Knowledge journal / Edition 1 / 2022


From predictive toxicology with algorithms to groundwater drainage in New Orleans

This is the fourteenth edition of Water Matters, the knowledge magazine of H2O. This edition consists eleven articles on diverse subjects, written by Dutch water professionals based on thorough research.

When assessing the articles, the editorial board consisting of experts from the sector, made a selection, looking for a clear relationship with daily practice in the water sector, which is the purpose of Water Matters. Research, results and findings form the basis for articles that describe new knowledge, insights and technologies with a view to practical application.
This edition again covers a wide range of topics. From the role of the groundwater balance on water availability to monitoring biocides in the water chain.
But also: Lessons on drought, mitigation and adaptation for Dutch water boards; far-reaching removal of pharmaceuticals and phosphorus with PAC + cloth filtration; advanced oxidation and dune filtration for the removal of organic micro-pollutants; system approach for the improvement of the ecological quality of Dutch waters; automatic interpretation of drinking water pipeline inspections; predictive toxicology with algorithms; the role of the groundwater balance on water availability; an area-oriented approach to the emission of nutrients in agriculture; spreading of groundwater contamination by open geothermal energy systems and groundwater drainage in New Orleans.
Water Matters is, just like the magazine H2O, an initiative of the Royal Dutch Water Network (KNW), the independent knowledge network for and by Dutch water professionals.
The publication of Water Matters is made possible by leading players in the Dutch water sector. These Founding Partners are Deltares, KWR Watercycle Research Institute, Royal HaskoningDHV and Stichting Toegepast Onderzoek Waterbeheer (STOWA). With the publication of Water Matters the participating institutions want to make new, applicable water knowledge accessible.

You can also read the Dutch version of Water Matters digitally on H2O-online (

Would you like to respond? Please let us know via

Monique Bekkenutte Publisher (Royal Dutch Network)
Huib de Vriend Chairman editorial board Water Matters


Knowledge journal / Edition 1 / 2022

A balanced approach to the groundwater balance

"The groundwater system is out of balance." "More is abstracted than recharged”. Or "we’re going to balance groundwater demand with groundwater supply." These are all statements that have recurred regularly in reports and press releases over the past year. There is still much misunderstanding about how a groundwater balance works. Using North Brabant as an example, we address three misunderstandings and show what role a groundwater balance can play when developing policies for water availability in areas of water scarcity.

The province of North Brabant with its thick sandy aquifers full of groundwater is special; not only is there a lot of groundwater, but the province is also very dependent on it. All drinking water is produced from groundwater and agriculture uses a relatively large amount of groundwater; it is estimated that half of the groundwater irrigation in the Netherlands is done on Brabant farmland. This article is based on two studies for North Brabant using a regional groundwater model and water balance calculations. The principles described also apply to other sandy provinces.

Basic principles of groundwater balance in this article

• The groundwater balance describes the inflows and outflows of groundwater.
• It is conclusive by definition. Water cannot disappear.
• The balance is defined in space and time.
• It covers the fresh groundwater up to the impermeable geohydrological base.
• The balance varies. In winter, groundwater recharge takes place, causing groundwater levels to rise (and groundwater volume to increase); in summer, the opposite takes place. This averages out over several years, provided that external factors such as climate change or water use do not change in trend.
• Groundwater recharge is equal to the amount of precipitation minus evaporation and surface water drainage. In winter this is a positive number, in summer a negative number.

Misunderstanding 1: there is more abstraction than recharge

A good balance between abstraction and recharge is a logical precondition for a sustainable groundwater system and is a requirement of the EU Water Framework Directive (Article 4.1.b.ii). If more is abstracted than recharged, the groundwater system will eventually become exhausted ('mining'). Due to overuse, deep water tables and groundwater levels continue to fall, and the saltwater interface rises. This is happening on a large scale worldwide, including in California, northern India and around Beijing.

Figure 1. Groundwater balance (in millions of m3 per year) for North Brabant for an average year (top) and a further simplified groundwater balance to deep groundwater (bottom)

Figure 1 shows a highly simplified groundwater balance for North Brabant. Groundwater recharge (1) is by far the largest item in the water balance (1,686 million m3 per year). Groundwater abstraction for drinking water, industry and agriculture (6) mainly takes place from the thicker aquifers and amounts to 256 million m3 per year. Therefore, there is a large amount of groundwater being extracted, but the recharge is more than six times greater than the abstraction.
Theoretically, it is even possible to meet the entire need for groundwater in the Netherlands by abstracting water from the subsoil of North Brabant, taking into account a balance between recharge and abstraction. The result: extremely low groundwater heads, permanently dry streams, hardly any water in the Meuse, but still an equilibrium situation (Verhagen, et al., 2017). Abstracting more water leads to a new equilibrium with lower groundwater levels and thus less groundwater drainage to surface water. Simply put, the surplus of precipitation is divided between surface water drainage and groundwater abstraction. With more groundwater abstraction, the surface water drainage decreases.
A balance between abstraction and recharge does not always mean a sustainable situation. The consequences of the current groundwater abstractions are permanent low groundwater heads in deep aquifers, sometimes by as much as several meters. This has consequences for the entire water system. Areas that used to be fed by deep seepage have now become infiltration areas, groundwater levels have been lowered over a large area. This in turn affects the supply of groundwater to these areas, the amount of water drained by surface water and the amount of water available for nature and agriculture.

Misunderstanding 2: a balanced water balance is the key to sustainable water management

It is said: "There is more than enough precipitation in Brabant, but about 80 percent is drained, through ditches, canals and streams via the Meuse to the sea. The remaining water infiltrates into the soil and recharges the aquifers. But all this water is then withdrawn for agricultural irrigation, as drinking and industrial water and for watering gardens, sports fields and other urban green spaces."
Of the 1,686 million m3 a year of groundwater recharge, a net amount (drainage minus infiltration of surface water in water supply areas) of 1,488 million m3 per year is drained away via ditches, brooks and rivers (Figure 1). This leaves 198 million m3 a year for infiltration into the deeper groundwater. From that deeper groundwater, another 256 million m3 a year are extracted, which is 60 million m3 a year more than the deep groundwater recharge. This quantity is brought in horizontally from across the provincial border.
However, suppose the abstraction for drinking water, industry and agriculture is reduced by 60 million m3 a year. Groundwater levels will then rise and most of this volume will benefit from additional surface water drainage. A new situation arises, a new balance of the water system. Therefore, the recharge to the deep groundwater also decreases by about the same amount. In numbers: A reduction of 60 million m3 a year in abstraction also means almost 60 million m3 a year more drainage to surface water, partly from inflowing deep seepage water. In other words, there is no unique ratio whereby the recharge to deep groundwater, the abstraction of groundwater and the discharge by surface water are in balance.

Misunderstanding 3: surface water flow must be prevented

Stream flow is an essential part of the water cycle and is needed, for example, to provide streams with sufficient base flow (also a requirement of the Water Framework Directive).
Every drop of water is eventually drained or extracted, providing it does not evaporate. Surface water drainage can occur quickly or slowly; within a few days when a raindrop enters surface water via drains, or over many hundreds (to tens of thousands) of years when a raindrop eventually emerges as seepage via groundwater flow. Drainage therefore has many varieties; where does it come from and how long was its journey? The low-lying areas receive seepage water and naturally discharge more (surface) water than the core infiltration areas that lack surface water.
A general ratio between groundwater recharge and surface water discharge can therefore not be defined, because this is area-dependent. Managing the water balance is therefore primarily a distribution issue. A more robust water balance implies smaller differences between winter and summer seasons and longer times in the groundwater system (Stuurman et al., 2020). Ideally, sufficient water is stored in the spring, to provide stream flow in the summer.

Figure 2. The groundwater balance of North Brabant for summer and winter.

How to proceed?

The conclusion is that a general consideration of recharge and abstraction gives little indication for defining good groundwater status. There is no fixed ratio for abstraction and recharge. A further analysis per (catchment) area is needed to gain more insight into the requirements for groundwater levels, seepage and discharge. These requirements differ for each water and nature system. For example, upper stretches of streams are highly dependent on a sufficient and constant supply of groundwater, the base flow. Other areas require a sufficient seepage flow throughout the year with suitable groundwater quality.
An integrated assessment of groundwater and surface water is a requirement of the WFD. A systems approach including a water balance helps here. Areas of concern are the influence of groundwater management on the surface water system (the degree of drainage) and the influence of groundwater or groundwater-dependent nature areas (both quantity and quality). By defining the required availability of sufficient water of good quality in streams and nature, also in time, the water system can be better designed. Insight into the water balance and the size of the water balance items over time helps (Figure 2) and shows the following on the scale of North Brabant:
- Evaporation is a major issue in summer. Total evaporation exceeds the amount of precipitation in winter; small changes in groundwater recharge in urban and rural areas can significantly contribute to more groundwater recharge over large areas;
- The horizontal supply of groundwater from outside the province is small compared to the groundwater recharge within the province;
- Irrigation is a smaller item than abstraction for drinking water. The essential difference is that this water disappears from the water system through crop evaporation, while the drinking water eventually enters the Brabant surface water system as sewage effluent;
- The seasonal differences are great. Almost all evaporation and irrigation take place in the summer half-year. In the dry summer months, more groundwater is abstracted daily for irrigation than for drinking water production;
- The groundwater system is already being used as a large buffer for storing groundwater in the winter for reuse in the summer;
- The item drainage to surface water is large and aggregated to a single number. The timing and location of drainage vary and have a major impact on water availability.

Floris Verhagen
(Royal HaskoningDHV)
Roelof Stuurman


Taking North Brabant as an example, we address three misunderstandings about the groundwater balance and show what role a groundwater balance can play when developing policies for water availability in areas of water scarcity. Areas of concern are the influence of groundwater management on drainage via surface water and on groundwater-dependent nature areas. By determining the required amount of clean water for streams and nature and other functions, also in time, the water system can be better designed. Understanding the water balance helps.


- Floris Verhagen, Tom van Steijn, Joachim Hunink, Roelof Stuurman. Draagkracht grondwater Noord-Brabant. Royal HaskoningDHV rapport WATBF3125R003WM. 21 december 2017

- Roelof Stuurman, Floris Verhagen, Arjan van Wachtendonk, Han Runhaar. Een verkenning naar de Watervraag van de Noord-Brabantse Natuur. Deltares rapport 11203929-002-BGS-0002. 7 oktober 2020

^ Back to start


Role groundwater balance

Knowledge journal / Edition 1 / 2022

Can Dutch Water Authorities learn how to cope better with drought by looking at other countries?

What lessons on drought mitigation and adaptation can Dutch Water Authorities learn from experiences, scientific articles and drought plans from countries with frequent prolonged droughts? Hoogheemraadschap de Stichtse Rijnlanden (HDSR), the Water Authority for Utrecht and South Holland, commissioned a study on this to complement the Blue Deal, a programme of Dutch Water Authorities with seventeen long-term partnerships in fifteen countries.

The HDSR area is subject to various drought risks. In the high sandy soils of the ice pushed moraine, also known as the Utrechtse Heuvelrug, groundwater is sinking deeper and deeper, while in the low-lying peatlands drought has negative effects on both agriculture and CO2 emissions due to peat subsidence. In urban areas, rainwater cannot infiltrate due to paving, foundations suffer from low groundwater levels and heat islands are created. Moreover, there is a growing demand in the Western Netherlands to receive water via the major rivers in the HDSR area (also known as the “KWA+ project” for capacity expansion of climate-resilient water supply).
For a long time, drought received relatively little attention in the Netherlands. However, in this century, 2003 and 2011 have seen major droughts, and the recent dry summers of 2018-2020 have shown the need for actively anticipating the risks of water shortage. The National Delta Decision on Freshwater (Deltabeslissing Zoetwater) contains the general policy to ensure that the Netherlands is resilient to freshwater shortages by 2050. Many other countries are more familiar with the high occurrence and duration of water shortages, so perhaps the Water Authorities do not need to reinvent the wheel and can be inspired by the different existing perspectives on risks and solutions.
An exploratory literature research of English-language management plans and scientific articles yielded a range of drought management measures. Through focus-group workshops at HDSR, these measures have been assessed based on the following criteria:
1. Does the drought measure match the drought problem in the Water Authority area?
2. Does the measure fit within the responsibilities of the Water Authority?
3. Does the measure suit the culture and norms of the Water Authority and the residents?
4. Does the measure suit the existing water system?
5. Does the measure coincide with the national objectives and the agreements with other Water Authorities?
6. Does the measure match the urgency of the risk respectively to other risks?
A number of drought measures that are new and relevant for Water Authorities in the Netherlands are discussed below.

