  • Those case studies with focus on nitrate pollution do not dispose pesticide indicators and vice versa.
  • ADWIs and the data to calculate them may be available on plot, farm or regional level.
  • There are far more indicators and data in use which are related to nitrogen than to pesticides.
  • Indicators in use for pesticide pollution are combined/compound indicators.

Questions on confidentiality of farm data aroused in conjunction with the survey. This is due to uncertainties related to the new regulation on data protection (EU 2016/679), but also due to a tightening of fertiliser legislation in some Member States.

3. First step in prioritistion of indicators in FAIRWAY

From the other articles in this section of FAIRWAYiS, the following aspects for a further prioritisation of ADWI can be deduced:

  • ADWI are useful on all levels: at farm level as an aid in farmer’s consultation, at local or even national level as an evaluation and monitoring tool for administration work and for policy-makers.
  • Regarding the two kinds of pollutants – nitrate and pesticides – frame conditions are quite different:
    - Nitrate is one single substance, being mobilised and immobilised, leached, transported by runoff and emitted. It is essential for plant growth and omnipresent, even under “natural” conditions.
    - On the contrary, around 250 so called “active substances” of pesticides are authorised by EFSA. Placement on the market of pesticide product needs national approvement. They may only consist of the registered active substances registered on EU-level, pure or in mixture, and of additives, for a better handling of the pesticide. Pesticides are supposed to be – to the greatest possible extent - harmless. They are supposed to degrade or at least to be absorbed by the soil matrix, but not to leach into groundwaters. Improper handling may however lead to runoff or drift and therefore to pollution of surface waters.

In »Agri-drinking water quality indicators at farm and drinking water levels, possible ADWIs are listed and explained. The ADWIs include those being subject of the survey among the case studies, including those indicators the case study leaders were proposing to be included in a further evaluation. Additionally, indicators used for pesticide monitoring/risk assessment were included, the range of pesticide indicators used in case studies was limited (see Annex 2 in »FAIRWAY Project Deliverable 3.1).

From the number of indicators listed, it can be deduced that indicators which act in the agricultural sector as driving forces and as pressure indicators, are far more numerous than state respectively impact indicators. In this sense, the relation being visualised in for AEI related to water quality on European level is mirrored for the frame conditons of the FAIRWAY project. The large number of driving forces and pressure indicators which stand for agricultural activity also explains, that from this part of the DPSLIR-model, many factors may influence water pollution. State indicators which are used for the evaluation of the water quality are on the contrary far more standardised, like the water quality standards they are supposed to monitor.

We have introduced the new concept of Link indicators within the DPSLIR-model, in order to explain the time lag between agricultural activity and water pollution and to elicit, which farm management practices would at all lead to water pollution.

A prioritisation of ADWI is therefore above all necessary for the driving forces and pressure indicators in the agricultural sector, in order to focus on the most

  • significant,
  • prevalent
  • effective and
  • easy to use indicators.

The survey on ADWIs already used in case studies and the most promising indicators discussed in »Agri-drinking water quality indicators at farm and drinking water levels lead to a first weighting of indicators. The result is listed in Table 8.1. On the right part of the table, three columns were added, which show the evaluation of a survey among FAIRWAY case studies about data availability in order to calculate ADWIs. Answers would also indicate the resolution in space, in which data can be delivered from the case studies (at plot, farm or regional/larger scale).

Table 8.1: Ranking of ADWI according to significance and prevalence based on a survey carried out in FAIRWAY

Subindicator of ADWIs Prevalance: evaluation of data availability in case studies (number of times mentioned)
  Plot scale Farm scale Regional scale
Land use/land cover 6 2 5
Land use change      
Precipitation/evapotranspiration 2 2 12
Soil type 5 1 4
Organic carbon      
Organic/conventional 1 7 1
(Average) crop yield  1 7  1
Cropping patterns      
Method of soil cultivation/tillage practice      
Soil cover      
Livestock density (LU/ha /yr on an area of reference) 3 7 4
Livestock excretion (kg N/ha/yr on an area of reference) 1 5 1
Organic fertilisation/ha; organic fertilisation/crop*ha 2 6 0
Mineral fertilisation/ha; mineral fertilisation/crop*ha 4 4 6
Total fertilisation/ha; total fertilisation/crop*ha 2 7 2
Type of Pesticides      
Chemical properties      
Consumption of pesticides      
Application of pesticides/ha (active substances; frequently used; most persistent or toxic) 2 6 0
Application of pesticides/ha*crop (active substances; frequently used; most persistent or toxic)      
Timing of pesticide application      
Splitting/frequency of pesticide application      
Nitrates in soil water 4 1 2
Pesticides in soil water      
Nitrogen leaching risk indicators      
Pesticide leaching risk indicators      
Surface transport of nitrogen and pesticides (with soil/fertiliser particles)      
Pesticide drift      
Volatile N-compounds      
Nitrate: grazing animals near surface waters, farmyard, storage facilities      
Pesticides: farmyard, pesticide storage facilities      
Annual average nitrate concentration (mg NO3/l) 4 1 8
Concentration trend analysis      
Frequency of exceedance quality standards (%) 2 0 8
Nitrogen maximal concentration in drinking water collection points 3 0 8
Catchment typology and dominant flowpath      
N stable isotopes      
Number of substances that exceed water quality standards at least once the year 4 0 7
Maximum concentration by substance (if >0.1 µg/l) in drinking water collection points 4 0 7
Frequency of exceedance quality standards in the drinking water (percentage of the number of samples where the 'drinking water' standard is exceeded) by substance 4 0 6
Vulnerability assessment maps of aquifer and surface water 2 0 7