Drought measures are used against water shortages (socio-economic drought), which depends on meteorological, hydrological and geophysical factors. Drought measures can increase (+) or decrease (-) water availability. More drought leads to more measures.

Better water distribution

In Australia, water rights have been decoupled from land rights to facilitate the trade of water (within a regulated 'market'). For example, water rights can be traded amongst farmers, encouraging a change in production or diversification of cropping systems. Water rights can also be granted to an authority that ensures protection and provision of ecologic needs. Such markets have also been set up in countries such as South Africa, the United Kingdom, Chile, Canada and the United States. This may inspire Dutch water authorities to expand the infrastructure of laws and regulations on the abstraction of surface water and groundwater to improve water distribution during droughts. Such a drastic reorganisation of the control of water requires institutional change at national level and a lot of additional enforcement. Water rights trading is most suitable for water authorities with large water shortages due to abstraction and salinisation.

Guaranteed water for fish

The State of Indiana (USA) protects fish and their habitat through legally guaranteed inflow rights: the minimum streamflow of rivers must be sufficient for the fish habitat. Therefore, abstraction from a brook is only recommended as long as the discharge is higher than the minimum flow for fish survival. In the Netherlands, water quality and ecology for fish welfare are included in water level decisions, so this measure fits within the current culture and standards. Inflow rights are best applied to rivers and brooks that flow freely by gravity. Since the emphasis of water authorities is now mostly on preventing duckweed and algae growth in smaller and stagnant bodies of waters, inflow rights for fish welfare offer an innovative perspective to this practice.

Proactive in case of drought: water consumption prioritization schemes

In preparation of the risk of longer droughts, local prioritization schemes have been implemented in Indiana (USA) and British Columbia (Canada). Public water suppliers distinguish different successive emergency phases in their water shortage contingency plans (WSCP). For each phase, triggers and measures are described that can be activated and deactivated by the water suppliers and water users. These WSCPs integrate the use of surface water, groundwater and alternative water sources and are interesting examples of locally applicable pre-crisis additions to the national displacement sequence. Targets for reduction of water demand are set in percentages, distinguishing between an objective of being prepared, voluntary conservation, mandatory restrictions and statutory action. Surveys are applied to monitor the effectiveness of the reduction targets. The WSCPs are publicly available whereby all residents are made aware of expectations and the ‘myth of water abundance’ fades.
Such water consumption prioritization schemes would refine the role of water authorities as mediators in times of water shortage. The agreements brokered by the water authorities in the summer of 2018 between water users can form the basis for publicly accessible schedules of agreements and measures for the various phases of drought. They offer perspectives for action in the risk dialogue. Integral management of groundwater and surface water is also becoming more important, bringing the responsibilities of water companies, water authorities and provinces closer together.

Weather index insurance

In North Africa, South-East Asia and South America, the principles of Conservation Agriculture are widely practised. One measure thereof is the weather index insurance. It covers deviations from a predetermined threshold of a weather index (such as precipitation or soil moisture) that can be monitored with a local weather station. These are indexes with a high correlation with the farmer's yield. Instead of receiving compensation for a reduced yield, farmers receive compensation for, for example, a lack of precipitation. This means that an insured farmer still has the same economic incentives to smartly manage his crop as an uninsured farmer and the need for insurers to assess damages in the field disappears. The disadvantage is that such insurance is a reactive way of dealing with drought. Dutch water authorities and farmers are looking for more proactive and sustainable measures. Yet it remains an interesting approach.

Early warning

In the U.S., drought is monitored and tracked in the U.S. Drought Monitor. It quantifies and visualises a composite indicator: a combination of different hydrometeorological indicators for meteorological drought, hydrological drought, soil moisture deficit and groundwater level. Composite indicators are a good addition to monitoring and early warning systems. As the monitoring of soils and water systems in the Netherlands is of a high standard, such an elaboration is a logical next step. In addition, water authorities can use water balance data to model and visualise many other drought management indicators, such as water scarcity caused by climate change and the (growing) water demand of stakeholders. Which warning system is most useful will depend on the water authority or the area of operation.
For example, rural Australia has developed a Spatial Simulation project, Aussie GRASS, to provide early warning of drought episodes that lead to large problems for grazing stock and feed availability. Moreover, there are models in use specifically for tracking water use for sewage and for water quality maintenance.

Drought and mental health

Because drought affects not only physical but also mental health, the sharing of citizens' experiences with drought through narrative research is applied in several countries, such as Australia, South Africa and the UK. The Drought Risk and You (DRY) programme in the UK collects stories at festivals and river walks through micro-interviews (audio) and short videos. These stories can clarify problems that people have with drought. In the UK for example, these include water shortages for gardening and access to nature, and problems in agriculture. In South Africa, future drought storylines are co-created based on scenarios from hydrological models. Through workshops with relevant storylines, set in a local context, participants can create personal narratives of the future with which they can explore (as yet) unknown drought events. This will increase social awareness and bottom-up preparedness, which may enable people to mitigate the effects of drought and adapt. Such methods can be a good supplement to the Water Authorities' Water Bosses Campaign to raise water awareness.

Discussion and conclusion

This study shows that Dutch water authorities can be inspired by other countries to create drought measures with a new perspective of the risks and responsibilities. This makes new solutions possible. Drought measures are often about counteracting water shortages, which is not only determined by meteorological, hydrological and geophysical factors, but often also by society's dependence on water. Therefore, drought measures from countries with different climates than the Netherlands are interesting for the Netherlands, as measures are targeted at drought risks that are largely determined by socio-economic and political factors.
If we learn more explicitly from our international projects, such as the Blue Deal, they can also contribute to our national water challenges. In this way, we turn working abroad into learning abroad.

Charlotte Offringa
(Utrecht University)

Opening photo
A dry lakebed, Laagravense Plas, Utrecht. Photo Charlotte Offringa


Hoogheemraadschap de Stichtse Rijnlanden (HDSR) is participating in the Blue Deal, a Dutch Water Authorities programme involving seventeen partnerships in fifteen countries. So far, the Netherlands has mainly been a supplier of expertise in such projects. When it comes to drought, however, it is obvious that Dutch water authorities can learn a lot from other countries. New perspectives can lead to innovative solutions, some of which this article discusses. If we start to learn more explicitly from our international projects, such as the Blue Deal, they can also contribute to our national water challenges.


- KNMI (Royal Netherlands Meteorological Institute) (2021, 25 October). KNMI Klimaatsignaal’21. consulted on 8 April 2022.

- Offringa, C. M. (2021). Getting Inspired: Applying other countries' drought measures to the Netherlands (Master's thesis). Utrecht University.

- Sivapalan, M., Savenije, H. H. G., & Blöschl, G. (2011). Socio-hydrology: A new science of people and 47 water. Hydrological Processes.

- Union of Water Authorities (undated). Blue Deal. - consulted on 8 April 2022.

- Van Kesteren, K. J. (2020, 10 August). Water authorities help dry countries, but do they learn from it? H2O/Waternetwerk. - consulted on 8 April 2022.

^ Back to start


Mitigation and adaptation

Knowledge journal / Edition 1 / 2022

Monitoring of biocides in the water system

Biocides do what they are supposed to do: control harmful or undesirable organisms. Depending on the nature of their application, however, they can enter surface water (and possibly drinking water) via different pathways. Here they are harmful to desirable organisms. What do we know about the emission pathways of biocides and the risks they pose to people and nature? And how do we develop efficient and effective monitoring strategies?

Biocide literally means life-killing. Legislation defines biocides as products containing one or more active substances to control (destroy, repel or render harmless) harmful or unwanted organisms. Therefore, they typically contain toxic substances. This makes them potentially harmful to organisms that do not need to be controlled, even after emission into the (water) environment.

In the Knowledge Impulse for Water Quality (KIWK), the national government, provinces, water boards, drinking water utilities and knowledge institutes work together to gain more insight into the quality of groundwater and surface water and the factors that influence this quality. The KIWK Ketenverkenner project has, amongst others, aggregated knowledge about biocides and their effects on water quality. The study which we describe in this article, investigated what is known about the presence of biocides in the water system and the risks associated with this presence, as well as how this knowledge can be used in developing monitoring approaches.

Emission pathways are diverse

By monitoring chemicals in the water, we know whether emissions are causing problems for water quality. But biocides are hardly monitored, possibly because little is known about the extent of use and emission pathways. Based on the nature of their application, the European Biocides Regulation distinguishes four main groups of biocidal products: disinfectants, preservatives, pest control agents and others. In total, the four groups comprise 22 product types [1]. These include antifouling agents for ships, preservatives in building materials, insect repellents, agents to prevent biofilm in cooling towers or algae in artificial turf.

As the applications of biocides are very diverse, so are the routes of spreading to water. Disinfectants often end up in water via the sewage treatment plant. This also applies to preservatives, but these can also end up in surface water directly from, for example, building materials or wood or in groundwater via the soil. By discharge of cooling water or the use of antifoulants, direct emission to surface water is possible. Pest control agents can reach the water through all these routes.

Model-based regulation

The active substances in biocidal products must have received European approval or be in the review programme. Subsequently, the biocides are tested before they can be sold on the Dutch market. This is done by the Dutch Board for the Authorisation of Plant Protection Products and Biocides (Ctgb). The review examines whether the product is sufficiently effective and has no significant adverse effect on humans and the environment. An authorised biocidal product has binding instructions for use. It states where and how it may be used, and how waste and treated material should be dealt with after use.

The assessment of biocidal products is based on estimations and models. If actual use deviates or the models do not describe all situations well, differences arise between assumed and actual emissions and environmental concentrations. Only by monitoring in the water system can these differences come to light. However, there are around 270 substances in use as biocides. It is impossible to measure them all everywhere.

Biocide measurements in the water system

Which biocides are currently being monitored in surface and groundwater? An analysis of monitoring data from the Water Quality Portal (data from 2019) shows that 18% of the authorised active substances are actually being measured. For groundwater (with fewer measurements in the Water Quality Portal and therefore analysed for the period 2000-2018), this percentage is 15%. Respectively 13% and 10% of the permitted substances are actually found (Figure 1).

Figure 1. Percentage of active substances in biocidal products monitored and detected in surface water (2019) and in groundwater (2000-2018) Source: Water quality portal.

At most groundwater locations, all measured substances were below the reporting limit. In surface water, biocides are widespread: they were found above the reporting limit at two-thirds of the sites. One third of the monitored substances were found (once or several times) above the detection limit, both in groundwater as well as in surface water [2]. (Signal value: when the concentration of a new (emerging) substance in intake water for drinking water production structurally exceeds 0.1 µg/l for 3 years, the health risks are investigated).

Drawing up measurement strategies for biocides

Which substances should be included in a monitoring programme? Ideally this decision is based on 1) the probability of their occurrence and 2) the impact they have on water quality. The probability of detection is determined by the substance properties, the selection of the monitoring sites and the timing of the monitoring. Looking at the current data, it appears that targeted monitoring for biocides is either lacking or not performed in the most relevant locations. Measured data on biocides are almost always by-catch of measurement programmes aimed at other substance groups, such as plant protection products [2]. In addition, many active substances in biocidal products also have other uses; therefore, it is unclear whether a substance found in a biocidal product originates from its use as a biocide. To be able to say something about the risks that substances pose to people and nature, more measurement data is needed.

Substance properties of biocidal products

In the KIWK project Ketenverkenner, the substance properties of biocides were collected [3] and linked to the monitoring data from the Water Quality Portal (Figure 2). Biocides that are poorly biodegradable appear to be over-represented in the measurements [4], while highly volatile biocides (with high vapour pressure; VP) are hardly observed at all. This can be explained by the fact that easily biodegradable and volatile biocides disappear quickly. Surprisingly, substances that strongly adsorb to soil or sediment (with a high sorption coefficient of organic carbon, Koc) are found in high concentrations in water. In this it’s certainly a factor that these substances are over-represented in environmental research because their isolation and analysis is easier [5].

Figure 2. The occurrence (frequency) of biocides with a certain substance property in the water system in relation to the ,i>total occurrence of biocides with those substance properties. Displayed properties: biodegradability, vapour pressure (log VP) and organic carbon sorption coefficient (log Koc).

Of the 116 biocides that, based on their properties, could have meaningful emissions to the water system, for 83 there is no known measurement method [2]. These substances are therefore not monitored, even if they are or may be relevant to water quality. In order to get a better picture, it is important to develop measuring methods and to actually start monitoring in a targeted manner.

Developing robust and sensitive measurement methods and integrating them into monitoring programmes is not always possible or easy. Therefore, in order to draw up an effective monitoring strategy, a good selection of substances to be monitored is necessary [6]. The Ketenverkenner project has developed a decision tree for this purpose (Figure 3) [4].

The decision tree distinguishes four steps for the selection. In step 1, the likelihood of exposure to (finding) a biocide is determined for the local water system. This is based on knowledge of use, application, substance properties (see box) and hydrology. In step 2, the potential impact on drinking water quality or ecological surface water quality is determined. Together these determine the risk of the substance: that is step 3. In step 4, for the relevant ('prioritised') substances, it is then examined whether a suitable measurement method is available (with a sufficiently low detection limit) or whether it should be developed. When a method is available (or developed), inventory measurements are taken at selected locations. Based on the results, it will be decided in step 5 whether the substances in question will be included in regular monitoring.

Figure 3: Decision tree for the inclusion or exclusion of a substance in a measurement strategy.

In order to maintain an adequate monitoring programme, it is essential to continue surveying. This is to avoid measuring only what we already know and keeping new substances out of sight - the so-called confirmation bias - and to test assumptions about exposure. If a substance is structurally not found, monitoring can be discontinued [7]. Without such continuous evaluation, the monitoring programme loses its effectiveness in the long run.

Thomas ter Laak
(KWR Water Research Institute)
Tessa Pronk
(KWR Water Research Institute)
Joep van den Broeke
(KWR Water Research Institute)
Joke Wezenbeek
Els Smit
Ivo Roessink
(Wageningen Environmental Research)
Sanne van den Berg
(Wageningen Environmental Research)
Bas Buddendorf
(Wageningen Environmental Research)


The applications of biocides are very diverse. There are therefore many chains with producers, users and related emission pathways into the water system. This study – part of the Ketenverkenner project of the Knowledge Impulse for Water Quality (KIWK) – shows what is known about the presence of biocides in the water system, but also, and above all, that much is unknown. Among other things, it appears that there is no targeted monitoring of biocides (in contrast to, for example, plant protection products). In order to draw up an efficient and effective monitoring strategy, a five-step decision tree is presented, based on the available knowledge. The first step is to estimate the exposure (probability of detection). The physical and chemical properties of the substance in question play an important role here, as the study also shows.


- Ctgb (2022), website consulted on 11-04-2022.

- Pronk, T.E., Wezenbeek, J., & Buddendorf, B. (2022). Biocides Excel Table (Version v2) [Data set]. Zenodo.

- Wezenbeek, J., Roessink, I., van den Berg, S., ter Laak, T., Pronk, T. (2021). Deltafact Biocides. STOWA:

- Pronk, T.E., Roessink, I., Smit, E. (2022). MEETING STRATEGY BIOCIDES Considerations and criteria, STOWA report no. 2022-07.

- Reemtsma, T.; Berger, U.; Arp, H. P. H.; Gallard, H.; Knepper, T. P.; Neumann, M.; Benito Quintana, J.; Voogt, P. (2016). Mind the gap: persistent and mobile organic compounds - water contaminants that slip though. Environmental Science & Technology 50(19), 10308 - 10315.

- Pohl, K., et al. (2015). Environmental monitoring of biocides in Europe - compartment-specific strategies. Workshop Report (June 25-26, 2015 in Berlin).

- von der Ohe, P. C. and V. Dulio (2013). NORMAN Prioritisation framework for emerging substances. ISBN: 978-2-9545254-0-2.

^ Back to start


Emission pathways and risks

Knowledge journal / Edition 1 / 2022

Removal of organic micro-pollutants with advanced oxidation and dune infiltration

Drinking water utility Dunea (The Hague and surroundings) produces drinking water from surface water by means of dune infiltration. The quality of the surface water will be under increasing pressure in the next decades due to an increase in organic micropollutants (OMPs), such as pharmaceuticals, X-ray contrast agents, pesticides and industrial compounds, combined with periodically low river runoffs. OMPs can have negative effects on health, but are difficult to remove completely. This is a serious concern for drinking water production. Does a combination of advanced oxidation and dune infiltration offer a solution?

Drinking water utility Dunea already demonstrated ten years ago that a wide range of OMPs is removed from river water when rapid sand filtration is followed by an advanced oxidation process (AOP) with hydrogen peroxide, ozone and UV. AOP ensures the conversion of organic material into smaller carbon chains, including more easily 'assimilable organic carbon' (AOC). A disadvantage of this, is that AOC is associated with increased bacterial growth in distribution networks. Another known disadvantage of AOP is the formation of undesirable by-products (such as nitrite and bromate) and transformation products when complete mineralization of OMPs does not occur. Naturally, Dunea wants to limit this. Among other things, the company uses dune infiltration in the dunes of South Holland for this purpose.
The hypothesis of this study is that a combination of AOP and dune infiltration offers advantages:
1. Because AOP breaks down organic compounds such as micropollutants (OMPs), fewer OMPs are infiltrated into the dunes. In addition, the AOC that is formed can stimulate natural degradation processes in the dunes.
2. The transformation products and by-products of AOP are removed by natural degradation processes during dune infiltration.
To investigate this hypothesis, Dunea started a large-scale study in 2018.

The purification process

The river water taken in at Bergambacht undergoes pre-purification by rapid sand filtration.
Afterwards, the infiltration water is transported to, among others, the Berkheide dune area (near Katwijk).
In this study, a side stream at Bergambacht was submitted to an extra pre-treatment using an advanced oxidation process (AOP) followed by activated carbon filtration. An in-house AOP method has been developed for this, 'GOBAM' (the Dutch acronym for BergAMbacht Advanced Oxidation), which works with a combination of hydrogen peroxide, ozone and UV (with low-pressure lamps).
Next, the main flow (rapid sand filtrate) is mixed with the GOBAM-treated water (30-40%) and, after transport, infiltrated into the dunes of Berkheide. First into infiltration ponds with short residence times, then into ponds with long residence times (see opening photo). The infiltrated water is recovered by horizontal drains and pumping wells at a certain distance, and is post-treated to produce drinking water (Figure 1).

Figure 1. Diagram of drinking water treatment by Dunea. Surface water is taken in (1) and pre-treated with rapid sand filtration (2) and then partially (30-40%) with advanced oxidation (3). This mixed, pre-treated water is transported to the dune infiltration system (4) and infiltrated (5), after which water is extracted (6) for drinking water production (7).

Sampling and measurements

The effects of this approach on the water quality in the dune infiltration system and on the final drinking water were investigated. The baseline measurements (2016-2017, no GOBAM) and the effect measurements (2018-2020, with GOBAM) used the same sampling points, distributed over the entire treatment process. Here the hydrological, microbiological, organic and inorganic chemical and toxicological effects were monitored. The following sampling points were used:
- before and after pre-treatment with rapid sand filtration (and AOP in 2018-2020),
- in the transportation pipeline (just before arriving at the dunes),
- in the short-residence time infiltration ponds,
- in the long-residence time ponds,
- after approximately one meter of dune infiltration (in a monitoring well in each of the ponds),
- in the raw water collected from the entire Berkheide infiltration area,
- and in the post-purified drinking water.

The effect of GOBAM was assessed on the conversion of OMPs into polar transformation products and the formation of by-products (including bromate and nitrite) and how these behaved during dune infiltration. For this purpose, changes in travelling times, nutrients, redox-sensitive components, heavy metals, OMP transformation products, microbiology and toxicology were monitored during dune infiltration and in the final drinking water.

Hydrological and hydrogeochemical conditions

First, it was determined whether the dunes in the periods with and without GOBAM were comparable in their potential to remove OMPs. For this, the redox conditions and the travel times above and below ground are important parameters. At all sampling points, the travel time was determined from peaks in the (semi)natural tracers chloride, electrical conductivity and temperature. In order to be able to detect any shifts in redox conditions, the redox-sensitive main components O2, NO3-, SO43-, Fe, Mn and NH4+ were analyzed regularly. The differences in these parameters between the two periods (with and without GOBAM) were found to be negligible. Changes in water quality between the two periods can therefore be attributed to other external factors, such as the quality of the intake water and seasonal effects, or to the effects of GOBAM.

OMP removal

The removal of OMPs by GOBAM was, as expected, very effective. Before the GOBAM step, 48 OMPs were found to be above the reporting limit at least once. Of these, 31 were more than 80% removed, and another six showed a reduction of 50-80%. GOBAM thus had a net positive effect on the quality of the water to be infiltrated. However, after dune infiltration of the 37 (partially) removed substances, three compounds were still found incidentally and six substances regularly above the reporting limits. The latter were amidotrizoic acid and iopamidol (x-ray contrast agents), EDTA and urotropine (industrial compounds), acesulfame (sweetener) and primidone (pharmaceutical). This was partly due to a delay in the effect of GOBAM due to long residence times, and because only part of the water was treated with GOBAM. In the period shortly after the introduction of GOBAM, water that had not undergone GOBAM was extracted. However, after a longer monitoring period, the elimination in 2020 for the X-ray contrast agents and urotropine had improved significantly (Figure 2).

Figure 2. Concentrations (µg/L) of amidotrizoic acid, iopamidol, and urotropine over 5 years, in intake water (IN; black dots), infiltration water (INF; open dots) and drinking water (DW; blue dots). The vertical line separates the periods before and after the introduction of GOBAM. The reporting limit is shown as a grey bar.

Effects of AOP on dune infiltration area

Elevated concentrations of AOC were found in the effluent of the AOP (GOBAM). In the transportation pipeline from Bergambacht to the dunes, these decreased to the same level as during the baseline measurement, probably due to (limited) microbial degradation and mixing with untreated rapid sand filtrate. Microbial growth also has not so far led to an increase in pipe resistance. Thus, the AOC formed by AOP decreased during transport and did not reach the first infiltration ponds, thus not stimulating the activity of microorganisms in the dunes. If it is decided to increase the GOBAM side stream and AOC ends up in the dunes, this effect can be further investigated.

In AOP, various (mostly unwanted) by-products are formed. In the first infiltration ponds, slightly elevated concentrations of bromate, bromoform and nitrite were measured. In one of those ponds, there was minor mutagenic activity. This was no longer measurable in infiltration ponds located further away or during the subsequent dune infiltration. The drinking water showed no differences between the baseline measurement and the effect measurement for the above by-products.

A parallel ecological study in the same infiltration area also showed that implementation of AOP had no effect on WFD classification (Penders et al).


These results show that AOP and dune infiltration complement each other well as a purification process. AOP is therefore suitable to improve OMP removal without negative effects on drinking water production or the environment (ecology). The direct result of GOBAM was a reduced OMP load in the dunes and a faster decrease of the concentrations in drinking water. It is expected that the use of GOBAM will, in the long term, result in a further decrease in the concentrations of the aforementioned OMPs in drinking water. These beneficial effects may be applicable in more infiltration areas and in other underground and/or biological treatment processes. Follow-up research in which AOP would be applied to 100% of the infiltration water could provide even more insight into the effects of AOP on dune infiltration (the organic and inorganic chemistry, toxicology, biological stability and activity, and the microbial populations) and its applicability.

The extensive results of this research are described by Timmers et al. As a follow-up to this project, a geochemical and microbiological measurement program and modelling of the behavior of OMPs in the infiltration ponds and underground at Berkheide was carried out, in order to better understand and predict the removal mechanisms of priority compounds in the subsurface (Van de Grift et al. and Stuyfzand).

Peer Timmers
(KWR Water Research)
Astrid Reus
(KWR Water Research)
Tineke Slootweg
(Het Waterlaboratorium)
Aleksandra Knezev
(Het Waterlaboratorium)
Jamal El Majjaoui
Karin Lekkerkerker-Teunissen
Pieter Stuyfzand
(KWR, Stuyfzand Hydroconsult)

Background picture:
Berkheide infiltration area. After being transported via a pipeline, pre-treated water from the Meuse flows into the infiltration ponds. The infiltrated water is then recovered and treated to produce drinking water.


Drinking water utility Dunea produces drinking water from surface water by means of dune infiltration. The increase of organic micropollutants (OMPs) in surface water is a serious concern for drinking water quality.

This article shows that a combination of advanced oxidation and dune infiltration is suitable to improve the removal of OMPs, without negative effects on drinking water production or on the infiltration area.


- Timmers et al., 2021. ‘Effects of advanced oxidation on water quality and microbiology in dune infiltration', KWR report 2021.047.

- Timmers et al., 2022, Improved drinking water quality after adding advanced oxidation for organic micropollutant removal to pretreatment of river water undergoing dune infiltration near The Hague, Netherlands, Journal of Hazardous Materials, online 26 January 2022, 128346,

- Van de Grift B., Timmers P.H.A., Stuyfzand P. 2022. Geochemical and microbiological measurements around ponds 38 and 40-2 at Berkheide, KWR report 2022.026.

- Stuyfzand P.J. 2021. Modelling of organic micropollutants, with effects of AOP and silt, in Berkheide dune infiltration system. Report Stuyfzand Hydroconsult+, SH+2021.008, 53p.