In Table 8.1, ADWI, for which data can be supplied by the case studies are marked in orange. ADWI for which data can (possibly) not be supplied by case studies are marked in blue. This may be the case because these data are not used in certain or all case studies, or because in the data survey carried out in the beginning of FAIRWAY, we did not ask for the specific information. This applies to background information (e. g. climate, topography, rock types) about the case study sites, which may be critical for leaching risk assessment and catchment typology; therefore, in the data compliation stage, we will collect such data as well. It also refers to specific information particularly on pesticide use. Case studies do not seem to collect specific data on the use of single active substances. But from sum parameters and general indices, no link can be drawn to the parameter at sink level (e. g. pesticide analyses of raw water).

Indicators, for which data are not readily available in the case studies may be calculated if these data are free available from other data sources. Annex 3 in »FAIRWAY Project Deliverable 3.1 lists data sources for free available data in order to calculate ADWIs. One example is the use of pesticides, which may be deduced from local cropping patterns and from usage data reported from the Member States according to Regulation (EC) No 1185/2009. The next step towards priorisation will be done in FAIRWAY using data of cathments in the case studies (see »Further prioritisation and evaluation of agri-drinking water quality indicators). For this reason, data are requested from the case studies.


Note: For full references to papers quoted in this article see

» References


Main authors: Susanne Klages, Nicolas Surdyk, Christophoros Christophoridis, Birgitte Hansen, Claudia Heidecke, Abel Henriot, Hyojin Kim, Sonja Schimmelpfennig
FAIRWAYiS Editor: Jane Brandt
Source document: »Klages, S. et al. 2018. Review report of Agri-Drinking Water quality Indicators and IT/sensor techniques, on farm level, study site and drinking water source. FAIRWAY Project Deliverable 3.1, 180 pp


Contents table
1. ADWI pressure indicators in the French case study 
2. ADWI state indicators in the French case study 
3. Linkages between ADWI
4. Future on work on indicators and database
5. Main insight of the approach

In order to further drive forward the proiritisation of the selected ADWIs, we intend to connect ADWIs from the agricultural and the water work side, using statistical methods.

We also intend to further investigate on the Link indicator, especially how this ADWI fits in between the other indicators. We intend to examine

  • the feasibility of indicators calculation,
  • the link between indicators and
  • the relevance of some indicators, as statistical calculations give the mathematical expression for the link that exists between them.

For this purpose, a database of ADWI-data on catchment level will be collected from the FAIRWAY-case studies. Preparatory work has been carried out, in order to specify the data request to the case studies. In ththis article, trial calculations being conducted with a small database are further explained.

The implemented database – here called ‘draft database’ – contains French data, provided by ‘Eau de Paris’ (for more information, see »La Voulzie Case study description).

1. ADWI pressure indicators in the French case study

French Ministry of Agriculture (RGA: General agricultural survey) provides, for each municipality, the Utilized Agricultural Area (UAA) and the distribution of this area between crops (ex: wheat, maize, rapeseed). These data come from farm surveys achieved in 1970, 1979, 1988, 2000 and 2010. A time step transformation (interpolation) has been carried out to obtain yearly data, for each major crop (fallow, meadow, sunflower, peas, maize, oilseed rape, sugarbeet, spring barley, winter wheat). These data were integrated in the draft database and now graphs can be plotted to illustrate evolutions over time.

It should be highlighted that over the studied period, cereals are most important in terms of proportion of land use.

Nitrogen fluxes were calculated using a soil surface budget method (Oenema et al. 2003) implemented in an online tool (Cassis-N, Poisvert et al, 2017).