- Penders E.J.M. Water Framework Directive evaluation of AOP (In Dutch: KaderRichtlijnWater beoordeling GOBAM studie), project report 202104, Het Waterlaboratorium.

^ Back to start


Oxidation and dune infiltration

Knowledge journal / Edition 1 / 2022

PAC + cloth filtration: far-reaching removal of pharmaceuticals and phosphorus from waste water is possible

The effort required to meet the requirements of the Water Framework Directive (e.g. far-reaching phosphorus removal) and the increasing focus on the removal of pharmaceuticals call for technologies that serve both purposes. The PAC + cloth filtration technology offers this solution. At the Vinkel waste water treatment plant, this technology was investigated for the first time in the Netherlands. Within the STOWA (Foundation for Applied Water Research) Innovation Programme Micropollutant Removal from Wastewater, a pilot study was successfully carried out under the supervision of six water authorities.

The removal of micropollutants such as pharmaceuticals from wastewater is a hot topic that many water authoritiesare working on. A promising technology is the combination of powdered activated carbon (PAC) with cloth filtration. This PAC + cloth filtration technology has already been successfully applied in some wastewater treatment plants in Germany. Powdered activated carbon, metallic salt and polyelectrolyte (PE) are added to the effluent. This binds pharmaceuticals and phosphorus. Sedimentation and cloth filtration remove the bound compounds from the water.
Based on the experience in Germany, Royal HaskoningDHV carried out a feasibility study into the technology within the STOWA Innovation Programme Micropollutant Removal from Wastewater (IPMV) in 2018 (STOWA 2020-21). This showed that the technology is also promising for Dutch wastewater practice. In the IPMV it was therefore decided to give the green light for further research on a pilot scale in the Netherlands. The pilot study was carried out by the Aa and Meuse water authority, ELIQUO and Royal HaskoningDHV under the supervision of STOWA, the Ministry of Infrastructure and Water Management and six water authorities that financed the pilot study.

The technology

The technology focuses on the further treatment of the WWTP effluent, at the end of the active sludge process. PAC (5-15 mg/L) and metal salt (3-10 mol Fe/mol P) are added in the post-treatment PAC + cloth filtration step. Pharmaceuticals bind to the PAC and phosphate binds to the metallic salt. The sludge that is formed is separated from the water by means of a settling and cloth filtration step, which requires the addition of a small amount of PE (0.3 - 0.5 mg/L). The clean water can then be discharged; the sludge must be disposed via the regular sludge disposal or as a separate waste stream.
Similar to other activated carbon technologies, pharmaceuticals are removed from wastewater by adsorption using activated carbon. By varying the PAC dosage, the type of PAC, the PAC concentration in the contact tank and the PAC residence time in the system, the removal efficiency was adjusted and the process optimised. Metal salt is dosed as a coagulant to bind phosphate and for the coagulation of PAC. For the latter, the dosage of PE is additionally required.
Separation of the sludge (mixture of PAC, metal salt and PE) and water takes place in a two-stage separation configuration. In the first step, the bulk of the sludge is settled in a lamella separator by gravitational separation. The settled sludge is largely returned to the contact tank. A small part of the settled sludge is discharged as surplus. In the second step, the remaining sludge, which is still present in the sedimentation step, is separated by cloth filtration. The cloth filter is cleaned periodically to remove the captured sludge. This flow is returned to the contact tank or discharged as surplus sludge.

Design of the pilot study

The pilot study was carried out between March and August 2021 at the Vinkel WWTP of the Aa and Meuse water authority. The pilot plant was not a one-on-one copy of the technology already proven in Germany. An alternative configuration was chosen with a lamella separator instead of a traditional sedimentation tank with the aim of a smaller space requirement, more flexibility in the design and realisation and possibly also a cost saving.
It is known from Germany that the configuration with a traditional sedimentation tank is capable of both good removal of pharmaceuticals (approx. 80%) as well as a low total phosphorus concentration (approx. 0.1 mg/L). However, this has not been investigated in the Netherlands before. The central research question of the pilot study was therefore: Is it possible to use a single technology on a Dutch WWTP to remove both pharmaceuticals and phosphorus to a large extent using lamella separation as the sedimentation step? A second research question was whether the technology works without dosing PE.
In the pilot plant, 1 to 5 m3/hour of WWTP effluent was treated, which is only a small portion of the total effluent of the Vinkel WWTP with an average flow of 600 m3/hr. The installation was set up outside and consisted of the process steps: 1) contact tank, 2) lamella separator and 3) cloth filter (Figure 2). During the first months, a stable operating mode was sought in which a good effluent quality was achieved (concentration of total suspended solids <10 mg/L). In particular, the influence of the sludge concentration in the contact tank on the settling and filtration process and the influence of PAC, metal salt and PE on this process were investigated. The main outcome is that dosage of PE is needed. Without PE, the settling of the sludge mixture of PAC and metal salt is moderate, which hinders proper separation. By dosing a small amount of PE (0.3 - 0.5 mg/L), the sedimentation rate increases significantly, resulting in a very high separation efficiency of >99.9%. With this knowledge applied, the process was run for 2 months without making any major changes to the pilot settings. In this stable period, the most important research results have been obtained.

Figure 2: Schematic diagram of the pilot plant

Results of the pilot study

The investigated configuration of the PAC + cloth filtration technology worked well: good sludge separation took place, pharmaceuticals were well removed and low concentrations of total phosphorus were achieved. In the lamella separator the bulk (>99%) of the sludge was separated, in the cloth filter the last sludge particles were removed (to <3 NTU).
The results with regard to the improvement of water quality (pharmaceuticals and phosphorus) are very positive. Far-reaching removal of pharmaceuticals has been achieved with a relatively low PAC dosage. At PAC doses of 5, 10 and 15 mg/L, removal efficiencies of 67, 92 and 95%, respectively, were achieved on average for 7 of the 11 guide substances (see Figure 3).
The national target for pharmaceutical removal is a 70% removal efficiency of 7 of the 11 guidance substances over the entire WWTP (STOWA IPMV and Ministry of I&W contribution regulation). Taking into account the removal efficiency of the activated sludge process itself (approx. 30%) and the capacity of a post-treatment technology that is usually 1.5 times the dry weather flow (by-pass for storm weather flow), a PAC dosage of 8 mg/L is sufficient. Compared to the best available PAC technology, Powdered Activated Carbon in Activated Sludge (PACAS), which doses approximately 15 mg PAC/L, the PAC dosage of the PAC + cloth filtration and the associated CO2 footprint of PAC + cloth filtration is almost halved.
In addition to the removal of pharmaceuticals, the reduction of ecotoxicity was also investigated. A battery of biological effect measurements with in vivo and in vitro bioassays (various CALUX tests, the Daphnia Immobilisation test and the Microtox test) has shown that the effects measured in the various tests have decreased by more than 50%, and have dropped below alert values (i.e. after PAC + cloth filtration technology the water is no longer toxic).

Figure 3: Average removal efficiency of pharmaceuticals over the 6 measuring days (date) at different PAC dosages (mg/L).


Phosphorus has also been successfully removed to a large extent. Over the period of two months of stable operation, the concentration of total phosphorus in the wastewater effluent was reduced from an average of 0.96 mg/L to 0.18 mg/L. During a two-week period of very stable operation, an average total phosphorus concentration of less than 0.1 mg/L was achieved, with sampling days of less than 0.05 mg/L on individual measuring days. With these low concentrations of total phosphorus, the WFD requirements for total phosphorus in surface water of 0.15 mg/L are met. The metal salt dosage required to achieve this translated into a ratio of 4 to 9 moles of iron per mole of phosphorus (molar ratio Fe:P).
Zoomed in into the different fractions of which total phosphorus is derived, the largest removal is visible in the ortho-phosphate fraction. This fraction, with an average concentration of 0.81 mg/L, accounts for approximately 85% of the incoming phosphorus and was largely removed to an average concentration of 0.05 mg/L. It is noteworthy that the organicdissolved phosphorus fraction, which is usually very difficult to remove, was also largely removed.
The concentration in the wastewater effluent of this fraction of 0.09 mg/L was reduced to 0.02 mg/L on average. This removal is thought to result from the presence of PAC to which the organic dissolved phosphorus binds. The main phosphorus fraction in the pilot plant's run-off is the metal-bound phosphorus. These are (very) fine particles that did not coagulate into larger flocs and therefore pass the settling and filtration step. Optimisation of the coagulation process can further reduce these emissions on a practical scale, so that (even) less metal-bound phosphorus is released. A more detailed description of the pilot tests (set-up, implementation and results) is currently being set out in a STOWA report. It is expected to be published at the end of Q2 2022.

Relevance to practice

The results of the pilot study are positive; a high level of removal of pharmaceuticals and a low concentration of phosphorus were achieved. The pilot scheme showed that with the PAC + cloth filtration technology it is possible to remove pharmaceuticals and phosphorus from wastewater in a single treatment step. The results of the pilot study therefore pave the way for application of the technology on a practical scale. The performance of the technology and the points of attention for its operation and management have been made clear by the pilot study. This allows us to move on to full-scale realisation. Application on a practical scale fits in seamlessly with the current tasks in the water cycle. The far-reaching removal of phosphorous (WFD requirement) and pharmaceuticals (Ministry of I&W incentive scheme) are high on the agenda of many water authorities.

Karin Bertens Zorzano
(Aa and Meuse Water Authority)
Devon Dekkers
(Aa and Meuse Water Authority)
Bart Verberkt
(Aa and Meuse Water Authority)
Tonke van der Pol
Xian Riedijk
(Royal HaskoningDHV)
Arnoud de Wilt
(Royal HaskoningDHV)

Background picture:
Figure 1. Set-up of the PAC + cloth filtration pilot plant at Vinkel WWTP


A technology that is new to the Netherlands for the far-reaching removal of micropollutants (pharmaceuticals, etc.) and phosphorus from wastewater effluent has been studied on a pilot scale: PAC + cloth filtration. In this technology, powdered activated carbon (PAC) is used for the absorption of pharmaceuticals. Phosphorus is removed with a metal salt. The technology was tested at a scale of 5 m3/hr during a six-month pilot study at the Vinkel WWTP. The results of the pilot study are very positive. Pharmaceuticals have been removed by 95%. Phosphorus has been removed to concentrations less than 0.05 mg/L. Compared to the currently best available technology, PAC consumption is almost halved. This greatly reduces the CO2 footprint.

Reference list

– STOWA 2020-21 Feasibility study PAC + cloth filtration

^ Back to start


Pharmaceuticals and phosphorus

Knowledge journal / Edition 1 / 2022

Innovative monitoring and modelling supports spatially targeted approach to reduce nutrient losses from agricultural fields

In order to reduce nutrient losses from agricultural fields to ground- and surfacewater, the low-hanging fruit has long ago been picked. To support an effective spatially targeted approach, an innovative, intensive monitoring method was set up in two areas: the catchment areas of the Vuursteentocht (Flevopolder, clay soil) and of the Vinkenloop (North Brabant, sandy soil). What new knowledge does this provide and what action perspectives does it open up for a further spatially targeted approach to nutrient losses?

In order to comply with the Nitrates Directive and WFD targets for ground and surface water, inputs of nitrogen and phosphorus to groundwater and surface water must be further reduced. In addition to generic rules for manure policy, the national government is committed to a spatially targeted approach. Many measures can be used for this purpose, but which ones and where are they applicable and effective?

Area-based approach

To find effective measures for this approach, the research programme of the Water Quality Knowledge Impulse set up an intensive and innovative monitoring network in two catchments (Vinkenloop and Vuursteentocht, 7 and 13 km2). The monitoring network is intended to provide a better understanding of the sources and routes of nutrient flows. Where are hotspots, and when and why do hot moments occur? How does the soil and water system function, how are nutrient losses related to weather, fertiliser application and land use? The study is also intended to demonstrate the added value of combining different measurement techniques.
The monitoring has been operational since the end of 2020. This was followed by a year and a half of measuring and subsequent modelling.

Plot and catchment scale monitoring

Different measuring techniques have been combined to monitor at both catchment and plot scale. Sensors were deployed at various locations to detect hot moments and hotspots of nitrogen and phosphorus flows. In addition, manual measurements were taken to identify spatial patterns, with underlying sources and processes. The table gives an overview of the monitoring techniques and of the points for attention and results that they produced.

Table 1. Evaluation of monitoring results at catchment area level


With a routing, the concentrations and other parameters in the ditches and larger watercourses are measured on one day, so that a spatial picture emerges. Use is made of sensors and a nitrate app attached to a canoe or fishing rod. In both areas (Vinkenloop and Vuursteentocht), the temperature, conductivity and nitrate concentrations were measured at five different times. This provided a relatively cheap picture of the main routes and hotspots for nitrate leaching in the areas.
In the Vinkenloop, it was repeatedly found that the highest nitrate concentrations (approx. 30 mg/l NO3-N) came from a lily field; in spring this was even the only source of nitrate in the catchment area. Around the Vuursteentocht, nitrate was transported to the canals via the drains and ditches at concentrations of around 8 mg/l NO3-N, with some peaks of around 20 mg/l NO3-N. No nitrate was found in surface run-off, but in the leachate from uncovered manure piles the concentration was above 50 mg/l NO3-N.
The temperature measurements from the routings show which ditches are fed mainly from the shallow groundwater (relatively cold) and which ditches are fed more from the deep groundwater (relatively warm). See Figure 1 for the situation in the Vinkenloop area.

Figure 1. Results of routings February 2022. On the left the temperature (difference from reference (˚C) and on the right nitrate concentrations. This provides a picture of where relatively high levels of seepage occur (left, orange-coloured ditches and streams) and areas where high nitrate leaching occurs (right).

Depth profiles of nitrate concentrations

Sampling from two crop plots in the Vinkenloop (lilies and chicory, Figure 2) shows that the top metre of groundwater at both locations is high in nitrate, and that concentrations decrease with depth. At one measuring point, the groundwater is nitrate-free from a depth of 5 metres, whereas at another measuring point it is nitrate-free from a depth of 3-4 metres. The differences between the measuring points are caused both by the type of cultivation and by differences in denitrification (conversion of nitrate into harmless nitrogen gas) due to the presence of organic matter. The dampening effect of denitrification on the leaching of nitrate is also visible in relatively low nitrate concentrations in the drains.
The depth profiles also indicate that the ammonium concentration in the deeper groundwater is not directly related to fertilisation. It is likely that this ammonium is released through mineralisation of organic matter or through conversion of nitrate during soil passages. Isotope analyses in groundwater and surface water indicate that discharge of deep groundwater containing ammonium is largely responsible for the background concentration of ammonium in the Vinkenloop. Continuous measurements of ammonium at the outflow point confirm this picture.

Figure 2. Nitrate depth profiles at different times under the lily plot and the adjacent chicory plot. The nitrate standard of 50 mg/L is shown with a vertical dotted line.

Data-driven modelling

Various data-driven modelling techniques were used to process and interpret the measurement data, for example to check measurement data from high-frequency monitoring and to fill in gaps in measurement series. The relationships found in these models between hydrology and water quality are also useful for extrapolating measurement series to other weather years, and for roughly estimating the effect on concentrations at the outflow point when dust loads change in the emission pathways. The results of these data-driven models can also be used for input and validation of dynamic process models.

Process models are needed to quantify the origin of nutrients in water leaching from fields and the effects of agricultural measures. For both pilot areas, a first version of detailed process modelling was set up with the calculation codes of SWAP and ANIMO. These simulate the crop uptake of moisture and nutrients and the moisture and nutrient balance in the soils, including run-off into groundwater and surface water. The concept of the process model, after reducing the maximum flow depth and lowering shallow resistances, fits well with the measurements.


The results of the measurements, together with the different models, provide a clear picture of
• how the soil and water system functions,
• how the nutrient fluxes react to weather (precipitation), manure application and land use,
• where hotspots are and when and why hot moments occur.

The insights gained confirm the expectations from previous research for the catchment and field level, and provide experimental justification and quantification.

The monitoring also provides new insights. The most important are listed below:
• Nitrate leaching occurs more quickly after a rainy period than expected from older modelling studies.
• Ammonium in groundwater is a determining factor for background concentrations in surface waters in both pilot areas.
• In the Vuursteentocht pilot area, seepage water only drains to the deeper canals and not to the drains and ditches;
• The routes to the drains and ditches (the groundwater flow patterns) are shallower than always assumed in previous modelling studies.
• In the Vinkenloop, a transition from oxidised groundwater with high nitrate concentrations to anoxic groundwater without nitrate was observed at relatively shallow depths. Water flowing through the reduced zone to drains and ditches has a lower nitrate concentration than water flowing shallowly to the drains.
This provides guidance for finding applicable and effective measures.

The combination of measurement and modelling provides better information than the mere collection of measurement data or the mere application of models. With models, measurement data can be better interpreted and, conversely, the measurements contribute to greater model reliability. As it is not feasible to set up such an intensive area-based monitoring in all problem areas, building blocks from the Delta Fact 'Guide to area-based monitoring of nutrient losses from agriculture' may help to determine which monitoring strategy is appropriate for the area.
It is strongly recommended to continue monitoring in both pilot areas over a long period of time, preferably permanently. It will then remain an ideal testing ground for new measuring techniques, for the testing of models and for deriving effects of agricultural measures in a real field situation.

Peter Schipper
(Wageningen Environmental Research)
Piet Groenendijk
(Wageningen Environmental Research)
Saskia Lukacs
(National Institute for Public Health and the Environment (RIVM))
Arnaut van Loon
(KWR Water Research Institute)
Joachim Rozemeijer

Background picture:
Vinkenloop outflow point: trailer with measuring equipment (temp, EC, TotP, NH4, NO3)


Dutch fertilizer policy aims to improve water quality by reducing (especially in agricultural areas) the run-off of nitrogen and phosphorus into ground and surface water. In addition to generic rules, the national government is committed to an area-specific approach. An intensive and innovative monitoring network has been set up in two catchment areas (one on sandy soils, one on clay) to gain a better understanding of the sources and routes of nutrient flows. Where are the hotspots, and when and why do hot moments occur? The combination of measurement and modelling provides more detailed and reliable information than the mere collection of measurement data or the mere application of models. The study shows that this new approach provides guidance for an area-specific approach to nutrient emissions in agriculture.


- Knoben, R.N. Evers, A. Jacks, J. Rost, N. Schoffelen, M. de Haan, B. van Spronsen, F.L. Verhagen, H. Evenblij and B. van Velthoven, 2021. ‘Ex Ante Analysis of Water Quality' . Report Royal Haskoning DHV 28-9-2021.

- Introductory dossier for members of the Ministry of Agriculture, Nature and Food Quality (2022),

- Parliamentary letter 7th action programme Nitrate Directive, Parliamentary paper 26-11-21,

- Schipper, P. Groenendijk, L. van Gerven, A. van Loon, S. Lukács, J. Rozemeijer 2022. Monitoring and modelling in two pilot areas for area-based approaches'. STOWA (KIWK) report 2022-22.

^ Back to start


Intensive monitoring method

Knowledge journal / Edition 1 / 2022

Predicting the formation of transformation products during drinking water treatments and their potential toxicity

Water treatment processes are essential for drinking water production but may produce unintended transformation products. Identifying these products and their potential toxicity (hazards) is necessary to guarantee safe drinking water. Predictive toxicology can speed up and improve the assessment of these substances.

Drinking water production involves a series of processes to remove pathogens and other contaminants from the water. Chemical contaminants in drinking water sources, especially surface waters, include substances produced by humans such as pharmaceuticals, plant protection products, and biocides. These substances can be transformed (WHO, 2022) during drinking water production. Transformation product (TPs) formation is, to some extent, unavoidable (Anagnostopoulou, 2022), and the physicochemical and toxicological properties of most TPs are unknown (Gassman, 2021). Analytical techniques can identify TPs, but these techniques are expensive and time-consuming. When identified, concentrations of individual TPs and other chemical contaminants are usually below health-related thresholds. Since exposure to TPs generally occurs in the form of low-level mixtures and may last a lifetime, it is vital that we learn more about TPs' formation and toxicological activities. Predictive toxicology can help in this respect.

New horizons for the assessment of TPs

Predictive toxicology is an innovative approach that uses computerized (in silico) tools, that have shown a promising ability to predict the properties of substances (Raies, 2016). in silico tools are based on algorithms that can predict the formation and toxicity (hazard) of compounds based on their chemical structure. They can contribute to filling data gaps on contaminants, such as TPs, that may be present in drinking water at low concentrations. in silico approaches provide a relevant and cost-effective contribution to the exposure and risk assessment for substances that lack toxicological data, including TPs. In this way, predictive toxicology can direct water quality monitoring, prioritize further experiments and research needs, and reduce the costs and time needed. This article gives an overview of the applicability of in silico tools to the evaluation of TPs in drinking water.

Formation of TPs

The first step in assessing TPs in drinking water using in silico tools is identifying which compounds are expected to be formed. The composition of the mixture of TPs that can occur in drinking water depends on chemical and biological reactions related to the water treatments. Transformations in the environment and organisms, including humans, are not considered here.
In Europe, the most common water treatment processes are chlorination, ozonation, UV treatment, rapid sand filtration, filtration over biologically active carbon, and advanced oxidation. Different combinations of these techniques are used to remove pathogens and other contaminants, depending on the quality of the water sources (WHO, 2022).
Based on knowledge from experimental data (reaction libraries), it is possible to predict TPs potentially formed due to the water treatments. This provides a starting point for further chemical analysis, which can save costs and time when assessing water quality. Simulation by computerized methods can thus be a valuable addition to data on TPs actually found in water. Additional drinking water analyses can validate the predictions and verify the actual presence of TPs.

Toxicity of TPs

The second step is the assessment of the toxicity of TPs (hazard assessment). The toxicity of substances can be assessed not only in vivo (animal experiments) and in vitro (biochemical and cell testing), but also in silico (computerized tools). All these approaches have their fundamental limitations that result in uncertainties in the outcome.
in vivo experiments, by definition, inherently consider the adsorption, distribution, metabolism, and excretion (ADME) of substances in an organism but are generally expensive and time-consuming. Their use should be reduced, replaced, or refined from an ethical point of view.
In vitro experiments examine specific interactions of a substance with biological structures, which may, for example, lead to DNA damage. ADME aspects are only partially addressed, leading to uncertainties in translating the results into effects in humans.
In silico tools work with algorithms that predict toxicity based on chemical structure. They are time and cost-effective, but they are based on and strictly linked to the availability and quality of experimental data. Predictive toxicology only gives reliable predictions of chemicals’ in vitro and in vivo responses if the algorithms are based on high-quality data. A specific in silico model is reliably applicable only for substances that are comparable to experimentally analyzed molecules whose data were used to build the algorithm. In addition, the quality of the prediction depends on the specific toxicological effect (endpoint) evaluated (Figure 1).
The most relevant endpoints for assessing drinking water quality include mutagenicity/genotoxicity (DNA damage), carcinogenicity (tumor formation), reproductive toxicity, developmental toxicity, and endocrine disruption (disturbance of the hormone balance). The better the mechanism of action responsible for the toxicity is understood, and the higher the quality of the available data, the more reliable the in silico models will be. The state of knowledge is now such that in silico tools guarantee good predictions for assessing mutagenicity/genotoxicity. The challenge is significantly greater for endpoints with less clear mechanisms of action, primarily due to the scarcity of experimental data and differences between experimental protocols. This is especially true for complex endpoints such as carcinogenicity, and reproductive and developmental toxicology. In these cases, predictive toxicology is less suitable for assessing drinking water quality. Nevertheless, in silico tools can also contribute to understanding specific effects for these endpoints and can play a role in an integrated approach with other information sources.

Figure 1. Different strategies for the hazard assessment, their relations, and advantages (+) and disadvantages (-).

In silico toxicity assessment of TPs

In silico approaches for toxicity assessment are based on (but not limited at) (Figure 1):
(a) the quantitative structure-activity relationship (QSAR): the recognition of chemical substructures that are predictive of specific toxicological effects (endpoints);
b) the 'read-across' approach: the extrapolation of available information on toxicologically known substances to chemicals with a similar structure for which no data are available;
(c) 'expert judgment': essential for assessing and adequately interpreting the reliability of the predictions.

Even though most software is user-friendly, the output must be critically evaluated to detect outliers, inconsistencies, or errors in the model. The prediction should be backed up by reasoning and mechanistic interpretations as much as possible to increase the reliability of the results.
Several in silico tools for the prediction and toxicity assessment of chemicals are freely available and there are others that require purchase (Figure 2)
The potential of these techniques is increasingly recognized. Authorities such as the European Food Safety Authority (EFSA), the European Chemicals Agency (ECHA), and the Organization for Economic Cooperation and Development (OECD) suggest the use of these techniques to replace and complement animal testing. Nevertheless, authorities have not yet agreed upon an internationally recognized methodology for applying in silico tools specifically for the evaluation of TPs.

Figure 2. in silico approaches and examples of in silico tools for the prediction of TPs formed during different treatment processes, and specific hazards. 1. Chlorination; 2. Ozonation; 3. UV treatments; 4. Biodegradation; 5. Oxidation processes; 6. Genotoxicity; 7. Carcinogenicity; 8. Reproductive and developmental toxicology; 9. Endocrine disruption.