In this method, data of different scales are used (and combined/homogenised in the tool itself):

  • Regional (SAA: annual agricultural statistics, and UNIFA: Union of Industries of fertilisation),
  • Departmental (SAA and UNIFA),
  • Municipal (RGA: general agricultural survey).

Providing the spatial delimitation of the studied catchment (shapefile format), calculations are performed and expressed in relation to this catchment leading to the following outputs:

  • Mineral fertilisation,
  • Organic fertilisation,
  • Atmospheric deposition,
  • Fixation,
  • Output (plant N consumption),
  • Surplus (i. e. budget calculation).

 D3.1 fig09.01
Figure 9.1
D3.1 fig09.02
Figure 9.2
D3.1 fig09.03
Figure 9.3

In addition, a N soil surface budget calculation is performed using a N budget = [mineral fertilisation + organic fertilisation + fixation + atmospheric deposition]- [Output (plant N consumption)]. Plant N consumption included grazing (grass N comsuption) but there very little animal production in this catchment. All these data were then released at annual time steps and integrated to the draft database. This budget is refered as a surplus in Cassis-N method.

Most of the hydrologic data were collected by Eau de Paris at the “La Voulzie” spring and are transmitted to the BRGM. The hydraulic head is measured at the BRGM piezometer located in the city of Bauchery Saint-Martin (6 km, north-east). All these data are available at irregular time steps varying from daily to monthly. Bulk data were integrated to the draft database and a homogenisation of the time steps was performed.

2. ADWI state indicators in the French case study

The concentrations of nitrate were collected by Eau de Paris at the “La Voulzie” spring and were transmitted to the BRGM. All major chemical compounds are analysed, but nitrate only will be used in the present project. Nitrate time series starts in the 50’s with a concentration close to 20 mg/L and exhibit a nearly continuous rise up to the year 2000. After that date nitrate concentrations seems to be more stable. Reasons of stabilisation could be found in climatic driving forces (more humid period for example), but part of the stabilisation could also be due to the implementation of action plans at the national (nitrates directives) and local (Fertimieux) level.

D3.1 fig09.04
Figure 9.4

3. Linkages between ADWI

ADWI calculation and linkage

In order to find a link between pressure and state indicators it is intended to investigate relations that exist between nitrogen input on a study site and the observed effect on water quality. When needed, information on the hydrogeological/hydrological functioning is shown.

For that reason, statistical links between indicators that were integrated in the draft database were computed and tested.

Pre-processing of the available data had been carried out, mainly in order to homogenise time steps for all data. In a general manner, main processing consists in scaling down data to annual values by averaging or summing data. Annual data are attributed to the first July of the running year.

Impact of recharge on spring discharge

At a yearly scale, linkage between recharge and spring discharge can be evaluated using a cross correlation function (ccf). The intensity of the correlation returned by ccf(x, y) varies between 0 and 1, and being 1 at best. For a range of lag (k) the correlation is calculated between x[t+k] and y[t] and then plotted on a bar chart. Each bar on the chart is then separated from another by 12 months (i. e. one year).

From this analysis, it can be seen that the best correlation is found for a lag k= -12 months, which means, that the spring discharge time series can be rather well explained by the evolution of the recharge of the year before. This analysis is of importance, since nitrate is measured in water from a spring, whereas nitrogen inputs are measured at the soil/root zone. Transfers from the soil/root zone to water should then be described in simple terms.

D3.1 fig09.05
Figure 9.5
D3.1 fig09.06
Figure 9.6

Using information coming from the cross correlation analysis, recharge shifted by 12 months and discharge are plotted on the same graph. It highlights the similarity of the time evolution. One should remark that main evolutions are similar, but small variations (peaks mainly) diverge in some cases.

Exploring the link between recharge and NO3-

Annual variation in nitrate concentration could be due to variations of the flow regime. As discharge flow rate is mainly governed by recharge rate, the relation between recharge and nitrate concentration could be directly investigated. Using the same cross correlation method as before, best correlation between annual recharge and nitrate concentration is found at lag k= - 36 months (correlation is the highest when evaluated between y[t] and x[t-36]), which correspond to 3 years. In that case, the relation is weak (CCF=0.20 at best) in comparison of the previous case. The trends in the nitrate time series is one of the more evident explanation.

  D3.1 fig09.07
Figure 9.7
D3.1 fig09.08
Figure 9.8

Further tests showed, that the two time series do not have the same cyclicalities. Discharge has an annual cyclicality (12 months), while the time series of concentrations has a cyclicality of 8 years. Recharge (blue dashed line) shifted by 3 years and annual mean nitrate concentration. Due to the increasing trend in nitrates, similarity between the curves is not as good as for the recharge/discharge approach.