An integrated approach

Suppose that one scientific source of information does not provide sufficient answers about the toxicity of TPs. In this case, an integrated approach that considers all available evidence may be the solution. On one hand, an (internal) statistical validation of each methodology separately is needed to justify their integration into the approach and, on the other hand, an (external) comparison between different methods to assess the significance and concordance of the results has to be conducted.
In a recent study, Hensen et al. (2020) propose a tiered approach to evaluate TPs' toxicological effects derived from pesticides. Their first step was a combination of literature review on experimental data and in silico methods, followed by targeted in vitro and in vivo experiments to verify the previous results. This study showed that most TPs (94%) were correctly predicted. Despite the researchers' warning that this methodology needs further evaluation and development, this is a promising result.


This article was produced thanks to research by the Dutch and Flemish drinking water companies (BTO). The authors thank Remi Hoencamp for his contribution.

Ferrario A.S.
(KWR Water Research Institute, Utrecht University)
M.M.L. Dingemans
(KWR Water Research Institute, Utrecht University)
Reus A.
Hofman-Caris R.
(KWR, Wageningen University)


Water treatment processes are essential for drinking water production but may inadvertently produce transformation products (TPs). Predictive toxicology with algorithms (in silico tools) can speed up and improve the identification and assessment of these substances. in silico tools are practical and user-friendly and are a valuable alternative or addition to existing experimental methods. They can contribute to evaluating water treatment processes with models for the formation and toxicity of chemicals, including TPs. Therefore, predictive toxicology is a promising and cost-effective tool for assessing contaminants in drinking water.


- Anagnostopoulou, K., Nannou, C., Evgenidou, E., & Lambropoulou, D. (2022). Overarching issues on relevant pesticide transformation products in the aquatic environment: A review. Science of the Total Environment. Elsevier B.V.

- Gassmann, M. (2021). Modelling the Fate of Pesticide Transformation Products From Plot to Catchment Scale—State of Knowledge and Future Challenges. Frontiers in Environmental Science. Frontiers Media S.A.

- Hensen, B., Olsson, O., & Kümmerer, K. (2020). A strategy for an initial assessment of the ecotoxicological effects of transformation products of pesticides in aquatic systems following a tiered approach. Environment International, 137.

- Raies, A. B., & Bajic, V. B. (2016). in silico toxicology: computational methods for the prediction of chemical toxicity. Wiley Interdisciplinary Reviews: Computational Molecular Science, 147–172.

- WHO. (2022). Guidelines for drinking-water quality, 4th edition, incorporating the 1st and 2nd addendum. Geneva: World Health Organization. Geneva, Switzerland. World Health Organization

^ Back to start


Predictive toxicology

Knowledge journal / Edition 1 / 2022

Which restoration measures are ecologically effective?

Since the implementation of the Water Framework Directive, a large number of restoration measures have been carried out in Dutch waters to improve the ecological water quality. To what extent are these measures effective? How can we increase its success? And how can we evaluate its effects more easily? Looking for answers in available studies on the effects of restoration.

The ecological water quality in many waters is not yet at the desired level. Many restoration projects have already been carried out to enhance the status of these waters. In order to evaluate the effectiveness of the applied measures, information on the approach and implementation of the restoration projects over time is needed, as well as biological monitoring information. The Knowledge Impulse for Water Quality, Ecology section, examined what lessons can be learned about the design and monitoring of restoration projects. For 261 reports on restoration projects in the Netherlands over the period 2006-2019 the restoration outcome of measure-effect-combinations is evaluated[1].

Effectiveness measures based on evaluated reports on restoration projects

For standing waters, the majority of reports describe measures reducing the nutrient load and nature-friendly bank construction. For running waters, channel reprofiling and remeandering are common measures.

In terms of effects recorded, macroinvertebrates is the most well-studied organism group, followed by fish and aquatic plants.
Despite the large number of studies evaluated, insight into the effectiveness of the measures is limited. For each measure, contrasting effects were reported, and remarkably often no conclusion was drawn (Figure 1). However, for almost every measure there is at least one project where a positive impact has been reported. In addition, the difference in effects between projects indicates a strong context-dependence of the outcomes. Factors such as: what exactly has been done, on what spatial scale, and which other stressors play a role at a site, determine the final outcome.

Figure 1: Evaluation of the reported effects of restoration measures in projects in the Netherlands for standing (left panel) and running (right panel) waters [1].

The inability to identify consistent effects has two main causes. Firstly, the monitoring design of many of the evaluated studies is not suitable for evaluation of the effectiveness, and secondly, statistical testing of the monitoring data is often lacking. This hinders a a well-founded effectiveness assessment.

Monitoring design
The international standard for evaluating restoration measures is the Before-After, Control-Impact (BACI) monitoring design [2]. This means measurements are needed before (before) and after (after) implementation of the measures , both at the restored location (impact site) and at a control location where no measures have been taken (control site). This approach was applied in only 9% of the cases. Incomplete BACI studies, i.e. with only a comparison of the impact location with a control (CI, 29%) or only a time series at the impact location (BA, 20%) represent a larger proportion of the cases.

Statistical testing
Reported effects were often based only on visual interpretations of, for example, graphs, while only 25% of the cases were statistically tested. None of the studies met the statistical requirements for a solid BACI study: at least three years of data before and at least four years of data after implementation of restoration measures, with consistent monitoring over the years. Figure 2 explains why this set-up is important.

Figure 2: To determining the effects of restoration measures a BACI design is most effective, which overcomes a number of issues which might obscure the effects. This example of the application of dead woody debris in a stream shows why. When measuring only once before and once after implementation (A), the observed effectmay be part of a natural fluctuation (B), a trend already in progress (C), which may also occur upstream (triangles) where the measure has not been implemented (D)A BACI design corrects for these effects [3].

Analysis of local restoration measures

Subsequently, the effects of three types of restoration measures were determined for a subset of 40 waters for which sufficient biological data was available [3]. The effects on macrofauna of hydromorphological stream restoration and the construction of nature-friendly banks were examined, as well as the effects of nutrient-reducing measures on algal blooms.

Based on the macroinvertebrate data, no direct effects on and changes in trends in the community composition could be identified that could be traced back to the hydromorphological restoration measures carried out in the streams. Also, the difference between the macroinvertebrate communities of nature-friendly and traditional banks was not unambiguous. Algae blooms decreased due to a reduction in nutrient levels, but not to the desired level and often only temporarily.

Do the measures address all the relevant bottlenecks within the studied systems? In order to examine this, an overview was made of the bottlenecks present at the sites based on the stressors described in the fact sheets associated with the WFD river basin management plans. This analysis showed that all investigated waters were multistressed, e.g. several stressors negatively affected the waters, while the implemented measures only addressed part of these stressors. Furthermore, measures often did not target the source of the stress but only mitigated its effect.

Towards a basin-wide approach

Finally, the effectiveness of measures taken at a larger spatial scale (landscape/watershed level) was examined: what is the scope of large-scale interventions and the effect on the aquatic system [4]? In order to investigate the effectiveness of combinations of measures on a watershed scale, waters were selected where several measures had been implemented in space and time and where sufficient monitoring had been carried out as case studies.. Only four running water systems proved suitable: Tongelreep, Groote Molenbeek, Tungelroyse beek and Vlootbeek.

These cases showed an overruling effect of large-scale changes in the system on local interventions. A positive example is the general improvement in water quality through improved wastewater treatment upstream in the Tongelreep, which positively affected the results of other more local measures downstream. A negative example is the major negative impact of the 2018-summer drought on the hydromorphologically restored sections of the Vlootbeek. However, when multiple measures are taken in combination the individual effect of local single measures is often difficult to identify; it was often not possible to separate the effects of individual measures. However, in the Tongelreep, a distinct ecological recovery at the system level occurred when multiple stressors were addressed. A combination of improving the chemical water quality, large-scale re-meandering, cessation of vegetation mowing, introduction of dead woody debris and the construction of gravel beds led to a strong ecological improvement.

A watershed approach as implemented in the Tongelreep illustrates the importance of a thorough understanding of all ecologically relevant stressors in a water body. One approach to diagnose stressors on a watershed scale is the system-oriented ecological stress analysis [5]. This system analysis method quantifies biologically relevant environmental stressors on the catchment scale and relates this to the underlying causes related to the human activities within the watershed.


Currently it is often unclear how effective restoration measures are in the Dutch surface waters. Important reasons for this are insufficient monitoring and/or unsuitable monitoring designs. In addition, in most systems, combinations of stressors have a negative impact on the aquatic communities, whilst restoration measures often only address part of these stressors. The stream Tongelreep case study shows that when restoration measures address multiple stressors simultaneously, ecological recovery becomes visible at the system level.


Based on the results of the study, the following actions are recommended:

• Prior to taking measures, carry out a system-oriented ecological stress analysis (SESA) in order to quantify all relevant stressors. This provides insight into which measures should be taken to effectively improve the ecological water quality.
• Move from carrying out individual measures to implementing combinations of measures. This integrality is needed in the multi-stress context of Dutch waters, as it is the only way to simultaneously address all the bottlenecks that hinder ecological recovery. A large spatial scale is desirable here.
• Monitor the effects of restoration measures according to a BACI design instead of simply extending the time series of nearby existing biological monitoring points. This way, it is easier to determine which measures are effective (and which are not) in specific situations.
• Paying attention to cooperation and knowledge sharing. A BACI research programme is relatively labour-intensive and time-consuming, requiring a considerable investment in comparison to regular monitoring schemes However, this approach produces solid and much more generalisable results, which could therefore be used much more widely by water managers.

Ralf Verdonschot
(Wageningen Environmental Research)
Gea van der Lee
(Wageningen Environmental Research)
Jip de Vries
(Wageningen Environmental Research)
Anne-Marie van Noord
(Wageningen Environmental Research)
Annalieke Bakker
(Wageningen Environmental Research)
Piet Verdonschot
(Wageningen Environmental Research)

Background picture:
A meandering stretch of the Tongelreep in the forest with high ecological value


To improve the ecological quality of Dutch waters, many restoration projects have been carried out in recent decades. For many of these projects, it is unclear whether they have been ecologically effective, as monitoring was lacking or the design not suitable to answer this question. Also, restoration measures often only target part of the stressors present in a watershed, whilst the remaining ones hinder ecological recovery. To improve the effectiveness of the measures and to meet the ecological goals it is necessary to target all ecologically relevant stressors on a suitable spatial scale using combinations of measures. To identify these measures, a system-oriented ecological stress-analysis could be used. Furthermore, specific monitoring of the effects (before and after the intervention, at the location where measures have been taken and at a control location) is required to determine the effectiveness of the measures applied, which facilitates its implementation in future projects.


- Van Noord, A., de Vries, J., Verdonschot, P.F.M., R.C.M. (2022). Effectiveness of single measures: Evaluation of documented restoration projects in the Netherlands from 2008 to 2019. Notitie Kennisimpuls waterwaliteit (KIWK), Zoetwatererecosystemen, Wageningen Environmental Research, Wageningen UR, Wageningen.

- Underwood A.J. (1994). On Beyond BACI: Sampling designs that might reliably detect environmental disturbances. Ecological Applications 4, 3-15.

- Van der Lee G.H. Bakker, A., Verdonschot, R.C.M. and Verdonschot P.F.M. (2022). Quantification of bottlenecks and analysis of the effectiveness of remedial measures in surface waters. Notitie KIWK, Zoetwatererecosystemen, Wageningen Environmental Research, Wageningen UR, Wageningen.

- Van der Lee G.H., Bakker, A., Verdonschot R.C.M., P.F.M. (2022). Impact of local stream restoration on the ecological quality of the entire catchment area. An analysis of four river basins. Notitie Kennisimpuls waterwaliteit (KIWK), Zoetwatererecosystemen, Wageningen Environmental Research, Wageningen UR, Wageningen.

- Verdonschot P.F.M. & Verdonschot R.C.M. (2021). Ecological systems approach and ecological systems analysis. Report Knowledge Impulse Water Quality (KIWK), Freshwater Ecosystems, Wageningen Environmental Research, Wageningen UR, Wageningen.

^ Back to start


Effects of restoration

Knowledge journal / Edition 1 / 2022

Groundwater drainage in New Orleans

Greater New Orleans in Louisiana consists of three parishes: Orleans Parish, Jefferson Parish and St. Bernard Parish. They are almost completely surrounded by water: Lake Pontchartrain to the north, the Mississippi River to the south, Lake Borgne to the east and wetlands to both the east and west. These wet borders and the fact that a large part of New Orleans lies below sea level create major flooding and drainage problems. Due to the bowl-like shape of these parishes, almost all rain that falls in the area, and seepage that rises, can only be drained off with extensive pumping systems or by evaporation. This drainage system, together with the wastewater system and the drinking water network, play a crucial role in groundwater drainage and recharge.

Groundwater and rainwater drainage

New Orleans is equipped with an extensive network of pumping stations (Figure 1) to drain rainwater. The network is designed to drain a maximum of 1 inch (2.5 cm) in the first hour when it rains, followed by 0.5 inch (1.3 cm) per subsequent hour. In total, this pumping system can pump water from the city at a rate of more than 45,000 cubic feet (1,300 m3) per second. However, the pumping stations also drain groundwater, in both wet and dry periods, through the storm drainage system. Due to the age of the storm drainage system and to Hurricane Katrina, this system has been severely damaged. There are many cracks in the pipes and leaks at the pipe connections. Rainwater is drained into catch basins, from where it is transported by gravity through pipes to the nearest drainage pumping station. When it is not raining, smaller “constant duty” pumps remove “dry weather flow” to empty the cisterns of the pumping stations. These constant duty pumps are also turned on when major storms are forecast. They are also burdened by irrigation runoff, domestic runoff from gardens and drained groundwater. In dry periods, the low water level in the pipes and culverts provides groundwater drainage, and in wet periods, the rapid rise in groundwater levels increases the pressure on the rainwater drainage pipes, thereby causing additional groundwater drainage through the storm system.

Figure 1. New Orleans drainage system with many pumping stations (red).

Groundwater, wastewater and drinking water network

There is also an interaction between groundwater and other water infrastructures such as the wastewater system and the drinking water network. The wastewater system is a gravity collection system. Throughout the system, there are a number of wastewater pumping stations that lift the wastewater to a higher level for continuous gravity flow to the wastewater treatment plants (WWTPs). Ideally, the wastewater system should transport only industrial and domestic wastewater and not rainwater or groundwater. However, as mentioned, the system in New Orleans is in poor condition, so the WWTPs receive much more influent than the volume of domestic and industrial wastewater. During wet periods, the WWTPs treat considerably more water than during dry periods due to the increased groundwater pressure on the wastewater pipes.
The current drinking water network in New Orleans has components that are more than 100 years old and therefore, like the other two systems, suffers from leaking pipes. However, the interaction between groundwater and drinking water network is the opposite of the interaction with the other systems. Through leaking pipes, drinking water leaks into the ground, causing it to be lost. This water replenishes the groundwater system, after which most of it is drained by the closely situated draining rainwater and wastewater systems.
In a desk study based on data and information from various parties, the magnitude and consequences of the total groundwater drainage and leakage in New Orleans through the stormwater drainage system, the wastewater system and the drinking water system were quantified.

Groundwater drainage

To quantify the amount of water drained through the storm drainage system, information from the Sewerage & Water Board of New Orleans (SWBNO) was used. The data were time series for daily precipitation and pump station discharge. An impulse response analysis carried out during the study showed that a significant proportion of the water discharged could not be explained by precipitation. This indicates the presence of drained groundwater in the pump station discharge.
To quantify the amount of drained groundwater in the wastewater system, the total daily WWTP influent (East Bank WWTP, Lower 9th Ward) was collected from SWBNO and a distinction was made between wastewater flow, dry weather groundwater flow (the “base flow”) and wet weather groundwater flow. The base flow was quantified using an existing model developed by Stantec. The wastewater flow and the total drained groundwater were then found using the monitored WWTP influent time series. The influent quantity was found to fluctuate markedly, indicating a strong response of the WWTP influent to precipitation. During wet periods, the water in the wastewater system increases mainly due to additional groundwater pressure on the wastewater pipes.
Monitored groundwater levels from borehole data and groundwater monitoring collected by Deltares, Batture LLC and Tornqvist confirm the influence of the stormwater drainage system and wastewater system on groundwater drainage. After rainfall, the groundwater level drops to a constant level that coincides with the depth of the water infrastructure pipes (Figure 2).

Figure 2: (Above) A diagram of the groundwater behaviour before and after rainfall. After rainfall, we see the groundwater sinking back to the depth of the rainwater drainage system (location Mirabeau, Gentilly New Orleans, data summer 2016).

Finally, a water audit by Freeman LLC was used to quantify drinking water losses. The total drinking water loss turns out to consist of two components: the apparent losses and the actual losses. The apparent losses include inaccuracies in customer metering, water theft, illegal connections, data processing problems and billing system errors. The real losses consist of distribution leaks and connection leaks. This study only looked at the real losses, as this is the water that seeps into the ground.

Conclusions on the water balance

The following conclusions were drawn from the desk study: 1) the rainwater drainage system accounts for most of the total groundwater drainage (58 per cent); 2) 50 per cent of the influent of the WWTP is groundwater, which is a large unnecessary load for the treatment process; and 3) 55 per cent of the drinking water produced infiltrates the soil during distribution, which means that the drinking water losses are a larger groundwater replenishment than the annual precipitation surplus (see Figure 3).

Figure 3: The proportions and quantities of the various fluxes in the period 2018-2020 (precipitation 1,782 mm/year and evaporation about 1,289 mm/year).


The huge amount of groundwater that is drained or infiltrated into the soil has major physical and financial implications. Groundwater drainage through the rainwater drainage system and the wastewater system leads to lower groundwater levels. This leads to subsidence through oxidation in peat soils (about 12 per cent of the case study area). In mineral soils such as clay (51 per cent of the case study area), a damaging shrink-swell process occurs whereby the subsoil dries out in dry periods and shrinks and swells again in wet periods. The more the land subsides, the more groundwater has to be drained to prevent groundwater flooding, resulting in further subsidence - a vicious circle.
Since subsidence occurs at different rates depending on soil properties and groundwater conditions, there are many fractures, cracks and sinkholes in roads, car parks and patios, leading to high renovation and repair costs. Another financial consequence concerns drinking water losses. More than half of the drinking water produced is lost during transport.
As mentioned above, 50 per cent of the influent of the WWTP is drained groundwater which does not need to be treated. This places an enormous burden on WWTP, is bad for water quality and costs a lot of money. During big storms, this leads to direct wastewater discharge into the Mississippi River because the WWTP cannot cope with the amount of influent plus drained groundwater.
Repairing these systems is essential to solving all these physical and financial problems, but it is expensive. Not all pipes can be replaced at once. It is important to identify the most critical locations (oldest pipelines or pipelines in areas worst affected by Hurricane Katrina) and to repair them first. Furthermore, additional research is needed before repairs can be started at a location, in order to locally predict the influence of different drainage and recharge fluxes. Restoration of the rainwater drainage system can reduce groundwater drainage, which can lead to groundwater flooding. Or vice versa: the restoration of the drinking water network decreases the groundwater recharge, which may accelerate soil subsidence.
This desk study focused on New Orleans, but groundwater drainage and recharge through faulty pipes is a major problem in many places, in both developed and developing countries. This research demonstrates the value of integrated urban water analysis. Before a groundwater model can be created, it is important to thoroughly study and evaluate the available data to understand what is happening in the field. Only then can a model be created that accurately represents what is happening in reality.


Thanks to Adam Kay and Tyler Antrup of SWBNO for their data and explanations. And to Bob Mora and Jennifer Snape of Batture LLC for their monitoring well data; they also put us in touch with Sean Walsh of Eustis Engineering LLC. Finally, thanks to Christopher Sanchez from Stantec for his help with the wastewater model and Tor Tornqvist for the Irish Channel groundwater data.

Laura Nougues
Roelof Stuurman

Background picture:
During a dry spell, water from a Jefferson Parish storm drain pipe flows directly into the surface water. Chemical analyses show that drinking water is leaking here.


Greater New Orleans is surrounded by the Mississippi River, two lakes and wetlands. Due to the bowl-like shape of a large part of the city, excess rain can only be drained off with extensive pumping systems or by evaporation. These pumping systems, along with the wastewater system, have been found to also pump out a lot of groundwater and leaking drinking water, of which more than half of the produced drinking water is lost during transport. This leads to land subsidence, water quality degradation and financial losses. Wastewater treatmentplants (WWTPs) are being unnecessarily burdened with large quantities of relatively clean drained groundwater (half of the influent). During big storms, wastewater is directly discharged into the Mississippi River as the WWTP cannot cope with the influent volume. Repairing these systems is essential to solve all these problems, but is expensive and can result in critical changes to the urban groundwater system.


- National Resources Conservation Service. (2021). Web Soil Survey.

- Nougues. Laura (2021): Groundwater drainage in New Orleans. An internship (TU Delft) at Deltares

^ Back to start


Groundwater drainage

Knowledge journal / Edition 1 / 2022

Image processing for automatic interpretation of drinking water mains inspections

After the construction of a drinking water network, the most important thing is to monitor the maintenance condition of the pipes. The amount of data collected for this purpose is growing enormously, especially when it comes to video inspections. But how do you find useful information as a drinking water utility in such a growing mountain of data? In other sectors, image processing techniques are already widely used for this purpose. Could this also be a solution for the drinking water sector?

Network operators usually rely on 'manual' estimates from statistical analyses and asset data to predict the condition of the pipes. In recent years, however, mobile inspection platforms have also been developed. Such devices use cameras and other sensors to assess the network from the inside. Further developments in the field are expected to lead to the collection of ever larger amounts of data. Video inspections, for example, will provide the drinking water industry with millions of images of the inside of pipes.
For water utilities, registering the components of the pipe system – such as pipe joints – is one valuable step in mapping out the condition of the pipe network in more detail. However, continuing the example, there are millions of connections in the pipeline network, making it impossible for humans to search for them among inspection data. In this article, we explore and demonstrate the potential of automatic processing of video inspections by applying it to detect the location of connections. The work is part of KWR Water Research Institute's project 'The pipeline network in view'.


The images in this study were made with two different devices that are already in use at Dutch water utilities. One is an inspection robot (IBAK PANORAMO). The other is a camera mounted on a 'pig’, moving through the pipe (Quasset). Because the devices are different, they produce different images. Image processing algorithms will have to be able to cope with this. An example of a shot from the IBAK robot is shown in Figure 1A.
The conventional image processing process starts with a pre-processing step that improves the colour contrast. Segmentation is then applied, which allows different elements in the image to be distinguished on the basis of characteristics such as brightness and colour. Then more abstract features, such as shape, centre or diameter, are extracted from these segments. These characteristics are fundamental data for computers to recognise or detect objects. Finally, the useful information extracted from the images is summarised and translated so that system operators can easily understand the state of the system and make decisions based on this information.
These steps are explained below for the inspection of connections.

Figure 1. The original camera image (A) is converted with some pre-processing steps (B, C, D) into data in which a computer can search for patterns.


The first step in the image processing algorithm is to convert the data into grey values. Each pixel is assigned a grey value for the amount of reflected light (Figure 1B). The move to greyscale images makes further data processing easier.
Based on the differences in light intensity, a Canny edge detector algorithm can then identify the edges between objects (Figure 1C). The edges are then artificially enhanced (exaggeratedly dilated) to show a better representation of the objects found. (Figure 1D).
In Figure 1D, several lines can be distinguished, such as the circle marking the connection, as well as the diagonal lines representing the track left by the inspection robot. To enable a computer to detect only the connection, an algorithm is then applied to the dilated edges that is ideally suited for detecting circular patterns: the “Circular Hough Transform” (CHT). Figure 2A shows which circles the algorithm was able to identify in the previous steps. To remove the irrelevant circles, as a final step in the algorithm, a filter is added to exclude those whose centres are not at the centre of the image. Thus, only the circle corresponding to the connection remains, as shown in Figure 2B.
The above algorithm can be applied to any image in a video of the pipe. In some of these images, there is too much noise to be able to detect circles. However, one connection can be seen in several consecutive images; each connection is ultimately detected on the basis of the relationship between neighbouring images.

Figure 2. The circles detected in the camera image with Circular Hough Transform (CHT) (A) and the connection filtered from them with an additional dedicated algorithm.

Mapping the components detected

The algorithm described can indicate in a video which images show a connection and which do not. In order to record the connections in the pipe information system, it is also necessary to know the corresponding location of the robot in the pipe network for each individual image. In order to automatically determine the exact location of pipeline components, in addition to an algorithm for recognising connections and valves, an adequate camera positioning system is required. For other possible inspection data, it is also essential to link the measurement to the relevant location and pipe. Only then can water utilities sharpen their decisions about specific pipelines on the basis of the inspections.