Linkage between Cassis-N surplus and NO3- concentration

Cassis-N calculates the nitrogen surplus and therefore is similar to the GNB, but does not follow exactly the calculations intended by OECD rules (OECD 2007) as, for instance, volatile N-losses from manure are not considered. This indicator could be considered as representative for the amount of nitrates actually inflowing in the aquifer. So, a relationship between the nitrogen budget and the nitrate concentration is expected. Using a similar approach, the best correlation between nitrogen budget and nitrate concentration was searched for.

The best correlation was found at lag k= -12 months, which means, that the strongest correlation between x (GNB) and y (NO3) is found for x[t-12] and y[t]. This signifies, that nitrogen surplus data of the year before (12 months before) are the most appropriate to explain a linkage between nitrogen surplus and NO3--concentration. Nevertheless, the correlation is relatively weak (but significant).

D3.1 fig09.09
Figure 9.9
D3.1 fig09.10
Figure 9.10

Plotting the two time series on the same graph highlights the rather poor correlation between this two time series. The increasing trend of the NO3-, whereas the Cassis-N surplus is decreasing starting from 1990, can explain this lack of strong relationship. On the following plot, NO3- - concentrations are in green (shifted by 12 months) and the Cassis-N surplus is plotted in purple (dashed lines).

Linkage between mineral fertilisation and NO3- concentration

Cassis-N also provides the mineral fertilisation. Since there is very little organic fertilisation on the catchment (Figure 9.2), this indicator could represent by itself the load of nitrogen entering the system. A relationship between the mineral fertilisation and the nitrate concentration in the water of the well is expected even though processes of nitrogen consumption by crop would regulate the inflow in the aquifer.

Since mineral fertilisation is a much more direct indicator than the surplus, it represents a much simplier opportunity to test the input/output link into the system.

In that case, the correlation is much better (0.8) and represents a rather good explanation of the variation of nitrate concentration. A disturbing result is the absence of lag for the peak of correlation. Nevertheless, giving the objectives of this first approach, one can invoke several processes than can affect the statistical analysis and explain the results. Investigating these processes remains out of the scope of this study.

D3.1 fig09.11
Figure 9.11
D3.1 fig09.12
Figure 9.12

Plotting the two time series on the same graph highlights the rather good correlation between this two time series. Tthe correlation (0.8) is better than with surplus (0.2).

4. Future on work on indicators and database

Main ideas for databases

The first attempt to build this database enabled the first calculations of indicators as well as the first links between pressure indicators and state indicators. These links have shown that the most integrated and aggregated indicators (GNB or Cassis-N surplus) are not the ones that yield the best results. Thus, the study of the cross-correlation between N-surplus and concentration of nitrates in the spring water does not show the best results. Additional data could be integrated to better account for water and solute transfers in the unsaturated zone, but there is a risk of obtaining a too complex indicator or a too complex model to explain links between indicators. The integration of many explanatory factors in order to obtain a model has been realised by the BRGM on the source of the Voulzy, this type of work deviates from the will to simplify the FAIRWAY project.

First test in order to implement the data base at the project scale

The first tests were carried out with the French case study data. It shows that diversity of time steps are important to take into account. An appropriate approach could be developed. Integration of other case studies data is a work to achieve in order to deliver a final and definitive database by the end of the project.

Again, the topic “data availability respectively accessibility” will be of relevance. Linking data and developing algorihms will bring us to questions on :

  • data resolution in the time and space and
  • data quality and uncertainty.

The most effective data according to the simulations on cathment level should be further tested in the case studies on relevant target groups, in order to identiy the most feasible indicators.

In the meantime, other indicators will be calculated and compared with each other in order to eventually link the pressure indicators and the state indicators.

5. Main insight of the approach

First attempts to integrate case studies data in a rather simple database and to explore link between this indicators shows that:

  • the data base modelling (i. e. the establishment of the database) is of importance to have an internal platform for the discussion of what data/indicators may represent and who they can be linked,
  • the issue of time steps has to be taken into account,
  • the issue of spatial definition of data also has to be taken into account,
  • the search of links between pressure and state indicators can be time consuming and could be strongly dependent of the data available for each case study. A major risk of invoking rather complex relations between indicators is identified, which lead to consider probably more conceptual (i.e. flowcharts) than statistical links between indicators.

Evaluation of the indicators

After meaningful indicators have been identified, they need to be further evaluated. Evaluation of indicators can be effectuated by different methods (Bockstaller et al., 2008).