Limitations of 'classical' image processing

In the case study above, we can see that these conventional image processing techniques can already support us enormously in processing large amounts of inspection data. However, there are also limitations.
The successful application of image processing in this case is due to the fact that the algorithm is focused on the precise characteristics of the target: a circular joint in the centre of the image. However, the algorithm is not flexible enough to allow it to be used for other things that are worth detecting, such as the locations of drillings, incipient cracks or deterioration. The algorithm for connections uses the fact that all connections generally look the same. The shapes of cracks and other signs of degradation are much more diverse, which also makes it more difficult to design a targeted algorithm.
An even more important limitation is that an algorithm can only be focused on a detectable aspect if we know what exactly we are looking for. This means that we cannot detect unknown abnormalities in a targeted way. Since video inspections of drinking water pipes are not yet widely used, we do not yet have a good picture of all the issues that may be worth detecting. Thus, a lot of "manual" experience is still needed before image processing algorithms can take over tasks.

Potential of 'modern' artificial intelligence

More modern artificial intelligence (AI) techniques are better suited for detecting more diverse or unknown objects. With AI, a self-learning system can be taught to recognise objects on the basis of examples (machine learning) and then also recognise new variants of these objects. In other sectors, it has already been demonstrated that such algorithms can eventually become even better than humans at detecting objects. It is important to emphasise, however, that thousands or even millions of examples must be provided to such algorithms before they can function on their own. Although artificial intelligence can be a very powerful tool for detecting unknown or diverse objects, it is not a substitute for human experience. The drinking water sector will first have to explain to the computers themselves what they have to watch out for.

Next steps

This article shows by way of an example that image processing techniques can be used to automatically recognise objects in video images of water pipes. The Circular Hough Transform method was used to detect connections - this method has proven effective in identifying circular objects. The images of the detected connections must also be linked to the measurement location. Such forms of automated data processing are crucial for the widespread use of video inspection of pipes.
The first important step towards using computers to process inspection data in the future is to gain experience and find out exactly what we want to get out of our inspections. On the basis of these insights, classical or more modern image processing techniques can then be set up and adjusted. In follow-up research, therefore, it is important to have more inspection images searched through by real people.

Mollie Mary Torello
(KWR Water Research Institute)
Xin Tian
(KWR Water Research Institute)
Peter van Thienen
(KWR Water Research Institute)
Karel van Laarhoven
(KWR Water Research Institute)


Once a drinking water pipe is in use, it becomes a 'black box'. It is difficult to keep checking the condition of the pipes. However, information on the condition of the pipes is vital for the proper management and maintenance of the pipeline network. For example, is there corrosion or are there cracks in the pipe walls? Where exactly is the broken connection? Previously, these questions could only be answered by carrying out an on-site investigation. Today, new inspection robots, often equipped with cameras, make it possible to 'see' the inside of the pipe. This innovation leads to a new challenge: the interpretation of an abundance of data. In this study, we demonstrate the potential of automated processing of inspection data using an example: the automatic detection of connections in camera images of the pipeline network.


- KWR Water Research Institute, website consulted on 12-05-2022.

^ Back to start


Image processing

Knowledge journal / Edition 1 / 2022

Effect of source location on the spread of groundwater contamination by open geothermal energy systems

Due to the growing demand for sustainable energy, the number of open geothermal energy systems (also known as aquifer thermal energy storage (ATES) systems) is increasing. However, there are many contaminants, especially in cities, due to past industrialisation. If ATES systems are installed there, the contamination may spread and dilute, as these systems simultaneously extract and infiltrate groundwater. This is especially the case if the shallow aquifer is used for ATES systems. Because spreading and dilution of contamination is not desirable, it is important to understand the mechanisms of spreading and any possible measures to limit or control the spreading.

The aim of this research is to quantify the effect of scaling up ATES systems on the spread of contaminants in the groundwater based on a case study. In addition, we provide insight into the extent to which control measures can be used to prevent spreading under practical conditions. The research is part of the WarmingUP research programme on the scaling up of ATES systems, which is needed for the large-scale application of aquathermy - the recovery of thermal energy from water.

Spreading of groundwater contamination by ATES systems

ATES systems influence the spread of contaminants in an aquifer in various ways (Bloemendal et al., 2022):
1. Mixing and homogenisation of contaminant concentrations over the depth of the aquifer;
2. Mixing of contaminated water with the surrounding groundwater by transport from hot to cold sources and vice versa;
3. Dispersion at the edges of the injected volume into the surrounding groundwater;
4. Changes in groundwater flow (direction and/or magnitude);
5. Incomplete recovery of the injected volume due to background flow or imbalance of the ATES systems;
6. If the hydraulic radius of different ATES systems overlap, short-circuit flow occurs between different ATES systems. This effect is stronger when ATES systems are unbalanced.

In areas with many ATES systems, short-circuit flow is a major cause of rapid spreading. Usually ATES sources are installed so that each system is outside the thermal radius of another. However, the hydraulic radius, defined here as the capture zone, is 1.5 to 2 times larger than the thermal radius, and contaminants move within the hydraulic radius.
When a contaminant is present in the dissolved phase, mixing contaminated groundwater with (clean) surrounding groundwater will dilute the concentration of the contamination over a larger area. In the presence of pure product (DNAPL) mixing can cause an increase in the dissolution from the pure product (Zuurbier et al., 2013).

Model study

Using the city of Eindhoven as a case study, three scenarios have been modelled in which the numbers and layout of ATES systems have been varied:
1. The current situation of the existing /licensed ATES systems (n=48);
2. ‘Business as Usual' (BaU) or the standard layout, where the distance between the sources is three times the thermal radius (n=154);
3. High Density (HD): in the area of highest demand, the sources were optimised for thermal efficiency. Sources of the same temperature were clustered to accommodate more systems, reducing the distance between sources and increasing the average distance between wells of different temperatures. This resulted in more systems (n=190).

The simulations cover a period of 10 years. For the two future scenarios (BaU and HD), two management options were tested to limit the spread of contamination:
A. Net extraction: the 6 ATES sources closest to the contamination were selected for this purpose. From these wells, 90 per cent of the pumped volume was re-injected, assuming that the rest would be treated above ground and discharged separately.
B. Treatment & reinjection: for the same 6 sources, 50 per cent is treated and reinjected. This removal is relatively high and is intended to simulate a best-case scenario.

The contaminant is simulated as a conservative tracer with no degradation or adsorption, which represents a worst-case scenario for spreading. Two situations were simulated: a plume zone and a pure product zone, the latter covering half the area of the former. Both simulations were performed with a dimensionless concentration of 100. They were evaluated according to the total spread (area of the 10-4 contour in m2) and the degree of contamination, defined as total mass of contamination in relation to the total annual pumped volume of the ATES systems.

The three scenarios compared

When comparing the three source placement scenarios without control measures for groundwater contamination, the scenario with only the existing systems gives the largest spreading in relative terms (Figure 1). This is because in this scenario the (fewer) ATES sources are further apart than in the BaU and HD scenarios. When short-circuit flow between different ATES systems occurs, the distance over which the contamination spreads is therefore much greater than in the BaU and HD scenarios.
The absolute spreading increases with the number of sources, and the contaminated area is therefore the largest in the HD scenario. In the HD scenario, the groundwater contamination zone after 10 years is generally 30-57% larger than in the BaU scenario. However, the number of wells in the HD scenario is also 23% higher (N.B. when the degree of spreading for the BaU and HD scenarios is compared to the total volume pumped, there is little difference (+/- 1%)).
Therefore, in practice, there is a trade-off between the spread of contamination, the improved thermal efficiency resulting from clustering wells of the same temperature and the resulting optimisation of the use of the subsurface for ATES systems.
When the contamination is present in the form of a pure product, the total degree of spreading does not change much compared to the plume for each scenario, though the area of high concentration are larger compared to the plume scenario.

Figure 1. The spread of the contamination after 10 years for the existing (left), BaU (middle) and HD (right) scenarios for plume (top) and pure product contamination (bottom). Each subplot shows the concentration in a 3x3 kilometer area. Each black dot is an ATES well, and the red dot indicates the centre of mass of the plume.

Measures against spreading

As might be expected, a net withdrawal of contaminated water from ATES systems leads to a reduction in the total mass of contamination (Figure 2). The reduction in the BaU and HD scenarios is 22 and 28% for the plume area and 68 and 77% for the total mass, respectively. Treatment with reinjection is slightly less effective but still gives a reduction of 14-18% of the plume area and 49-61% of the total mass. The better performance of the net withdrawal was likely due to an increase in the total withdrawal rate, which was done to ensure that the system still met the thermal requirements.

Figure 2. The effect of different control measures on the spread of a contaminant plume for the two future scenarios, BaU (left) and HD (right). Each subplot shows the concentration of the contaminant in a 3x3 kilometer area, the contours indicate the dilution level of the contaminant. Each black dot is an ATES source, and the red dot indicates the centre of mass of the groundwater contamination.

The extraction of an additional 10% for treatment from ATES systems wells closest to the contamination proved to be the most effective option for limiting the spread of contamination. Unsurprisingly it was also found that the control measures for net withdrawal and treatment and reinjection were most effective in the first years of application. Of the total contaminant mass removed, 90% was removed in the first three years of the simulation. Moreover, only a few wells were responsible for removing most of the contamination, which means that a measure such as treatment must be optimised in time and space. For example, a tiered treatment can be applied, treating the volumes with high concentrations in the first years, followed by a treatment that is more effective for the treatment of low concentration flows.


The modelling study shows that the locations of ATES sources and the type of contamination (plume or pure product) influence the extent to which contaminants spread in the groundwater. The choice and implementation of measures to prevent the spread of contamination therefore also depends on local conditions. Nevertheless, the results make it clear that it is not possible to prevent the spread of groundwater contamination in an area with a lot of ATES systems. The higher the density of ATES sources, the greater the spreading. In practice, the best way to limit spreading is through early assessment of local contamination, appropriate choice of measures and careful selection of ATES sources to apply the measures. Moreover, it is important to apply management measures from the beginning, as they are most effective in the first years of application.

Alex Hockin
(KWR Water Research Institute)
Martin Bloemendal
(KWR, TU Delft)
Niels Hartog
(KWR, Utrecht University)

Background picture:
Inner city Eindhoven: simulated spread of a fictitious contamination by existing and planned open soil energy systems and background flow.


In cities, the first aquifer is often contaminated. The use of open geothermal energy systems (also known as aquifer thermal energy storage (ATES) systems,) systems can spread such contamination. In a modelled case study, the extent of spreading was quantified for the city centre of Eindhoven, and two control measures were tested. What emerges is that with many ATES systems it is practically impossible to prevent the spread and dilution of contaminants. Although spreading was greatest at the highest density (190 installations per 9 km2), the differences with the scenario with fewer ATES systems (154) were not great when compared with the total pumped volume of the systems. Net withdrawal and above-ground treatment can limit spreading and are most effective in the first few years after ATES systems comes into operation. In practice, this means that it is important to locate local contamination before ATES systems are started and to carefully select the ATES sources on which to apply the measures.


- Bloemendal, M., Hockin, A., J., V., Hoekstra, N., & van Ree, D. (2022). Effects of aquathermy-linked geothermal systems on the subsurface. 3B Effects, role and regulation. WarmingUp Innovative Sustainable Heat Collective.

- Zuurbier, K. G., Hartog, N., Valstar, J., Post, V. E. A., & Van Breukelen, B.M. (2013). The impact of low-temperature seasonal aquifer thermal energy storage (ATES) systems on chlorinated solvent contaminated groundwater: Modelling of spreading and degradation. Journal of Contaminant Hydrology, 147, 1–13.

^ Back to start


Groundwater contamination


Edition 2/2021

Edition 1/2021

Edition 2/2020

Edition 1/2020

Edition 2/2019

Edition 1/2019

Edition 1/2018

Edition 2/2017

Edition 1/2017

Edition 2/2016

Edition 1/2016

Edition 2/2015

Edition 1/2015


Previous editions

Knowledge journal / Edition 1 / 2022


The knowledge section Water Matters of H2O is an initiative of

Royal Dutch Waternetwerk
Independent knowledge networking organisation for and by Dutch water professionals.

Water Matters is supported by

Independent institute for applied research in the field of water, subsurface and infrastructure. Throughout the world, we work on smart solutions, innovations and applications for people, environment and society.

KWR Watercycle Research Institute
Institute that assists society in optimally organising and using the water cycle by creating knowledge through research; building bridges between science, business and society; promoting societal innovation by applying knowledge.

Royal HaskoningDHV
Independent international engineering and project management consultancy that contributes to a sustainable environment in cooperation with clients and partners.

Foundation for Applied Water Research (STOWA)
Knowledge centre of the regional water managers (mostly the Water Boards) in the Netherlands. Its mission is to develop, collect, distribute and implement applied knowledge, which the water managers need in order to adequately carry out the tasks that their work supports.