Sensitivity analysis: Aim of a sensitivity analysis is to test how the indicator outputs react to input variables: the information can be used to select the most effective indicators.

Evaluation of the quality of the indicator: Bockstaller et al. (2008) report, that the classical approach for evaluating the accuracy of a model prediction is to compare calculated and measured data. However, this approach may be difficult to apply to simplified indicators Rigby et al., 2001, Reus et al., 2002) and propose a three-step methodological framework for the evaluation of environmental indicators.

Evaluation of the indicator design: This evaluation could be effectuated by an expert panel, alternatively, a peer-reviewed article is suitable. Aim is to identify need of improvement.

Evaluation of the indicator output: This step is based on the comparison of the indicator output with measured data. While indicators based on a simulation model can be directly evaluated using the measured data, simplified indicators need a modified approach for their evaluation.

Evaluation by end-users: A consistent and comprehensive set of indicators is userful for the correct interpretation of complex systems. Taking into account the interactions of indicators enables the investigator to map appropriately the main structure and processes of the system (Binder et al., 2010).

With a selection process based on the concept of a causal network, interactions between environmental themes and indicators can be taken into account. To assess the environmental impact of nitrogen fertilisation on surface water ecosystems, selection is made by considering a network composed of causal chains related to crop production, socioeconomic issues, air, soil, and water (Niemeijer and de Groot 2008).

The identification of correlations between indicators is recommended in the process of selecting a minimal, consistent, and sufficient set of indicators. By way of illustration, Thomassen and de Boer (2005) showed for dairy farms in the Netherlands, that there was a correlation between nitrogen surplus and eutrophication potential onfarm. This means, that the nitrogen surplus is relevant “to a moderate extent” to assess the environmental impact of eutrophication with more easily available data.

Evaluation by end-users: The purpose of this last step is to evaluate the acceptance of the end-users for the new indicator. The develover can at this stage collect feed back, especially suggestions for improvements.



Note: For full references to papers quoted in this article see

» References


Main authors: Susanne Klages, Nicolas Surdyk, Christophoros Christophoridis, Birgitte Hansen, Claudia Heidecke, Abel Henriot, Hyojin Kim, Sonja Schimmelpfennig
FAIRWAYiS Editor: Jane Brandt
Source document: »Klages, S. et al. 2018. Review report of Agri-Drinking Water quality Indicators and IT/sensor techniques, on farm level, study site and drinking water source. FAIRWAY Project Deliverable 3.1, 180 pp


Contents table
1. Indicators at farm level in the DPSLIR framework 
2. Indicators at drinking water level in the DPSLIR framework
3. Indicators for linking farm and drinking water levels in the DPSLIR framework

In Annex 1 in »FAIRWAY Project Deliverable 3.1, all 28 AEI according to COM final 0508/2006 and applied on EU-level are listed (COM 2006, eurostat 2018). The AEI are allocated to domains and subdomains of the DPSIR framework.

In the table below we have compiled all agri-drinking water quality indicators (ADWIs), which were reported to us during a survey among the in the FAIRWAY case studies and we supplemented the table with indicators according to a literature review. They are grouped according to whether they are relevant at farm (driving force and pressure indicators) or drinking water level (state/impact indicators) or if they provide a link between the two.

1. Indicators at farm level in the DPSLIR framework

Domain Sub-domain Indicator category Indicator
Impact Societal and economic demands  
  • Demands for clean drinking water *)
  • Population density *)
  • Cost for drinking water production *)

*) Indicator not discussed in this report

Driving forces               Resource management and planning   Land use planning
Agricultural preconditions
Farm management          Farming standards
Farming intensity
Farm management
Pesticide application
Pressure  Leaching Leaching travel time
    Leaching quantity
    Nitrogen in soil water
  Surface water pollution  
    Pesticides in surface water
  Point sources  
  Aerial emission  
  Nitrogen efficiency Nitrogen budgets


2. Indicators at drinking water level in the DPSLIR framework

Domain Sub-domain Indicator category Indicator
State/Impact Water quality  
  Regulatory compliances  


3. Indicators for linking farm and drinking water levels in the DPSLIR framework

Domain Sub-domain Indicator category Indicator
Link Catchment typology   
  Lag time  
  Source identification  
  Vulnerability of the hydrogeologic system Nitrate vulnerability assessment
    Pesticide vulnerability assessment
  Environmental risk Nitrogen loss indicators - overview Nitrogen loss
    Pesticide risk indicators - overview


Note: For full references to papers quoted in this article see

» References


Go To Top